Single Variable Calculus Student Solutions Manual

c 2008 by W. H. Freeman and Company ISBN-13: 978-0-7167-9866-8 ISBN-10: 0-7167-9866-2 All rights reserved Printed in ...

Author: Jon Rogawski

801 downloads 4923 Views 7MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

c 2008 by W. H. Freeman and Company ISBN-13: 978-0-7167-9866-8 ISBN-10: 0-7167-9866-2 All rights reserved Printed in the United States of America

W. H. Freeman and Company, 41 Madison Avenue, New York, NY 10010 Houndmills, Basingstoke RG21 6XS, England www.whfreeman.com

Student’s Solutions Manual to accompany Jon Rogawski’s

Single Variable

CALCULUS BRIAN BRADIE Christopher Newport University With Chapter 12 contributed by

GREGORY P. DRESDEN Washington and Lee University and additional contributions by

Art Belmonte Cindy Chang-Fricke Benjamin G. Jones Kerry Marsack Katherine Socha Jill Zarestky Kenneth Zimmerman

W. H. Freeman and Company New York

This page intentionally left blank

CONTENTS Chapter 1 Precalculus Review 1.1 1.2 1.3 1.4 1.5

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9

4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8

5.1 5.2 5.3 5.4 5.5

1

Real Numbers, Functions, and Graphs Linear and Quadratic Functions The Basic Classes of Functions Trigonometric Functions Technology: Calculators and Computers Chapter Review Exercises

1 8 13 16 23 26

Chapter 2 Limits

31

Limits, Rates of Change, and Tangent Lines Limits: A Numerical and Graphical Approach Basic Limit Laws Limits and Continuity Evaluating Limits Algebraically Trigonometric Limits Intermediate Value Theorem The Formal Definition of a Limit Chapter Review Exercises

31 36 45 48 55 59 65 68 72

Chapter 3 Differentiation

79

Definition of the Derivative The Derivative as a Function Product and Quotient Rules Rates of Change Higher Derivatives Trigonometric Functions The Chain Rule Implicit Differentiation Related Rates Chapter Review Exercises

79 90 99 105 112 118 123 132 141 148

Chapter 4 Applications of the Derivative

159

Linear Approximation and Applications Extreme Values The Mean Value Theorem and Monotonicity The Shape of a Graph Graph Sketching and Asymptotes Applied Optimization Newton’s Method Antiderivatives Chapter Review Exercises

159 165 173 180 188 202 216 221 228

Chapter 5 The Integral

237

Approximating and Computing Area The Definite Integral The Fundamental Theorem of Calculus, Part I The Fundamental Theorem of Calculus, Part II Net or Total Change as the Integral of a Rate

237 250 259 266 273

5.6

6.1 6.2 6.3 6.4 6.5

7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9

8.1 8.2 8.3 8.4 8.5 8.6

Substitution Method Chapter Review Exercises

276 283

Chapter 6 Applications of the Integral

292

Area Between Two Curves 292 Setting Up Integrals: Volume, Density, Average Value 300 Volumes of Revolution 308 The Method of Cylindrical Shells 316 Work and Energy 324 Chapter Review Exercises 331

Chapter 7 The Exponential Function

338

Derivative of b x and the Number e Inverse Functions Logarithms and Their Derivatives Exponential Growth and Decay Compound Interest and Present Value Models Involving y = k ( y − b) ˆ L’Hopital’s Rule Inverse Trigonometric Functions Hyperbolic Functions Chapter Review Exercises

338 344 349 357 362 365 369 377 386 392

Chapter 8 Techniques of Integration

404

Numerical Integration Integration by Parts Trigonometric Integrals Trigonometric Substitution The Method of Partial Fractions Improper Integrals Chapter Review Exercises

404 415 426 435 449 469 487

Chapter 9 Further Applications of the Integral and Taylor Polynomials 503 9.1 9.2 9.3 9.4

10.1 10.2 10.3 10.4

Arc Length and Surface Area Fluid Pressure and Force Center of Mass Taylor Polynomials Chapter Review Exercises

503 513 518 526 538

Chapter 10 Introduction to Differential Equations

545

Solving Differential Equations Graphical and Numerical Methods The Logistic Equation First-Order Linear Equations Chapter Review Exercises

545 558 563 568 578

iv

11.1 11.2 11.3 11.4 11.5 11.6 11.7

C A L C U L U S CONTENTS

Chapter 11 Infinite Series

587

Sequences Summing an Infinite Series Convergence of Series with Positive Terms Absolute and Conditional Convergence The Ratio and Root Tests Power Series Taylor Series Chapter Review Exercises

587 598 608 622 628 636 646 660

12.1 12.2 12.3 12.4 12.5

Chapter 12 Parametric Equations, Polar Coordinates, and Conic Sections

674

Parametric Equations Arc Length and Speed Polar Coordinates Area and Arc Length in Polar Coordinates Conic Sections Chapter Review Exercises

674 690 698 711 721 732

P R E CALCULUS 1 REVIEW 1.1 Real Numbers, Functions, and Graphs Preliminary Questions 1. Give an example of numbers a and b such that a |b|. SOLUTION

Take a = −3 and b = 1. Then a 1 = |b|.

2. Which numbers satisfy |a| = a? Which satisfy |a| = −a? What about | − a| = a? SOLUTION

The numbers a ≥ 0 satisfy |a| = a and | − a| = a. The numbers a ≤ 0 satisfy |a| = −a.

3. Give an example of numbers a and b such that |a + b| < |a| + |b|. SOLUTION

Take a = −3 and b = 1. Then |a + b| = | − 3 + 1| = | − 2| = 2,

but

|a| + |b| = | − 3| + |1| = 3 + 1 = 4.

Thus, |a + b| < |a| + |b|. 4. What are the coordinates of the point lying at the intersection of the lines x = 9 and y = −4? SOLUTION

The point (9, −4) lies at the intersection of the lines x = 9 and y = −4.

5. In which quadrant do the following points lie? (a) (1, 4) (c) (4, −3)

(b) (−3, 2) (d) (−4, −1)

SOLUTION

(a) Because both the x- and y-coordinates of the point (1, 4) are positive, the point (1, 4) lies in the first quadrant. (b) Because the x-coordinate of the point (−3, 2) is negative but the y-coordinate is positive, the point (−3, 2) lies in the second quadrant. (c) Because the x-coordinate of the point (4, −3) is positive but the y-coordinate is negative, the point (4, −3) lies in the fourth quadrant. (d) Because both the x- and y-coordinates of the point (−4, −1) are negative, the point (−4, −1) lies in the third quadrant. 6. What is the radius of the circle with equation (x − 9)2 + (y − 9)2 = 9? SOLUTION

The circle with equation (x − 9)2 + (y − 9)2 = 9 has radius 3.

7. The equation f (x) = 5 has a solution if (choose one): (a) 5 belongs to the domain of f . (b) 5 belongs to the range of f . SOLUTION

The correct response is (b): the equation f (x) = 5 has a solution if 5 belongs to the range of f .

8. What kind of symmetry does the graph have if f (−x) = − f (x)? SOLUTION

If f (−x) = − f (x), then the graph of f is symmetric with respect to the origin.

Exercises 1. Use a calculator to find a rational number r such that |r − π 2 | < 10−4 . r must satisfy π 2 − 10−4 < r < π 2 + 10−4 , or 9.869504 < r < 9.869705. r = 9.8696 = 12337 1250 would be one such number. SOLUTION

In Exercises express interval of an inequality involving absolute value. Let a3–8, = −3 and b the = 2. Whichinofterms the following inequalities are true? (a) a2] 0 1 1 (f) < a b

2

CHAPTER 1

PRECALCULUS REVIEW SOLUTION The midpoint of the interval is c = (0 + 4)/2 = 2, and the radius is r = (4 − 0)/2 = 2; therefore, (0, 4) can be expressed as |x − 2| < 2.

7. [1, 5] [−4, 0] SOLUTION The midpoint of the interval is c = (1 + 5)/2 = 3, and the radius is r = (5 − 1)/2 = 2; therefore, the interval [1, 5] can be expressed as |x − 3| ≤ 2. In Exercises write the inequality in the form a < x < b for some numbers a, b. (−2, 9–12, 8) 9. |x| < 8 SOLUTION

−8 < x < 8

11. |2x + 1| < 5 |x − 12| < 8 SOLUTION −5 < 2x + 1 < 5 so −6 < 2x < 4 and −3 < x < 2 In Exercises |3x −13–18, 4| < 2express the set of numbers x satisfying the given condition as an interval. 13. |x| < 4 SOLUTION

(−4, 4)

15. |x − 4| < 2 |x| ≤ 9 SOLUTION The expression |x − 4| < 2 is equivalent to −2 < x − 4 < 2. Therefore, 2 < x < 6, which represents the interval (2, 6). 17. |4x − 1| ≤ 8 |x + 7| < 2 7 9 SOLUTION The expression |4x − 1| ≤ 8 is equivalent to −8 ≤ 4x − 1 ≤ 8 or −7 ≤ 4x ≤ 9. Therefore, − 4 ≤ x ≤ 4 , 7 9 which represents the interval [− 4 , 4 ]. In Exercises |3x +19–22, 5| < 1describe the set as a union of finite or infinite intervals. 19. {x : |x − 4| > 2} SOLUTION

x − 4 > 2 or x − 4 < −2 ⇒ x > 6 or x < 2 ⇒ (−∞, 2) ∪ (6, ∞)

21. {x : |x 2 − 1| > 2}

{x : |2x + 4| > 3} √ √ x 2 − 1 > 2 or x 2 − 1 < −2 ⇒ x 2 > 3 or x 2 < −1 (this will never happen) ⇒ x > 3 or x < − 3 ⇒ √ √ (−∞, − 3) ∪ ( 3, ∞). SOLUTION

23. Match the inequalities (a)–(f) with the corresponding statements (i)–(vi). {x : |x 2 + 2x| > 2} 1 (a) a > 3 (b) |a − 5| < 3 1 (c) a − < 5 (d) |a| > 5 3 (e) |a − 4| < 3 (f) 1 < a < 5 (i) (ii) (iii) (iv) (v) (vi)

a lies to the right of 3. a lies between 1 and 7. The distance from a to 5 is less than 13 . The distance from a to 3 is at most 2. a is less than 5 units from 13 . a lies either to the left of −5 or to the right of 5.

SOLUTION

(a) On the number line, numbers greater than 3 appear to the right; hence, a > 3 is equivalent to the numbers to the right of 3: (i). (b) |a − 5| measures the distance from a to 5; hence, |a − 5| < 13 is satisfied by those numbers less than 13 of a unit from 5: (iii). (c) |a − 13 | measures the distance from a to 13 ; hence, |a − 13 | < 5 is satisfied by those numbers less than 5 units from 1 : (v). 3 (d) The inequality |a| > 5 is equivalent to a > 5 or a < −5; that is, either a lies to the right of 5 or to the left of −5: (vi). (e) The interval described by the inequality |a − 4| < 3 has a center at 4 and a radius of 3; that is, the interval consists of those numbers between 1 and 7: (ii). (f) The interval described by the inequality 1 < x < 5 has a center at 3 and a radius of 2; that is, the interval consists of those numbers less than 2 units from 3: (iv).

S E C T I O N 1.1

Real Numbers, Functions, and Graphs

3

25. Show that if a > b, then b−1 > a −1, provided that a and b have the same sign. What happens if a > 0 and b < 0? x Describe the set x : < 0 as an interval. b 1 1 SOLUTION Case 1a: If a and b are x+ 1 both positive, then a > b ⇒ 1 > a ⇒ b > a . b Case 1b: If a and b are both negative, then a > b ⇒ 1 < a (since a is negative) ⇒ b1 > a1 (again, since b is negative). Case 2: If a > 0 and b < 0, then a1 > 0 and b1 < 0 so b1 < a1 . (See Exercise 2f for an example of this). 27. Show that if |a − 5||x m, then 2n 5n r = a5n−m or r = and once again r has 10n a finite decimal expansion. SOLUTION

81. is symmetrical with respect vertical line x = a if f (a − x) = f (a + x). be an integer with digits p1 , . . .to, pthe that Let pA=function p1 . . . psf (x) s . Show (a) Draw the graph of a function that is symmetrical with respect to x = 2. p p . . . p = 0. s 1 (b) Show that if f (x) is symmetrical with respect to = 1a, then g(x) = f (x + a) is even. 10xs − SOLUTION

2 Use this find the decimal expansion of r may = be . Note that (a) There aretomany possibilities, two of which 11 r=

18 2 = 2 11 10 − 1

8

CHAPTER 1

PRECALCULUS REVIEW y

2 1

−1

x 1

2

3

4

5

y = | x − 2|

(b) Let g(x) = f (x + a). Then g(−x) = f (−x + a) = f (a − x) = f (a + x)

symmetry with respect to x = a

= g(x) Thus, g(x) is even. Formulate a condition for f (x) to be symmetrical with respect to the point (a, 0) on the x-axis.

1.2 Linear and Quadratic Functions Preliminary Questions 1. What is the slope of the line y = −4x − 9? SOLUTION

The slope of the line y = −4x − 9 is −4, given by the coefficient of x.

2. Are the lines y = 2x + 1 and y = −2x − 4 perpendicular? SOLUTION The slopes of perpendicular lines are negative reciprocals of one another. Because the slope of y = 2x + 1 is 2 and the slope of y = −2x − 4 is −2, these two lines are not perpendicular.

3. When is the line ax + by = c parallel to the y-axis? To the x-axis? SOLUTION

The line ax + by = c will be parallel to the y-axis when b = 0 and parallel to the x-axis when a = 0.

4. Suppose y = 3x + 2. What is y if x increases by 3? SOLUTION

Because y = 3x + 2 is a linear function with slope 3, increasing x by 3 will lead to y = 3(3) = 9.

5. What is the minimum of f (x) = (x + 3)2 − 4? SOLUTION

−4.

Because (x + 3)2 ≥ 0, it follows that (x + 3)2 − 4 ≥ −4. Thus, the minimum value of (x + 3)2 − 4 is

6. What is the result of completing the square for f (x) = x 2 + 1? SOLUTION

Because there is no x term in x 2 + 1, completing the square on this expression leads to (x − 0)2 + 1.

Exercises In Exercises 1–4, find the slope, the y-intercept, and the x-intercept of the line with the given equation. 1. y = 3x + 12 Because the equation of the line is given in slope-intercept form, the slope is the coefficient of x and the y-intercept is the constant term: that is, m = 3 and the y-intercept is 12. To determine the x-intercept, substitute y = 0 and then solve for x: 0 = 3x + 12 or x = −4. SOLUTION

3. 4x + 9y = 3 y =4−x SOLUTION To determine the slope and y-intercept, we first solve the equation for y to obtain the slope-intercept form. This yields y = − 49 x + 13 . From here, we see that the slope is m = − 49 and the y-intercept is 13 . To determine the x-intercept, substitute y = 0 and solve for x: 4x = 3 or x = 34 .

In Exercises 5–8,1find the slope of the line. y − 3 = 2 (x − 6) 5. y = 3x + 2 SOLUTION

m=3

7. 3x + 4y = 12 y = 3(x − 9) + 2 3 SOLUTION First solve the equation for y to obtain the slope-intercept form. This yields y = − 4 x + 3. The slope of 3 the line is therefore m = − 4 . 3x + 4y = −8

S E C T I O N 1.2

Linear and Quadratic Functions

9

In Exercises 9–20, find the equation of the line with the given description. 9. Slope 3, y-intercept 8 SOLUTION

Using the slope-intercept form for the equation of a line, we have y = 3x + 8.

11. Slope 3, passes through (7, 9) Slope −2, y-intercept 3 SOLUTION Using the point-slope form for the equation of a line, we have y − 9 = 3(x − 7) or y = 3x − 12. 13. Horizontal, passes through (0, −2) Slope −5, passes through (0, 0) SOLUTION A horizontal line has a slope of 0. Using the point-slope form for the equation of a line, we have y − (−2) = 0(x − 0) or y = −2. 15. Parallel to y = 3x − 4, passes through (1, 1) Passes through (−1, 4) and (2, 7) SOLUTION Because the equation y = 3x − 4 is in slope-intercept form, we can readily identify that it has a slope of 3. Parallel lines have the same slope, so the slope of the requested line is also 3. Using the point-slope form for the equation of a line, we have y − 1 = 3(x − 1) or y = 3x − 2. 17. Perpendicular to 3x + 5y = 9, passes through (2, 3) Passes through (1, 4) and (12, −3) SOLUTION We start by solving the equation 3x + 5y = 9 for y to obtain the slope-intercept form for the equation of a line. This yields 3 9 y=− x+ , 5 5 from which we identify the slope as − 35 . Perpendicular lines have slopes that are negative reciprocals of one another, so the slope of the desired line is m ⊥ = 53 . Using the point-slope form for the equation of a line, we have y − 3 = 53 (x − 2) or y = 53 x − 13 . 19. Horizontal, passes through (8, 4) Vertical, passes through (−4, 9) SOLUTION A horizontal line has slope 0. Using the point slope form for the equation of a line, we have y − 4 = 0(x − 8) or y = 4. 21. Find the equation of the perpendicular bisector of joining (1, 2) and (5, 4) (Figure 11). Hint: The the segment Slope 3, x-intercept 6 a+c b+d midpoint Q of the segment joining (a, b) and (c, d) is , . 2 2 y

Perpendicular bisector (5, 4) Q (1, 2) x

FIGURE 11 SOLUTION

The slope of the segment joining (1, 2) and (5, 4) is m=

4−2 1 = 5−1 2

and the midpoint of the segment (Figure 11) is midpoint =

1+5 2+4 , 2 2

= (3, 3)

The perpendicular bisector has slope −1/m = −2 and passes through (3, 3), so its equation is: y − 3 = −2(x − 3) or y = −2x + 9. 23. Find the equation of the line with x-intercept x = 4 and y-intercept y = 3. Intercept-intercept Form Show that if a, b = 0, then the line with x-intercept x = a and y-intercept y = b y x SOLUTION From Exercise has equation (Figure 12) 22, 4 + 3 = 1 or 3x + 4y = 12. y x + cy = 1 25. Determine whether there exists a constant c such thatx the line 1 (3, y) lies on the line. A line of slope m = 2 passes through (1, 4). Find y+such=that a b Passes through (3, 1) (a) Has slope 4 (b) (c) Is horizontal (d) Is vertical SOLUTION

10

CHAPTER 1

PRECALCULUS REVIEW

(a) Rewriting the equation of the line in slope-intercept form gives y = − xc + 1c . To have slope 4 requires − 1c = 4 or c = − 14 . (b) Substituting x = 3 and y = 1 into the equation of the line gives 3 + c = 1 or c = −2. (c) From (a), we know the slope of the line is − 1c . There is no value for c that will make this slope equal to 0. (d) With c = 0, the equation becomes x = 1. This is the equation of a vertical line. 27. Materials expand when heated. Consider a metal rod of length L 0 at temperature T0 . If the temperature is changed Assume that the number N of concert tickets which can be sold at a price of P dollars per ticket is a linear by an amount T , then the rod’s length changes by L = α L 0 T , where α is the thermal expansion coefficient. For function N (P) for 10 ≤ P ≤ 40. Determine N (P) (called the demand function) if N (10) = 500 and N (40) = 0. steel, α = 1.24 × 10−5 ◦ C−1 . What is the decrease N in the number of tickets sold if the price is increased by P = 5 dollars? (a) A steel rod has length L 0 = 40 cm at T0 = 40◦ C. What is its length at T = 90◦ C? (b) Find its length at T = 50◦ C if its length at T0 = 100◦ C is 65 in. (c) Express length L as a function of T if L 0 = 65 in. at T0 = 100◦ C. SOLUTION

(a) With T = 90◦ C and T0 = 40◦ C, T = 50◦ C. Therefore, L = α L 0 T = (1.24 × 10−5 )(40)(50) = .0248

and

L = L 0 + L = 40.0248 cm.

(b) With T = 50◦ C and T0 = 100◦ C, T = −50◦ C. Therefore, L = α L 0 T = (1.24 × 10−5 )(65)(−50) = −.0403

and

L = L 0 + L = 64.9597 in.

(c) L = L 0 + L = L 0 + α L 0 T = L 0 (1 + α T ) = 65(1 + α (T − 100)) 29. Find b such that (2, −1), (3, 2), and (b, 5) lie on a line. Do the points (0.5, 1), (1, 1.2), (2, 2) lie on a line? SOLUTION The slope of the line determined by the points (2, −1) and (3, 2) is 2 − (−1) = 3. 3−2 To lie on the same line, the slope between (3, 2) and (b, 5) must also be 3. Thus, we require 3 5−2 = = 3, b−3 b−3 or b = 4. 31. The period T of a pendulum is measured for pendulums of several different lengths L. Based on the following data, Find an expression for the velocity v as a linear function of t that matches the following data. does T appear to be a linear function of L? t (s) 0 2 4 6 L (ft) 2 3 4 5 v (m/s) 39.2 58.6 78 97.4 T (s) 1.57 1.92 2.22 2.48 SOLUTION

Examine the slope between consecutive data points. The first pair of data points yields a slope of 1.92 − 1.57 = .35, 3−2

while the second pair of data points yields a slope of 2.22 − 1.92 = .3, 4−3 and the last pair of data points yields a slope of 2.48 − 2.22 = .26 5−4 Because the three slopes are not equal, T does not appear to be a linear function of L. 33. Find the roots of the quadratic polynomials: Show that f (x) is linear of slope m if and only if (a) 4x 2 − 3x − 1 (b) x 2 − 2x − 1 f (x + h) − f (x) = mh (for all x and h) SOLUTION √ √ 3 ± 9 − 4(4)(−1) 3 ± 25 1 (a) x = = = 1 or − 2(4) 8 4 √ √ √ 2 ± 4 − (4)(1)(−1) 2± 8 (b) x = = =1± 2 2 2

Linear and Quadratic Functions

S E C T I O N 1.2

11

In Exercises 34–41, complete the square and find the minimum or maximum value of the quadratic function. 35. y = x 2 −26x + 9 y = x + 2x + 5 SOLUTION y = (x − 3)2 ; therefore, the minimum value of the quadratic polynomial is 0, and this occurs at x = 3. 37. y = x 2 + 6x2+ 2 y = −9x + x SOLUTION y = x 2 + 6x + 9 − 9 + 2 = (x + 3)2 − 7; therefore, the minimum value of the quadratic polynomial is −7, and this occurs at x = −3. 39. y = −4x 2 2+ 3x + 8 y = 2x − 4x − 7 3 9 9 3 137 SOLUTION y = −4x 2 + 3x + 8 = −4(x 2 − 4 x + 64 ) + 8 + 16 = −4(x − 8 )2 + 16 ; therefore, the maximum 3 value of the quadratic polynomial is 137 16 , and this occurs at x = 8 . 2 41. y = 4x − 12x y = 3x 2 + 12x − 5 1 ) + 1 = −12(x − 1 )2 + 1 ; therefore, the maximum value of the SOLUTION y = −12(x 2 − x3 ) = −12(x 2 − x3 + 36 3 6 3 quadratic polynomial is 13 , and this occurs at x = 16 .

43. Sketch the graph of y = x 2 +24x + 6 by plotting the minimum point, the y-intercept, and one other point. Sketch the graph of y = x − 6x + 8 by plotting the roots and the minimum point. SOLUTION y = x 2 + 4x + 4 − 4 + 6 = (x + 2)2 + 2 so the minimum occurs at (−2, 2). If x = 0, then y = 6 and if x = −4, y = 6. This is the graph of x 2 moved left 2 units and up 2 units. 10 8 6 4 2 –4

–3

–2

–1

45. ForIfwhich valuesAofand c does (x) cystic = x2 + cx + 1gene haveoccur a double No real roots? the alleles B off the fibrosis in a root? population with frequencies p and 1 − p (where p is a fraction between 0 and 1), then carriers (carriers SOLUTION A double root occurs whenthe c2frequency − 4(1)(1)of =heterozygous 0 or c2 = 4. Thus, c= ±2. with both alleles) is 2 p(1 − p). Which of proots giveswhen the largest frequency There arevalue no real c2 − 4(1)(1) < 0oforheterozygous c2 < 4. Thus,carriers? −2 < c < 2. 1 2 for all x > 0. Hint: Consider (x 1/2 − x −1/2 )2 . 47. ProveLet that fx(x) + be ≥ x a quadratic function and c a constant. Which is correct? Explain graphically. (a) There is a unique value of c such that y = f (x) − c has a double root. SOLUTION Let x > 0. Then (b) There is a unique value of c such that y = f (x − c) has a double root. 2

1 x 1/2 − x −1/2 = x − 2 + . x Because (x 1/2 − x −1/2 )2 ≥ 0, it follows that x −2+

1 ≥0 x

x+

or

1 ≥ 2. x

49. If objects of weights x and w1 are suspended from the balance in Figure 13(A), the cross-beam is horizontal if √ a+b bx = aw If the a and are geometric known, wemean may useabthis determine an unknown selecting Hint: Use a is equation not largertothan the arithmetic meanweight x. by Let b >lengths 0. Show thatb the 1 . a, 2 First balance x w1 so that the cross-beam is horizontal. If a and b are not known precisely, we might proceed as follows. variation of the hint given in Exercise 47. by w1 on the left as in (A). Then switch places and balance x by w2 on the right as in (B). The average x¯ = 12 (w1 + w2 ) gives an estimate for x. Show that x¯ is greater than or equal to the true weight x. a

a

b

w1

x (A)

b

x

w2 (B)

FIGURE 13

12

CHAPTER 1

PRECALCULUS REVIEW SOLUTION

First note bx = aw1 and ax = bw2 . Thus, x¯ = = = ≥ =

1 (w1 + w2 ) 2 1 bx ax + 2 a b x b a + 2 a b x (2) by Exercise 47 2 x

51. Find a pair of numbers whose sum and product are both equal to 8. Find numbers x and y with sum 10 and product 24. Hint: Find a quadratic polynomial satisfied by x. SOLUTION Let x and y be numbers whose sum and product are both equal to 8. Then x + y = 8 and x y = 8. From the second equation, y = 8x . Substituting this expression for y in the first equation gives x + 8x = 8 or x 2 − 8x + 8 = 0. By the quadratic formula, √ √ 8 ± 64 − 32 = 4 ± 2 2. x= 2 √ If x = 4 + 2 2, then √ √ 8 8 4−2 2 y= √ = √ · √ = 4 − 2 2. 4+2 2 4+2 2 4−2 2 √ On the other hand, if x = 4 − 2 2, then √ √ 8 8 4+2 2 y= √ = √ · √ = 4 + 2 2. 4−2 2 4−2 2 4+2 2 √ √ Thus, the two numbers are 4 + 2 2 and 4 − 2 2. 2 consists of all points P such that d = d , where d is the distance 1 2 1 from P to (0, 4 ) and d2 is the distance from P to the horizontal line y = − 14 (Figure 14). 53. Show that if f (x) and g(x) are linear, then so is f (x) + g(x). Is the same true of f (x)g(x)?

Show that the graph of the parabola y = x Further Insights and Challenges 1 SOLUTION

If f (x) = mx + b and g(x) = nx + d, then f (x) + g(x) = mx + b + nx + d = (m + n)x + (b + d),

which is linear. f (x)g(x) is not generally linear. Take, for example, f (x) = g(x) = x. Then f (x)g(x) = x 2 . overfthe ] is=not a constant, but=depends 55. Show thatthat the ifratio y/x forare thelinear function f (x) = x 2 that 1 , xf 2(1) Show f (x) and g(x) functions such (0) interval = g(0) [x and g(1), then f (x) g(x). on the interval. Determine the exact dependence of y/x on x 1 and x 2 . SOLUTION

For x 2 ,

x 2 − x 12 y = 2 = x2 + x1 . x x2 − x1

57. Let a, c = 0. Show that the roots of ax 2 + bx + c = 0 and cx 2 +2 bx + a = 0 are reciprocals of each other. Use Eq. (2) to derive the quadratic formula for the roots of ax + bx + c = 0. SOLUTION Let r1 and r2 be the roots of ax 2 + bx + c and r3 and r4 be the roots of cx 2 + bx + a. Without loss of generality, let −b + b2 − 4ac 2a 1 −b − b2 − 4ac r1 = = ⇒ · 2a r1 −b + b2 − 4ac −b − b2 − 4ac −b − b2 − 4ac 2a(−b − b2 − 4ac) = = r4 . = 2c b2 − b2 + 4ac Similarly, you can show

1 = r3 . r2

59. Prove Viète’s Formulas, which state that the quadratic polynomial with given numbers α and β as roots is x 2 + square to show that. the parabolas y = ax 2 + bx + c and y = ax 2 have the same shape (show that bx + c, Complete where b =the −α − β and c = αβ the first parabola is congruent to the second by a vertical and horizontal translation). SOLUTION If a quadratic polynomial has roots α and β , then the polynomial is (x − α )(x − β ) = x 2 − α x − β x + αβ = x 2 + (−α − β )x + αβ . Thus, b = −α − β and c = αβ .

S E C T I O N 1.3

The Basic Classes of Functions

13

1.3 The Basic Classes of Functions Preliminary Questions 1. Give an example of a rational function. SOLUTION

One example is

3x 2 − 2 . 7x 3 + x − 1

2. Is |x| a polynomial function? What about |x 2 + 1|? SOLUTION

|x| is not a polynomial; however, because x 2 + 1 > 0 for all x, it follows that |x 2 + 1| = x 2 + 1, which is

a polynomial. 3. What is unusual about the domain of f ◦ g for f (x) = x 1/2 and g(x) = −1 − |x|? SOLUTION Recall that ( f ◦ g)(x) = f (g(x)). Now, for any real number x, g(x) = −1 − |x| ≤ −1 < 0. Because we cannot take the square root of a negative number, it follows that f (g(x)) is not defined for any real number. In other words, the domain of f (g(x)) is the empty set. x 4. Is f (x) = 12 increasing or decreasing? SOLUTION

function.

The function f (x) = ( 12 )x is an exponential function with base b = 12 < 1. Therefore, f is a decreasing

5. Give an example of a transcendental function. SOLUTION

One possibility is f (x) = e x − sin x.

Exercises In Exercises 1–12, determine the domain of the function. 1. f (x) = x 1/4 SOLUTION

x≥0

3. f (x) = x 3 + 3x − 4 g(t) = t 2/3 SOLUTION All reals 1 5. g(t)h(z) = = z 3 + z −3 t +2 SOLUTION

t = −2

1 7. G(u) = 2 1 f (x) u= −2 4 x +4 SOLUTION u = ±2 √(x − 1)−3 9. f (x) = x −4 + x f (x) = 20, 1 SOLUTION x = x −9 √ −1 11. g(y) = 10 y+y s F(s) = sin SOLUTION y > 0 s + 1 In Exercises 13–24, identify each of the following functions as polynomial, rational, algebraic, or transcendental. x + x −1 f (x) =3 2 − 8+ 4) +− 9x3)(x 13. f (x) = 4x (x Polynomial √ 15. f (x) = x f (x) = x −4 SOLUTION Algebraic SOLUTION

x2 17. f (x)f (x) = = 1 − x2 x + sin x SOLUTION

Transcendental

2x 3 x+ 3x 19. f (x)f (x) = =2 2 9 − 7x SOLUTION

Rational

21. f (x) = sin(x 2 ) 3x − 9x −1/2 f (x) = 9 − 7x 2

14

CHAPTER 1

PRECALCULUS REVIEW SOLUTION

Transcendental

23. f (x) = x 2 + 3xx −1 f (x) = √ SOLUTION Rational x +1 2

x a transcendental 25. Is f f(x) function? x) (x)==2sin(3 SOLUTION Yes.

In Exercises 27–34, calculate2 the composite functions f ◦ g and g ◦ f , and determine their domains. Show that f (x) = x + 3x −1 and g(x) = 3x 3 − 9x + x −2 are rational functions (show that each is a quotient √ polynomials). 27. of f (x) = x, g(x) = x + 1 √ √ SOLUTION f (g(x)) = x + 1; D: x ≥ −1, g( f (x)) = x + 1; D: x ≥ 0 29. f (x) = 2x , 1 g(x) = x 2 −4 f (x) = , g(x) = 2x SOLUTION f (g(x)) x = 2x ; D: R,

g( f (x)) = (2x )2 = 22x ; D: R

3 2 31. f (θf)(x) = cos θ , g(x) = |x|, g(θ )==xsin+θ x 3 SOLUTION f (g(x)) = cos(x + x 2 ); D: R,

g( f (θ )) = cos3 θ + cos2 θ ; D: R

1 33. f (t) = √ , g(t) 1 = −t 2 , g(x) = x −2 f (x) =t 2 x +1 1 SOLUTION f (g(t)) = ; D: Not valid for any t, −t 2

2 g( f (t)) = − √1 = − 1t ; D: t > 0 t

kt √ (in millions) of 35. The population 3 a country as a function of time t (years) is P(t) = 30 · 2 , with k = 0.1. Show f (t) = t, g(t) = 1 − t that the population doubles every 10 years. Show more generally that for any nonzero constants a and k, the function g(t) = a2kt doubles after 1/k years. SOLUTION

Let P(t) = 30 · 20.1t . Then P(t + 10) = 30 · 20.1(t+10) = 30 · 20.1t+1 = 2(30 · 20.1t ) = 2P(t).

Hence, the population doubles in size every 10 years. In the more general case, let g(t) = a2kt . Then 1 = a2k(t+1/k) = a2kt+1 = 2a2kt = 2g(t). g t+ k Hence, the function g doubles after 1/k years. x +1 is R. f (x) = 2 x + 2cx + 4 In Exercises 37–43, we define the first difference δ f of a function f (x) by δ f (x) = f (x + 1) − f (x).

Further Insights Challenges Find all valuesand of c such that the domain of

37. Show that if f (x) = x 2 , then δ f (x) = 2x + 1. Calculate δ f for f (x) = x and f (x) = x 3 . f (x) = x 2 : δ f (x) = f (x + 1) − f (x) = (x + 1)2 − x 2 = 2x + 1 f (x) = x: δ f (x) = x + 1 − x = 1 f (x) = x 3 : δ f (x) = (x + 1)3 − x 3 = 3x 2 + 3x + 1

SOLUTION

39. Show that for any two functions f and g, δ ( f + g) = δ f + δ g and δ (c f ) = cδ ( f ), where c is any constant. Show that δ (10x ) = 9 · 10x , and more generally, δ (b x ) = c · b x for some constant c. SOLUTION δ ( f + g) = ( f (x + 1) + g(x + 1)) − ( f (x) − g(x)) = ( f (x + 1) − f (x)) + (g(x + 1) − g(x)) = δ f (x) + δ g(x)

δ (c f ) = c f (x + 1) − c f (x) = c( f (x + 1) − f (x)) = cδ f (x). x(x + 1) 41. First show that P(x)can = be used tosatisfies δ P = (x for + 1). 40 toSuppose conclude First differences derive formulas theThen sum apply of theExercise kth powers. wethat can find a function 2 P(x) such that δ P = (x + 1)k and P(0) = 0. Prove that P(1) = 1k , P(2) = 1k + 2k , and more generally, for every n(n + 1) whole number n, 1 + 2 + 3 + ··· + n = 2 P(n) = 1k + 2k + · · · + n k SOLUTION Let P(x) = x(x + 1)/2. Then

δ P(x) = P(x + 1) − P(x) =

(x + 1)(x + 2) x(x + 1) (x + 1)(x + 2 − x) − = = x + 1. 2 2 2

Also, note that P(0) = 0. Thus, by Exercise 40, with k = 1, it follows that P(n) =

n(n + 1) = 1 + 2 + 3 + · · · + n. 2

Calculate δ (x 3 ), δ (x 2 ), and δ (x). Then find a polynomial P(x) of degree 3 such that δ P = (x + 1)2 and P(0) = 0. Conclude that

S E C T I O N 1.3

The Basic Classes of Functions

15

43. This exercise combined with Exercise 40 shows that for all k, there exists a polynomial P(x) satisfying Eq. (1). The solution requires proof by induction and the Binomial Theorem (see Appendix C). (a) Show that

δ (x k+1 ) = (k + 1) x k + · · · where the dots indicate terms involving smaller powers of x. (b) Show by induction that for all whole numbers k, there exists a polynomial of degree k + 1 with leading coefficient 1/(k + 1): P(x) =

1 x k+1 + · · · k+1

such that δ P = (x + 1)k and P(0) = 0. SOLUTION

(a) By the Binomial Theorem: n+1 n+1 δ (x n+1 ) = (x + 1)n+1 − x n+1 = x n+1 + xn + x n−1 + · · · + 1 − x n+1 1 2 n+1 n+1 xn + x n−1 + · · · + 1 = 1 2 Thus,

δ (x n+1 ) = (n + 1) x n + · · · where the dots indicate terms involving smaller powers of x. (b) For k = 0, note that P(x) = x satisfies δ P = (x + 1)0 = 1 and P(0) = 0. Now suppose the polynomial P(x) =

1 k x + pk−1 x k−1 + · · · + p1 x k

which clearly satisfies P(0) = 0 also satisfies δ P = (x + 1)k−1 . We try to prove the existence of Q(x) =

1 x k+1 + qk x k + · · · + q1 x k+1

such that δ Q = (x + 1)k . Observe that Q(0) = 0. If δ Q = (x + 1)k and δ P = (x + 1)k−1 , then

δ Q = (x + 1)k = (x + 1)δ P = x δ P(x) + δ P By the linearity of δ (Exercise 39), we find δ Q − δ P = x δ P or δ (Q − P) = x δ P. By definition, 1 1 Q−P= x k+1 + qk − x k + · · · + (q1 − p1 )x, k+1 k so, by the linearity of δ ,

δ (Q − P) =

1 1 δ (x k+1 ) + qk − δ (x k ) + · · · + (q1 − p1 ) = x(x + 1)k−1 k+1 k

By part (a),

δ (x k+1 ) = (k + 1)x k + L k−1,k−1 x k−1 + . . . + L k−1,1 x + 1 δ (x k ) = kx k−1 + L k−2,k−2 x k−2 + . . . + L k−2,1 x + 1 .. .

δ (x 2 ) = 2x + 1 where the L i, j are real numbers for each i, j. To construct Q, we have to group like powers of x on both sides of (1). This yields the system of equations 1

(k + 1)x k = x k k+1

(1)

16

CHAPTER 1

PRECALCULUS REVIEW

1 1 L k−1,k−1 x k−1 + qk − kx k−1 = (k − 1)x k−1 k+1 k .. .

1 1 + qk − + (qk−1 − pk−1 ) + · · · + (q1 − p1 ) = 0. k+1 k The first equation is identically true, and the second equation can be solved immediately for qk . Substituting the value of qk into the third equation of the system, we can then solve for qk−1 . We continue this process until we substitute the values of qk , qk−1 , . . . q2 into the last equation, and then solve for q1 .

1.4 Trigonometric Functions Preliminary Questions 1. How is it possible for two different rotations to define the same angle? SOLUTION Working from the same initial radius, two rotations that differ by a whole number of full revolutions will have the same ending radius; consequently, the two rotations will define the same angle even though the measures of the rotations will be different.

2. Give two different positive rotations that define the angle π4 . π SOLUTION The angle π /4 is defined by any rotation of the form 4 + 2π k where k is an integer. Thus, two different positive rotations that define the angle π /4 are

9π π + 2π (1) = 4 4

and

41π π + 2π (5) = . 4 4

3. Give a negative rotation that defines the angle π3 . SOLUTION The angle π /3 is defined by any rotation of the form π 3 + 2π k where k is an integer. Thus, a negative rotation that defines the angle π /3 is

5π π + 2π (−1) = − . 3 3 4. The definition of cos θ using right triangles applies when (choose the correct answer): π (a) 0 < θ < (c) 0 < θ < 2π (b) 0 < θ < π 2 The correct response is (a): 0 < θ < π2 . 5. What is the unit circle definition of sin θ ?

SOLUTION

SOLUTION Let O denote the center of the unit circle, and let P be a point on the unit circle such that the radius O P makes an angle θ with the positive x-axis. Then, sin θ is the y-coordinate of the point P.

6. How does the periodicity of sin θ and cos θ follow from the unit circle definition? SOLUTION Let O denote the center of the unit circle, and let P be a point on the unit circle such that the radius O P makes an angle θ with the positive x-axis. Then, cos θ and sin θ are the x- and y-coordinates, respectively, of the point P. The angle θ + 2π is obtained from the angle θ by making one full revolution around the circle. The angle θ + 2π will therefore have the radius O P as its terminal side. Thus

cos(θ + 2π ) = cos θ

and

sin(θ + 2π ) = sin θ .

In other words, sin θ and cos θ are periodic functions.

Exercises 1. Find the angle between 0 and 2π that is equivalent to 13π /4. SOLUTION Because 13π /4 > 2π , we repeatedly subtract 2π until we arrive at a radian measure that is between 0 and 2π . After one subtraction, we have 13π /4 − 2π = 5π /4. Because 0 < 5π /4 < 2π , 5π /4 is the angle measure between 0 and 2π that is equivalent to 13π /4.

3. Convert from radians to degrees: Describe the angle θ = π6 by 5 π an angle of negative radian measure. (c) (a) 1 (b) 3 12

(d) −

3π 4

Trigonometric Functions

S E C T I O N 1.4 SOLUTION

180◦ 180◦ ≈ 57.1◦ = π π 5 180◦ 75◦ (c) ≈ 23.87◦ = 12 π π

180◦ = 60◦ π 3π 180◦ (d) − = −135◦ 4 π

(a) 1

(b)

π 3

5. Find the lengths of the arcs subtended by the angles θ and φ radians in Figure 20. Convert from degrees to radians: (b) 30◦ (c) 25◦ (a) 1◦

(d) 120◦

4 q = 0.9 f =2

FIGURE 20 Circle of radius 4. SOLUTION

s = r θ = 4(.9) = 3.6; s = r φ = 4(2) = 8

7. Fill in the remaining values of (cos θ , sin θ ) for the points in Figure 22. Calculate the values of the six standard trigonometric functions for the angle θ in Figure 21. 3p 4

2p 3

p 2

p 3

5p 6

( 12 , 23 ) p 2 2 , 4 ( 2 2 ) p 3 1 , ( 6 2 2)

p

0 (0, 0)

7p 6

11p 6 5p 4 4p 3

7p 5p 4 3

3p 2

FIGURE 22 SOLUTION

θ

π 2

(cos θ , sin θ )

(0, 1)

2π 3 √

−1 , 3 2 2

θ

5π 4

4π 3

(cos θ , sin θ )

√

√ − 2, − 2 2 2

√

3π 4

√ − 2, 2 2 2 3π 2

√ −1 , − 3 2 2

5π 6

√ − 3, 1 2 2 5π 3 √

1, − 3 2 2

(0, −1)

(−1, 0)

7π 6

√ − 3 , −1 2 2

7π 4

11π 6

π

√

√ 2 − 2 2 , 2

π θsatisfying the given condition. In Exercises use Figure 22 to find alltrigonometric angles between 0 and 2at Find 9–14, the values of the six standard functions = 11π /6. 9. cos θ = 12

θ = π3 , 53π 11. tan θ = −1 tan θ = 1 3π 7π SOLUTION θ = 4 , 4 √ 3 13. sin csc x =θ = 2 2 SOLUTION

x = π3 , 23π 15. Fill in the following table of values: sec t = 2 SOLUTION

θ tan θ sec θ SOLUTION

π 6

π 4

π 3

π 2

2π 3

3π 4

5π 6

√

3 −1 2 , 2

17

18

CHAPTER 1

PRECALCULUS REVIEW

π 6

π 4

tan θ

1 √ 3

1

sec θ

2 √ 3

θ

√

2

π 3 √ 3 2

π 2

3π 4

5π 6

und

2π 3 √ − 3

−1

1 −√ 3

und

−2

√ − 2

2 −√ 3

17. Show that if tan = c and 0table ≤ θof<signs π /2, for then θ = 1/ 1 +functions: c2 . Hint: Draw a right triangle whose opposite and Complete theθfollowing thecos trigonometric adjacent sides have lengths c and 1. SOLUTION Because 0 ≤ θ < π /2, definition the trigonometric in terms of right triangles. θ we can use the sin cos of tan cot sec functions csc tan θ is the ratio of the length of the side opposite the angle θ to the length of the adjacent side. With c = 1c , we label π the length of the opposite side as c 0and length of the θ< + adjacent + side as 1 (see the diagram below). By the Pythagorean < the theorem, the length of the hypotenuse is 12+ c2 . Finally, we use the fact that cos θ is the ratio of the length of the adjacent side to the length of the hypotenuse to obtain π 5? SOLUTION

Yes. If limx→5 f (x) exists, then both one-sided limits must exist and be equal.

8. Which of the following pieces of information is sufficient to determine whether lim f (x) exists? Explain. x→5

(a) (b) (c) (d) (e)

The values of The values of The values of The values of f (5)

f (x) for all x f (x) for x in [4.5, 5.5] f (x) for all x in [4.5, 5.5] other than x = 5 f (x) for all x ≥ 5

SOLUTION To determine whether lim x→5 f (x) exists, we must know the values of f (x) for values of x near 5, both smaller than and larger than 5. Thus, the information in (a), (b) or (c) would be sufficient. The information in (d) does not include values of f for x < 5, so this information is not sufficient to determine whether the limit exists. The limit does not depend at all on the value f (5), so the information in (e) is also not sufficient to determine whether the limit exists.

Exercises In Exercises 1–4, fill in the tables and guess the value of the limit. x3 − 1 . 1. lim f (x), where f (x) = 2 x→1 x −1 f (x)

x

f (x)

x

1.002

0.998

1.001

0.999

1.0005

0.9995

1.00001

0.99999

SOLUTION

x

0.998

0.999

0.9995

0.99999

1.00001

1.0005

1.001

1.002

f (x)

1.498501

1.499250

1.499625

1.499993

1.500008

1.500375

1.500750

1.501500

38

CHAPTER 2

LIMITS

The limit as x → 1 is 32 . 2− y−2 ycos t −1 3. limlim f (y), where f (y) = h(t), where h(t) =y 2 + 2y − 6. .Note that h(t) is even, that is, h(t) = h(−t). y→2 t→0 t

t

y±0.002 f (y) y ±0.0001

f (y) ±0.00005

h(t) 2.002

1.998

2.001

1.999

2.0001

1.9999

±0.00001

SOLUTION

y

1.998

1.999

1.9999

2.0001

2.001

2.02

f (y)

0.59984

0.59992

0.599992

0.600008

0.60008

0.601594

The limit as y → 2 is 35 . 5. Determine lim f (x) for the function f (x) shown in Figure 8. sin θ − θ x→0.5 lim f (θ ), where f (θ ) = . θ →0 θ3 y

θ f (θ )

±0.002

±0.0001

±0.00005

±0.00001

1.5

x .5

FIGURE 8 SOLUTION

The graph suggests that f (x) → 1.5 as x → .5.

In Exercises 7–8, of evaluate limit. functions in Figure 9 appear to approach a limit as x → 0? Do either the twothe oscillating 7. lim x x→21

SOLUTION

As x → 21, f (x) = x → 21. You can see this, for example, on the graph of f (x) = x.

√ verify each limit using the limit definition. For example, in Exercise 9, show that |2x − 6| can be made In Exercises 9–18, lim 3 as smallx→4.2 as desired by taking x close to 3. 9. lim 2x = 6 x→3

SOLUTION

|x − 3| small.

|2x − 6| = 2|x − 3|. |2x − 6| can be made arbitrarily small by making x close enough to 3, thus making

11. lim (4x + 3) = 11 x→2lim 4 = 4 x→3

|(4x + 3) − 11| = |4x − 8| = 4|x − 2|. Therefore, if you make |x − 2| small enough, you can make |(4x + 3) − 11| as small as desired. SOLUTION

13. lim (−2x) = −18 x→9lim (5x − 7) = 8 x→3

We have |−2x − (−18)| = |(−2)(x − 9)| = 2 |x − 9|. If you make |x − 9| small enough, you can make 2 |x − 9| = |−2x − (−18)| as small as desired. SOLUTION

x 2 =(10 − 2x) = 11 15. lim lim x→0 x→−5

As x → 0, we have |x 2 − 0| = |x + 0||x − 0|. To simplify things, suppose that |x| < 1, so that |x + 0||x − 0| = |x||x| < |x|. By making |x| sufficiently small, so that |x + 0||x − 0| = x 2 is even smaller, you can make |x 2 − 0| as small as desired. SOLUTION

+ 3) = 3 17. lim (x 2 + 2x x→0lim (2x 2 + 4) = 4 x→0

As x → 0, we have |x 2 + 2x + 3 − 3| = |x 2 + 2x| = |x||x + 2|. If |x| < 1, |x + 2| can be no bigger than 3, so |x||x + 2| < 3|x|. Therefore, by making |x − 0| = |x| sufficiently small, you can make |x 2 + 2x + 3 − 3| = |x||x + 2| as small as desired. SOLUTION

lim (x 3 + 9) = 9

x→0

Limits: A Numerical and Graphical Approach

S E C T I O N 2.2

In Exercises 19–32, estimate the limit numerically or state that the limit does not exist. √ x −1 19. lim x→1 x − 1 SOLUTION

x

.9995

.99999

1.00001

1.0005

f (x)

.500063

.500001

.49999

.499938

x

1.999

1.99999

2.00001

2.001

f (x)

1.666889

1.666669

1.666664

1.666445

x

−0.01

−0.005

0.005

0.01

f (x)

1.999867

1.999967

1.999967

1.999867

The limit as x → 1 is 12 . x2 + x − 6 2x 2 − 18 x→2 lim x2 − x − 2 x→−3 x + 3

21. lim

SOLUTION

The limit as x → 2 is 53 . sin 2x 23. lim x 3 − 2x 2 − 9 x→0lim x x→3 x 2 − 2x − 3 SOLUTION

The limit as x → 0 is 2. sin x 25. lim sin 5x x→0limx 2 x→0 x SOLUTION

x

−.01

−.001

−.0001

.0001

.001

.01

f (x)

−99.9983

−999.9998

−10000.0

10000.0

999.9998

99.9983

The limit does not exist. As x → 0−, f (x) → −∞; similarly, as x → 0+, f (x) → ∞. 1 27. lim cos cos θ − 1 h→0lim h θ θ →0 SOLUTION

h

±0.1

±0.01

±0.001

±0.0001

f (h)

−0.839072

0.862319

0.562379

−0.952155

The limit does not exist since cos (1/ h) oscillates infinitely often as h → 0. 2h − 1 29. limlim sin h cos 1 h h→0 h h→0 SOLUTION

h

−.05

−.001

.001

.05

f (h)

.681273

.692907

.693387

.705298

The limit as x → 0 is approximately 0.693. (The exact answer is ln 2.) 5h −x25 x 2 −3 31. lim lim h→2 h − 2 x x→0 SOLUTION

39

40

CHAPTER 2

LIMITS

x

1.95

1.999

2.001

2.05

f (x)

38.6596

40.2036

40.2683

41.8992

The limit as h → 2 is approximately 40.2. (The exact answer is 25 ln 5.) 33. Determine x lim f (x) and lim f (x) for the function shown in Figure 10. lim |x| x→2+ x→2− x→0

y

2

1

x 2

FIGURE 10 SOLUTION

The left-hand limit is lim f (x) = 2, whereas the right-hand limit is lim f (x) = 1. Accordingly, the x→2−

x→2+

two-sided limit does not exist. 35. The greatest integer function is defined by [x] = n, where n is the unique integer such that n ≤ x < n + 1. See Determine the one-sided limits at c = 1, 2, 4, 5 of the function g(t) shown in Figure 11 and state whether the Figure 12. limit exists at these points. (a) For which values of c does lim [x] exist? What about lim [x]? x→c−

x→c+

(b) For which values of c does lim [x] exist? x→c

y 2 1 x

−1

1

2

3

FIGURE 12 Graph of y = [x]. SOLUTION

(a) The one-sided limits exist for all real values of c. (b) For each integer value of c, the one-sided limits differ. In particular, lim [x] = c − 1, whereas lim [x] = c. (For x→c−

x→c+

noninteger values of c, the one-sided limits both equal [c].) The limit lim [x] exists when x→c

lim [x] = lim [x] ,

x→c−

x→c+

namely for noninteger values of c: n < c < n + 1, where n is an integer. In Exercises 37–39, determine the xone-sided limits numerically. −1 and use it to determine the one-sided limits lim f (x) and lim f (x). Draw a graph of f (x) = |x − 1| x→1+ x→1− sin x 37. lim |x| x→0± SOLUTION

x

−.2

−.02

.02

.2

f (x)

−.993347

−.999933

.999933

.993347

The left-hand limit is lim f (x) = −1, whereas the right-hand limit is lim f (x) = 1. x→0−

x→0+

x − sin(|x|) 39. limlim |x|1/x 3 x→0± x→0± x SOLUTION

x

−.1

−.01

.01

.1

f (x)

199.853

19999.8

.166666

.166583


S E C T I O N 2.2

The left-hand limit is lim f (x) = ∞, whereas the right-hand limit is lim f (x) = x→0−

x→0+

1 . 6

41. Determine the one-sided limits of f (x) at c = 2 and c = 4, for the function shown in Figure 14. Determine the one- or two-sided infinite limits in Figure 13. y 15 10 5 x 2

4

−5

FIGURE 14 SOLUTION

• For c = 2, we have lim f (x) = ∞ and lim f (x) = ∞. x→2− x→2+ • For c = 4, we have lim f (x) = −∞ and lim f (x) = 10. x→4− x→4+

In Exercises 43–46,the draw the graph of atwo-sided function with theingiven Determine infinite one- and limits Figurelimits. 15. 43. lim f (x) = 2, lim f (x) = 0, lim f (x) = 4 x→1

x→3−

x→3+

SOLUTION y 6 4 2

x 1

45.

2

3

4

lim f (x) = f (2) = 3, lim f (x) = −1, lim f (x) = 2 = f (4)

lim f (x) = ∞, lim x→2− f (x) = 0, lim fx→4 (x) = −∞ x→2+ x→1 x→3− x→3+

SOLUTION y 3 2 1 x −1

1

2

3

4

5

Inlim Exercises 47–52, use = the−∞ graph to estimate the value of the limit. f (x) = ∞, graph lim fthe (x)function = 3, limandf (x) x→1+

sin 3θ 47. lim θ →0 sin 2θ

x→1−

x→4

SOLUTION y 1.500 y= 1.495 1.490 1.485

The limit as θ → 0 is 32 . 2x − 1 x→0 4x − 1 lim

sin 3 sin 2

41

42

CHAPTER 2

LIMITS

2x − cos x x x→0

49. lim

SOLUTION y 0.6940 0.6935 y=

2x

− cos x x

0.6930

0.6925 0.6920

The limit as x → 0 is approximately 0.693. (The exact answer is ln 2.) cos 3θ − cos 4θ 51. lim sinθ224θ θ →0lim θ →0 cos θ − 1 SOLUTION

y 3.500

y=

cos 3 − cos 4 2

3.475 3.450 3.425

The limit as θ → 0 is 3.5. cos 3θ − cos 5θ FurtherlimInsights and Challenges 2

θ →0 θ 53. Light waves of frequency λ passing through a slit of width a produce a Fraunhofer diffraction pattern of light and dark fringes (Figure 16). The intensity as a function of the angle θ is given by sin(R sin θ ) 2 I (θ ) = I m R sin θ

where R = π a/λ and Im is a constant. Show that the intensity function is not defined at θ = 0. Then check numerically that I (θ ) approaches Im as θ → 0 for any two values of R (e.g., choose two integer values).

a Incident light waves Slit

Viewing screen

Intensity pattern

FIGURE 16 Fraunhofer diffraction pattern. SOLUTION

If you plug in θ = 0, you get a division by zero in the expression

sin R sin θ ; R sin θ

thus, I (0) is undefined. If R = 2, a table of values as θ → 0 follows:

θ

−0.01

−0.005

0.005

0.01

I (θ )

0.998667 Im

0.9999667 Im

0.9999667 Im

0.9998667 Im

The limit as θ → 0 is 1 · Im = Im . If R = 3, the table becomes:

θ

−0.01

−0.005

0.005

0.01

I (θ )

0.999700 Im

0.999925 Im

0.999925 Im

0.999700 Im

S E C T I O N 2.2


43

Again, the limit as θ → 0 is 1Im = Im . bx − 1 sin nlim θ 55. Show numerically that for b = 3, 5 appears to equal ln 3, ln 5, where ln x is the natural logarithm. Then for several values of n and then guess the value in general. Investigate lim x x→0numerically θ θ →0 make a conjecture (guess) for the value in general and test your conjecture for two additional values of b (these results are explained in Chapter 7). SOLUTION

•

−.1

−.01

−.001

.001

.01

.1

1.486601

1.596556

1.608144

1.610734

1.622459

1.746189

−.1

−.01

−.001

.001

.01

.1

1.040415

1.092600

1.098009

1.099216

1.104669

1.161232

x 5x − 1 x We have ln 5 ≈ 1.6094. •

x 3x − 1 x We have ln 3 ≈ 1.0986. • We conjecture that lim x→0

bx − 1 = ln b for any positive number b. Here are two additional test cases. x −.1

−.01

−.001

.001

.01

.1

−.717735

−.695555

−.693387

−.692907

−.690750

−.669670

x 1 x 2

−1

x

We have ln 12 ≈ −0.69315. x

−.1

−.01

−.001

.001

.01

.1

7x − 1 x

1.768287

1.927100

1.944018

1.947805

1.964966

2.148140

We have ln 7 ≈ 1.9459. xn − 1 for (m, n) equal to + (2,nx) 1),1/x (1,. 2), (2, 3), and (3, 2). Then guess the value of the limit in 57. Investigate lim number Let n be x→1 any lim (1 x m − 1 and consider x→0 general and check your guess for at least three additional pairs. (a) Give numerical evidence that for n = 1, the limit is e [where e = exp(1) can be found on any scientific SOLUTION calculator]. •(b) Estimate the limit for n = 2, 3 and try to express the results in terms of e. Then form a conjecture for the value of the limit in general. Is it valid for noninteger n? x .99 .9999 1.0001 1.01 x −1 x2 − 1

.502513

.500025

.499975

.497512

x

.99

.9999

1.0001

1.01

x2 − 1 x −1

1.99

1.9999

2.0001

2.01

The limit as x → 1 is 12 .

The limit as x → 1 is 2. x

.99

.9999

1.0001

1.01

x2 − 1 x3 − 1

0.670011

0.666700

0.666633

0.663344

44

CHAPTER 2

LIMITS

The limit as x → 1 is 23 . x

.99

.9999

1.0001

1.01

x3 − 1 x2 − 1

1.492513

1.499925

1.500075

1.507512


xn − 1 n • For general m and n, we have lim = . m x→1 x m − 1 • x

.99

.9999

1.0001

1.01

x −1 x3 − 1

.336689

.333367

.333300

.330022

x

.99

.9999

1.0001

1.01

x3 − 1 x −1

2.9701

2.9997

3.0003

3.0301

x

.99

.9999

1.0001

1.01

x3 − 1 x7 − 1

.437200

.428657

.428486

.420058


The limit as x → 1 is 3.

The limit as x → 1 is 37 ≈ 0.428571. 2x − 8 2 Sketch a graph of f (x) = with a graphing calculator. Observe that f (3) is not defined. x − 3 k such that lim sin(sin x) exists. Find by experimentation the positive integers k x→0 x (a) Zoom in on the graph to estimate L = lim f (x).

59.

x→3

(b) Observe that the graph of f (x) is increasing. Explain how this implies that f (2.99999) ≤ L ≤ f (3.00001) Use this to determine L to three decimal places. SOLUTION

(a)

y 5.565 5.555 5.545

y=

2x − 8 x−3

5.535 5.525 x=3

(b) It is clear that the graph of f rises as we move to the right. Mathematically, we may express this observation as: whenever u < v, f (u) < f (v). Because 2.99999 < 3 = lim f (x) < 3.00001, x→3

it follows that f (2.99999) < L = lim f (x) < f (3.00001). x→3

With f (2.99999) ≈ 5.54516 and f (3.00001) ≈ 5.545195, the above inequality becomes 5.54516 < L < 5.545195; hence, to three decimal places, L = 5.545. 21/x − 2−1/x The function f (x) = 1/x is defined for x = 0. −1/x

S E C T I O N 2.3

Basic Limit Laws

45

sin x is equal to the slope of a secant line through the origin and the point (x, sin x) on x the graph of y = sin x (Figure 17). Use this to give a geometric interpretation of lim f (x). Show that f (x) =

61.

x→0

y (x, sin x)

1 sin x

x x

FIGURE 17 Graph of y = sin x.

sin x sin x − 0 = . Since x −0 x f (x) is the slope of the secant line, the limit lim f (x) is equal to the slope of the tangent line to y = sin x at x = 0.

SOLUTION

The slope of a secant line through the points (0, 0) and (x, sin x) is given by x→0

2.3 Basic Limit Laws Preliminary Questions 1. State the Sum Law and Quotient Law. SOLUTION

Suppose limx→c f (x) and limx→c g(x) both exist. The Sum Law states that lim ( f (x) + g(x)) = lim f (x) + lim g(x).

x→c

x→c

x→c

Provided limx→c g(x) = 0, the Quotient Law states that lim

f (x)

x→c g(x)

=

limx→c f (x) . limx→c g(x)

2. Which of the following is a verbal version of the Product Law? (a) The product of two functions has a limit. (b) The limit of the product is the product of the limits. (c) The product of a limit is a product of functions. (d) A limit produces a product of functions. SOLUTION

The verbal version of the Product Law is (b): The limit of the product is the product of the limits.

3. Which of the following statements are incorrect (k and c are constants)? (b) lim k = k (a) lim k = c x→c

x→c

(c) lim x = c2 x→c2

(d) lim x = x x→c

SOLUTION Statements (a) and (d) are incorrect. Because k is constant, statement (a) should read lim x→c k = k. Statement (d) should be limx→c x = c.

4. Which of the following statements are incorrect? (a) The Product Law does not hold if the limit of one of the functions is zero. (b) The Quotient Law does not hold if the limit of the denominator is zero. (c) The Quotient Law does not hold if the limit of the numerator is zero. SOLUTION Statements (a) and (c) are incorrect. The Product Law remains valid when the limit of one or both of the functions is zero, and the Quotient Law remains valid when the limit of the numerator is zero.

Exercises In Exercises 1–22, evaluate the limits using the Limit Laws and the following two facts, where c and k are constants: lim x = c,

x→c

1. lim x x→9

SOLUTION

lim x = 9.

x→9

lim x

x→−3

lim k = k

x→c

46

CHAPTER 2

LIMITS

3. lim 14 x→9

SOLUTION

5.

lim 14 = 14.

x→9

lim (3x + 4)

lim 14 x→−3 x→−3

SOLUTION

We apply the Laws for Sums, Products, and Constants: lim (3x + 4) = lim 3x + lim 4

x→−3

x→−3

x→−3

= 3 lim x + lim 4 = 3(−3) + 4 = −5. x→−3

7.

x→−3

lim (y + 14)

lim 14y y→−3 y→−3

SOLUTION

lim (y + 14) = lim y + lim 14 = −3 + 14 = 11.

y→−3

y→−3

y→−3

9. lim (3t − 14) t→4 lim y(y + 14) y→−3

SOLUTION

lim (3t − 14) = 3 lim t − lim 14 = 3 · 4 − 14 = −2.

t→4

t→4

t→4

11. lim (4x + 1)(2x − 1) x→ 21 lim (x 3 + 2x) x→−5

SOLUTION

lim (4x + 1)(2x − 1) = 4 lim x + lim 1 2 lim x − lim 1

x→1/2

x→1/2

x→1/2

x→1/2

x→1/2

1 1 = 4 +1 2 − 1 = 3 · 0 = 0. 2 2

13. lim x(x + 1)(x + 2) x→2 lim (3x 4 − 2x 3 + 4x) x→−1 SOLUTION We apply the Product Law and Sum Law: lim (x + 1) lim (x + 2) lim x(x + 1)(x + 2) = lim x x→2

x→2

=2

x→2

x→2

lim x + lim 1 lim x + lim 2

x→2

x→2

x→2

= 2(2 + 1)(2 + 2) = 24 t 15. lim lim (x + 1)(3x 2 − 9) t→9 t + 1 x→2

SOLUTION

lim t t 9 9 t→9 = = = . lim t + lim 1 9+1 10 t→9 t + 1 lim

t→9

t→9

1−x 17. lim 3t − 14 1+x x→3lim t→4 t + 1 lim 1 − lim x 1−3 −2 1 1−x x→3 x→3 SOLUTION lim = = = =− . lim 1 + lim x 1+3 4 2 x→3 1 + x x→3

x→3

19. lim t −1 x t→2 lim x→−1 x 3 + 4x SOLUTION

We apply the definition of t −1 , and then the Quotient Law. lim 1 1 1 t→2 lim t −1 = lim = = . lim t 2 t→2 t→2 t t→2

21. lim (x 2 +−29x −3 ) x→3lim x x→5

x→2


Basic Limit Laws

47

We apply the Sum, Product, and Quotient Laws. The Product Law is applied to the exponentiations x 2 =

x · x and x 3 = x · x · x. lim (x 2 + 9x −3 ) = lim x 2 + lim 9x −3 =

x→3

x→3

⎛

= 9 + 9⎝

x→3

x→3

⎞

lim 1

x→3

( lim x)3

2 lim x

⎠=9+9

1 27

+9

=

1 x→3 x 3

lim

28 . 3

x→3

23. Use the Quotient Law to prove that if lim f (x) exists and is nonzero, then x→c z −1 + z lim z→1 z + 1 1 1 lim = x→c f (x) lim f (x) x→c

SOLUTION

Since lim f (x) is nonzero, we can apply the Quotient Law: x→c

lim

x→c

1 f (x)

lim 1

1 x→c = =

. lim f (x) lim f (x) x→c x→c

In Exercises 25–28, the=limit assuming Assume that evaluate lim f (x) 4 and compute:that lim f (x) = 3 and lim g(x) = 1. x→−4

x→6

25. (a) lim limf (x)g(x) f (x)2 x→−4

(b) lim

1

(c) lim x f (x)

x→6 f (x)

x→6

SOLUTION

x→−4

x→6

lim f (x)g(x) = lim f (x) lim g(x) = 3 · 1 = 3.

x→−4

x→−4

x→−4

g(x) 27. limlim 2(2 f (x) + 3g(x)) x→−4 x x→−4 SOLUTION

Since lim x 2 = 0, we may apply the Quotient Law, then applying the Product Law (from x 2 = x · x): x→−4

lim g(x) g(x) x→−4 = = x→−4 x 2 lim x 2

1

lim

x→−4

lim x

1 2 = 16 .

x→−4

sin x 29. Can the Quotient ? Explain. f (x) +Law 1 be applied to evaluate lim x→0 x lim x→−4 3g(x) − 9 sin x SOLUTION The limit Quotient Law cannot be applied to evaluate lim since lim x = 0. This violates a condition x→0 x x→0 of the Quotient Law. Accordingly, the rule cannot be employed. 31. Give an example where lim ( f (x) + g(x)) exists but neither lim f (x) nor lim g(x) exists. Show that the Productx→0 Law cannot be used to evaluate the x→0 lim (x − π /2) tan x. x→0 x→π /2

SOLUTION

Let f (x) = 1/x and g(x) = −1/x. Then lim ( f (x) + g(x)) = lim 0 = 0 However, lim f (x) = x→0

lim 1/x and lim g(x) = lim −1/x do not exist.

x→0

x→0

x→0

x→0

x→0

Use the Limit Laws and the result lim lim x n = cn for all whole numbers n. If you are a x −x→c 1 x = c to show that x→c exists and that lim a x = 1 for all a > 0. Prove that L ab = L a + L b Assume that the limit L a = lim familiar with induction, give a formalx→0 proof by x induction. x→0 for a, b > 0. Hint: (ab)x − 1 = a x (b x − 1) + (a x − 1). Verify numerically that L 12 = L 3 + L 4 . SOLUTION Correct answers can vary. An example is given: Let P[n] be the proposition : lim x n = cn , and proceed by induction.

33.

x→c

P[1] is true, as lim x = c. Suppose that P[n] is true, so that lim x n = cn . We must prove P[n + 1], that is, that x→c

x→c

lim x n+1 = cn+1 .

x→c

Applying the Product Law and P[n], we see: lim x n+1 = lim x n · x =

x→c

x→c

lim x n

x→c

lim x = cn c = cn+1 .

x→c

Therefore, P[n] true implies that P[n + 1] true. By induction, the limit is equal to cn for all n. Extend Exercise 33 to negative integers.

48

CHAPTER 2

LIMITS

Further Insights and Challenges 35. Show that if both lim f (x) g(x) and lim g(x) exist and lim g(x) is nonzero, then lim f (x) exists. Hint: Write x→c

x→c

x→c

f (x) = ( f (x) g(x))/g(x) and apply the Quotient Law. SOLUTION

x→c

Given that lim f (x)g(x) = L and lim g(x) = M = 0 both exist, observe that x→c

x→c

lim f (x)g(x) f (x)g(x) L x→c = = x→c g(x) lim g(x) M

lim f (x) = lim

x→c

x→c

also exists. h(t) = 5,=then lim h(t) = 15.exists and equals 4. 37. Prove thatthat if lim 12, then lim g(t) Show if lim tg(t) t→3t→3t t→3 t→3 SOLUTION

h(t) = 5, observe that lim t = 3. Now use the Product Law: t→3 t t→3 h(t) h(t) lim = lim t = 3 · 5 = 15. lim h(t) = lim t t t→3 t→3 t→3 t→3 t

Given that lim

f (x) 39. Prove that if lim f (x) = Lf (x) = 0 and lim g(x) = 0, then lim does not exist. x→cthat lim x→c g(x) Assuming = 1, x→c which of the following statements is necessarily true? Why? x→0 x f (x) (a) f (0) Suppose =0 (b) lim f (x) = 0 SOLUTION that lim exists. Then x→c g(x) x→0 L = lim f (x) = lim g(x) · x→c

x→c

f (x) f (x) f (x) = lim g(x) · lim = 0 · lim = 0. x→c x→c g(x) x→c g(x) g(x)

But, we were given that L = 0, so we have arrived at a contradiction. Thus, lim

f (x)

x→c g(x)

does not exist.

There is a Limit Law for composite functions but it is not stated in the text. Which of the following is the correct statement? Give an intuitive explanation. 2.4 (a)Limits (g(x))Continuity = lim f (x) lim f and x→c

x→c

(b) lim f (g(x)) = lim f (x), where L = lim g(x) Preliminary Questions x→c

x→c

x→L

f (g(x)) where lim f (x) (c) lim 1. Which property of = f (x)lim = g(x), x 3 allows us Lto= conclude that lim x 3 = 8? x→c

x→c

x→L

x→2

Use the correct version to evaluate SOLUTION We can conclude that lim x→2 x 3 = 8 because the function x 3 is continuous at x = 2. π lim sin(g(x)), where lim 1 ? g(x) = f (x) = 2. What can be said about f (3) if f is continuous and lim 6 x→2 x→2 2 x→3

SOLUTION

If f is continuous and limx→3 f (x) = 12 , then f (3) = 12 .

3. Suppose that f (x) < 0 if x is positive and f (x) > 1 if x is negative. Can f be continuous at x = 0? SOLUTION

Since f (x) < 0 when x is positive and f (x) > 1 when x is negative, it follows that lim f (x) ≤ 0

x→0+

and

lim f (x) ≥ 1.

x→0−

Thus, limx→0 f (x) does not exist, so f cannot be continuous at x = 0. 4. Is it possible to determine f (7) if f (x) = 3 for all x < 7 and f is right-continuous at x = 7? SOLUTION No. To determine f (7), we need to combine either knowledge of the values of f (x) for x < 7 with left-continuity or knowledge of the values of f (x) for x > 7 with right-continuity.

5. Are the following true or false? If false, state a correct version. (a) f (x) is continuous at x = a if the left- and right-hand limits of f (x) as x → a exist and are equal. (b) f (x) is continuous at x = a if the left- and right-hand limits of f (x) as x → a exist and equal f (a). (c) If the left- and right-hand limits of f (x) as x → a exist, then f has a removable discontinuity at x = a. (d) If f (x) and g(x) are continuous at x = a, then f (x) + g(x) is continuous at x = a. (e) If f (x) and g(x) are continuous at x = a, then f (x)/g(x) is continuous at x = a. SOLUTION

(a) False. The correct statement is “ f (x) is continuous at x = a if the left- and right-hand limits of f (x) as x → a exist and equal f (a).”

S E C T I O N 2.4

Limits and Continuity

49

(b) True. (c) False. The correct statement is “If the left- and right-hand limits of f (x) as x → a are equal but not equal to f (a), then f has a removable discontinuity at x = a.” (d) True. (e) False. The correct statement is “If f (x) and g(x) are continuous at x = a and g(a) = 0, then f (x)/g(x) is continuous at x = a.”

Exercises 1. Find the points of discontinuity of the function shown in Figure 14 and state whether it is left- or right-continuous (or neither) at these points. y 5 4 3 2 1 x 1

2

3

4

5

6

FIGURE 14 SOLUTION

• The function f is discontinuous at x = 1; it is left-continuous there. • The function f is discontinuous at x = 3; it is neither left-continuous nor right-continuous there. • The function f is discontinuous at x = 5; it is left-continuous there.

In Exercises 2–4, refer to the function f (x) in Figure 15. y 5 4 3 2 1 x 1

2

3

4

5

6

FIGURE 15

3. At which point c does f (x) have a removable discontinuity? What value should be assigned to f (c) to make f Find the points of discontinuity of f (x) and state whether f (x) is left- or right-continuous (or neither) at these continuous at x = c? points. SOLUTION Because limx→3 f (x) exists, the function f has a removable discontinuity at x = 3. Assigning f (3) = 4.5 makes f continuous at x = 3. 5. (a) For the function shown in Figure 16, determine the one-sided limits at the points of discontinuity. Find the point c1 at which f (x) has a jump discontinuity but is left-continuous. What value should be assigned (b) Which discontinuities is removable to f (c1of ) tothese make f right-continuous at x = and c1 ? how should f be redefined to make it continuous at this point? y

6

x

−2

2

4

FIGURE 16 SOLUTION

(a) The function f is discontinuous at x = 0, at which lim f (x) = ∞ and lim f (x) = 2. The function f is also x→0−

discontinuous at x = 2, at which lim f (x) = 6 and lim f (x) = 6. x→2−

x→0+

x→2+

(b) The discontinuity at x = 2 is removable. Assigning f (2) = 6 makes f continuous at x = 2. In Exercises Laws of Continuity Let f7–14, (x) beuse thethe function (Figure 17): and Theorems 2–3 to show that the function is continuous. ⎧ 2 ⎪ ⎨x + 3

for x < 1

50

CHAPTER 2

LIMITS

7. f (x) = x + sin x Since x and sin x are continuous, so is x + sin x by Continuity Law (i).

SOLUTION

9. f (x) = 3x + 4 sin x f (x) = x sin x SOLUTION Since x and sin x are continuous, so are 3x and 4 sin x by Continuity Law (iii). Thus 3x + 4 sin x is continuous by Continuity Law (i). 1 11. f (x)f (x) = =2 3x 3 + 8x 2 − 20x x +1 SOLUTION

• Since x is continuous, so is x 2 by Continuity Law (ii). • Recall that constant functions, such as 1, are continuous. Thus x 2 + 1 is continuous. • Finally,

1 x2 + 1

is continuous by Continuity Law (iv) because x 2 + 1 is never 0.

3x 13. f (x) = x 2x− cos x f (x) 1=+ 4 3 + cos x SOLUTION The functions 3 x , 1 and 4 x are each continuous. Therefore, 1 + 4 x is continuous by Continuity Law (i). 3x Because 1 + 4x is never zero, it follows that is continuous by Continuity Law (iv). 1 + 4x In Exercises 15–32, determine the points at which the function is discontinuous and state the type of discontinuity: x f (x) = 10x cos removable, jump, infinite, or none of these. 15. f (x) = SOLUTION

1 x The function 1/x is discontinuous at x = 0, at which there is an infinite discontinuity.

x −2 17. f (x)f (x) = = |x| |x − 1| SOLUTION

The function

x −2 is discontinuous at x = 1, at which there is an infinite discontinuity. |x − 1|

19. f (x) = [x] x −2 f (x) = SOLUTION This function has a jump discontinuity at x = n for every integer n. It is continuous at all other values of |x − 2| x. For every integer n, lim [x] = n

x→n+

since [x] = n for all x between n and n + 1. This shows that [x] is right-continuous at x = n. On the other hand, lim [x] = n − 1

x→n−

since [x] = n − 1 for all x between n − 1 and n. Thus [x] is not left-continuous. 1 21. g(t) = 2 1 f (x)t =− 1 x 2 1 1 = SOLUTION The function f (t) = is discontinuous at t = −1 and t = 1, at which there are 2 (t − 1)(t + 1) t −1 infinite discontinuities. 23. f (x) = 3x 3/2 − 19x 3 x+ f (x) = SOLUTION The f (x) = 3x 3/2 − 9x 3 is continuous for x > 0. At x = 0 it is right-continuous. (It is not 4xfunction −2 defined for x < 0.) 1 − 2z 25. h(z)g(t) = =2 3t −3/2 − 9t 3 z −z−6 1 − 2z 1 − 2z The function f (z) = 2 is discontinuous at z = −2 and z = 3, at which there = (z + 2)(z − 3) z −z−6 are infinite discontinuities. SOLUTION

x 2 1−−3x2z+ 2 27. f (x) = h(z) = |x2 − 2| z +9

S E C T I O N 2.4

SOLUTION


51

(x − 2)(x − 1) x 2 − 3x + 2 = . For x > 2, the function |x − 2| |x − 2| f (x) =

For x < 2, f (x) =

(x − 2)(x − 1) (x − 2)(x − 1) = = x − 1. |x − 2| (x − 2)

(x − 2)(x − 1) = −(x − 1). This function has a jump discontinuity at x = 2. 2−x

29. f (x) == csctan x 22t g(t)

1 is discontinuous whenever sin(x 2 ) = 0; i.e., whenever x 2 = n π 2) sin(x √ or x = ± n π , where n is a positive integer. At every such value of x there is an infinite discontinuity. SOLUTION

The function f (x) = csc(x 2 ) =

31. f (x) = tan(sin x) 1 f (x) = cosfunction f (x) = tan(sin x) is continuous everywhere. Reason: sin x is continuous everywhere and SOLUTION The x tan u is continuous on − π2 , π2 —and in particular on −1 ≤ u = sin x ≤ 1. Continuity of tan(sin x) follows by the continuity of composite functions. In Exercises the domain of the function and prove that it is continuous on its domain using the Laws of f (x)33–46, = x − determine |x| Continuity and the facts quoted in this section. 33. f (x) = 9 − x 2 √ SOLUTION The domain of 9 − x 2 is all x such that 9 − x 2 ≥ 0, or |x| ≤ 3. Since x and the polynomial 9 − x 2 are both continuous on this domain, so is the composite function 9 − x 2 . √ sin x 35. f (x) = x √ √ f (x) = x 2 + 9 SOLUTION This function is defined as long as x ≥ 0. Since x and sin x are continuous, so is x sin x by Continuity Law (ii). 37. f (x) = x 2/3 2x 2 x 2/3 x 2/3 f (x) = SOLUTION The domain x+ x 1/4 of x 2 is all real numbers as the denominator of the rational exponent is odd. Both x x 2/3 x and 2 are continuous on this domain, so x 2 is continuous by Continuity Law (ii). 39. f (x) = x −4/3 f (x) = x 1/3 + x 3/4 SOLUTION This function is defined for all x = 0. Because the function x 4/3 is continuous and not equal to zero for x = 0, it follows that 1 x −4/3 = 4/3 x is continuous for x = 0 by Continuity Law (iv). 41. f (x) = tan2 x 3 f (x) = cos x SOLUTION The domain of tan2 x is all x = ±(2n − 1)π /2 where n is a positive integer. Because tan x is continuous on this domain, it follows from Continuity Law (ii) that tan2 x is also continuous on this domain. 43. f (x) = (x 4 + 1)3/2 f (x) = cos(2x ) SOLUTION The domain of (x 4 + 1)3/2 is all real numbers as x 4 + 1 > 0 for all x. Because x 3/2 and the polynomial x 4 + 1 are both continuous, so is the composite function (x 4 + 1)3/2 . cos(x 2 ) 45. f (x)f (x) = =2 3−x 2 x −1 The domain for this function is all x = ±1. Because the functions cos x and x 2 are continuous on this domain, so is the composite function cos(x 2 ). Finally, because the polynomial x 2 − 1 is continuous and not equal to zero cos(x 2 ) for x = ±1, the function 2 is continuous by Continuity Law (iv). x −1 SOLUTION

47. Suppose that tan f (x) = 2 for x > 0 and f (x) = −4 for x < 0. What is f (0) if f is left-continuous at x = 0? What is =9 x f (0) if ff (x) is right-continuous at x = 0? SOLUTION

Let f (x) = 2 for positive x and f (x) = −4 for negative x.

• If f is left-continuous at x = 0, then f (0) = limx→0− f (x) = −4. • If f is right-continuous at x = 0, then f (0) = limx→0+ f (x) = 2.

Sawtooth Function Draw the graph of f (x) = x − [x]. At which points is f discontinuous? Is it left- or right-continuous at those points?

52

CHAPTER 2

LIMITS

In Exercises 49–52, draw the graph of a function on [0, 5] with the given properties. 49. f (x) is not continuous at x = 1, but lim f (x) and lim f (x) exist and are equal. x→1+

x→1−

SOLUTION y 4 3 2 1 x 1

2

3

4

5

51. f (x) has a removable discontinuity at x = 1, a jump discontinuity at x = 2, and f (x) is left-continuous but not continuous at x = 2 and right-continuous but not continuous at x = 3. lim f (x) = 2 lim f (x) = −∞, x→3−

x→3+

SOLUTION y 4 3 2 1 x 1

2

3

4

5

53. Each of the following statements is false. For each statement, sketch the graph of a function that provides a counf (x) is right- but not left-continuous at x = 1, left- but not right-continuous at x = 2, and neither left- nor terexample. right-continuous at x = 3. (a) If lim f (x) exists, then f (x) is continuous at x = a. x→a

(b) If f (x) has a jump discontinuity at x = a, then f (a) is equal to either lim f (x) or lim f (x). x→a−

x→a+

(c) If f (x) has a discontinuity at x = a, then lim f (x) and lim f (x) exist but are not equal. x→a−

x→a+

(d) The one-sided limits lim f (x) and lim f (x) always exist, even if lim f (x) does not exist. x→a−

x→a

x→a+

Refer to the four figures shown below. (a) The figure at the top left shows a function for which lim f (x) exists, but the function is not continuous at x = a

SOLUTION

x→a

because the function is not defined at x = a. (b) The figure at the top right shows a function that has a jump discontinuity at x = a but f (a) is not equal to either lim f (x) or lim f (x). x→a−

x→a−

(c) This statement can be false either when the two one-sided limits exist and are equal or when one or both of the one-sided limits does not exist. The figure at the top left shows a function that has a discontinuity at x = a with both one-sided limits being equal; the figure at the bottom left shows a function that has a discontinuity at x = a with a one-sided limit that does not exist. (d) The figure at the bottom left shows a function for which lim f (x) does not exist and one of the one-sided limits x→a

also does not exist; the figure at the bottom right shows a function for which lim f (x) does not exist and neither of the x→a

one-sided limits exists.

S E C T I O N 2.4


53

y

y

a

x

y

a

x

y

a

x

a

x

In Exercises 55–70,toevaluate using the method. According the Lawstheoflimit Continuity, if substitution f (x) and g(x) are continuous at x = c, then f (x) + g(x) is continuous at x =2 c. Suppose that f (x) and g(x) are discontinuous at x = c. Is it true that f (x) + g(x) is discontinuous at 55. lim x xx→5 = c? If not, give a counterexample. lim x 2 = 52 = 25.

SOLUTION

57.

x→5

lim (2x 33 − 4) lim x x→−1 x→8

lim (2x 3 − 4) = 2(−1)3 − 4 = −6.

SOLUTION

x→−1

x +9 59. limlim (5x − 12x −2 ) x −9 x→0 x→2 0+9 x +9 = = −1. x − 9 0−9 x→0

x 61. lim sin x + − 2π x→πlim 2 x→3 x 2 + 2x x SOLUTION lim sin( − π ) = sin(− π ) = −1. lim

SOLUTION

x→π

2

2

63. lim tan(3x) x→ π4lim sin(x 2 ) x→π

lim tan(3x) = tan(3 · π4 ) = tan( 34π ) = −1

SOLUTION

x→ π4

65. lim x −5/21 x→4lim x→π cos x 1 . SOLUTION lim x −5/2 = 4−5/2 = 32 x→4 67.

− 8x 3 )3/2 lim (1 x 3 + 4x

x→−1 lim x→2

lim (1 − 8x 3 )3/2 = (1 − 8(−1)3 )3/2 = 27.

SOLUTION

x→−1

69. lim 10x −2x 7x + 2 2/3 x→3lim x→2 4 − x 2 2 SOLUTION lim 10x −2x = 103 −2(3) = 1000. 2

x→3

In Exercises 71–74, sketch the graph of the given function. At each point of discontinuity, state whether f is left- or lim 3sin x π right-continuous. x→ 2 for x ≤ 1 x2 71. f (x) = 2 − x for x > 1

54

CHAPTER 2

LIMITS SOLUTION y 1

x

−1

1

2

3

−1

The function f is continuous everywhere. |x⎧− 3| for x ≤ 3 73. f (x) = ⎨x + 1 for x < 1 for x > 3 f (x) =x − 13 ⎩ for x ≥ 1 SOLUTION x y 4

2

x 2

4

6

The function f is continuous everywhere. ⎧ find the value of c that makes the function continuous. In Exercises 75–76, ⎪x 3 + 1 for −∞ < x ≤ 0 ⎨ f (x) =x 2 − −xc + 1for x < 5 for 0 < x < 2 ⎪ 75. f (x) = ⎩ 2 10xx − −x2c +for 4x + ≥15 5 for x ≥ 2 As x → 5−, we have x 2 − c → 25 − c = L. As x → 5+, we have 4x + 2c → 20 + 2c = R. Match the limits: L = R or 25 − c = 20 + 2c implies c = 53 . SOLUTION

77. Find all constants a, b such that the following function has no discontinuities: 2x + 9 for x ≤ 3 ⎧ π f (x) = ⎪ ⎪ −4x + c for x > 3 ⎨ax + cos x for x ≤ 4 f (x) = ⎪ π ⎪ ⎩bx + 2 for x > 4 As x → π4 −, we have ax + cos x → a4π + √1 = L. As x → π4 +, we have bx + 2 → b4π + 2 = R. 2 √ Match the limits: L = R or a4π + √1 = b4π + 2. This implies π4 (a − b) = 2 − √1 or a − b = 8−2π 2 . SOLUTION

2

2

In of 1993, the amount T (x) ofwould federalbeincome tax owed on an income of x dollars the 79. Which the following quantities represented by continuous functions of time was and determined which wouldbyhave formula one or more discontinuities? ⎧ (a) Velocity of an airplane during a flight ⎪0.15x for 0 ≤ x < 21,450 ⎨ (b) Temperature in aTroom under ordinary conditions (x) = 3,217.50 + 0.28(x − 21,450) for 21,450 ≤ x < 51,900 ⎪ ⎩ interest paid yearly (c) Value of a bank account with 11,743.50 + 0.31(x − 51,900) for x ≥ 51,900 (d) The salary of a teacher Sketch graph of T (x)ofand if it has any discontinuities. Explain why, if T (x) had a jump discontinuity, it (e)the The population the determine world might be advantageous in some situations to earn less money. SOLUTION T (x), the amount of federal income tax owed on an income of x dollars in 1993, might be a discontinuous function depending upon how the tax tables are constructed (as determined by that year’s regulations). Here is a graph of T (x) for that particular year. y 15,000 10,000 5,000 x 20,000 40,000 60,000

S E C T I O N 2.5

Evaluating Limits Algebraically

55

If T (x) had a jump discontinuity (say at x = c), it might be advantageous to earn slightly less income than c (say c − ) and be taxed at a lower rate than to earn c or more and be taxed at a higher rate. Your net earnings may actually be more in the former case than in the latter one.

Further Insights and Challenges 81. Give an example of functions f (x) and g(x) such that f (g(x)) is continuous but g(x) has at least one discontinuity. If f (x) has a removable discontinuity at x = c, then it is possible to redefine f (c) so that f (x) is continuous SOLUTION Answers vary. The simplest at x = c. Can f (c) may be redefined in more thanexamples one way?are the functions f (g(x)) where f (x) = C is a constant function, and g(x) is defined for all x. In these cases, f (g(x)) = C. For example, if f (x) = 3 and g(x) = [x], g is discontinuous at all integer values x = n, but f (g(x)) = 3 is continuous. 83. Let f (x) = 1 if x is rational and f (x) = −1 if x is irrational. Show that f (x) is discontinuous at all points, whereas Continuous at Only One Point Let f (x) = x for x rational, f (x) = −x for x irrational. Show that f (x)2 isFunction continuous at all points. f is continuous at x = 0 and discontinuous at all points x = 0. SOLUTION lim f (x) does not exist for any c. If c is irrational, then there is always a rational number r arbitrarily x→c

close to c so that | f (c) − f (r )| = 2. If, on the other hand, c is rational, there is always an irrational number z arbitrarily close to c so that | f (c) − f (z)| = 2. f (x)2 , on the other hand, is a constant function that always has value 1, which is obviously continuous.

2.5 Evaluating Limits Algebraically Preliminary Questions 1. Which of the following is indeterminate at x = 1? x2 + 1 , x −1

x2 − 1 , x +2

√

x2 − 1 x +3−2

,

√

x2 + 1 x +3−2

2 At x = 1, √x −1

is of the form 00 ; hence, this function is indeterminate. None of the remaining functions x+3−2 2 +1 2 and √x +1 are undefined because the denominator is zero but the numerator is not, is indeterminate at x = 1: xx−1 x+3−2 2 −1 while xx+2 is equal to 0. SOLUTION

2. Give counterexamples to show that each of the following statements is false: (a) If f (c) is indeterminate, then the right- and left-hand limits as x → c are not equal. (b) If lim f (x) exists, then f (c) is not indeterminate. x→c

(c) If f (x) is undefined at x = c, then f (x) has an indeterminate form at x = c. SOLUTION 2 −1 (a) Let f (x) = xx−1 . At x = 1, f is indeterminate of the form 00 but

x2 − 1 x2 − 1 = lim (x + 1) = 2 = lim (x + 1) = lim . x→1− x − 1 x→1− x→1+ x→1+ x − 1 lim

2 −1 (b) Again, let f (x) = xx−1 . Then

x2 − 1 = lim (x + 1) = 2 x→1 x − 1 x→1

lim f (x) = lim

x→1

but f (1) is indeterminate of the form 00 . (c) Let f (x) = 1x . Then f is undefined at x = 0 but does not have an indeterminate form at x = 0. 3. Although the method for evaluating limits discussed in this section is sometimes called “simplify and plug in,” explain how it actually relies on the property of continuity. SOLUTION If f is continuous at x = c, then, by definition, lim x→c f (x) = f (c); in other words, the limit of a continuous function at x = c is the value of the function at x = c. The “simplify and plug-in” strategy is based on simplifying a function which is indeterminate to a continuous function. Once the simplification has been made, the limit of the remaining continuous function is obtained by evaluation.

56

CHAPTER 2

LIMITS

Exercises In Exercises 1–4, show that the limit leads to an indeterminate form. Then carry out the two-step procedure: Transform the function algebraically and evaluate using continuity. x 2 − 25 x→5 x − 5

1. lim

2 −25 When we substitute x = 5 into xx−5 , we obtain the indeterminate form 00 . Upon factoring the numerator and simplifying, we find

SOLUTION

x 2 − 25 (x − 5)(x + 5) = lim = lim (x + 5) = 10. x −5 x→5 x − 5 x→5 x→5 lim

3. lim

2t − 14

x +1

5t − 35 t→7 lim x→−1 x 2 + 2x + 1

0 When we substitute t = 7 into 2t−14 5t−35 , we obtain the indeterminate form 0 . Upon dividing out the common factor of t − 7 from both the numerator and denominator, we find SOLUTION

lim

2t − 14

t→7 5t − 35

= lim

2(t − 7)

t→7 5(t − 7)

= lim

2

t→7 5

=

2 . 5

In Exercises 5–32, the limit or state that it does not exist. h +evaluate 3 lim 2 h→−3 x 2 −h64− 9 5. lim x→8 x − 8 SOLUTION

x 2 − 64 (x + 8)(x − 8) = lim = lim (x + 8) = 16. x −8 x→8 x − 8 x→8 x→8 lim

x 2 −23x + 2 7. lim x − 64 x→2lim x − 2 x→8 x − 9 x 2 − 3x + 2 (x − 1)(x − 2) = lim = lim (x − 1) = 1. SOLUTION lim x −2 x −2 x→2 x→2 x→2 x −2 9. lim 3 x 3 − 4x x→2lim x − 4x x→2 x − 2 1 x −2 x −2 1 SOLUTION lim 3 = lim = . = lim 8 x→2 x − 4x x→2 x(x − 2)(x + 2) x→2 x(x + 2) (1 +3h)3 − 1 11. lim x − 2x h h→0lim x→2 x − 2 SOLUTION

(1 + h)3 − 1 1 + 3h + 3h 2 + h 3 − 1 3h + 3h 2 + h 3 = lim = lim h h h h→0 h→0 h→0 lim

= lim 3 + 3h + h 2 = 3 + 3(0) + 02 = 3. h→0

3x 2 − 4x −2 4 13. lim (h + 2) − 9h x→2lim 2x 2 − 8 h−4 h→4 3x 2 − 4x − 4 (3x + 2)(x − 2) 3x + 2 8 = lim SOLUTION lim = lim = = 1. 8 x→2 x→2 2(x − 2)(x + 2) x→2 2(x + 2) 2x 2 − 8 (y − 2)3 15. lim lim3 2t + 4 y→2 y − 5y + 22 t→−2 12 − 3t (y − 2)3 (y − 2)3 (y − 2)2 = lim = lim = 0. SOLUTION lim 3 2 2 y→2 y − 5y + 2 y→2 (y − 2) y + 2y − 1 y→2 y + 2y − 1 1√ 1 x−− 4 3 + 3 lim 17. limx→16 hx − 16 h h→0

S E C T I O N 2.5

1

Evaluating Limits Algebraically

57

1

−3 1 3 − (3 + h) −1 1 lim 3+h = lim = lim =− . h 9 h→0 h→0 h 3(3 + h) h→0 3(3 + h)

SOLUTION

√

2+h−2 1 1 − h 2 4 (h + 2) √ lim hh+2−2 h→0 SOLUTION lim does not exist. h h→0 √ √ √ h + 2 − 2 ( h + 2 + 2) h+2−2 h−2 • As h → 0+, we have = √ = √ → −∞. h h( h + 2 + 2) h( h + 2 + 2) √ √ √ h + 2 − 2 ( h + 2 + 2) h+2−2 h−2 • As h → 0−, we have = √ → ∞. = √ h h( h + 2 + 2) h( h + 2 + 2)

19. lim

h→0

x −2 √ 21. lim √ x√− 6 − 2 x→2limx − 4 − x x − 10 x→10 SOLUTION

√ √ √ √ x −2 (x − 2)( x + 4 − x) (x − 2)( x + 4 − x) lim √ = lim √ = lim √ √ √ √ x − (4 − x) x→2 x − 4 − x x→2 ( x − 4 − x)( x + 4 − x) x→2 √ √ √ √ (x − 2)( x + 4 − x) (x − 2)( x + 4 − x) = lim = lim 2x − 4 2(x − 2) x→2 x→2 √ √ √ √ ( x + 4 − x) 2+ 2 √ = lim = = 2. 2 2 x→2

√ 2−1− √ x√ x +1 1 + x − 1−x 23. lim x −3 x→2lim x x→0 √ √ √ x2 − 1 − x + 1 3− 3 SOLUTION lim = = 0. x −3 −1 x→2 4 √1 25. lim √ 5 − x−− 1 x→4lim x − 2 √ x − 4 x→4 2 − x √ √ 1 x +2−4 x −2 4 1 √ = lim √ √ = . SOLUTION lim √ − = lim √ x −4 4 x −2 x→4 x→4 x→4 x −2 x +2 x −2 x +2 27. lim

cot x

1 x→0 csc x 1 lim √ − 2 x x→0+ cot x x + x cos x

SOLUTION

29. lim

lim

x→0 csc x

= lim

x→0 sin x

· sin x = cos 0 = 1.

sin x − cos x cot θ

x→ π4limπtan x − 1 θ → 2 csc θ

SOLUTION

√ sin x − cos x cos x π 2 (sin x − cos x) cos x · = lim = cos = . cos x sin x − cos x 4 2 x→ π4 tan x − 1 x→ π4 lim

2 x + 3 cos x− 2 2 cos 31. limlimπ sec θ − tan θ 2 cos x − 1 x→θπ3→ 2 SOLUTION

5 2 cos2 x + 3 cos x − 2 π (2 cos x − 1) (cos x + 2) = lim = lim cos x + 2 = cos + 2 = . π π π 2 cos x − 1 2 cos x − 1 3 2 x→ 3 x→ 3 x→ 3 lim

x −2 Use to estimate lim f (x) to two decimal places. Compare with the answer √ cosaθplot − 1of f (x) = √ lim x→2 x − 4−x θ →0 sin θ obtained algebraically in Exercise 21.

33.

√ Let f (x) = √ x−2 . From the plot of f (x) shown below, we estimate lim f (x) ≈ 1.41; to two decimal x− 4−x x→2 √ places, this matches the value of 2 obtained in Exercise 21. SOLUTION

58

CHAPTER 2

LIMITS y 1.414 1.410 1.406 x

1.402 1.6

1.8

2.0

2.2

2.4

In Exercises 35–40, use the identity 1a 3 − b3 =4 (a − b)(a 2 + ab + b2 ). to evaluate lim f (x) numerically. Compare with the answer obtained Use a plot of f (x) = √ − x −2 x −4 x→4 x3 − 1 in Exercise 25. 35. algebraically lim x→1 x − 1

2 +x +1

3 (x − 1) x x −1 SOLUTION lim = lim = lim x 2 + x + 1 = 3. x −1 x→1 x − 1 x→1 x→1 x 2 −33x + 2 37. lim x −8 x→1lim x 3 − 1 x→2 x 2 − 4 x 2 − 3x + 2 (x − 1)(x − 2) x −2 1 = lim SOLUTION lim = lim =− . 3 x→1 x→1 (x − 1) x 2 + x + 1 x→1 x 2 + x + 1 x3 − 1 x4 − 1 3 39. lim 3 x −8 x→1lim x −1 x→2 x 2 − 6x + 8 SOLUTION

4 x4 − 1 (x 2 − 1)(x 2 + 1) (x − 1)(x + 1)(x 2 + 1) (x + 1)(x 2 + 1) = lim = lim = lim = . 3 x→1 x 3 − 1 x→1 (x − 1)(x 2 + x + 1) x→1 (x − 1)(x 2 + x + 1) x→1 (x 2 + x + 1) lim

In Exercises 41–50, evaluate the limits in terms of the constants involved. x − 27 lim 1/3 41. limx→27 (2a +x x) − 3 x→0

lim (2a + x) = 2a.

SOLUTION

x→0

43. lim (4t − 2at + 3a) t→−1lim (4ah + 7a) h→−2

SOLUTION

lim (4t − 2at + 3a) = −4 + 5a.

t→−1

2(x + h)2 −22x 2 2 (3a + h) − 9a h x→0lim h h→0 2(x + h)2 − 2x 2 4hx + 2h 2 = lim = lim (4x + 2h) = 2h. SOLUTION lim h h x→0 x→0 x→0 √ √ x− a 2 47. lim (x + a) − 4x 2 x→alim x − a x→a √ √ x√ −a √ 1 x− a x− a 1 SOLUTION lim = lim √ √ = √ . √ √ √ = lim √ x→a x − a x→a x→a x + a 2 a x− a x+ a 45. lim

√a)3 − a 3 √ (x + 49. lim a + 2h − a x x→0lim h h→0 (x + a)3 − a 3 x 3 + 3x 2 a + 3xa 2 + a 3 − a 3 SOLUTION lim = lim = lim (x 2 + 3xa + 3a 2 ) = 3a 2 . x x x→0 x→0 x→0 1 1 Further Insights and Challenges −

h a lim 51–52, In Exercises find all values of c such that the limit exists. h→a h − a x 2 − 5x − 6 x→c x −c

51. lim

x 2 − 5x − 6 will exist provided that x − c is a factor of the numerator. (Otherwise there will be an x→c x −c infinite discontinuity at x = c.) Since x 2 − 5x − 6 = (x + 1)(x − 6), this occurs for c = −1 and c = 6. SOLUTION

lim

lim

x 2 + 3x + c

S E C T I O N 2.6

Trigonometric Limits

59

53. For which sign ± does the following limit exist?

1 1 lim ± x(x − 1) x→0 x

SOLUTION

(x − 1) + 1 1 1 1 + = lim = lim = −1. x x(x − 1) x(x − 1) x − 1 x→0 x→0 1 1 • The limit lim − does not exist. x(x − 1) x→0 x

• The limit lim x→0

1 1 (x − 1) − 1 x −2 − = = → ∞. x x(x − 1) x(x − 1) x(x − 1) 1 1 (x − 1) − 1 x −2 – As x → 0−, we have − = = → −∞. x x(x − 1) x(x − 1) x(x − 1)

– As x → 0+, we have

For which sign ± does the following limit exist?

2.6 Trigonometric Limits

1 1 ± |x| x→0+ x

lim

Preliminary Questions 1. Assume that −x 4 ≤ f (x) ≤ x 2 . What is lim f (x)? Is there enough information to evaluate lim f (x)? Explain. x→ 12

x→0

Since limx→0 −x 4 = limx→0 x 2 = 0, the squeeze theorem guarantees that limx→0 f (x) = 0. Since 1 = 1 = lim 4 limx→ 1 −x = − 16 x 2 , we do not have enough information to determine limx→ 1 f (x). 4 x→ 21 2 2 SOLUTION

2. State the Squeeze Theorem carefully. SOLUTION

Assume that for x = c (in some open interval containing c), l(x) ≤ f (x) ≤ u(x)

and that lim l(x) = lim u(x) = L. Then lim f (x) exists and x→c

x→c

x→c

lim f (x) = L .

x→c

3. Suppose that f (x) is squeezed at x = c on an open interval I by two constant functions u(x) and l(x). Is f (x) necessarily constant on I ? SOLUTION

Yes. Because f (x) is squeezed at x = c on an open interval I by the two functions u(x) and l(x), we know

that lim u(x) = lim l(x).

x→c

x→c

Combining this relation with the fact that u(x) and l(x) are constant functions, it follows that u(x) = l(x) for all x on I . Finally, to have l(x) ≤ f (x) ≤ u(x) = l(x), it must be the case that f (x) = l(x) for all x on I . In other words, f (x) is constant on I . sin 5h , it is a good idea to rewrite the limit in terms of the variable (choose one): h→0 3h 5h (b) θ = 3h (c) θ = 3

4. If you want to evaluate lim (a) θ = 5h SOLUTION

To match the given limit to the pattern of sin θ , θ →0 θ lim

it is best to substitute for the argument of the sine function; thus, rewrite the limit in terms of (a): θ = 5h.

60

CHAPTER 2

LIMITS

Exercises 1. In Figure 6, is f (x) squeezed by u(x) and l(x) at x = 3? At x = 2? y

u(x) f(x) l(x)

1.5

x 1

2

3

4

FIGURE 6 SOLUTION

Because there is an open interval containing x = 3 on which l(x) ≤ f (x) ≤ u(x) and lim l(x) = lim u(x), x→3

x→3

f (x) is squeezed by u(x) and l(x) at x = 3. Because there is an open interval containing x = 2 on which l(x) ≤ f (x) ≤ u(x) but lim l(x) = lim u(x), f (x) is trapped by u(x) and l(x) at x = 2 but not squeezed. x→2

x→2

3. What information about f (x) does the Squeeze Theorem provide if we assume that the graphs of f (x), u(x), and What is lim f (x) if f (x) satisfies cos x ≤ f (x) ≤ 1 for x in (−1, 1)? l(x) are related as in Figure 7 and that lim u(x) = lim l(x) = 6? Note that the inequality f (x) ≤ u(x) is not satisfied x→0 x→7

x→7

for all x. Does this affect the validity of your conclusion? y f(x)

6 u(x) l(x) x 7

FIGURE 7 SOLUTION The Squeeze Theorem does not require that the inequalities l(x) ≤ f (x) ≤ u(x) hold for all x, only that the inequalities hold on some open interval containing x = c. In Figure 7, it is clear that l(x) ≤ f (x) ≤ u(x) on some open interval containing x = 7. Because lim u(x) = lim l(x) = 6, the Squeeze Theorem guarantees that lim f (x) = 6.

x→7

x→7

x→7

In Exercises use the Theorem tosufficient evaluate the limit. to determine lim State 5–8, whether the Squeeze inequality provides information

x→1 f (x) and, if so, find the limit. (a) 4x − 51≤ f (x) ≤ x 2 5. lim x cos 2 (b) x→02x − 1x≤ f (x) ≤ x 2+2 (c) 2x + We 1 ≤prove f (x) the ≤ xlimit SOLUTION to the right first, and then the limit to the left. Suppose x > 0. Since −1 ≤ cos 1x ≤ 1, multiplication by x (positive) yields −x ≤ x cos 1x ≤ x. The squeeze theorem implies that lim −x ≤ lim x cos 1x ≤ x→0+

x→0+

lim x, so 0 ≤ lim x cos 1x ≤ 0. This gives us lim x cos 1x = 0. On the other hand, suppose x < 0. −1 ≤ cos 1x ≤ 1

x→0+

x→0+

x→0+

multiplied by x gives us x ≤ x cos 1x ≤ −x, and we can apply the squeeze theorem again to get lim x cos 1x = 0. x→0−

Therefore, lim x cos 1x = 0. x→0

1 1 x 7. lim cos 2 sin x→0lim x xsin x x→0

1 ≤ 1, observe that − sin x ≤ cos 1x sin x ≤ sin x. Since lim sin x = 0 = SOLUTION Since −1 ≤ cos x x→0 lim − sin x, the Squeeze Theorem shows that x→0

lim cos

x→0

1 sin x = 0. x

In Exercises 9–16, evaluate π the limit using Theorem 2 as necessary. lim (x − 1) sin x −1 x→1 sin x cos x 9. lim x x→0 sin x cos x sin x SOLUTION lim = lim · lim cos x = 1 · 1 = 1. x x→0 x→0 x x→0 sin2√t 11. lim t + 2 sin t t→0lim t t t→0

S E C T I O N 2.6

SOLUTION

sin2 t sin t sin t = lim sin t = lim · lim sin t = 1 · 0 = 0. t→0 t t→0 t t→0 t t→0 lim

x2 13. limlim 2tan x x→0 sin xx x→0 SOLUTION

x2

lim

x→0 sin2 x

1 1 1 1 1 = lim sin x sin x = lim sin x · lim sin x = · = 1. 1 1 x→0 x→0 x→0 x x x x

sin t 15. lim 1 − cos t πlimt t→ 4 π t t→ 2 sin t π SOLUTION is continuous at t = . Hence, by substitution t 4 √ √ 2 2 2 sin t 2 . = π = lim π t→ π4 t 4

sin 10x 17. Let L =cos limt − cos2 t . x x→0 lim t t→0 sin θ (a) Show, by letting θ = 10x, that L = lim 10 . θ θ →0 (b) Compute L. SOLUTION

θ . θ → 0 as x → 0, so we get: Since θ = 10x, x = 10

sin θ sin θ sin θ sin 10x = lim · = lim 10 = 10 lim · = 10. x θ x→0 θ →0 (θ /10) θ →0 θ →0 θ lim

In Exercises 19–40, evaluate sin 2h 2 sin 2h 5h sin 2h the limit. . Hint: = . Evaluate lim sin 6h h→0 sin 5h sin 5h 5 2h sin 5h 19. lim h→0 h SOLUTION

lim

h→0

sin 6h sin 6h = lim 6 = 6. h 6h h→0

sin 6h 21. lim 6 sin h h→0lim6h h→0 6h sin 6h sin x SOLUTION Substitute x = 6h. As h → 0, x → 0, so: lim = lim = 1. h→0 6h x→0 x sin 7x 23. lim sin h x→0lim3x h→0 6h sin 7x 7 sin 7x 7 = lim = . SOLUTION lim 3 x→0 3x x→0 3 7x 25. lim

tan 4x

x

x→0lim9x x→ π4 sin 11x

SOLUTION

27. lim

lim

x→0

tan 4t

4 4 tan 4x 1 sin 4x = lim · · = . 9x 4x cos 4x 9 x→0 9

x

t sec t t→0lim x→0 csc 25x

SOLUTION

tan 4t 4 sin 4t 4 cos t sin 4t = lim = lim · = 4. 4t t→0 t sec t t→0 4t cos(4t) sec(t) t→0 cos 4t lim

sin(z/3) sin 2h sin 3h z→0limsin z h→0 h2 sin(z/3) z/3 1 z sin(z/3) 1 SOLUTION lim · = lim · · = . z/3 z→0 3 sin z z/3 3 z→0 sin z

29. lim

31. lim

tan 4x sin(−3θ )

tan 9x x→0lim θ →0 sin(4θ )

SOLUTION

lim

lim

tan 4x

x→0 tan 9x

csc 8t

t→0 csc 4t

= lim

cos 9x

x→0 cos 4x

·

sin 4x 4 9x 4 · · = . 4x 9 sin 9x 9


61

62

CHAPTER 2

LIMITS

33. lim

sin 5x sin 2x

x→0 sin 3x sin 5x

SOLUTION

lim

sin 5x sin 2x

x→0 sin 3x sin 5x

= lim

x→0

sin 2x 2 3x 2 · · = . 2x 3 sin 3x 3

1 − cos 2h 35. lim sin 3x sin 2x h→0lim h x→0 x sin 5x 1 − cos 2h 1 − cos 2h 1 − cos 2h SOLUTION lim = lim 2 = 2 lim = 2 · 0 = 0. h 2h 2h h→0 h→0 h→0 1 − cos t sin(2h)(1 − cos h) t→0limsin t h h→0 t SOLUTION A single multiplication by t turns this limit into the quotient of two familiar limits.

37. lim

lim

t→0

1 − cos t t (1 − cos t) t (1 − cos t) t (1 − cos t) = lim = lim = lim lim = 1 · 0 = 0. sin t t sin t t t t→0 t→0 sin t t→0 sin t t→0

1 − cos 3h 39. lim cos 2θ − cos θ h h→ π2lim θ θ →0 π SOLUTION The function is continuous at 2 , so we may use substitution: 1 − cos 3 π2 1−0 2 1 − cos 3h = π = . = π π h π h→ 2 2 2 lim

sin x sin x 41. Calculate1 (a) lim and (b) lim . − cos 4θ lim x→0+ |x| x→0− |x| θ →0 sin 3θ SOLUTION

sin x = 1. x sin x sin x (b) lim = −1. = lim x→0− |x| x→0− −x (a)

lim

sin x

x→0+ |x|

= lim

x→0+

1 ≤ |tan x|. Then evaluate lim tan x cos 1 using the Squeeze Theorem. 43. Show thatthe −|tan x| ≤ tan x cosstated x x Prove following result in Theorem 2: x→0

1 1 − cos θ1 SOLUTION Since | cos x | ≤ 1, it follows that | tan lim x cos x =| 0≤ | tan x|, which is equivalent to −| tan x| ≤

θ θ →0 tan x cos 1x ≤ | tan x|. Because limx→0 −| tan x| = limx→0 | tan x| = 0, the Squeeze Theorem gives 1 − cos θ 1 1 − cos2 θ Hint: = · . 1 θ 1 + cos θ θ = 0. lim tan x cos x x→0

45. Plot the graphs of u(x) = 1 + |x − π2 | and l(x) = sin x on the same set of axes. What can you say about 1 using the Squeeze π tan x cos by sinl(x) Evaluate lim f (x) if f (x)lim is squeezed x and u(x) at x = 2 ?Theorem. x→ π2

x→0

SOLUTION y u(x) = 1 + | x −

1

/2 |

l(x) = sin x x /2

lim u(x) = 1 and lim l(x) = 1, so any function f (x) satisfying l(x) ≤ f (x) ≤ u(x) for all x near π /2 will satisfy

x→π /2

lim

x→π /2

f (x) = 1.

x→π /2

Use the Squeeze Theorem to prove that if lim | f (x)| = 0, then lim f (x) = 0. x→c

x→c

S E C T I O N 2.6


63

Further Insights and Challenges 1 − cos h numerically (and graphically if you have a graphing utility). Then prove that the h2 limit is equal to 12 . Hint: see the hint for Exercise 42.

47.

Investigate lim

h→0

SOLUTION

•

h 1 − cos h h2

−.1

−.01

.01

.1

.499583

.499996

.499996

.499583

The limit is 12 . •

y 0.5 0.4 0.3 0.2 0.1 −2

•

−1

x 1

2

1 1 − cos h 1 − cos2 h 1 sin h 2 = . = lim = lim h 1 + cos h 2 h→0 h→0 h 2 (1 + cos h) h→0 h2 lim

cos 3h − 1 49. Evaluate lim cos 3h −. 1 cos 2h − 1 . Hint: Use the result of Exercise 47. Evaluate h→0lim h→0 h2 SOLUTION

cos 3h − 1 cos 3h − 1 cos 3h + 1 cos 2h + 1 cos2 3h − 1 cos 2h + 1 · = lim · · = lim h→0 cos 2h − 1 h→0 cos 2h − 1 cos 3h + 1 cos 2h + 1 h→0 cos2 2h − 1 cos 3h + 1 lim

= lim

− sin2 3h

h→0 − sin2 2h

=

·

9 2h 2h cos 2h + 1 sin 3h sin 3h cos 2h + 1 = lim · · · cos 3h + 1 4 h→0 3h 3h sin 2h sin 2h cos 3h + 1

2 9 9 ·1·1·1·1· = . 4 2 4

51. Using a diagram of the unit circle and the Pythagorean Theorem, show that Use the result of Exercise 47 to prove that for m = 0, sin2 θ ≤ (1 − cos θ )2 + sin2 θ ≤ θ2 m2 cos mx − 1 lim =− Conclude that sin2 θ ≤ 2(1 − cos θ ) ≤ θ 2 and use this to give 2 proof of Eq. (7) in Exercise 42. Then give an x→0 x 2 an alternate alternate proof of the result in Exercise 47. SOLUTION

• Consider the unit circle shown below. The triangle B D A is a right triangle. It has base 1 − cos θ , altitude sin θ , and

hypotenuse h. Observe that the hypotenuse h is less than the arc length AB = radius · angle = 1 · θ = θ . Apply the Pythagorean Theorem to obtain (1 − cos θ )2 + sin2 θ = h 2 ≤ θ 2 . The inequality sin2 θ ≤ (1 − cos θ )2 + sin2 θ follows from the fact that (1 − cos θ )2 ≥ 0.

B A O

D

• Note that

(1 − cos θ )2 + sin2 θ = 1 − 2 cos θ + cos2 θ + sin2 θ = 2 − 2 cos θ = 2(1 − cos θ ).

64

CHAPTER 2

LIMITS

Therefore, sin2 θ ≤ 2(1 − cos θ ) ≤ θ 2 . • Divide the previous inequality by 2θ to obtain

1 − cos θ sin2 θ θ ≤ ≤ . 2θ θ 2 Because 1 1 sin2 θ sin θ = lim · lim sin θ = (1)(0) = 0, 2 θ →0 θ 2 θ →0 2θ θ →0 lim

and lim

θ

h→0 2

= 0, it follows by the Squeeze Theorem that lim

θ →0

1 − cos θ = 0. θ

• Divide the inequality

sin2 θ ≤ 2(1 − cos θ ) ≤ θ 2 by 2θ 2 to obtain sin2 θ 1 − cos θ 1 ≤ ≤ . 2 2θ 2 θ2 Because sin θ 2 sin2 θ 1 1 1 = = (12 ) = , lim 2 2 θ →0 θ 2 2 θ →0 2θ lim

and lim

1

h→0 2

=

1 , it follows by the Squeeze Theorem that 2 lim

θ →0

1 1 − cos θ = . 2 θ2

Let A(n) be sin the xarea of a regular n-gon inscribed in a unit circle (Figure 8). − sin c numerically for the five values c = 0, π , π , π , π . (a) Investigate lim1 6 4 3 2 x −2cπ . (a) Prove that A(n) x→c = n sin n for general c? (We’ll see why in Chapter 3.) (b) Can you guess 2the answer (b) Intuitively, might we expect A(n) to converge the values area ofof thec.unit circle as n → ∞? (c) Check why if your answer to part (b) works for two to other (c) Use Theorem 2 to evaluate lim A(n). 53.

n→∞

y

x 1

FIGURE 8 Regular n-gon inscribed in a unit circle. SOLUTION

(a) A(n) = n · a(n), where a(n) is the area of one section of the n-gon as shown. The sections into which the n-gon is divided are identical. By looking at the section lying just atop the x-axis, we can see that each section has a base length of 1 and a height of sin θ , where θ is the interior angle of the triangle.

θ=

2π , n

so a(n) =

2π 1 sin . 2 n

S E C T I O N 2.7

This means

Intermediate Value Theorem

65

2π 1 2π 1 = n sin . A(n) = n sin 2 n 2 n

(b) As n increases, the difference between the n-gon and the unit circle shrinks toward zero; hence, as n increases, A(n) approaches the area of the unit circle. (c) Substitute x = n1 . n = 1x , so A(n) = 12 1x sin(2π x). x → 0 as n → ∞, so lim A(n) = lim

1 1 sin(2π x) = (2π ) = π . x 2

x→0 2

n→∞

2.7 Intermediate Value Theorem Preliminary Questions 1. Explain why f (x) = x 2 takes on the value 0.5 in the interval [0, 1]. Observe that f (x) = x 2 is continuous on [0, 1] with f (0) = 0 and f (1) = 1. Because f (0) < 0.5 < f (1), the Intermediate Value Theorem guarantees there is a c ∈ [0, 1] such that f (c) = 0.5. SOLUTION

2. The temperature in Vancouver was 46◦ at 6 AM and rose to 68◦ at noon. What must we assume about temperature to conclude that the temperature was 60◦ at some moment of time between 6 AM and noon? SOLUTION

We must assume that temperature is a continuous function of time.

3. What is the graphical interpretation of the IVT? SOLUTION If f is continuous on [a, b], then the horizontal line y = k for every k between f (a) and f (b) intersects the graph of y = f (x) at least once.

4. Show that the following statement is false by drawing a graph that provides a counterexample: If f (x) is continuous and has a root in [a, b], then f (a) and f (b) have opposite signs. SOLUTION

f (a) f (b) a

b

Exercises 1. Use the IVT to show that f (x) = x 3 + x takes on the value 9 for some x in [1, 2]. SOLUTION Observe that f (1) = 2 and f (2) = 10. Since f is a polynomial, it is continuous everywhere; in particular on [1, 2]. Therefore, by the IVT there is a c ∈ [1, 2] such that f (c) = 9. 3. Show that g(t) = t 2 tan tt takes on the value 12 for some t in 0, π4 . Show that g(t) = takes on the value 0.499 for some t in [0, 1]. t + 1π π2 π 1 π2 SOLUTION g(0) = 0 and g( 4 ) = 16 . g(t) is continuous for all t between 0 and 4 , and 0 < 2 < 16 ; therefore, by the IVT, there is a c ∈ [0, π4 ] such that g(c) = 12 .

5. Show that cos x = x has a solution in the interval [0, 1]. Hint: Show that f (x) = x − cos x has a zero in [0, 1]. x2 on the value Show that SOLUTION Let f (x) = = x7 − costakes x. Observe that 0.4. f is continuous with f (0) = −1 and f (1) = 1 − cos 1 ≈ .46. x +1 Therefore, by the IVT there is a c ∈ [0, 1] such that f (c) = c − cos c = 0. Thus c = cos c and hence the equation cos x = x has a solution c in [0, 1]. In Exercises 7–14, use the IVT to prove each of the following statements. 1 3 √ Use√the IVT to find an interval of length 2 containing a root of f (x) = x + 2x + 1. 7. c + c + 1 = 2 for some number c.

66

CHAPTER 2

LIMITS

√ √ Let f (x) = x + x + 1 − 2. Note that f is continuous on 14 , 1 with f ( 14 ) = 14 + 54 − 2 ≈ −.38 √ √ √ and f (1) = 2 − 1 ≈ .41. Therefore, by the IVT there is a c ∈ 14 , 1 such that f (c) = c + c + 1 − 2 = 0. Thus √ √ √ √ c + c + 1 = 2 and hence the equation x + x + 1 = 2 has a solution c in 14 , 1 . √ 9. 2For exists. Hint: Consider = xx 2for . some x ∈ [0, π ]. all integers n, sin nxf (x) = cos 2 SOLUTION Let f (x) = x . Observe that f is continuous with f (1) = 1 and f (2) = 4. Therefore, by the IVT there is √ a c ∈ [1, 2] such that f (c) = c2 = 2. This proves the existence of 2, a number whose square is 2. SOLUTION

11. ForAallpositive positivenumber integersck,has there x such cos x =integers xk. an exists nth root for that all positive n. (This fact is usually taken for granted, but it k SOLUTION For each positive integer k, let f (x) = x − cos x. Observe that f is continuous on 0, π2 with f (0) = −1 requires proof.) k and f ( π2 ) = π2 > 0. Therefore, by the IVT there is a c ∈ 0, π2 such that f (c) = ck − cos(c) = 0. Thus cos c = ck and hence the equation cos x = x k has a solution c in the interval 0, π2 . 13. 2x =x b has a solution for all b > 0 (treat b ≥ 1 first). 2 = bx has a solution if b > 2. SOLUTION Let b ≥ 1. Let f (x) = 2 x . f is continuous for all x. f (0) = 1, and f (x) > b for some x > 0 (since f (x) increases without bound). Hence, by the IVT, f (x) = b for some x > 0. On the other hand, suppose b < 1. Since limx→−∞ 2x = 0, there is some x < 0 such that f (x) = 2x < b. f (0) = 1 > b, so the IVT states that f (x) = b for some x < 0. 15. (a) (b) (c)

Carry of themany Bisection Method for f (x) = 2x − x 3 as follows: tanout x =three x hassteps infinitely solutions. Show that f (x) has a zero in [1, 1.5]. Show that f (x) has a zero in [1.25, 1.5]. Determine whether [1.25, 1.375] or [1.375, 1.5] contains a zero.

SOLUTION

Note that f (x) is continuous for all x.

(a) f (1) = 1, f (1.5) = 21.5 − (1.5)3 < 3 − 3.375 < 0. Hence, f (x) = 0 for some x between 0 and 1.5. (b) f (1.25) ≈ 0.4253 > 0 and f (1.5) < 0. Hence, f (x) = 0 for some x between 1.25 and 1.5. (c) f (1.375) ≈ −0.0059. Hence, f (x) = 0 for some x between 1.25 and 1.375. 17. Find an interval of length 14 in [0,31] containing a root of x 5 − 5x + 1 = 0. Figure 4 shows that f (x) = x − 8x − 1 has a root in the interval [2.75, 3]. Apply the Bisection Method twice SOLUTION f (x)of=length x 5 −15xcontaining + 1. Observe that f is continuous with f (0) = 1 and f (1) = −3. Therefore, this root. to find an Let interval 16 by the IVT there is a c ∈ [0, 1] such that f (c) = 0. f (.5) = −1.46875 < 0, so f (c) = 0 for some c ∈ [0, .5]. f (.25) = −.0.249023 < 0, and so f (c) = 0 for some c ∈ [0, .25]. This means that [0, .25] is an interval of length 0.25 containing a root of f (x). In Exercises 19–22, draw the graph of a function f (x) on [0, 4] with the given property. Show that tan3 θ − 8 tan2 θ + 17 tan θ − 8 = 0 has a root in [0.5, 0.6]. Apply the Bisection Method twice to find interval of lengthat0.025 root. the conclusion of the IVT. 19. an Jump discontinuity x = 2containing and does this not satisfy SOLUTION The function graphed below has a jump discontinuity at x = 2. Note that while f (0) = 2 and f (4) = 4, there is no point c in the interval [0, 4] such that f (c) = 3. Accordingly, the conclusion of the IVT is not satisfied. y 4 3 2 1 x 1

2

3

4

21. Infinite one-sided limits at x = 2 and does not satisfy the conclusion of the IVT. Jump discontinuity at x = 2, yet does satisfy the conclusion of the IVT on [0, 4]. SOLUTION The function graphed below has infinite one-sided limits at x = 2. Note that while f (0) = 2 and f (4) = 4, there is no point c in the interval [0, 4] such that f (c) = 3. Accordingly, the conclusion of the IVT is not satisfied. y 6 5 4 3 2 1 x −1

1

2

3

4

S E C T I O N 2.7

Intermediate Value Theorem

67

Corollary 2 islimits not foolproof. Letdoes f (x) = x 2the−conclusion 1 and explain why 23. Infinite one-sided at x = 2, yet satisfy of the IVTtheoncorollary [0, 4]. fails to detect the roots x = ±1 if [a, b] contains [−1, 1]. If [a, b] contains [−1, 1], f (a) > 0 and f (b) > 0. If we were using the corollary, we would guess that f (x) has no roots in [a, b]. However, we can easily see that f (1) = 0 and f (−1) = 0, so that any interval [a, b] containing [−1, 1] contains two roots. SOLUTION

Further Insights and Challenges f (x) is continuous and that ≤ f (x) ≤ 1 on forit0 anywhere. ≤ x ≤ 1 (see Figure Show that f (c) =c 25. Take Assume any mapthat (e.g., of the United States) and0draw a circle Prove that 5). at any moment in time for some in [0,a1]. therecexists pair of diametrically opposite points on that circle corresponding to locations where the temperatures at that moment are equal. Hint: Let θ be an angular coordinate along the circle and let f (θ ) be the difference in y temperatures at the locations corresponding to θ and θ + π . y = x 1

y = f (x)

x c

1

FIGURE 5 A function satisfying 0 ≤ f (x) ≤ 1 for 0 ≤ x ≤ 1. SOLUTION If f (0) = 0, the proof is done with c = 0. We may assume that f (0) > 0. Let g(x) = f (x) − x. g(0) = f (0) − 0 = f (0) > 0. Since f (x) is continuous, the Rule of Differences dictates that g(x) is continuous. We need to prove that g(c) = 0 for some c ∈ [0, 1]. Since f (1) ≤ 1, g(1) = f (1) − 1 ≤ 0. If g(1) = 0, the proof is done with c = 1, so let’s assume that g(1) < 0. We now have a continuous function g(x) on the interval [0, 1] such that g(0) > 0 and g(1) < 0. From the IVT, there must be some c ∈ [0, 1] so that g(c) = 0, so f (c) − c = 0 and so f (c) = c. This is a simple case of a very general, useful, and beautiful theorem called the Brouwer fixed point theorem.

27. Figure 6(B) shows a slice of ham a piece of bread. Prove thatthat it is to θ slice thisθ open-faced Ham Sandwich Theorem Figure 6(A)onshows a slice of ham. Prove forpossible any angle (0 ≤ ≤ π ), it is θ ≤ π a line sandwich so that each part has equal amounts of ham and bread. Hint: By Exercise 26, for all 0 ≤ givenisby the possible to cut the slice of ham in half with a cut of incline θ . Hint: The lines of inclination θ are there L(θ )equations of incliney θ=(which we+assume is unique) into twoline equal pieces. = 0into or πtwo , letpieces B(θ ) (tan θ )x b, where b varies that fromdivides −∞ tothe ∞.ham Each such divides the For hamθslice denote theofamount of bread to theLet leftA(b) of L(beθ )the minus the of amount theleft right. Notice that L( π )amount = L(0) θ = (one which may be empty). amount ham totothe of the line minus the to since the right and0 and let θ =A πbegive the same line, but B( π ) = −B(0) since left and right get interchanged as the angle moves from 0 to the total area of the ham. Show that A(b) = −A if b is sufficiently large and A(b) = A if b is sufficiently π π . Assume that B( θ ) is continuous and apply the IVT. (By a further extension of this argument, one can prove the full negative. Then use the IVT. This works if θ = 0 or 2 . If θ = 0, define A(b) as the amount of ham above the “Ham Sandwich Theorem,” which states thatHow if you knifethe to cut at a slant, then itwhen is possible cut a sandwich line y = b minus the amount below. canallow you the modify argument to work θ = πto 2 (in which case consisting of a slice of ham and two slices of bread so that all three layers are divided in half.) tan θ = ∞)? y

( )

L 2

y

L( )

L (0) = L( )

x (A) Cutting a slice of ham at an angle .

x (B) A slice of ham on top of a slice of bread.

FIGURE 6 SOLUTION For each angle θ , 0 ≤ θ < π , let L(θ ) be the line at angle θ to the x-axis that slices the ham exactly in half, as shown in figure 6. Let L(0) = L(π ) be the horizontal line cutting the ham in half, also as shown. For θ and L(θ ) thus defined, let B(θ ) = the amount of bread to the left of L(θ ) minus that to the right of L(θ ). To understand this argument, one must understand what we mean by “to the left” or “to the right”. Here, we mean to the left or right of the line as viewed in the direction θ . Imagine you are walking along the line in direction θ (directly right if θ = 0, directly left if θ = π , etc). We will further accept the fact that B is continuous as a function of θ , which seems intuitively obvious. We need to prove that B(c) = 0 for some angle c. Since L(0) and L(π ) are drawn in opposite direction, B(0) = −B(π ). If B(0) > 0, we apply the IVT on [0, π ] with B(0) > 0, B(π ) < 0, and B continuous on [0, π ]; by IVT, B(c) = 0 for some c ∈ [0, π ]. On the other hand, if B(0) < 0, then we apply the IVT with B(0) < 0 and B(π ) > 0. If B(0) = 0, we are also done; L(0) is the appropriate line.

68

CHAPTER 2

LIMITS

2.8 The Formal Definition of a Limit Preliminary Questions 1. Given that lim cos x = 1, which of the following statements is true? x→0

(a) (b) (c) (d)

If |cos x − 1| is very small, then x is close to 0. There is an > 0 such that |x| < 10−5 if 0 < |cos x − 1| < . There is a δ > 0 such that |cos x − 1| < 10−5 if 0 < |x| < δ . There is a δ > 0 such that |cos x| < 10−5 if 0 < |x − 1| < δ .

SOLUTION

The true statement is (c): There is a δ > 0 such that |cos x − 1| < 10−5 if 0 < |x| < δ .

2. Suppose that for a given and δ , it is known that | f (x) − 2| < if 0 < |x − 3| < δ . Which of the following statements must also be true? (a) | f (x) − 2| < if 0 < |x − 3| < 2δ (b) | f (x) − 2| < 2 if 0 < |x − 3| < δ δ (c) | f (x) − 2| < if 0 < |x − 3| < 2 2 δ (d) | f (x) − 2| < if 0 < |x − 3| < 2 SOLUTION Statements (b) and (d) are true.

Exercises 1. Consider lim f (x), where f (x) = 8x + 3. x→4

(a) Show that | f (x) − 35| = 8|x − 4|. (b) Show that for any > 0, | f (x) − 35| < if |x − 4| < δ , where δ = 8 . Explain how this proves rigorously that lim f (x) = 35. x→4

SOLUTION

(a) | f (x) − 35| = |8x + 3 − 35| = |8x − 32| = |8(x − 4)| = 8 |x − 4|. (Remember that the last step is justified because 8 > 0). (b) Let > 0. Let δ = /8 and suppose |x − 4| < δ . By part (a), | f (x) − 35| = 8|x − 4| < 8δ . Substituting δ = /8, we see | f (x) − 35| < 8/8 = . We see that, for any > 0, we found an appropriate δ so that |x − 4| < δ implies | f (x) − 35| < . Hence lim f (x) = 35. x→4

3. Consider lim x 2 = 4 (refer to Example 2). Consider x→2lim f (x), where f (x) = 4x − 1. x→2 (a) Show that |x 2 − 4| < 0.05 if 0 < |x − 2| < 0.01.

(a) Show that | f (x) − 7| < 4δ if |x − 2| < δ . (b) Show thata|xδ 2such − 4|that: < 0.0009 if 0 < |x − 2| < 0.0002. (b) Find (c) Find a value of δ such that |x 2 − 4| is less than 10−4 if 0 < |x − 2| < δ . | f (x) − 7| < 0.01 if |x − 2| < δ SOLUTION |x −rigorously (a) If 2| < δ = that .01, then and x 2 − 4 = |x − 2||x + 2| ≤ |x − 2| (|x| + 2) < 5|x − 2| < .05. (c)0 < Prove lim f|x| (x) 0 such that With regard to the limit lim x2 = 25, x→5 1 1 −4 if Write 0< −25| 3| < < 10 2− x < 6. Hint: |x |x = δ|x + 5| · |x − 5|. (a) Show that |x 2 − 25| < 11|x −x 5|−if34 < (b) Find a δ such that |x 2 − 25| < 10−3 if |x − 5| < δ . SOLUTION The Example shows that for any > 0 we have2 (c) Give a rigorous proof of the limit by showing that |x − 25| < if 0 < |x − 5| < δ , where δ is the smaller of 1 11 and 1. − 1 ≤ if |x − 3| < δ x 3 where δ is the smaller of the numbers 6 and 1. In our case, we may take δ = 6 × 10−4 .

S E C T I O N 2.8

The Formal Definition of a Limit

69

Plot f (x) =√tan x together with the horizontal lines y = 0.99 and y = 1.01. Use this plot to find a value of 7. π | 0 such that | 2x − 1 − 3| < 0.1 if |x − 5| < δ . π SOLUTION From the plot below, we see that δ = 0.005 will guarantee that |tan x − 1| < 0.01 whenever |x − 4 | ≤ δ . y 1.02 1.01 1 0.99 0.98

x 0.775

0.78

0.785

0.79

0.795

1 9. Consider lim . 4 . Observe that the limit L = lim f (x) has the value L = f ( 12 ) = 16 Let x→2 f (x) x= 2 5 . Let = 0.5 and, x x→ 12 (a) Show that if |x − 2| 0 such 5 2 1 − 1 < 1 |x − 2| x 2 2 (b) Let δ be the smaller of 1 and 2. Prove the following statement: 1 − 1 < if 0 < |x − 2| < δ x 2 (c) Find a δ > 0 such that 1x − 12 < 0.01 if |x − 2| < δ . (d) Prove rigorously that lim

1

x→2 x

=

1 . 2

SOLUTION

(a) Since |x − 2| < 1, it follows that 1 < x < 3, in particular that x > 1. Because x > 1, then 1 − 1 = 2 − x = |x − 2| < 1 |x − 2|. x 2 2x 2x 2

1 < 1 and x

(b) Let δ = min{1, 2} and suppose that |x − 2| < δ . Then by part (a) we have 1 − 1 < 1 |x − 2| < 1 δ < 1 · 2 = . x 2 2 2 2 1 1 1 (c) Choose δ = .02. Then − < δ = .01 by part (b). x 2 2 (d) Let > 0 be given. Then whenever 0 < |x − 2| < δ = min {1, 2}, we have 1 − 1 < 1 δ ≤ . x 2 2 Since was arbitrary, we conclude that lim

1

x→2 x

=

1 . 2

11. Based on the information conveyed in Figure 5(A), find values of L, , and δ > 0 such that the following statement √ +< 3. δ . lim holds: | Consider f (x) − L|x→1 < ifx|x| √ √ SOLUTION forces < Hint: 4.7. Rewritten, thisinequality means that 0|3< implies that x + 3 3 + 2 > 3. This gives us 1 √ < |x − 4| 1 . | x − 2| = |x − 4| √ 3 x + 2

SOLUTION

Let δ = min(1, 3). If |x − 4| < δ ,

1 √ < |x − 4| 1 < δ 1 < 3 1 = , | x − 2| = |x − 4| √ 3 3 3 x + 2

thus proving the limit rigorously. 19. lim x 3 = 12 x→1lim (3x + x) = 4 x→1

SOLUTION

Let > 0 be given. We bound x 3 − 1 by factoring the difference of cubes: 3 x − 1 = (x 2 + x + 1)(x − 1) = |x − 1| x 2 + x + 1 .

S E C T I O N 2.8

The Formal Definition of a Limit

71

Let δ = min(1, 7 ), and assume |x − 1| < δ . Since δ < 1, 0 < x < 2. Since x 2 + x + 1 increases as x increases for x > 0, x 2 + x + 1 < 7 for 0 < x < 2, and so 3 x − 1 = |x − 1| x 2 + x + 1 < 7|x − 1| < 7 = 7 and the limit is rigorously proven. 21. lim |x| = |c| x→c 1 lim x −2 = |x| + |c| 4 x→2 SOLUTION Let > 0 be given. First we bound |x| − |c| by multiplying by . |x| + |c| 2 2 2 2 |x| − |c| < |x| − |c| |x| + |c| = ||x| − |c| | = x − c = (x + c)(x − c) ≤ |x − c| |x + c| . |x| + |c| |x| + |c| |x| + |c| |x| + |c| |x| + |c| Let δ = . We apply the Triangle Inequality—for all a, b: |a + b| ≤ |a| + |b|. This yields: |x| − |c| ≤ |x − c| |x + c| ≤ |x − c| = , |x| + |c| and the limit is rigorously proven. x 23. Let f (x) = 1 . Prove rigorously that lim f (x) does not exist. Hint: Show that no number L qualifies as the limit x→0 lim x sin|x| = 0 x exists some x such that |x| < δ but | f (x) − L| ≥ 1 , no matter how small δ is taken. becausex→0 there always 2 Let L be any real number. Let δ > 0 be any small positive number. Let x = δ2 , which satisfies |x| < δ , and f (x) = 1. We consider two cases: SOLUTION

• (| f (x) − L| ≥ 1 ) : we are done. 2 • (| f (x) − L| < 1 ): This means 1 < L < 3 . In this case, let x = − δ . f (x) = −1, and so 3 < L − f (x). 2 2 2 2 2

In either case, there exists an x such that |x| < δ2 , but | f (x) − L| ≥ 12 . 25. Use the identity Prove rigorously that lim sin 1x does not exist. x+y x−y x→0 sin x + sin y = 2 sin cos 2 2 to verify the relation sin(a + h) − sin a = h

sin(h/2) h cos a + h/2 2

6

Conclude that |sin(a + h) − sin a| < |h| for all a and prove rigorously that lim sin x = sin a. You may use the inequality x→a sin x ≤ 1 for x = 0. x SOLUTION

We first write sin(a + h) − sin a = sin(a + h) + sin(−a).

Applying the identity with x = a + h, y = −a, yields:

a+h−a 2a + h cos 2 2 h h h h h sin(h/2) h = 2 sin cos a + =2 sin cos a + =h cos a + . 2 2 h 2 2 h/2 2

sin(a + h) − sin a = sin(a + h) + sin(−a) = 2 sin

Therefore,

sin(h/2) cos a + h . |sin(a + h) − sin a| = |h| h/2 2

sin θ < 1 and that |cos θ | ≤ 1, we have Making the substitution h = x − a, we see that this last Using the fact that θ relation is equivalent to |sin x − sin a| < |x − a|.

72

CHAPTER 2

LIMITS

Now, to prove the desired limit, let > 0, and take δ = . If |x − a| < δ , then |sin x − sin a| < |x − a| < δ = , Therefore, a δ was found for arbitrary , and the proof is complete. Use Eq. (6) to prove rigorously that Further Insights and Challenges sin(a + h) − sin 27. Uniqueness of the Limit Show that a function converges to aat most one limiting value. In other words, use the = cos a lim limit definition to show that if lim f (x) = L 1h→0 and lim f (x) h = L 2 , then L 1 = L 2 . x→c

SOLUTION

x→c

Let > 0 be given. Since lim f (x) = L 1 , there exists δ1 such that if |x − c| < δ1 then | f (x) − L 1 | < . x→c

Similarly, since lim f (x) = L 2 , there exists δ2 such that if |x − c| < δ2 then | f (x) − L 2 | < . Now let |x − c| < x→c min(δ1 , δ2 ) and observe that |L 1 − L 2 | = |L 1 − f (x) + f (x) − L 2 | ≤ |L 1 − f (x)| + | f (x) − L 2 | = | f (x) − L 1 | + | f (x) − L 2 | < 2. So, |L 1 − L 2 | < 2 for any > 0. We have |L 1 − L 2 | = lim |L 1 − L 2 | < lim 2 = 0. Therefore, |L 1 − L 2 | = 0 →0

and, hence, L 1 = L 2 .

→0

In Exercises 28–30, prove the statement using the formal limit definition. 29. The Squeeze Theorem. (Theorem 1 in Section 2.6, p. 89) The Constant Multiple Law [Theorem 1, part (ii) in Section 2.3, p. 64] SOLUTION Proof of the Squeeze Theorem. Suppose that (i) the inequalities h(x) ≤ f (x) ≤ g(x) hold for all x near (but not equal to) a and (ii) lim h(x) = lim g(x) = L. Let > 0 be given. x→a

x→a

• By (i), there exists a δ1 > 0 such that h(x) ≤ f (x) ≤ g(x) whenever 0 < |x − a| < δ1 . • By (ii), there exist δ2 > 0 and δ3 > 0 such that |h(x) − L| < whenever 0 < |x − a| < δ2 and |g(x) − L| < whenever 0 < |x − a| < δ3 . • Choose δ = min {δ1 , δ2 , δ3 }. Then whenever 0 < |x − a| < δ we have L − < h(x) ≤ f (x) ≤ g(x) < L + ;

i.e., | f (x) − L| < . Since was arbitrary, we conclude that lim f (x) = L. x→a

31. Let f (x) = 1 if x is rational and f (x) = 0 if x is irrational. Prove that lim f (x) does not exist for any c. The Product Law [Theorem 1, part (iii) in Section 2.3, p. 64]. Hint: x→c Use the identity SOLUTION Let c be any number, and let δ > 0 be an arbitrary small number. We will prove that there is an x such f (x)g(x) − L M = ( f (x) − L) g(x) + L(g(x) − M) that |x − c| < δ , but | f (x) − f (c)| > 12 . c must be either irrational or rational. If c is rational, then f (c) = 1. Since the irrational numbers are dense, there is at least one irrational number z such that |z − c| < δ . | f (z) − f (c)| = 1 > 12 , so the function is discontinuous at x = c. On the other hand, if c is irrational, then there is a rational number q such that |q − c| < δ . | f (q) − f (c)| = |1 − 0| = 1 > 12 , so the function is discontinuous at x = c.

Here is a function with strange continuity properties: ⎧ p 1 if x is the rational number in ⎪ ⎨ q q lowest terms f (x) = 1. A particle’s position at time t (s) is s(t) =⎪ ⎩ t 2 + 1 m. Compute its average velocity over [2, 5] and estimate its 0 if x is an irrational number instantaneous velocity at t = 2. (a) ShowLet thats(t) f (x) at c if velocity c is rational. exist irrational numbers arbitrarily close to c. SOLUTION = is discontinuous t 2 + 1. The average over Hint: [2, 5] There is (b) Show that f (x) is continuous at c if c is irrational. Let I be the interval {x : |x − c| < 1}. Show that for √ Hint: p√ q < Q. Conclude that there is a δ such that all any Q > 0, I contains at most finitely s(5) −many s(2) fractions 26 −q with 5 = ≈ 0.954 m/s. larger than Q. fractions in {x : |x − c| < δ } have a denominator 5−2 3

CHAPTER REVIEW EXERCISES

From the data in the table below, we estimate that the instantaneous velocity at t = 2 is approximately 0.894 m/s. interval average ROC

[1.9, 2]

[1.99, 2]

[1.999, 2]

[2, 2.001]

[2, 2.01]

[2, 2.1]

0.889769

0.893978

0.894382

0.894472

0.894873

0.898727

3. For a whole number n, let P(n) be the number of partitions of n, that is, the number of ways of writing n as a sum The price p ofnumbers. natural gas the United States dollars per 1,000 ft3 ) be on partitioned the first dayinoffive each monthways: in 2004 of one or more whole Forinexample, P(4) = 5(in since the number 4 can different 4, 3 per month) over the is listed in the table below. Compute the average rate of change of p (in dollars per 1,000 ft 3 + 1, 2 + 2, 2 + 1 + 1, 1 + 1 + 1 + 1. Treating P(n) as a continuous function, use Figure 1 to estimate the rate of quarterly periods April–June, and July–September. change of P(n) at n =January–March, 12. J

F

M

A

M

J

Chapter Review Exercises

73

SOLUTION The tangent line drawn in the figure appears to pass through the points (15, 140) and (10.5, 40). We therefore estimate that the rate of change of P(n) at n = 12 is

100 200 140 − 40 = = . 15 − 10.5 4.5 9 In Exercises 5–8, estimate the limit numerically to two decimal places or state that the limit does not exist.√ The average velocity v (m/s) of an oxygen molecule in the air at temperature T (◦ C) is v = 25.7 273.15 + T . 3 (x) What1is−the speed at T = 25◦ (room temperature)? Estimate the rate of change of average velocity with cosaverage 5. respect lim to temperature at T = 25◦ . What are the units of this rate? x→0 x2 SOLUTION

3x Let f (x) = 1−cos . The data in the table below suggests that 2 FIGURE 1 Graph of P(n).

x

1 − cos3 x ≈ 1.50. x→0 x2 lim

In constructing the table, we take advantage of the fact that f is an even function. x

±0.001

±0.01

±0.1

f (x)

1.500000

1.499912

1.491275

(The exact value is 32 .) xx − 4 7. limlim2 x 1/(x−1) x→2 x −4 x→1 SOLUTION

x Let f (x) = x 2 −4 . The data in the table below suggests that

x −4

xx − 4 ≈ 1.69. x→2 x 2 − 4 lim

x

1.9

1.99

1.999

2.001

2.01

2.1

f (x)

1.575461

1.680633

1.691888

1.694408

1.705836

1.828386

(The exact value is 1 + ln 2.) In Exercises 9–42, the limit if possible or state that it does not exist. 3x − evaluate 9 lim x 1/2 (3 +5x − 25 ) 9. limx→2 x→4 √ SOLUTION lim (3 + x 1/2 ) = 3 + 4 = 5. x→4

4 11. lim 35 − x 2 x→−2 limx x→1 4x + 7 4 4 1 SOLUTION lim = =− . 2 x→−2 x 3 (−2)3 x3 − x 2 3x + 4x + 1 x −1 x→1 lim x +1 x→−1 x3 − x x(x − 1)(x + 1) SOLUTION lim = lim = lim x(x + 1) = 1(1 + 1) = 2. x −1 x→1 x − 1 x→1 x→1 √ t −33 15. lim x − 2x t −9 t→9lim √ √ x→1 x − 1 1 1 t −3 t −3 1 SOLUTION lim = . = lim √ = √ = lim √ √ 6 t→9 t − 9 t→9 ( t − 3)( t + 3) t→9 t + 3 9+3 √ x +1−2 t −6 17. lim x −3 x→3lim √ t→9 t − 3 13. lim

SOLUTION

√ lim

x→3

x +1−2 = lim x −3 x→3

√

= lim √ x→3

√ x +1−2 x +1+2 (x + 1) − 4 = lim ·√ √ x −3 x + 1 + 2 x→3 (x − 3)( x + 1 + 2) 1 x +1+2

= √

1 3+1+2

=

1 . 4

74

CHAPTER 2

LIMITS

2 − 2a 2 2(a + h) 19. lim 1 − s2 + 1 h h→0lim s→0 s2 SOLUTION

2(a + h)2 − 2a 2 2a 2 + 4ah + 2h 2 − 2a 2 h(4a + 2h) = lim = lim = lim (4a + 2h) = 4a + 2(0) = 4a. h h h h→0 h→0 h→0 h→0 lim

1 21. lim 2 1 t→3 tlim −9 t→−1+ x + 1 SOLUTION Because the one-sided limits 1

lim

= −∞

t→3− t 2 − 9

and

lim

1

t→3+ t 2 − 9

= ∞,

are not equal, the two-sided limit lim

1

does not exist.

t→3 t 2 − 9

a 2 −33ab + 2b2 x − b3 a−b a→blim x→b x − b a 2 − 3ab + 2b2 (a − b)(a − 2b) SOLUTION lim = lim = lim (a − 2b) = b − 2b = −b. a−b a−b a→b a→b a→b 1 1 25. lim x 3−−x(x ax 2++3)ax − 1 x→0lim3x x − 1 x→1 1 (x + 3) − 3 1 1 1 1 SOLUTION lim − = lim = lim = = . x(x + 3) 3(0 + 3) 9 x→0 3x x→0 3x(x + 3) x→0 3(x + 3)

23. lim

27.

lim

[x]

x→1.5 limx x→1+

SOLUTION

1

1

− x− x2 − 1 1 [1.5] 2 [x]1 lim = = = . 1.5 1.5 3 x→1.5 x √

[x] t→0−limx

29. lim

1

x→4.3 x − [x]

SOLUTION

For x sufficiently close to zero but negative, [x] = −1. Therefore, lim

[x]

x→0− x

= lim

−1

x→0− x

= ∞.

31. lim sec θ [x] θ → π4 lim t→0+ x SOLUTION

lim sec θ = sec

θ → π4

√ π = 2. 4

cos θ − 2 33. lim lim θ sec θ θ →0 θ → π2 θ SOLUTION

Because the one-sided limits cos θ − 2 =∞ θ θ →0− lim

and

cos θ − 2 = −∞ θ θ →0+ lim

are not equal, the two-sided limit cos θ − 2 θ θ →0 lim

sin 4x 35. lim sin 5θ sin 3x θ →0lim θ →0 θ

does not exist.


75

SOLUTION

lim

sin 4θ

θ →0 sin 3θ

=

4 3θ 4 4 sin 4θ sin 4θ 3θ 4 · = lim · lim = (1)(1) = . lim 3 θ →0 4θ sin 3θ 3 θ →0 4θ 3 3 θ →0 sin 3θ

37. lim tan x x→ π2 sin2 t lim 3 t→0 Because t SOLUTION the one-sided limits lim tan x = ∞

lim tan x = −∞

and

x→ π2 −

x→ π2 +

are not equal, the two-sided limit lim tan x

does not exist.

x→ π2

√ 1 t cos1 39. lim lim cos t t→0+ t t→0 SOLUTION For t > 0, −1 ≤ cos so

1 ≤ 1, t

√ √ √ 1 ≤ t. − t ≤ t cos t

Because √ √ t = 0, lim − t = lim

t→0+

t→0+

it follows from the Squeeze Theorem that √ 1 t cos = 0. t t→0+ lim

41. lim

cos x − 1 sin 7x

x→0lim sin x x→0 sin 3x

SOLUTION

cos x − 1 cos x − 1 cos x + 1 − sin2 x sin x 0 = lim · = lim = − lim =− = 0. cos x + 1 x→0 sin x(cos x + 1) 1+1 x→0 sin x x→0 sin x x→0 cos x + 1 lim

43. Find the left- and right-hand limits of the function f (x) in Figure 2 at x = 0, 2, 4. State whether f (x) is left- or tan θ − sin θ right-continuous (or both) at these points. lim θ →0 sin3 θ y

2 1 x 1

2

3

4

5

FIGURE 2 SOLUTION

According to the graph of f (x), lim f (x) = lim f (x) = 1

x→0−

x→0+

lim f (x) = lim f (x) = ∞

x→2−

x→2+

lim f (x) = −∞

x→4−

lim f (x) = ∞.

x→4+

The function is both left- and right-continuous at x = 0 and neither left- nor right-continuous at x = 2 and x = 4.

76

CHAPTER 2

LIMITS

45. Sketch the graph of a function g(x) such that Sketch the graph of a function f (x) such that lim g(x) = −∞, lim g(x) = ∞ lim g(x) = ∞, = 1, lim f (x) = 3 lim f (x) x→−3− x→−3+ x→4 x→2−

x→2+

SOLUTION

and such that lim f (x) but does not equal f (4). x→4

y 10 5 −4

x

−2

2

4

6

−5 −10

47. Find a constant b such that h(x) is continuous at x = 2, where Find the points of discontinuity of g(x) and determine the type of discontinuity, where ⎧x + 1 for π x |x| < 2 h(x) = ⎪ ⎨cos 2 for |x| < 1 b − x 2 for |x| ≥ 2 g(x) = ⎪ ⎩ for |x| ≥ 1 With this choice of b, find all points of discontinuity. |x − 1| SOLUTION

To make h(x) continuous at x = 2, we must have the two one-sided limits as x approaches 2 be equal.

With lim h(x) = lim (x + 1) = 2 + 1 = 3

x→2−

x→2−

and lim h(x) = lim (b − x 2 ) = b − 4,

x→2+

x→2+

it follows that we must choose b = 7. Because x + 1 is continuous for −2 < x < 2 and 7 − x 2 is continuous for x ≤ −2 and for x ≥ 2, the only possible point of discontinuity is x = −2. At x = −2, lim

x→−2+

h(x) =

lim (x + 1) = −2 + 1 = −1

x→−2+

and lim

x→−2−

h(x) =

lim (7 − x 2 ) = 7 − (−2)2 = 3,

x→−2−

so h(x) has a jump discontinuity at x = −2. 49. Let f (x) and g(x) be functions such that g(x) = 0 for x = a, and let Calculate the following limits, assuming that B= lim g(x), L= = 4lim A = lim f (x), = x→a 6, lim g(x) lim f (x) x→a x→a x→3

x→3

f (x) g(x)

Prove(a)thatlim if the limits A, B, and L exist and L = 1, then A = (b) B. Hint: cannot use the Quotient Law if B = 0, so 2 f (x) ( f (x) − 2g(x)) lim xYou apply thex→3 Product Law to L and B instead. x→3 f (x) SOLUTION (d)Then lim (2g(x)3 − g(x)2 ) (c) lim Suppose the limits A, B, and L all exist and L = 1. x→3 g(x) + x x→3 f (x) f (x) = lim g(x) = lim f (x) = A. B = B · 1 = B · L = lim g(x) · lim x→a x→a g(x) x→a x→a g(x) 51. Let f (x) be a function defined for all real numbers. Which of the following statements must be true, might be true, In true. the notation of Exercise 49, give an example in which L exists but neither A nor B exists. or are never (a) lim f (x) = f (3). x→3

= 1, then f (0) = 0. (b) If lim f (x) x→0 x 1

1 . 8 (d) If lim f (x) = 4 and lim f (x) = 8, then lim f (x) = 6. (c) If lim f (x) = 8, then lim x→−7

x→5+

x→−7 f (x)

x→5−

(e) If lim f (x) = 1, then lim f (x) = 0 x→0 x x→0

=

x→5


77

(f) If lim f (x) = 2, then lim f (x)3 = 8. x→5

x→5

SOLUTION

(a) This statement might be true. If f (x) is continuous at x = 3, then the statement will be true; if f (x) is not continuous at x = 3, then the statement will not be true. (b) This statement might be true. If f (x) is continuous at x = 0, then the statement will be true; if f (x) is not continuous at x = 0, then the statement will not be true. (c) This statement is always true. (d) This statement is never true. If the two one-sided limits are not equal, then the two-sided limit does not exist. (e) This statement is always true. (f) This statement is always true. 1 , where [x] is the greatest integer function. LetSqueeze f (x) = xTheorem 53. x Use the to prove that the function f (x) = x cos 1 for x = 0 and f (0) = 0 is continuous at x

x = 0. the graph of f (x) on the interval [ 1 , 2]. (a) Sketch 4 (b) Show that for x = 0,

1 1 1 −1< ≤ x x x Then use the Squeeze Theorem to prove that 1 =1 x x→0 lim x

SOLUTION

(a) The graph of f (x) = x 1x over [ 14 , 2] is shown below. y 1 0.8 0.6 0.4 0.2 x 0.5

1

1.5

2

(b) Let y be any real number. From the definition of the greatest integer function, it follows that y − 1 < [y] ≤ y, with equality holding if and only if y is an integer. If x = 0, then 1x is a real number, so 1 1 1 −1< ≤ . x x x Upon multiplying this inequality through by x, we find 1−x < x

1 ≤ 1. x

Because lim (1 − x) = lim 1 = 1,

x→0

x→0

it follows from the Squeeze Theorem that 1 = 1. x x→0 lim x

1 . 55. Let f (x) = x+2 Let r1 and r2 be the roots of the quadratic polynomial f (x) = ax 2 − 2x + 20. Observe that f (x) “approaches” |x−2| 1 the linear as < a 1. →Hint: 0. Since r = that 10 is the+unique = 1.0, we might expect (a) Show that function |x 20 − 2| Observe |4(x 2)| > root 12 if of |x L(x) − 2| < f (x) − L(x) 4 < =12−2xif + of f (x) to approach 10 as a → 0 (Figure 3). Prove that the roots can be labeled so that at least one of the roots 1 (b) Find lim rδ1 > = 010such andthat lim rf 2(x) =− ∞.4 < 0.01 for |x − 2| < δ . a→0

a→0

(c) Prove rigorously that lim f (x) = 14 . x→2

SOLUTION

FIGURE 3 Graphs of f (x) = ax 2 − 2x + 20.

78

CHAPTER 2

LIMITS 1 . Then (a) Let f (x) = x+2

f (x) −

1 4 − (x + 2) |x − 2| 1 1 = − = = . 4 x + 2 4 4(x + 2) |4(x + 2)|

If |x − 2| < 1, then 1 < x < 3, so 3 < x + 2 < 5 and 12 < 4(x + 2) < 20. Hence, 1 1 f (x) − 1 < |x − 2| . < and |4(x + 2)| 12 4 12 (b) If |x − 2| < δ , then by part (a),

f (x) −

1 δ < . 4 12

Choosing δ = 0.12 will then guarantee that | f (x) − 14 | < 0.01. (c) Let > 0 and take δ = min{1, 12}. Then, whenever |x − 2| < δ , f (x) − 1 = 1 − 1 = |2 − x| ≤ |x − 2| < δ = . 4 x + 2 4 4|x + 2| 12 12 57. Prove rigorously that lim (4 + 8x) = −4. x→−1 Plot the function f (x) = x 1/3 . Use the zoom feature to find a δ > 0 such that |x 1/3 − 2| < 0.05 for δ . |x − 8| < SOLUTION Let > 0 and take δ = /8. Then, whenever |x − (−1)| = |x + 1| < δ , | f (x) − (−4)| = |4 + 8x + 4| = 8|x + 1| < 8δ = . 59. Use the IVT to prove that the curves y = x 2 and y = cos x intersect. Prove rigorously that lim (x 2 − x) = 6. SOLUTION Let f (x) = x 2x→3 − cos x. Note that any root of f (x) corresponds to a point of intersection between the curves y = x 2 and y = cos x. Now, f (x) is continuous over the interval [0, π2 ], f (0) = −1 < 0 and f ( π2 ) = π4 > 0. Therefore, by the Intermediate Value Theorem, there exists a c ∈ (0, π2 ) such that f (c) = 0; consequently, the curves y = x 2 and y = cos x intersect. 2

61. Use the Bisection Method to locate a root of x22 − 7 = 0 to two decimal places. +2 has a root in the interval [0, 2]. Use the IVT to prove that f (x) = x 3 − x x+2 SOLUTION Let f (x) = x 2 − 7. By trial andcos error, we find that f (2.6) = −0.24 < 0 and f (2.7) = 0.29 > 0. Because f (x) is continuous on [2.6, 2.7], it follows from the Intermediate Value Theorem that f (x) has a root on (2.6, 2.7). We approximate the root by the midpoint of the interval: x = 2.65. Now, f (2.65) = 0.0225 > 0. Because f (2.6) and f (2.65) are of opposite sign, the root must lie on (2.6, 2.65). The midpoint of this interval is x = 2.625 and f (2.625) < 0; hence, the root must be on the interval (2.625, 2.65). Continuing in this fashion, we construct the following sequence of intervals and midpoints. interval

midpoint

(2.625, 2.65) (2.6375, 2.65) (2.64375, 2.65) (2.64375, 2.646875) (2.6453125, 2.646875)

2.6375 2.64375 2.646875 2.6453125 2.64609375

At this point, we note that, to two decimal places, one root of x 2 − 7 = 0 is 2.65. Give an example of a (discontinuous) function that does not satisfy the conclusion of the IVT on [−1, 1]. Then show that the function ⎧ ⎨sin 1 for x = 0 x f (x) = ⎩ 0 for x = 0 satisfies the conclusion of the IVT on every interval [−a, a], even though f is discontinuous at x = 0.

3 DIFFERENTIATION 3.1 Definition of the Derivative Preliminary Questions 1. Which of the lines in Figure 10 are tangent to the curve? D B

C

A

FIGURE 10 SOLUTION

Lines A and D are tangent to the curve.

2. What are the two ways of writing the difference quotient? SOLUTION

The difference quotient may be written either as f (x) − f (a) x −a

or as f (a + h) − f (a) . h 3. For which value of x is f (x) − f (3) f (7) − f (3) = ? x −3 4 SOLUTION

With x = 7, f (7) − f (3) f (x) − f (3) = . x −3 4

4. What do the following quantities represent in terms of the graph of f (x) = sin x? sin 1.3 − sin 0.9 (a) sin 1.3 − sin 0.9 (b) 0.4 (c) f (1.3) Consider the graph of y = sin x. (a) The quantity sin 1.3 − sin .9 represents the difference in height between the points (.9, sin .9) and (1.3, sin 1.3). sin 1.3 − sin .9 (b) The quantity represents the slope of the secant line between the points (.9, sin .9) and (1.3, sin 1.3) .4 on the graph. (c) The quantity f (1.3) represents the slope of the tangent line to the graph at x = 1.3. SOLUTION

5. For which values of a and h is and (5, f (5)) on the graph of f (x)?

f (a + h) − f (a) equal to the slope of the secant line between the points (3, f (3)) h

f (a + h) − f (a) is equal to the slope of the secant line between the points h (3, f (3)) and (5, f (5)) on the graph of f (x). SOLUTION

With a = 3 and h = 2,

6. To which derivative is the quantity tan π4 + 0.00001 − 1 0.00001 a good approximation? SOLUTION

tan( π4 + 0.00001) − 1 0.00001

is a good approximation to the derivative of the function f (x) = tan x at x = π4 .

80

CHAPTER 3

D I F F E R E N T I AT I O N

Exercises 1. Let f (x) = 3x 2 . Show that f (2 + h) = 3h 2 + 12h + 12 Then show that f (2 + h) − f (2) = 3h + 12 h and compute f (2) by taking the limit as h → 0. With f (x) = 3x 2 , it follows that

SOLUTION

f (2 + h) = 3(2 + h)2 = 3(4 + 4h + h 2 ) = 12 + 12h + 3h 2 . Using this result, we find f (2 + h) − f (2) 12 + 12h + 3h 2 − 3 · 4 12 + 12h + 3h 2 − 12 12h + 3h 2 = = = = 12 + 3h. h h h h As h → 0, 12 + 3h → 12, so f (2) = 12. In Exercises 3–6, compute f (a) in two ways, using Eq. (1) and Eq. (2). Let f (x) = 2x 2 − 3x − 5. Show that the slope of the secant line through (2, f (2)) and (2 + h, f (2 + h)) is + 5. formula to compute the slope of: 9x, this a= 0 3. 2h f (x) = Then x 2 + use (a) The secant line through (2, f (2)) and (3, f (3)) SOLUTION Let f (x) = x 2 + 9x. Then (b) The tangent line at x = 2 (by taking a limit) f (0 + h) − f (0) (0 + h)2 + 9(0 + h) − 0 9h + h 2 = lim = lim = lim (9 + h) = 9. h h h h→0 h→0 h→0 h→0

f (0) = lim Alternately,

f (x) − f (0) x 2 + 9x − 0 = lim = lim (x + 9) = 9. x −0 x x→0 x→0 x→0

f (0) = lim

5. f (x) = 3x 2 + 4x + 2, a = −1 f (x) = x 2 + 9x, a = 2 SOLUTION Let f (x) = 3x 2 + 4x + 2. Then f (−1 + h) − f (−1) 3(−1 + h)2 + 4(−1 + h) + 2 − 1 = lim h h h→0 h→0

f (−1) = lim

3h 2 − 2h = lim (3h − 2) = −2. h h→0 h→0

= lim Alternately,

f (x) − f (−1) 3x 2 + 4x + 2 − 1 = lim x − (−1) x +1 x→−1 x→−1

f (−1) = lim = lim

x→−1

(3x + 1)(x + 1) = lim (3x + 1) = −2. x +1 x→−1

In Exercises 7–10, refer2 to the function whose graph is shown in Figure 11. f (x) = 9 − 3x , a = 0 7. Calculate the slope of the secant line through the points on the graph where x = 0 and x = 2.5. SOLUTION

f (0) ≈ 6 and f (2.5) ≈ 2, so the slope of the secant line connecting (0, f (0)) and (2.5, f (2.5)) is 2−6 −4 f (2.5) − f (0) ≈ = = −1.6. 2.5 − 0 2.5 2.5

f (2 + h) − f (2) Estimate for h = 0.5 and h = −0.5. Are these numbers larger or smaller than f (2)? f (2.5) − f (1) h . What does this quantity represent? Compute 2.5 − 1 Explain. 9.


Definition of the Derivative

81

f (2) ≈ 1. If h = .5, f (2 + h) = f (2.5) ≈ 2, so 2−1 f (2 + h) − f (2) ≈ = 2. h .5

This number is probably greater than f (2). Since the curve is bending upward slightly at (2, f (2)), the secant line lies inside and over the curve, showing it has higher slope than the tangent line. If h = −.5, f (2 + h) = f (1.5) ≈ .5, so .5 − 1 f (2 + h) − f (2) ≈ = 1. h −.5 This number is smaller than f (2), since the curve is bending upwards near (2, f (2)). In Exercises 11–14, let f (x) be the function whose graph is shown in Figure 12. Estimate f (2). 5 4 3 2 1

y

x

1 2 3 4 5 6 7 8 9

FIGURE 12 Graph of f (x).

11. Determine f (a) for a = 1, 2, 4, 7. SOLUTION Remember that the value of the derivative of f at x = a can be interpreted as the slope of the line tangent to the graph of y = f (x) at x = a. From Figure 12, we see that the graph of y = f (x) is a horizontal line (that is, a line with zero slope) on the interval 0 ≤ x ≤ 3. Accordingly, f (1) = f (2) = 0. On the interval 3 ≤ x ≤ 5, the graph of y = f (x) is a line of slope 12 ; thus, f (4) = 12 . Finally, the line tangent to the graph of y = f (x) at x = 7 is horizontal, so f (7) = 0.

13. Which is larger: f (5.5) or f (6.5)? Estimate f (6). SOLUTION The line tangent to the graph of y = f (x) at x = 5.5 has a larger slope than the line tangent to the graph of y = f (x) at x = 6.5. Therefore, f (5.5) is larger than f (6.5). In Exercises the limit definition to find the derivative of the linear function. (Note: The derivative does not Show15–18, that f use (3) does not exist. depend on a.) 15. f (x) = 3x − 2 SOLUTION

f (a + h) − f (a) 3(a + h) − 2 − (3a − 2) = lim = lim 3 = 3. h h h→0 h→0 h→0 lim

17. g(t) = 9 − t f (x) = 2 SOLUTION

g(a + h) − g(a) 9 − (a + h) − (9 − a) −h = lim = lim = lim (−1) = −1. h h h→0 h→0 h→0 h h→0 lim

1 1 1 1 Does or + ? Compute the difference quotient for f (x) at a = −2 19. Letk(z) f (x)==16z.+ 9 f (−2 + h) equal x −2 + h −2 h with h = 0.5. SOLUTION

Let f (x) = 1x . Then f (−2 + h) =

1 . −2 + h

With a = −2 and h = .5, the difference quotient is 1 − 1 f (−1.5) − f (−2) 1 f (a + h) − f (a) = = −1.5 −2 = − . h .5 .5 3

Let f (x) = with h = 1.

√

x. Does f (5 + h) equal

√

5 + h or

√

5+

√

h? Compute the difference quotient for f (x) at a = 5

82

CHAPTER 3


21. Let f (x) =

√

x. Show that 1 f (9 + h) − f (9) = √ h 9+h+3

Then use this formula to compute f (9) (by taking the limit). √ SOLUTION We multiply numerator and denominator by 9 + h + 3 to make a difference of squares. Thus, f (9 + h) − f (9) = h =

√ √ √ 9+h−3 ( 9 + h − 3)( 9 + h + 3) = √ h h( 9 + h + 3) 9+h−9 h 1 √ = √ = √ . h( 9 + h + 3) h( 9 + h + 3) 9+h+3

Applying this formula we find: f (9) = lim

h→0

1 f (9 + h) − f (9) 1 1 = . = lim √ = √ h 6 h→0 9 + h + 3 9+3

√ In Exercises derivative at x = the limit x at x of = the 4. tangent line. First 23–40, find thecompute slope andthethen an equation of atheusing tangent line todefinition the graphand of find f (x)an=equation 23. f (x) = 3x 2 + 2x,

a=2

Let f (x) = 3x 2 + 2x. Then

SOLUTION

f (2 + h) − f (2) 3(2 + h)2 + 2(2 + h) − 16 = lim h h h→0 h→0

f (2) = lim

12 + 12h + 3h 2 + 4 + 2h − 16 = lim (14 + 3h) = 14. h h→0 h→0

= lim At a = 2, the tangent line is

y = f (2)(x − 2) + f (2) = 14(x − 2) + 16 = 14x − 12. 25. f (x) = x 3 , a2 = 2 f (x) = 3x + 2x, a = −1 SOLUTION Let f (x) = x 3 . Then f (2 + h) − f (2) (2 + h)3 − 23 8 + 12h + 6h 2 + h 3 − 8 = lim = lim = lim (12 + 6h + h 2 ) = 12 h h h h→0 h→0 h→0 h→0

f (2) = lim

At a = 2, the tangent line is y = f (2)(x − 2) + f (2) = 12(x − 2) + 8 = 12x − 16. 27. f (x) = x 3 +3x, a = 0 f (x) = x , a = 3 SOLUTION Let f (x) = x 3 + x. Then

f (0 + h) − f (0) (0 + h)3 + (0 + h) − 0 h3 + h = lim = lim = lim h 2 + 1 = 1. h h h h→0 h→0 h→0 h→0

f (0) = lim

The tangent line at a = 0 is y = f (0) + f (0)(x − 0) = 0 + 1(x − 0) or y = x. 29. f (x) = x −1 , 3 a = 3 f (t) = 3t + 2t, a = 4 SOLUTION Let f (x) = x −1 . Then

1 − 1 3−3−h 1 f (3 + h) − f (3) −h 3+h 3 3(3+h) = lim = lim = lim =− f (3) = lim h h h 9 h→0 h→0 h→0 h→0 (9 + 3h)h The tangent at a = 3 is 1 1 1 2 y = f (3)(x − 3) + f (3) = − (x − 3) + = − x + . 9 3 9 3 31. f (x) = 9x − 4, a = −7 f (x) = x −2 , a = 1



83

Let f (x) = 9x − 4. Then f (−7) = lim

h→0

= lim

h→0

f (−7 + h) − f (−7) (9(−7 + h) − 4) − (−67) = lim h h h→0 (−63 + 9h − 4) + 67 9h = lim = lim 9 = 9. h h→0 h h→0

The tangent line at a = −7 is y = f (−7)(x + 7) + f (−7) = 9(x + 7) − 67 = 9x − 4. In general, we can take a shorter path. The tangent line to a line is the line itself. 1√ a =a−2 33. f (x)f (t) = = t ,+ 1, =0 x +3 SOLUTION

1 . Then Let f (x) = x+3 1 1 −1 −1 f (−2 + h) − f (−2) −h −1 = lim −2+h+3 = lim 1+h = lim = lim = −1. h h h h→0 h→0 h→0 h→0 h(1 + h) h→0 1 + h

f (−2) = lim

The tangent line at a = −2 is y = f (−2)(x + 2) + f (−2) = −1(x + 2) + 1 = −x − 1. 2 35. f (t) = x ,+ 1a = −1 , a=3 f (x)1=− t x −1 2 SOLUTION Let f (t) = 1−t . Then 2 f (−1 + h) − f (−1) 2 − (2 − h) 1 1 1−(−1+h) − 1 = lim = lim = lim = . h h 2 h→0 h→0 h→0 h(2 − h) h→0 2 − h

f (−1) = lim

At a = −1, the tangent line is 1 1 3 y = f (−1)(x + 1) + f (−1) = (x + 1) + 1 = x + . 2 2 2 37. f (t) = t −3 , 1a = 1 f (t) = , a=2 t +f (t) 9 = 1 . Then SOLUTION Let t3 f (1 + h) − f (h) = lim h h→0 h→0

f (1) = lim

1 −1 (1+h)3

h

= lim

h→0

−h 3+3h+h 2 3 (1+h)

h

−(3 + 3h + h 2 ) = −3. h→0 (1 + h)3

= lim

The tangent line at a = 1 is y = f (1)(t − 1) + f (1) = −3(t − 1) + 1 = −3t + 4. 1 39. f (x)f (t) == √ t 4, , aa==92 x SOLUTION

1 Let f (x) = √ . Then x f (9 + h) − f (9) = lim h h→0 h→0

f (9) = lim

√1 − 13 9+h

h

= lim

h→0

√ √ 3−√ 9+h 3+√9+h · 3 9+h 3+ 9+h

h

= lim

h→0

−1

1 = lim √ =− . 54 h→0 9 9 + h + 3(9 + h) At a = 9 the tangent line is 1 1 1 1 y = f (9)(x − 9) + f (9) = − (x − 9) + = − x + . 54 3 54 2 41. What is an equation of the tangent line at x = 3, assuming that f (3) = 5 and f (3) = 2? f (x) = (x 2 + 1)−1 , a = 0

√ 9−9−h 9 9+h+3(9+h)

h

84

CHAPTER 3

D I F F E R E N T I AT I O N SOLUTION By definition, the equation of the tangent line to the graph of f (x) at x = 3 is y = f (3) + f (3)(x − 3) = 5 + 2(x − 3) = 2x − 1.

43. Consider the “curve” y = 2x + 8. What is the tangent line at the point (1, 10)? Describe the tangent line at an Suppose that y = 5x + 2 is an equation of the tangent line to the graph of y = f (x) at a = 3. What is f (3)? arbitrary point. What is f (3)? SOLUTION Since y = 2x + 8 represents a straight line, the tangent line at any point is the line itself, y = 2x + 8. 1 1 Verify thatf (x) P = and L(x) = 12 + m(x − 1) for every on the f (x) = 2 + 5h. Suppose that is a 1, function thatgraphs f (2 + of h) both − f (2) = 3h 2 liessuch 1 + x2 slope(a) m. What Plot fis(x)f and (2)? L(x) on the same axes for several values of m. Experiment until you find a value of m for which y = (b) L(x)What appears tangent to the What(2, is your f (1)? is the slope of the graph secantof linef (x). through f (2))estimate and (6, for f (6))?

45.

SOLUTION

Let f (x) =

1 and L(x) = 12 + m(x − 1). Because 1 + x2 f (1) =

1 1 = 2 1 + 12

L(1) =

and

1 1 + m(1 − 1) = , 2 2

it follows that P = (1, 12 ) lies on the graphs of both functions. A plot of f (x) and L(x) on the same axes for several values of m is shown below. The graph of L(x) with m = − 12 appears to be tangent to the graph of f (x) at x = 1. We therefore estimate f (1) = − 12 . y 1 m = −1

0.8 0.6

m = −1/4

0.4 0.2 0.5

1

m = −1/2 x 2

1.5

Plot f (x) = xx x and the line y = x + c on the same set of axes for several values of c. Experiment until you 47. Plot f (x) that = x theforline 0 ≤is xtangent ≤ 1.5.to the graph. Then: find a value c0 such are there thatthat f (xf0 )(x=)0? (a) How many points x (a) Use your plot to estimate0 the value such x 0 such 0 = 1. plotting horizontal lines y = c together with the plot of f (x) until you find a value of c (b) Estimatef (x x 00 by + h) − f (xseveral 0) (b) Verify thatthe line is tangent to the is close for which graph.to 1 for h = 0.01, 0.001, 0.0001. h SOLUTION The figure below shows the graphs of the function f (x) = x x together with the lines y = x − 0.5, y = x, and y = x + 0.5. y 2 1.5 1 0.5 x 0.2

0.4

0.6

0.8

1

1.2

1.4

−0.5

(a) The graph of y = x appears to be tangent to the graph of f (x) at x = 1. We therefore estimate that f (1) = 1. (b) With x 0 = 1, we generate the table h

0.01

0.001

0.0001

( f (x 0 + h) − f (x 0 ))/ h

1.010050

1.001001

1.000100

(1)net Thefollowing vapor pressure water isa defined as the quotients atmospheric at whichf no takes 49. Use the table toof calculate few difference of fpressure (x). ThenP estimate by evaporation taking an average place. The following table and Figure 13 give P (in atmospheres) as a function of temperature T in kelvins. of difference quotients at h and −h. For example, take the average for h = 0.03 and −0.03, etc. Recall that the a +or b P (350)? Answer by referring to the graph. (a) Which is larger: P (300) average of a and b is . 2 303, 313, 323, 333, 343 using the table and the average of the difference quotients for (b) Estimate P (T ) for T = h = 10 and −10: x 1.03 1.02 1.01 1 P(T + 10) − P(T − 10) )≈ 4 f (x)P (T0.5148 0.5234 0.5319 0.5403 20 x 0.97 0.98 0.99

f (x)

0.5653

0.557

0.5486


S E C T I O N 3.1

T (K)

293

303

313

323

333

343

85

353

P (atm) 0.0278 0.0482 0.0808 0.1311 0.2067 0.3173 0.4754 P (atm) 1 0.75 0.5 0.25 293 313 333 353 373 T (K)

FIGURE 13 Vapor pressure of water as a function temperature T in kelvins. SOLUTION

(a) Consider the graph of vapor pressure as a function of temperature. If we draw the tangent line at T = 300 and another at T = 350, it is clear that the latter has a steeper slope. Therefore, P (350) is larger than P (300). (b) Using equation (4), P (303) ≈ P (313) ≈ P (323) ≈ P (333) ≈ P (343) ≈

P(313) − P(293) 20 P(323) − P(303) 20 P(333) − P(313) 20 P(343) − P(323) 20 P(353) − P(333) 20

= = = = =

.0808 − .0278 20 .1311 − .0482 20 .2067 − .0808 20 .3173 − .1311 20 .4754 − .2067 20

= .00265 atm/K; = .004145 atm/K; = .006295 atm/K; = .00931 atm/K; = .013435 atm/K

In Exercises 50–51, traffic speed S along a certain road (in mph) varies as a function of traffic density q (number of cars per mile on the road). Use the following data to answer the questions: q (density)

100

110

120

130

140

S (speed)

45

42

39.5

37

35

(q) when The Squantity V q==q120 S is cars called volume. why V is equal to the atnumber of cars 51. Estimate pertraffic mile using the Explain average of difference quotients h and −h as inpassing Exercisea (q) when q = 120. particular point per hour. Use the data to compute values of V as a function of q and estimate V 48. SOLUTION The traffic speed S has units of miles/hour, and the traffic density has units of cars/mile. Therefore, the traffic volume V = Sq has units of cars/hour. A table giving the values of V follows.

q

100

110

120

130

140

V

4500

4620

4740

4810

4900

h

−20

−10

10

20

V (120 + h) − V (120) h

12

12

7

8

To estimate d V /dq, we take the difference quotient.

The mean of the difference quotients is 9.75. Hence d V /dq ≈ 9.75 mph when q = 120. and a.the derivative is positive. In Exercises 53–58, of the 14, limits represents derivative f (a). For the grapheach in Figure determine theaintervals along the Find x-axisf (x) on which (5 + h)3 − 125 h h→0

53. lim

SOLUTION

The difference quotient

sin( π63 + h) − .5 x − 125 h h→0lim x→5 x − 5

55. lim

(5 + h)3 − 125 f (a + h) − f (a) has the form where f (x) = x 3 and a = 5. h h

86

CHAPTER 3


SOLUTION


sin( π6 + h) − .5 h

has the form

f (a + h) − f (a) where f (x) = sin x and a = π6 . h

52+h −1 − 25 x −4 h→0lim h 1 x→ 41 x − 4

57. lim

5(2+h) − 25 f (a + h) − f (a) has the form where f (x) = 5x and a = 2. h h 59. Sketch the graph of f (x) = sin x on [0, π ] and guess the value of f π2 . Then calculate the slope of the secant h π π 5 −1 line between lim x = 2 and x = 2 + h for at least three small positive and negative values of h. Are these calculations h h→0 consistent with your guess? SOLUTION


SOLUTION

Here is the graph of y = sin x on [0, π ]. y 1 0.8 0.6 0.4 0.2 x 0.5 1 1.5 2 2.5 3

At x = π2 , we’re at the peak of the sine graph. The tangent line appears to be horizontal, so the slope is 0; hence, f ( π2 ) appears to be 0. h

−.01

−.001

−.0001

.0001

.001

.01

sin( π2 + h) − 1

.005

.0005

.00005

−.00005

−.0005

−.005

h

These numerical calculations are consistent with our guess. 4 √ Let15(A) f (x) = .graph of f (x) = x. The close-up in (B) shows that the graph is nearly a straight line Figure shows the 1 + 2x near x = 16. Estimate the slope of this line and take it as an estimate for f (16). Then compute f (16) and compare (a) Plot f (x) over [−2, 2]. Then zoom in near x = 0 until the graph appears straight and estimate the slope f (0). with your estimate. (b) Use your estimate to find an approximate equation to the tangent line at x = 0. Plot this line and the graph on the same set of axes.

61.

SOLUTION

4 over [−2, 2]. The figure below at the right is a close-up (a) The figure below at the left shows the graph of f (x) = 1+2 x near x = 0. From the close-up, we see that the graph is nearly straight and passes through the points (−0.22, 2.15) and (0.22, 1.85). We therefore estimate

f (0) ≈

1.85 − 2.15 −0.3 = = −0.68 0.22 − (−0.22) 0.44

y

y

3 2.5

2.4 2.2

2 1.5 1 0.5 −2

−1

2.0 1.8 x 1

−0.2

2

−0.1

x 0.1

0.2

(b) Using the estimate for f (0) obtained in part (a), the approximate equation of the tangent line is y = f (0)(x − 0) + f (0) = −0.68x + 2. The figure below shows the graph of f (x) and the approximate tangent line. y 3 2.5 2 1.5 1 0.5 −2

−1

x 1

2


S E C T I O N 3.1

87

Let f (x) = cot x. Estimate f ( π2 ) graphically by zooming in on a plot of f (x) near x = π2 . Repeat Exercise 61 with f (x) = (2 + x)x . π SOLUTION The figure below shows a close-up of the graph of f (x) = cot x near x = 2 ≈ 1.5708. From the closeup, we see that the graph is nearly straight and passes through the points (1.53, 0.04) and (1.61, −0.04). We therefore estimate

π −0.04 − 0.04 −0.08 f ≈ = = −1 2 1.61 − 1.53 0.08 63.

y 1 0.05 1.58 1.6

x

1.54 1.56

−0.05 −1

65. Apply the method of Example 6 to f (x) = sin x to determine f π4 accurately to four decimal places. For each graph in Figure 16, determine whether f (1) is larger or smaller than the slope of the secant line SOLUTION between xWe = know 1 and that x = 1 + h for h > 0. Explain. √ f (π /4 + h) − f (π /4) sin(π /4 + h) − 2/2 f (π /4) = lim = lim . h h h→0 h→0 Creating a table of values of h close to zero: h

√

sin( π4 + h) − ( 2/2) h

−.001

−.0001

−.00001

.00001

.0001

.001

.7074602

.7071421

.7071103

.7071033

.7070714

.7067531

Accurate up to four decimal places, f ( π4 ) ≈ .7071. 67. Sketch the graph of f (x) = x 5/2 on [0, 6]. Apply the method of Example 6 to f (x) = cos x to determine f ( π5 ) accurately to four decimal places. Use a (a) Use theofsketch to explain justify the h > in 0: this case. graph f (x) to howinequalities the methodfor works f (4 + h) − f (4) f (4) − f (4 − h) ≤ f (4) ≤ h h (b) Use part (a) to compute f (4) to four decimal places. (c) Use a graphing utility to plot f (x) and the tangent line at x = 4 using your estimate for f (4). SOLUTION

(a) The slope of the secant line between points (4, f (4)) and (4 + h, f (4 + h)) is f (4 + h) − f (4) . h x 5/2 is a smooth curve increasing at a faster rate as x → ∞. Therefore, if h > 0, then the slope of the secant line is greater than the slope of the tangent line at f (4), which happens to be f (4). Likewise, if h < 0, the slope of the secant line is less than the slope of the tangent line at f (4), which happens to be f (4). (b) We know that f (4 + h) − f (4) (4 + h)5/2 − 32 = lim . h h h→0 h→0

f (4) = lim

Creating a table with values of h close to zero: h

−.0001

−.00001

.00001

.0001

(4 + h)5/2 − 32 h

19.999625

19.99999

20.0000

20.0000375

Thus, f (4) ≈ 20.0000. (c) Using the estimate for f (4) obtained in part (b), the equation of the line tangent to f (x) = x 5/2 at x = 4 is y = f (4)(x − 4) + f (4) = 20(x − 4) + 32 = 20x − 48.

88

CHAPTER 3

D I F F E R E N T I AT I O N y 80 60 40 20 x 1

− 20 − 40 − 60

2

3

4

5

6

69. The graph in Figure 18 (basedx on data collected by the biologist Julian Huxley, 1887–1975) gives the average antler of fas(x) = 2 is shown 17.the slope of the tangent line to the graph at t = 4. For which weight W The of agraph red deer a function of ageint. Figure Estimate (a)ofReferring to the explain thetoinequality values t is the slope ofgraph, the tangent linewhy equal zero? For which values is it negative? f (h) − f (0) Antler 8 f (0) ≤ Weight 7 h (kg) 6 5 holds for h > 0 and the opposite inequality holds for h < 0. 4 (b) Use part (a) to show that 0.69314 ≤ f3 (0) ≤ 0.69315. 2 (c) Use the same procedure to determine 1 f (x) to four decimal places for x = 1, 2, 3, 4. 0 0 x 2= 1,4 2, 3, 6 4. Can 8 10 14 an (approximate) you 12 guess (d) Now compute the ratios f (x)/ f (0) for Age (years) general x?

formula for f (x) for

FIGURE 18 SOLUTION Let W (t) denote the antler weight as a function of age. The “tangent line” sketched in the figure below passes through the points (1, 1) and (6, 5.5). Therefore

W (4) ≈

5.5 − 1 = 0.9 kg/year. 6−1

If the slope of the tangent is zero, the tangent line is horizontal. This appears to happen at roughly t = 10 and at t = 11.6. The slope of the tangent line is negative when the height of the graph decreases as we move to the right. For the graph in Figure 18, this occurs for 10 < t < 11.6. y 8 7 6 5 4 3 2 1 0

0

2

4

6

8

10

12

x 14

In Exercises 70–72, i(t) is the current (in amperes) at time t (seconds) flowing in the circuit shown in Figure 19. According to Kirchhoff’s law, i(t) = Cv (t) + R −1 v(t), where v(t) is the voltage (in volts) at time t, C the capacitance (in farads), and R the resistance (in ohms, ). i + v

R

−

C

FIGURE 19

71. Use the following table to estimate v (10). For a better estimate, take the average of the difference quotients for h Calculate the current at t = 3 if v(t) = 0.5t + 4 V, C = 0.01 F, and R = 100 . and −h as described in Exercise 48. Then estimate i(10), assuming C = 0.03 and R = 1,000. t v(t) SOLUTION

9.8

9.9

10

10.1

10.2

256.52

257.32

258.11

258.9

259.69

We generate a table of difference quotients at 10 using the following table: h

−0.2

−0.1

0.1

0.2

v(10 + h) − v(10) h

7.95

7.9

7.9

7.9


S E C T I O N 3.1

89

The mean of the difference quotients is 7.9125, so v (10) ≈ 7.9125 volts/s. Thus, i(10) = 0.03(7.9125) +

1 (258.11) = 0.495485 amperes. 1000

Assume that Rand = 200 but C is unknown. Use the following data to estimate v (4) as in Exercise 71 and deduce Further Insights Challenges an approximate value for the capacitance C. In Exercises 73–76, we define the symmetric difference quotient (SDQ) at x = a for h = 0 by 3.8f (a + 3.9 h) − f (a4− h) 4.1 2h 420 436.2 388.8 404.2

t v(t)

73. Explain how SDQ can be interpreted slope of a secant34.1 line. 34.98 i(t)as the 32.34 33.22 SOLUTION The symmetric difference quotient

4.2

5

452.8 35.86

f (a + h) − f (a − h) 2h is the slope of the secant line connecting the points (a − h, f (a − h)) and (a + h, f (a + h)) on the graph of f ; the difference in the function values is divided by the difference in the x-values. 75. Showusually that if gives f (x) is a quadratic polynomial, then the SDQthan at xdoes = athe (for any h = 0) is equal to f (a). The SDQ a better approximation to the derivative ordinary difference quotient. Let x Explain of this the result. = 0. Compute SDQ with h = 0.001 and the ordinary difference quotients with h = ±0.001. f (x)the=graphical 2 and a meaning Compare with the actual value of f (0) that, to eight decimal places, is 0.69314718. SOLUTION Let f (x) = px 2 + qx + r be a quadratic polynomial. We compute the SDQ at x = a. f (a + h) − f (a − h) p(a + h)2 + q(a + h) + r − ( p(a − h)2 + q(a − h) + r ) = 2h 2h pa 2 + 2 pah + ph 2 + qa + qh + r − pa 2 + 2 pah − ph 2 − qa + qh − r 2h 4 pah + 2qh 2h(2 pa + q) = = = 2 pa + q 2h 2h =

Since this doesn’t depend on h, the limit, which is equal to f (a), is also 2 pa + q. Graphically, this result tells us that the secant line to a parabola passing through points chosen symmetrically about x = a is always parallel to the tangent line at x = a. 77. Which of the two functions in Figure 20 satisfies the inequality Let f (x) = x −2 . Compute f (1) by taking the limit of the SDQs (with a = 1) as h → 0. f (a + h) − f (a) f (a + h) − f (a − h) ≤ 2h h for h > 0? Explain in terms of secant lines. y

y

x

x a

a (A)

(B)

FIGURE 20 SOLUTION

Figure (A) satisfies the inequality f (a + h) − f (a − h) f (a + h) − f (a) ≤ 2h h

since in this graph the symmetric difference quotient has a larger negative slope than the ordinary right difference quotient. [In figure (B), the symmetric difference quotient has a larger positive slope than the ordinary right difference quotient and therefore does not satisfy the stated inequality.]

90

CHAPTER 3


3.2 The Derivative as a Function Preliminary Questions 1. What is the slope of the tangent line through the point (2, f (2)) if f is a function such that f (x) = x 3 ? SOLUTION

The slope of the tangent line through the point (2, f (2)) is given by f (2). Since f (x) = x 3 , it follows

that f (2) = 23 = 8.

2. Evaluate ( f − g) (1) and (3 f + 2g) (1) assuming that f (1) = 3 and g (1) = 5. Can we evaluate ( f g) (1) using the information given and the rules presented in this section? SOLUTION ( f − g) (1) = f (1) − g (1) = 3 − 5 = −2 and (3 f + 2g) (1) = 3 f (1) + 2g (1) = 3(3) + 2(5) = 19. Using the rules of this section, we cannot evaluate ( f g) (1).

3. Which of the following functions can be differentiated using the rules covered in this section? Explain. (b) f (x) = 2x (a) f (x) = x 2 1 (c) f (x) = √ (d) f (x) = x −4/5 x (e) f (x) = sin x (f) f (x) = (x + 5)3 SOLUTION

(a) Yes. x 2 is a power function, so the Power Rule can be applied. (b) No. 2x is an exponential function (the base is constant while the exponent is a variable), so the Power Rule does not apply. 1 (c) Yes. √ can be rewritten as x −1/2 . This is a power function, so the Power Rule can be applied. x −4/5 (d) Yes. x is a power function, so the Power Rule can be applied. (e) No. The Power Rule does not apply to the trigonometric function sin x. (f) Yes. (x + 5)3 can be expanded to x 3 + 15x 2 + 75x + 125. The Power Rule, Rule for Sums, and Rule for Constant Multiples can be applied to this function. 4. Which algebraic identity is used to prove the Power Rule for positive integer exponents? Explain how it is used. SOLUTION

The algebraic identity x n − a n = (x − a)(x n−1 + x n−2 a + x n−3 a 2 + · · · + xa n−2 + a n−1 )

is used to prove the Power Rule for positive integer exponents. With this identity, the difference quotient x n − an x −a can be simplified to x n−1 + x n−2 a + x n−3 a 2 + · · · + xa n−2 + a n−1 . This expression is continuous at x = a, so the limit as x → a can be evaluated by substitution. √ 5. Does the Power Rule apply to f (x) = 5 x? Explain. √ SOLUTION Yes. 5 x = x 1/5 , which is a power function. 6. In which of the following two cases does the derivative not exist? (a) Horizontal tangent (b) Vertical tangent SOLUTION

The derivative does not exist when there is a vertical tangent. At a horizontal tangent, the derivative is zero.

Exercises In Exercises 1–8, compute f (x) using the limit definition. 1. f (x) = 4x − 3 SOLUTION

Let f (x) = 4x − 3. Then, f (x) = lim

h→0

2 3. f (x) = 1 − 2x f (x) = x 2 + x

f (x + h) − f (x) 4(x + h) − 3 − (4x − 3) 4h = lim = lim = 4. h h h→0 h→0 h

The Derivative as a Function


91

Let f (x) = 1 − 2x 2 . Then, f (x + h) − f (x) 1 − 2(x + h)2 − (1 − 2x 2 ) = lim h h h→0 h→0

f (x) = lim

−4xh − 2h 2 = lim (−4x − 2h) = −4x. h h→0 h→0

= lim 5. f (x) = x −1 3 f (x) = x SOLUTION Let f (x) = x −1 . Then,

x−x−h 1 − 1 f (x + h) − f (x) −1 1 x(x+h) x = lim x+h = lim = lim = − 2. h h h h→0 h→0 h→0 h→0 x(x + h) x

f (x) = lim

√ 7. f (x) = x x √ f (x) = SOLUTION Letx − f (x) 1 = x. Then, f (x + h) − f (x) = lim h h→0 h→0

f (x) = lim = lim

√

x +h− h

√

x

√ = lim

h→0

x +h− h

√

x

√ ·

√

√ x √ x +h+ x x +h+

1 (x + h) − x 1 √ √ = lim √ √ = √ . 2 x x + h + x) h→0 x + h + x

h→0 h(

In Exercises 9–16,−1/2 use the Power Rule to compute the derivative. f (x) = x d 4 9. x d x x=−2 d 4 d 4 3 x = 4x so x SOLUTION = 4(−2)3 = −32. dx d x x=−2 d 2/3 11. t d −4 dt x d x t=8x=3 d 2/3 2 1 d 2/3 2 −1/3 t SOLUTION so = (8)−1/3 = . t = t dt 3 dt 3 3 t=8 d 0.35 xd −2/3 dx t

dt d t=1 x 0.35 = 0.35(x 0.35−1 ) = 0.35x −0.65 . SOLUTION dx d √17 15. t d 14/3 dt x dx d √17 √ √17−1 SOLUTION = 17t t dt 13.

In Exercises compute f (a) and find an equation of the tangent line to the graph at x = a. 2 d −17–20, t π dt= x 5 , a = 1 17. f (x) Let f (x) = x 5 . Then, by the Power Rule, f (x) = 5x 4 . The equation to the tangent line to the graph of f (x) at x = 1 is

SOLUTION

y = f (1)(x − 1) + f (1) = 5(x − 1) + 1 = 5x − 4. √ 19. f (x) = 3 x + 8x, a = 9 f (x) = x −2 , a = 3 3 17 SOLUTION Let f (x) = 3x 1/2 + 8x. Then f (x) = 2 x −1/2 + 8. In particular, f (9) = 2 . The tangent line at x = 9 is 17 17 9 y = f (9)(x − 9) + f (9) = (x − 9) + 81 = x+ . 2 2 2 √ In Exercises a = 8 the derivative of the function. f (x)21–32, = 3 x,calculate 21. f (x) = x 3 + x 2 − 12

92

CHAPTER 3


SOLUTION

d 3 x + x 2 − 12 = 3x 2 + 2x. dx

23. f (x) = 2x 3 −310x −12 f (x) = 2x − 3x + 2x d 2x 3 − 10x −1 = 6x 2 + 10x −2 . SOLUTION dx 25. g(z) = 7z −3 5+ z 2 +2 5 f (x) = x − 7x + 10x + 9 d SOLUTION 7z −3 + z 2 + 5 = −21z −4 + 2z. dz √ √ 3 27. f (s) = 4 s + √ s 1 √ +√ h(t) =f6(s)t = 4√ SOLUTION st + 3 s = s 1/4 + s 1/3 . In this form, we can apply the Sum and Power Rules. 1 1 1 1 d 1/4 + s 1/3 = (s (1/4)−1 ) + (s (1/3)−1 ) = s −3/4 + s −2/3 . s ds 4 3 4 3 29. f (x) = (x + 1)43 (Hint: Expand) + 7y 2/3 W (y) = 6y d 3 d

SOLUTION (x + 1)3 = x + 3x 2 + 3x + 1 = 3x 2 + 6x + 3. dx dx 31. P(z) = (3z − 1)(2z + 1) R(s) = (5s + 1)2 d d 2 SOLUTION 6z + z − 1 = 12z + 1. ((3z − 1)(2z + 1)) = dz dz √ calculate the derivative indicated. In Exercises 33–38, q(t) = t(t + 1) 3 33. f (2), f (x) = 4 x SOLUTION

With f (x) = 3x −4 , we have f (x) = −12x −5 , so 3 f (2) = −12(2)−5 = − . 8

d T √ , T = 3Cx2/3 +1 dCy C=8 (16), y = x dT SOLUTION With T (C) = 3C 2/3 , we have = 2C −1/3 . Therefore, dC d T = 2(8)−1/3 = 1. dC

35.

C=8

ds d P , s = 4z − 16z72 dz z=2 , P= d V V =−2 V ds SOLUTION With s = 4z − 16z 2 , we have = 4 − 32z. Therefore, dz ds = 4 − 32(2) = −60. dz z=2

37.

39. Match the functions in graphs (A)–(D) with their derivatives (I)–(III) in Figure 11. Note that two of the functions R derivative. Explain have thedsame why. , R = Wπ d W W =1 y

y

y

y x

x

x x (A)

(B)

(C)

y

y

(D)

y x

x (I)

x (II)

(III)

FIGURE 11 SOLUTION

S E C T I O N 3.2


93

• Consider the graph in (A). On the left side of the graph, the slope of the tangent line is positive but on the right

side the slope of the tangent line is negative. Thus the derivative should transition from positive to negative with increasing x. This matches the graph in (III). • Consider the graph in (B). This is a linear function, so its slope is constant. Thus the derivative is constant, which matches the graph in (I). • Consider the graph in (C). Moving from left to right, the slope of the tangent line transitions from positive to negative then back to positive. The derivative should therefore be negative in the middle and positive to either side. This matches the graph in (II). • Consider the graph in (D). On the left side of the graph, the slope of the tangent line is positive but on the right side the slope of the tangent line is negative. Thus the derivative should transition from positive to negative with increasing x. This matches the graph in (III). Note that the functions whose graphs are shown in (A) and (D) have the same derivative. This happens because the graph in (D) is just a vertical translation of the graph in (A), which means the two functions differ by a constant. The derivative of a constant is zero, so the two functions end up with the same derivative. 41. Sketch the graph of f (x) for f (x) as in Figure 13, omitting points where f (x) is not differentiable. Assign the labels f (x), g(x), and h(x) to the graphs in Figure 12 in such a way that f (x) = g(x) and g (x) = h(x). y 5 4 3 2 1

x

1 2 3 4 5 6 7 8 9

FIGURE 13 Graph of f (x).

On the interval 0 ≤ x ≤ 3, the function is constant, so f (x) = 0. On the interval 3 ≤ x ≤ 5, the function is linear with slope 12 , so f (x) = 12 . From x = 5 to x = 7, the derivative should be positive but decreasing to zero at x = 7. Finally, for 7 < x ≤ 9, the derivative is negative and decreasing. A sketch of f (x) is given below. SOLUTION

y

2 1 x

−1 −2

1 2 3 4 5 6 7 8 9

43. Let R be a variable and r a constant. Compute the derivatives f (x). Does f (x) appear to increased or decrease as a function of x? d The table below lists values of a function d (a) Is (A) R or (B) in Figure 14 the graph of (b)f (x)?rExplain. (c) r 2 R3 dR dR dR SOLUTION

x 0 0.5 1 1.5 2 d (a) R = 1, since R is a linear of R with f (x) function 10 55 98 slope 139 1. 177 dR d r = 0, since r is a constant. (b) dR (c) We apply the Linearity and Power Rules:

2.5

3

3.5

4

210

237

257

268

d 3 d 2 3 r R = r2 R = r 2 3(R 2 ) = 3r 2 R 2 . dR dR 45. Find the points on the curve y = x 2 + 23x − 7 at which the slope of the tangent line is equal to 4. Sketch the graph of f (x) = x − 3x and find the values of x for which the tangent line is horizontal. 1 SOLUTION Let y = x 2 + 3x − 7. Solving d y/d x = 2x + 3 = 4 yields x = 2 . 47. Find all values of x where the tangent lines to y = x 3 and y = x 4 are parallel. Sketch the graphs of f (x) = x 2 − 5x + 4 and g(x) = −2x + 3. Find the value of x at which the graphs have SOLUTION Let f (x) = x 3 and let g(x) = x 4 . The two graphs have parallel tangent lines at all x where f (x) = g (x). parallel tangent lines. f (x) = g (x) 3x 2 = 4x 3 3x 2 − 4x 3 = 0 x 2 (3 − 4x) = 0 hence, x = 0 or x = 34 . Show that there is a unique point on the graph of the function f (x) = ax 2 + bx + c where the tangent line is horizontal (assume a = 0). Explain graphically.

94

CHAPTER 3


49. Determine coefficients a and b such that p(x) = x 2 + ax + b satisfies p(1) = 0 and p (1) = 4. SOLUTION Let p(x) = x 2 + ax + b satisfy p(1) = 0 and p (1) = 4. Now, p (x) = 2x + a. Therefore 0 = p(1) = 1 + a + b and 4 = p (1) = 2 + a; i.e., a = 2 and b = −3.

m > −3, there are precisely two points 51. Let f (x) = x 3 − 3x + 1. Show that f (x) ≥ −3 for all x, and that for every + 11x + 2 isfor steeper than of themtangent line values of x the such that theoftangent line toand thethe graph of y = 4x 2tangent (x) all where fFind = m. Indicate position these points corresponding lines one value in a sketch 3 to y = x . of the graph of f (x). SOLUTION

Let P = (a, b) be a point on the graph of f (x) = x 3 − 3x + 1.

• The derivative satisfies f (x) = 3x 2 − 3 ≥ −3 since 3x 2 is nonnegative. • Suppose the slope m of the tangent line is greater than −3. Then f (a) = 3a 2 − 3 = m, whence

m+3 a2 = >0 3

and thus

a=±

m+3 . 3

• The two parallel tangent lines with slope 2 are shown with the graph of f (x) here. y

4 2 −2

1 −1

x

2

−2

Hint: Show that 53. Compute the derivative of f (x) = x −2 using the limit1 definition. Show that if the tangent lines to the graph of y = 3 x 3 − x 2 at x = a and at x = b are parallel, then either a = b or a + b = 2. (x + h)2 − x 2 1 f (x + h) − f (x) =− 2 h h x (x + h)2 SOLUTION

1

1

x 2 −(x+h)2

− 2 f (x + h) − f (x) 1 (x+h)2 x x 2 (x+h)2 = = = h h h h

x 2 − (x + h)2 1

1 x 2 (x + h)2

1 (x + h)2 − x 2 =− 2 . h x (x + h)2 We now compute the limits as h → 0 of each factor. By continuity, we see that, if x = 0, lim

−1

h→0 x 2 (x + h)2

1 = − 4. x

The second factor can be recognized easily as the limit definition of the derivative of x 2 , which is 2x. Applying the Product Rule, we get that: 1 2 d −2 = − 4 (2x) = − 3 . x dx x x √ 55. The average speed (in meters per second) of a gas molecule is vavg = 8RT /(π M), where T is the temperature (in 3/2 limit definition. the derivative of f (x)and = xR =using kelvin),Compute M is the molar mass (kg/mol) 8.31. the Calculate dvavg /d THint: at T Show = 300that K for oxygen, which has a molar mass of 0.032 kg/mol. f (x + h) − f (x) (x + √ h)3 − x 3 1 1/2 = 1/2 , where M √ and R are constant, we use the SOLUTION Using the form vav = (8RT /(π M))= h h8R/(π M)T (x + h)3 + x 3 Power Rule to compute the derivative dvav /d T . d d 1/2 1 = 8R/(π M) (T (1/2)−1 ). 8R/(π M)T 1/2 = 8R/(π M) T dT dT 2 In particular, if T = 300◦ K, 1 d vav = 8(8.31)/(π (0.032)) (300)−1/2 = 0.74234 m/(s · K). dT 2 57. Some studies suggest that kidney mass K in mammals (in kilograms) is related to body mass m (in kilograms) by Biologists have observed that the pulse rate P (in beats per minute) in animals is related to body mass (in the approximate formula K = 0.007m 0.85 . Calculate d K /dm at m = 68. Then calculate the derivative with respect to kilograms) by the approximate formula P = 200m −1/4 . This is one of many allometric scaling laws prevalent in m of the relative kidney-to-mass ratio K /m at m = 68. biology. Is the absolute value |d P/dm| increasing or decreasing as m increases? Find an equation of the tangent line at the points on the graph in Figure 15 that represent goat (m = 33) and man (m = 68).

S E C T I O N 3.2


95

SOLUTION

dK = 0.007(0.85)m −0.15 = 0.00595m −0.15 ; dm hence, d K = 0.00595(68)−0.15 = 0.00315966. dm m=68 Because K m 0.85 = 0.007 = 0.007m −0.15 , m m we find d dm

K m

and d dm

= 0.007

d −0.15 m = −0.00105m −1.15 , dm

K = −8.19981 × 10−6 kg−1 . m m=68

59. Let L be a tangent line to the hyperbola x y = 1 at x = a, where a > 0. Show that the area of the triangle bounded The relation between the vapor pressure P (in atmospheres) of water and the temperature T (in kelvin) is given by L and the coordinate axes does not depend on a. by the Clausius–Clapeyron law: 1 1 SOLUTION Let f (x) = x −1 . The tangent line to f at x = a is y = f (a)(x − a) + f (a) = − 2 (x − a) + a . The a P dP 2 = k 2y = 0) is 2a. Hence the area of the triangle bounded y-intercept of this line (where x = 0) is a . Its x-intercept (where

dT T 1 1 by the tangent line and the coordinate axes is A = 2 bh = 2 (2a) a2 = 2, which is independent of a. where k is a constant. Use the table below and the approximation y dP P(T + 10) − P(T − 10) ≈ dT 20 1 y=

x

to estimate d P/d T for T = 303, 313, 323, 333, 343. Do your estimates seem to confirm the Clausius–Clapeyron 2 = (0, a )are the units of k? law? What is the approximate value of k?P What

( 1)

R = a, a

T

293

303

313

323

333

343

353

x

P 0.0278 0.0482 0.0808 0.1311 Q = (2a, 0)0.2067 0.3173 0.4754 61. Match the functions (A)–(C) with their derivatives (I)–(III) in Figure 16. In the notation of Exercise 59, show that the point of tangency is the midpoint of the segment of L lying in the first quadrant. y y

x

x

(A)

(I) y

y

x x

(B)

(II)

y

y

x x (C)

(III)

FIGURE 16

96

CHAPTER 3

D I F F E R E N T I AT I O N SOLUTION Note that the graph in (A) has three locations with a horizontal tangent line. The derivative must therefore cross the x-axis in three locations, which matches (III). The graph in (B) has only one location with a horizontal tangent line, so its derivative should cross the x-axis only once. Thus, (I) is the graph corresponding to the derivative of (B). Finally, the graph in (B) has two locations with a horizontal tangent line, so its derivative should cross the x-axis twice. Thus, (II) is the graph corresponding to the derivative of (C).

63. Make a rough sketch of the graph of the derivative of the function shown in Figure 17(A). Plot the derivative f (x) of f (x) = 2x 3 − 10x −1 for x > 0 (set the bounds of the viewing box appropriately) SOLUTION Thethat graph has>a 0. tangent negative slope onthe thegraph interval 3.6), andPlot has af (x) tangent and observe f (x) What line doeswith the positivity of f approximately (x) tell us about of (1, f (x) itself? and line with a positive slope elsewhere. This implies that the derivative must be negative on the interval (1, 3.6) and positive confirm. elsewhere. The graph may therefore look like this: y

x 1

65.

2

3

4

thederivative two functions and g inshown Figurein18, which is theomitting derivative of thewhere other? your is answer. GraphOfthe of the ffunction Figure 17(B), points theJustify derivative not defined. y f(x)

2

g(x) x

−1

1

FIGURE 18 SOLUTION g(x) is the derivative of f (x). For f (x) the slope is negative for negative values of x until x = 0, where there is a horizontal tangent, and then the slope is positive for positive values of x. Notice that g(x) is negative for negative values of x, goes through the origin at x = 0, and then is positive for positive values of x.

In Exercises 67–72, find is thethe points c (if in any) such19 that f (c) does not At which points function Figure discontinuous? Atexist. which points is it nondifferentiable? 67. f (x) = |x − 1| SOLUTION y 2 1.5 1 0.5 x

−1

1

2

3

Here is the graph of f (x) = |x − 1|. Its derivative does not exist at x = 1. At that value of x there is a sharp corner. 2/3 69. f (x)f (x) = x= [x]

SOLUTION Here is the graph of f (x) = x 2/3 . Its derivative does not exist at x = 0. At that value of x, there is a sharp corner or “cusp”. y 1.5 1

−2

−1

x 1

2

71. f (x) = |x 2 −3/2 1| f (x) = x SOLUTION Here is the graph of f (x) = x 2 − 1. Its derivative does not exist at x = −1 or at x = 1. At these values of x, the graph has sharp corners.

S E C T I O N 3.2


97

y 3 2 1

−2

x

−1

1

2

f (x) = |x −73–78, 1|2 zoom in on a plot of f (x) at the point (a, f (a)) and state whether or not f (x) appears to be In Exercises differentiable at x = a. If nondifferentiable, state whether the tangent line appears to be vertical or does not exist. 73. f (x) = (x − 1)|x|,

a=0

SOLUTION The graph of f (x) = (x − 1)|x| for x near 0 is shown below. Because the graph has a sharp corner at x = 0, it appears that f is not differentiable at x = 0. Moreover, the tangent line does not exist at this point. y x

− 0.2 −0.1

0.1

0.2

−0.1 −0.2 −0.3

75. f (x) = (x − 3)1/3 ,5/3a = 3 f (x) = (x − 3) , a = 3 SOLUTION The graph of f (x) = (x − 3)1/3 for x near 3 is shown below. From this graph, it appears that f is not differentiable at x = 3. Moreover, the tangent line appears to be vertical.

2.9

2.95

3

3.05

3.1

77. f (x) = | sin x|, a = 0 f (x) = sin(x 1/3 ), a = 0 SOLUTION The graph of f (x) = | sin x| for x near 0 is shown below. Because the graph has a sharp corner at x = 0, it appears that f is not differentiable at x = 0. Moreover, the tangent line does not exist at this point. y 0.1 0.08 0.04

−0.1 −0.05

79.

x 0.05

0.1

thatx|, a nondifferentiable function is not continuous? If not, give a counterexample. f (x)Is=it|xtrue − sin a=0

SOLUTION

This statement is false. Indeed, the function |x| is not differentiable at x = 0 but it is continuous.

Sketch the graph of y = x |x| and show that it is differentiable for all x Further Insights and Challenges x < 0, x = 0, and x > 0).

(check differentiability separately for

81. Prove the following theorem of Apollonius of Perga (the Greek mathematician born in 262 BCE who gave the parabola, ellipse, and hyperbola their names): The tangent to the parabola y = x 2 at x = a intersects the x-axis at the midpoint between the origin and (a, 0). Draw a diagram. SOLUTION

Let f (x) = x 2 . The tangent line to f at x = a is y = f (a)(x − a) + f (a) = 2a(x − a) + a 2 = 2ax − a 2 .

The x-intercept of this line (where y = 0) is a2 , which is halfway between the origin and the point (a, 0).

98

CHAPTER 3

D I F F E R E N T I AT I O N y

(a, a 2) y=

x2

(–,a2 0)

x

83. Plot the graph of f (x) = (4 − x 2/3 )3/2 (the “astroid”). Let L be a tangent line to a point on the graph in 3 Apollonius’s in Exercise canfirst be generalized. that the tangent the first quadrant. ShowTheorem that the portion of L 81 in the quadrant hasShow a constant length 8. to y = x at x = a intersects the x-axis at the point x = 23 a. Then formulate the general statement for the graph of y = x n and prove it. SOLUTION

• Here is a graph of the astroid. y 10

x

−10

10

−10

• Let f (x) = (4 − x 2/3 )3/2 . Since we have not yet encountered the Chain Rule, we use Maple throughout this

exercise. The tangent line to f at x = a is

4 − a 2/3 2/3 3/2 . y=− (x − a) + 4 − a a 1/3

The y-intercept of this line is the point P = 0, 4 4 − a 2/3 , its x-intercept is the point Q = 4a 1/3 , 0 , and the distance between P and Q is 8.

85. A vase is formed by rotating y = x 2 around the y-axis. If we drop in a marble, it will either the bottom point 2 fortouch −1marble ≤ x ≤be1 to and the Twoorsmall arches have thethe shape of parabolas. Thethefirst is given by21). f (x) = small 1 − xmust of the vase be suspended above bottom by touching sides (Figure How the touch 2 for 2 ≤ x ≤ 6. A board is placed on top of these arches so it rests on both (Figure second by g(x) = 4 − (x − 4) the bottom? 20). What is the slope of the board? Hint: Find the tangent line to y = f (x) that intersects y = g(x) in exactly one point. FIGURE 20

FIGURE 21

Suppose a circle is tangent to the parabola y = x 2 at the point (t, t 2 ). The slope of the parabola at this 1 (since it is perpendicular to the tangent line of the point is 2t, so the slope of the radius of the circle at this point is − 2t 1 (x − t) + t 2 crosses the y-axis. We can circle). Thus the center of the circle must be where the line given by y = − 2t 1 2 find the y-coordinate by setting x = 0: we get y = 2 + t . Thus, the radius extends from (0, 12 + t 2 ) to (t, t 2 ) and SOLUTION

r=

2 1 1 2 2 2 +t = +t −t + t 2. 2 4

This radius is greater than 12 whenever t > 0; so, if a marble has radius > 1/2 it sits on the edge of the vase, but if it has radius ≤ 1/2 it rolls all the way to the bottom. 87. Negative Exponents Let n be a whole number. Use the Power Rule for x n to calculate the derivative of f (x) = Let f (x) thatbe a differentiable function and set g(x) = f (x + c), where c is a constant. Use the limit definition to x −n by showing show that g (x) = f (x + c). Explain this result graphically, recalling that the graph of g(x) is obtained by shifting (x + − 0)f (x) the graph of f (x) c units to the fleft (ifh) c> or right (if −1 c < 0). (x + h)n − x n = n h x (x + h)n h

Product and Quotient Rules


99

Let f (x) = x −n where n is a positive integer.

• The difference quotient for f is 1

1

x n −(x+h)n

(x + h)−n − x −n f (x + h) − f (x) (x+h)n − x n x n (x+h)n = = = h h h h (x + h)n − x n −1 . = n x (x + h)n h • Therefore,

f (x + h) − f (x) −1 (x + h)n − x n = lim n n h h h→0 h→0 x (x + h)

f (x) = lim

= lim

−1

lim

h→0 x n (x + h)n h→0

(x + h)n − x n d n = −x −2n x . h dx

d n x = −x −2n · nx n−1 = −nx −n−1 . Since n is a positive integer, dx d −n d k d xk = x x = kx k−1 = −nx −n−1 = kx k−1 ; i.e. k = −n is a negative integer and we have dx dx dx for negative integers k.

• From above, we continue: f (x) = −x −2n

89. Infinitely Rapid Oscillations Define Verify the Power Rule for the exponent 1/n, where n is a positive integer, using the following trick: Rewrite the difference quotient for y = x 1/n at x = b in terms⎧ of u =1 (b + h)1/n and a = b1/n . ⎪ ⎨x sin if x = 0 x f (x) = ⎪ ⎩0 if x = 0 Show that f (x) is continuous at x = 0 but f (0) does not exist (see Figure 10).

if x = 0 x sin 1x SOLUTION Let f (x) = . As x → 0, 0 if x = 0 1 1 | f (x) − f (0)| = x sin − 0 = |x| sin →0 x x since the values of the sine lie between −1 and 1. Hence, by the Squeeze Theorem, lim f (x) = f (0) and thus f is continuous at x = 0. As x → 0, the difference quotient at x = 0,

x→0

x sin 1x − 0 1 f (x) − f (0) = = sin x −0 x −0 x does not converge to a limit since it oscillates infinitely through every value between −1 and 1. Accordingly, f (0) does not exist.

3.3 Product and Quotient Rules Preliminary Questions 1. (a) (b) (c)

Are the following statements true or false? If false, state the correct version. The notation f g denotes the function whose value at x is f (g(x)). The notation f /g denotes the function whose value at x is f (x)/g(x). The derivative of the product is the product of the derivatives.

SOLUTION

(a) False. The notation f g denotes the function whose value at x is f (x)g(x). (b) True. (c) False. The derivative of the product f g is f (x)g(x) + f (x)g (x). 2. Are the following equations true or false? If false, state the correct version. d (a) = f (4)g (4) − g(4) f (4) ( f g) dx x=4

100

CHAPTER 3


d f f (4)g (4) + g(4) f (4) = d x g x=4 (g(4))2 d (c) = f (0)g (0) + g(0) f (0) ( f g) dx x=0

(b)

SOLUTION

d = f (4)g (4) + g(4) f (4). ( f g) dx x=4 d = [g(4) f (4) − f (4)g (4)]/g(4)2 . (b) False. ( f /g) dx x=4 (c) True. (a) False.

3. What is the derivative of f /g at x = 1 if f (1) = f (1) = g(1) = 2, and g (1) = 4? d SOLUTION ( f /g)x=1 = [g(1) f (1) − f (1)g (1)]/g(1)2 = [2(2) − 2(4)]/22 = −1. dx 4. Suppose that f (1) = 0 and f (1) = 2. Find g(1), assuming that ( f g) (1) = 10. ( f g) (1) = f (1)g (1) + f (1)g(1), so 10 = 0 · g (1) + 2g(1) and g(1) = 5.

SOLUTION

Exercises In Exercises 1–4, use the Product Rule to calculate the derivative. 1. f (x) = x(x 2 + 1) Let f (x) = x(x 2 + 1). Then

SOLUTION

f (x) = x

3.

d 2 d (x + 1) + (x 2 + 1) x = x(2x) + (x 2 + 1) = 3x 2 + 1. dx dx

d y √ (t 2 + 1)(t (x), = y =x(1 − x 4 ) + 9) dt ft=3 Let y = (t 2 + 1)(t + 9). Then

SOLUTION

d d dy = (t 2 + 1) (t + 9) + (t + 9) (t 2 + 1) = (t 2 + 1) + (t + 9)(2t) = 3t 2 + 18t + 1. dt dt dt Therefore, d y = 3(3)2 + 18(3) + 1 = 82. dt t=3 In Exercises Rule to calculate the derivative. dh 5–8, use the Quotient −1/2 + 2x)(7 − x −1 ) , h(x) = (x dx x 5. f (x) =x=4 x −2 x . Then Let f (x) = x−2

SOLUTION

f (x) =

(x − 2) ddx x − x ddx (x − 2) (x − 2) − x −2 = = . (x − 2)2 (x − 2)2 (x − 2)2

dg t2 + 1 x+ , g(t) =4 2 (x) = dt ft=−2 x 2 + x +t 1− 1 t2 + 1 SOLUTION Let g(t) = . Then t2 − 1 7.

d (t 2 + 1) − (t 2 + 1) d (t 2 − 1) (t 2 − 1) dt (t 2 − 1)(2t) − (t 2 + 1)(2t) 4t dg dt = =− 2 . = dt (t 2 − 1)2 (t 2 − 1)2 (t − 1)2

Therefore, 4(−2) 8 dg =− = . dt t=−2 9 ((−2)2 − 1)2 dw

, w= √

z2

S E C T I O N 3.3


101

In Exercises 9–12, calculate the derivative in two ways: First use the Product or Quotient Rule, then rewrite the function algebraically and apply the Power Rule directly. 9. f (t) = (2t + 1)(t 2 − 2) SOLUTION

Let f (t) = (2t + 1)(t 2 − 2). Then, using the Product Rule, f (t) = (2t + 1)(2t) + (t 2 − 2)(2) = 6t 2 + 2t − 4.

Multiplying out first, we find f (t) = 2t 3 + t 2 − 4t − 2. Therefore, f (t) = 6t 2 + 2t − 4. x 3 + 2x 2 + 3x −1 11. g(x)f (x) = = x 2 (3 + x −1 ) x SOLUTION

3 2 −1 Let g(x) = x +2x x+3x . Using the quotient rule and the sum and power rules, and simplifying

g (x) =

x(3x 2 + 4x − 3x −2 ) − (x 3 + 2x 2 + 3x −1 )1 1 3 2 − 6x −1 = 2x + 2 − 6x −3 . 2x = + 2x x2 x2

Simplifying first yields g(x) = x 2 + 2x + 3x −2 , from which we calculate g (x) = 2x + 2 − 6x −3 . In Exercises 13–32, calculate the derivative using the appropriate rule or combination of rules. t2 − 1 h(t) = 4 1 2 + x + 1) 13. f (x) = (x t−−4)(x

SOLUTION Let f (x) = x 4 − 4 x 2 + x + 1 . Then f (x) = (x 4 − 4)(2x + 1) + (x 2 + x + 1)(4x 3 ) = 6x 5 + 5x 4 + 4x 3 − 8x − 4. 15.

d y 1 2 + 9)(x + x −1 ) (x), = y(x= d x fx=2 x +4

SOLUTION

1 . Using the quotient rule: Let y = x+4

1 (x + 4)(0) − 1(1) dy =− . = dx (x + 4)2 (x + 4)2 Therefore, d y 1 1 =− =− . d x x=2 36 (2 + 4)2 √ √ 17. f (x) = ( x + 1)( x − 1) x dz z == 2(√x + 1)(√x − 1). Multiplying through first yields f (x) = x − 1 for x ≥ 0. Therefore, SOLUTION Let , f (x) d x x=−1 x +1 f (x) = 1 for x ≥ 0. If we carry out the product rule on f (x) = (x 1/2 + 1)(x 1/2 − 1), we get 1 −1/2 1 −1/2 1 1 1 1 f (x) = (x 1/2 + 1) ) + (x 1/2 − 1) (x x = + x −1/2 + − x −1/2 = 1. 2 2 2 2 2 2 √ x4 − 4 d y 2 , y3 =x − (x) = d x fx=2 x2 − 5 x x4 − 4 SOLUTION Let y = . Then x2 − 5

x 2 − 5 4x 3 − x 4 − 4 (2x) 2x 5 − 20x 3 + 8x dy = . = 2 2 dx x2 − 5 x2 − 5

19.

Therefore, 2(2)5 − 20(2)3 + 8(2) d y = = −80. d x x=2 (22 − 5)2 21.

dz 1 4 + 2x + 1 , zx = d x fx=1 x3 + 1 (x) = x +1

102

CHAPTER 3

D I F F E R E N T I AT I O N SOLUTION

Let z =

1 . Using the quotient rule: x 3 +1

3x 2 (x 3 + 1)(0) − 1(3x 2 ) dz = − . = dx (x 3 + 1)2 (x 3 + 1)2 Therefore, 3(1)2 3 dz = − =− . 3 2 d x x=1 4 (1 + 1) t 23. h(t) = 4 3x 32 − 7x 2 + 2 (t + t )(t f (x) = √ + 1) x t t = SOLUTION Let h(t) = . Then t4 + t2 t7 + 1 t 11 + t 9 + t 4 + t 2

t 11 + t 9 + t 4 + t 2 (1) − t 11t 10 + 9t 8 + 4t 3 + 2t 10t 11 + 8t 9 + 3t 4 + t 2 =− h (t) = 2 2 . t 11 + t 9 + t 4 + t 2 t 11 + t 9 + t 4 + t 2 4 25. f (t) = 31/2 ·3/2 51/2 − 3x 2x √ f (x) = x √ + x −1/2 SOLUTION Let f (t) = 3 5. Then f (t) = 0, since f (t) is a constant function! 27. f (x) = (x + 3)(x − 1)(x − 5) h(x) = π 2 (x − 1) SOLUTION Let f (x) = (x + 3)(x − 1)(x − 5). Using the Product Rule inside the Product Rule with a first factor of (x + 3) and a second factor of (x − 1)(x − 5), we find f (x) = (x + 3) ((x − 1)(1) + (x − 5)(1)) + (x − 1)(x − 5)(1) = 3x 2 − 6x − 13. Alternatively,

f (x) = (x + 3) x 2 − 6x + 5 = x 3 − 3x 2 − 13x + 15. Therefore, f (x) = 3x 2 − 6x − 13. z2 − 1 z2 − 4 (x) = x(x 2 + 1)(x + 4) 29. g(z)f = z−1 z+2 SOLUTION

Hint: Simplify first.

Let g(z) =

z2 − 4 z−1

z2 − 1 z+2

=

(z + 2)(z − 2) (z + 1)(z − 1) z−1 z+2

= (z − 2)(z + 1)

for z = −2 and z = 1. Then, g (z) = (z + 1)(1) + (z − 2)(1) = 2z − 1. d xt − 4 d (x constant) 2 (ax (a, b constants) dt t 2 − x + b)(abx + 1) dx xt−4 SOLUTION Let f (t) = 2 . Using the quotient rule:

31.

t −x

f (t) =

(t 2 − x)(x) − (xt − 4)(2t) xt 2 − x 2 − 2xt 2 + 8t −xt 2 + 8t − x 2 = = . (t 2 − x)2 (t 2 − x)2 (t 2 − x)2

x the derivative over [−4, 4]. Use the graph to determine the intervals on which f (x) > of f (x) = 2 d Plot ax + b x +1 (a, b, c, d constants) x 0 for −1 < x < 1 and f (x) < 0 for |x| > 1. The original function is plotted in the figure below at the right. Observe that the graph of f (x) is increasing whenever f (x) > 0 and that f (x) is decreasing whenever f (x) < 0.


S E C T I O N 3.3 y

y 1 0.8 0.6 0.4 0.2

0.4 0.2

1

2

3

x

−4 −3 −2 −1 − 0.2

x −4 −3 −2 −1

103

1

2

3

4

−0.4

4

V2R 5. Calculate dP/dr, assuming that rthe is plot variable and R is whether constant. f (x) is positive 35. Let P Plot = f (x) =2 as xin Example (in a suitably bounded viewing box). Use to determine (R + r ) x 2 − 1 or negative on its {x :constant. x = ±1}.Let Then compute f (x) and confirm your conclusion algebraically. SOLUTION Note thatdomain V is also f (r ) =

V2R V2R = 2 . 2 (R + r ) R + 2Rr + r 2

Using the quotient rule: f (r ) =

(R 2 + 2Rr + r 2 )(0) − (V 2 R)(2R + 2r ) 2V 2 R(R + r ) 2V 2 R = − = − . (R + r )4 (R + r )4 (R + r )3

x at volts), x = 2. and resistance R (in ohms) in a circuit are 37. FindAccording an equation of the tangent line to Ithe y= to Ohm’s Law, current (ingraph amps), voltage V (in x + x −1 related by I = V /R. x . The equation of the tangent line to the graph of f (x) at x = 2 is y = f (2)(x − 2) + SOLUTION Let f (x) = d I x+x −1 , assuming that V has the constant value V = 24. Calculate f (2).(a) Using the quotient rule, we compute: d R R=6

dV 2 (x+ x −1 )(1) − (x)(1 − x −2 ) 1 −1 − x + x −1 = (b) Calculate , assuming that I has the=constant value Ix = 4. + x . f (x) = d R R=6 (x + x −1 )2 (x + x −1 )2 x(x + x −1 )2 2 = 4 . This makes the equation of the tangent line Therefore, f (2) = 25 25 2

y=

2 12 4 4 (x − 2) + 5 = x+ . 25 25 5 2

39. Let f (x) = g(x) = x. Show that ( f /g) = f /g . 1 = 0.ofOn The curve Agnesi (Figure 3) (after Maria Agnesi SOLUTION ( f /g)y = (x/x) =is 1,called so ( fthe /g)witch the other hand, f /g the ) =Italian (x /x mathematician ) = (1/1) = 1. We see that x2 + 1 0 = 1. (1718–1799) 3who wrote one of the first books on calculus. This strange name is the result of a mistranslation of the = 3 f 2 f . 41. Show that ( fla)versiera, Italian meaning “that( which Useword the Product Rule to show that f 2 ) =turns.” 2 f f . Find equations of the tangent lines at x = ±1. SOLUTION Let g = f 3 = f f f . Then

g = f 3 = [ f ( f f )] = f f f + f f + f f ( f ) = 3 f 2 f . In Exercises 42–45, use the following function values: f (4)

f (4)

g(4)

g (4)

2

−3

5

−1

43. Calculate F (4), where F(x) = x f (x). Find the derivative of fg and f /g at x = 4. SOLUTION Let F(x) = x f (x). Then F (x) = x f (x) + f (x), and F (4) = 4 f (4) + f (4) = 4(−3) + 2 = −10. x 45. Calculate H (4), where H (x) = = xg(x) f .(x). where G(x) Calculate G (4), g(x) f (x) x SOLUTION Let H (x) = . Then g(x) f (x)

g(x) f (x) · 1 − x g(x) f (x) + f (x)g (x) . H (x) = (g(x) f (x))2 Therefore, H (4) =

(5)(2) − 4 · ((5)(−3) + (2)(−1)) ((5)(2))2

=

78 39 = . 100 50

104

CHAPTER 3


47. Proceed as in Exercise 46 to calculate F (0), where Calculate F (0), where 3x 5 + 5x 4 + 5x + 1 9+ x 5/3 F(x) = 1 + x + x 4/3x+ x 8 + 4x 5 9− 7x 4 8x − 7x + 1 F(x) = 4 x − 3x 2 + 2x + 1 SOLUTION Write F(x) = f (x)(g(x)/ h(x)), where Hint: Do not calculate F (x). Instead, write F(x) = f (x)/g(x) and express F (0) directly in terms of f (0), f (0), g(0), g (0). f (x) = (1 + x + x 4/3 + x 5/3 ) g(x) = 3x 5 + 5x 4 + 5x + 1 and h(x) = 8x 9 − 7x 4 + 1. 1 2 Now, f (x) = 1 + 43 x 3 + 53 x 3 , g (x) = 15x 4 + 20x 3 + 5, and h (x) = 72x 8 − 28x 3 . Moreover, f (0) = 1, f (0) = 1, g(0) = 1, g (0) = 5, h(0) = 1, and h (0) = 0. From the product and quotient rules,

F (0) = f (0)

h(0)g (0) − g(0)h (0) 1(5) − 1(0) + f (0)(g(0)/ h(0)) = 1 + 1(1/1) = 6. 2 1 h(0)

Further Insights and Challenges 49. Prove the Quotient Rule using the limit definition of the derivative. Let f , g, h be differentiable functions. Show that ( f gh) (x) is equal to f . Suppose that f and g are differentiable at x = a and that g(a) = 0. Then SOLUTION Let p = g f (x)g(x)h (x) + f (x)g (x)h(x) + f (x)g(x)h(x) f (a + h) f (a + h)g(a) − f (a)g(a + h) f (a) Hint: Write f gh as f (gh). − p(a + h) − p(a) g(a + h) g(a) g(a + h)g(a) p (a) = lim = lim = lim h h h h→0 h→0 h→0 f (a + h)g(a) − f (a)g(a) + f (a)g(a) − f (a)g(a + h) hg(a + h)g(a) 1 f (a + h) − f (a) g(a + h) − g(a) = lim g(a) − f (a) h h h→0 g(a + h)g(a) 1 f (a + h) − f (a) g(a + h) − g(a) = lim g(a) lim − f (a) lim h h h→0 g(a + h)g(a) h→0 h→0

= lim

h→0

=

1 (g(a))2

In other words, p =

g(a) f (a) − f (a)g (a) g(a) f (a) − f (a)g (a) = (g(a))2

f g f − f g = . g g2

51. Derive the Quotient Rule using Eq. (6) and the Product Rule. Derivative of the Reciprocal Use the limit definition to show that if f (x) is differentiable and f (x) = 0, then f (x) 1 SOLUTION h(x) = g(x) 1/ f (x) is Let differentiable and. We can write h(x) = f (x) g(x) . Applying Eq. (6), 1 f (x) 1 1 d g−(x) − f (x)g (x) + f (x)g(x) f (x) = h (x) = f (x) . = − f (x) = + f (x) + 2 f 2(x) g(x) g(x) g(x) d x f (x) (g(x)) (g(x))2 Hint: Show that the difference quotient for 1/ f (x) is equal to 53. Carry out the details of Agnesi’s proof of the Quotient Rule from her book on calculus, published in 1748: Assume Show that Eq. (6) is a special case of the Quotient Rule. − f (x of + h) that f , g, and h = f /g are differentiable. Compute thef (x) derivative hg = f using the Product Rule and solve for h . h f (x) f (x + h) SOLUTION Suppose that f , g, and h are differentiable functions with h = f /g. • Then hg = f and via the product rule hg + gh = f .

f − hg • Solving for h yields h = = g

f − g

f g g

=

g f − f g . g2

In Exercises 55–56, let f (x) be a polynomial. A basic fact of algebra states that c is a root of f (x) if and only if The Power Rule Revisited If you are familiar with proof by induction, use induction to2prove the Power Rule h(x), where h(x) is a f (x) = (x − c)g(x) for some polynomial g(x). We say that c is a multiple root if f (x)n = (x − c) for all whole numbers n. Show that the Power Rule holds for n = 1, then write x as x · x n−1 and use the Product polynomial. Rule. 55. Show that c is a multiple root of f (x) if and only if c is a root of both f (x) and f (x).

S E C T I O N 3.4

Rates of Change

105

SOLUTION Assume first that f (c) = f (c) = 0 and let us show that c is a multiple root of f (x). We have f (x) = (x − c)g(x) for some polynomial g(x) and so f (x) = (x − c)g (x) + g(x). However, f (c) = 0 + g(c) = 0, so c is also a root of g(x) and hence g(x) = (x − c)h(x) for some polynomial h(x). We conclude that f (x) = (x − c)2 h(x), which shows that c is a multiple root of f (x). Conversely, assume that c is a multiple root. Then f (c) = 0 and f (x) = (x − c)2 g(x) for some polynomial g(x). Then f (x) = (x − c)2 g (x) + 2g(x)(x − c). Therefore, f (c) = (c − c)2 g (c) + 2g(c)(c − c) = 0.

57. Figure 4 55 is the graph of awhether polynomial at A, B, and of these is a multiple root? Explain Use Exercise to determine c = with −1 isroots a multiple root ofC. theWhich polynomials 5 4 3 2 your(a) reasoning using the result of Exercise 55. x + 2x − 4x − 8x − x + 2 (b) x 4 + x 3 − 5x 2 − 3x + 2 y

x A

B

C

FIGURE 4 SOLUTION A on the figure is a multiple root. It is a multiple root because f (x) = 0 at A and because the tangent line to the graph at A is horizontal, so that f (x) = 0 at A. For the same reasons, f also has a multiple root at C.

3.4 Rates of Change Preliminary Questions 1. What units might be used to measure the ROC of: (a) Pressure (in atmospheres) in a water tank with respect to depth? (b) The reaction rate of a chemical reaction (the ROC of concentration with respect to time), where concentration is measured in moles per liter? SOLUTION

(a) The rate of change of pressure with respect to depth might be measured in atmospheres/meter. (b) The reaction rate of a chemical reaction might be measured in moles/(liter·hour). 2. Suppose that f (2) = 4 and the average ROC of f between 2 and 5 is 3. What is f (5)? SOLUTION

The average rate of change of f between 2 and 5 is given by f (5) − 4 f (5) − f (2) = . 5−2 3

Setting this equal to 3 and solving for f (5) yields f (5) = 13. 3. Two trains travel from New Orleans to Memphis in 4 hours. The first train travels at a constant velocity of 90 mph, but the velocity of the second train varies. What was the second train’s average velocity during the trip? SOLUTION

Since both trains travel the same distance in the same amount of time, they have the same average velocity:

90 mph. 4. Estimate f (26), assuming that f (25) = 43 and f (25) = 0.75. SOLUTION

f (x) ≈ f (25) + f (25)(x − 25), so f (26) ≈ 43 + 0.75(26 − 25) = 43.75.

5. The population P(t) of Freedonia in 1933 was P(1933) = 5 million. (a) What is the meaning of the derivative P (1933)? (b) Estimate P(1934) if P (1933) = 0.2. What if P (1933) = 0? SOLUTION

(a) Because P(t) measures the population of Freedonia as a function of time, the derivative P (1933) measures the rate of change of the population of Freedonia in the year 1933. (b) P(1934) ≈ P(1933) + P (1933). Thus, if P (1933) = 0.2, then P(1934) ≈ 5.2 million. On the other hand, if P (1933) = 0, then P(1934) ≈ 5 million.

CHAPTER 3


Exercises 1. Find the ROC of the area of a square with respect to the length of its side s when s = 3 and s = 5. Let the area be A = f (s) = s 2 . Then the rate of change of A with respect to s is d/ds(s 2 ) = 2s. When s = 3, the area changes at a rate of 6 square units per unit increase. When s = 5, the area changes at a rate of 10 square units per unit increase. (Draw a 5 × 5 square on graph paper and trace the area added by increasing each side length by 1, excluding the corner, to see what this means.) SOLUTION

3. FindFind the the ROC of yof=the x −1 with respect to with x forrespect x = 1, to 10.the length of its side s when s = 3 and s = 5. ROC volume of a cube 1 . −2 SOLUTION The ROC of change is d y/d x = −x . If x = 1, the ROC is −1. If x = 10, the ROC is − 100 √ In Exercises 5–8,rate calculate the ROC. At what is the cube root 3 x changing with respect to x when x = 1, 8, 27? dV 5. , where V is the volume of a cylinder whose height is equal to its radius (the volume of a cylinder of height h dr and radius r is π r 2 h) SOLUTION

The volume of the cylinder is V = π r 2 h = π r 3 . Thus d V /dr = 3π r 2 .

4 3 7. ROC of the volume V of (the volume ROC of the volume V aofsphere a cubewith withrespect respecttotoitsitsradius surface area A of a sphere is V = 3 π r ) SOLUTION

The volume of a sphere of radius r is V = 43 π r 3 . Thus d V /dr = 4π r 2 .

In Exercises d A 9–10, refer to Figure 10, which shows the graph of distance (in kilometers) versus time (in hours) for2 a car , where A is the surface area of a sphere of diameter D (the surface area of a sphere of radius r is 4π r ) trip. dD Distance 150 (km) 100 50 0.5

1

1.5

2

2.5

3

Time (hours)

FIGURE 10 Graph of distance versus time for a car trip.

9. (a) Estimate the average velocity over [0.5, 1]. (b) Is average velocity greater over [1, 2] or [2, 3]? (c) At what time is velocity at a maximum? SOLUTION

(a) The average velocity over the interval [.5, 1] is 50 − 25 = 50 km/h. 1 − .5 (b) The average velocity over the interval [1, 2], 75 − 50 = 25 km/h, 2−1 which is less than that over the interval [2, 3], 150 − 75 = 75 km/h. 3−2 (c) The car’s velocity is maximum when the slope of the distance versus time curve is most positive. This appears to happen when t = 0.5 h, t = 1.25 h, or t = 2.5 h. 11. Figure 11 displays the voltage across a capacitor as a function of time while the capacitor is being charged. Estimate Match the description with the interval (a)–(d). the ROC of voltage at t = 20 s. Indicate the values in your calculation and include proper units. Does voltage change (i) Velocity increasing more quickly or more slowly as time goes on? Explain in terms of tangent lines. (ii) Velocity decreasing 4 (iii) Velocity negative 3 (iv) Average velocity of 50 kph 2 (a) [0, 0.5] 1 (b) [0, 1] (c) [1.5, 2] 10 20 30 40 (d) [2.5, 3] Time (s) Voltage (V)

106

FIGURE 11

S E C T I O N 3.4

Rates of Change

107

SOLUTION The tangent line sketched in the figure below appears to pass through the points (10, 3) and (30, 4). Thus, the ROC of voltage at t = 20 seconds is approximately

4−3 = 0.05 V/s. 30 − 10 As we move to the right of the graph, the tangent lines to it grow shallower, indicating that the voltage changes more slowly as time goes on. y 4 3 2 1

x 10

13. (a) (b) (c)

20

30

40

A stone is tossed vertically upward with an initial velocity of 25 ft/s from the top of a 30-ft building. Use Figure 12 to estimate d T /dh at h = 30 and 70, where T is atmospheric temperature (in degrees Celsius) What height the stone after 0.25iss?d T /dh equal to zero? and h is is the altitude (inofkilometers). Where Find the velocity of the stone after 1 s. When does the stone hit the ground?

1 SOLUTION We employ Galileo’s formula, s(t) = s0 + v0 t − 2 gt 2 = 30 + 25t − 16t 2 , where the time t is in seconds (s) and the height s is in feet (ft). (a) The height of the stone after .25 seconds is s(.25) = 35.25 ft. (b) The velocity at time t is s (t) = 25 − 32t. When t = 1, this is −7 ft/s. √ −25 ± 2545 2 or (c) When the stone hits the ground, its height is zero. Solve 30 + 25t − 16t = 0 to obtain t = −32 t ≈ 2.36 s. (The other solution, t ≈ −0.79, we discard since it represents a time before the stone was thrown.)

30tft.+Find 340 15. TheThe temperature an object (in degrees Fahrenheit) as a after function of time minutes) T (t) = 34 t 2−−15t height (inof feet) of a skydiver at time t (in seconds) opening his (in parachute is is h(t) = 2,000 for 0the ≤ tskydiver’s ≤ 20. At velocity what rateafter doesthe theparachute object cool after 10 min (give correct units)? opens. SOLUTION

Let T (t) = 34 t 2 − 30t + 340, 0 ≤ t ≤ 20. Then T (t) = 32 t − 30, so T (10) = −15◦ F/min.

2.99 × 1016 velocity centimetersforce per second) flowing through a capillary radius 0.008 cm is 17. TheThe earth exerts a(in gravitational of F(r ) of = a blood2 molecule (in Newtons) on an object with a of mass of 75 kg, where r r is the distance from the molecule to the center of the given by the formula v = 6.4 × 10−8 − 0.001r 2 , where r is the distance (inthe meters) the center the earth. Find thewhen ROC rof=force with capillary. Find ROCfrom of velocity as a of function of distance 0.004 cm.respect to distance at the surface of the earth, assuming the radius of the earth is 6.77 × 106 m. SOLUTION

The rate of change of force is F (r ) = −5.98 × 1016 /r 3 . Therefore, F (6.77 × 106 ) = −5.98 × 1016 /(6.77 × 106 )3 = −1.93 × 10−4 N/m.

2.25R 7 −1/2 19. TheThe power delivered byata abattery to ranmeters apparatus R (in ohms) is P= = W. Find rate of (2.82 m/s.the Calculate escape velocity distance fromof theresistance center of the earth is vesc 2 (R × + 10 0.5))r the of rate at which with respect surface of the earth. esc changes change power withvrespect to resistance for Rto=distance 3 and Rat=the 5 . SOLUTION

P (R) =

(R + .5)2 2.25 − 2.25R(2R + 1) . (R + .5)4

Therefore, P (3) = −0.1312 W/ and P (5) = −0.0609 W/. 21. By Faraday’s Law, if a conducting wire of length meters moves at velocity v m/s perpendicular to a magnetic field 2 − t + 10 cm. TheBposition of a aparticle line duringina 5-s s(t) = t that of strength (in teslas), voltagemoving of size in V a=straight −Bv is induced the trip wire.isAssume B = 2 and = 0.5. (a) What is the averagedvelocity (a) Find the rate of change V /dv. for the entire trip? (b) Isthe there at which instantaneous velocity to this average velocity? If so, find it. (b) Find rateaoftime change of V the with respect to time t if v is = equal 4t + 9. SOLUTION

(a) Assuming that B = 2 and l = 0.5, V = −2(.5)v = −v. Therefore, dV = −1. dv (b) If v = 4t + 9, then V = −2(.5)(4t + 9) = −(4t + 9). Therefore, ddtV = −4. The height (in feet) of a helicopter at time t (in minutes) is s(t) = −3t 3 + 400t for 0 ≤ t ≤ 10.


23. The population P(t) of a city (in millions) is given by the formula P(t) = 0.00005t 2 + 0.01t + 1, where t denotes the number of years since 1990. (a) How large is the population in 1996 and how fast is it growing? (b) When does the population grow at a rate of 12,000 people per year? SOLUTION

Let P(t) = (0.00005)t 2 + (0.01)t + 1 be the population of a city in millions. Here t is the number of years

past 1990. (a) In 1996 (t = 6 years after 1990), the population is P(6) = 1.0618 million. The rate of growth of population is P (t) = 0.0001t + 0.01. In 1996, this corresponds to a growth rate of P (6) = 0.0106 million per year or 10,600 people per year. (b) When the growth rate is 12,000 people per year (or 0.012 million per year), we have P (t) = 0.0001t + 0.01 = 0.012. Solving for t gives t = 20. This corresponds to the year 2010; i.e., 20 years past 1990. Ethan to finds that with hours of tutoring, he isI able answer correctly S(h) percent of thebyproblems on a 25. According Ohm’s Law,hthe voltage V , current , andtoresistance R in a circuit are related the equation (h)? Which would you expect to be larger: S (3) or S (30)? Explain. mathVexam. What is the the derivative = I R, where themeaning units areofvolts, amperes, Sand ohms. Assume that voltage is constant with V = 12 V. Calculate (specifying thederivative units): S (h) measures the rate at which the percent of problems Ethan answers correctly changes SOLUTION The The to average ROC of hours I withof respect to R the interval from R = 8 to R = 8.1 with (a) respect the number tutoring hefor receives. One of S(h) is shown in the Rfigure (b) possible The ROCgraph of I with respect to R when = 8 below on the left. This graph indicates that in the early hours of working withROC the tutor, makes to rapid progress (c) The of R Ethan with respect I when I = in 1.5learning the material but eventually approaches either the limit of his ability to learn the material or the maximum possible score on the exam. In this scenario, S (3) would be larger than S (30). An alternative graph of S(h) is shown below on the right. Here, in the early hours of working with the tutor little progress is made (perhaps the tutor is assessing how much Ethan already knows, his learning style, his personality, etc.). This is followed by a period of rapid improvement and finally a leveling off as Ethan reaches his maximum score. In this scenario, S (3) and S (30) might be roughly equal. Percentage correct

CHAPTER 3

Percentage correct

108

Hours of tutoring

Hours of tutoring

Table 2 gives the total month 1999 as determined Department of 27. Suppose θ (t) measures the U.S. anglepopulation between a during clock’seach minute andof hour hands. What is θ by (t)the at 3U.S. o’clock? Commerce. (a) Estimate P (t) for each of the months January–November. (b) Plot these data points for P (t) and connect the points by a smooth curve. (c) Write a newspaper headline describing the information contained in this plot. Total U.S. Population in 1999

TABLE 2

t

P(t) in Thousands

January February March April May June July August September October November December

271,841 271,987 272,142 272,317 272,508 272,718 272,945 273,197 273,439 273,672 273,891 274,076

The table in the text gives the growing population P(t) of the United States. (a) Here are estimates of P (t) in thousands/month for January–November. The estimates are computed using the estimate f (t + 1) − f (t). SOLUTION

t P (t)

Jan

Feb Mar Apr May Jun

146 155

175

191

210

Jul

227 252

Aug Sep Oct Nov 242

233 219

185

S E C T I O N 3.4

Rates of Change

109

(b) Here is a plot of these estimates (1 = Jan, 2 = Feb, etc.) P 250 200 150 100 50 t 2

4

6

8

10

(c) “U.S. Growth Rate Declines After Midsummer Peak” 29. According to a formula widely used by doctors√to determine drug dosages, a person’s body surface area (BSA) (in steeper as xh increases. At what rate do the and slopes tangent The tangent lines to of f BSA (x) ==x 2 grow hw/60, where is the height in centimeters w of thethe weight in meters squared) is given bythe thegraph formula lines increase? kilograms. Calculate the ROC of BSA with respect to weight for a person of constant height h = 180. What is this ROC for w = 70 and w = 80? Express your result in the correct units. Does BSA increase more rapidly with respect to weight at lower or higher body weights? √ √ 5 SOLUTION Assuming constant height h = 180 cm, let f (w) = hw/60 = 10 w be the formula for body surface area in terms of weight. The ROC of BSA with respect to weight is √ √ 5 1 −1/2 5 w = √ . f (w) = 10 2 20 w If w = 70 kg, this is f (70) =

√

√ 5 14 m2 ≈ 0.0133631 . √ = 280 kg 20 70

If w = 80 kg, √

5 1 1 m2 . √ = √ = 80 kg 20 80 20 16 √ Because the rate of change of BSA depends on 1/ w, it is clear that BSA increases more rapidly at lower body weights. f (80) =

31. What is the velocity of an object dropped from a height of 300 m when it hits the ground? A slingshot is used to shoot a pebble in the air vertically from ground level with an initial velocity 200 m/s. Find SOLUTION Wemaximum employ Galileo’s s(t) = s0 + v0 t − 12 gt 2 = 300 − 4.9t 2 , where the time t is in seconds (s) the pebble’s velocityformula, and height. and the height s is in meters (m). When the ball hits the ground its height is 0. Solve s(t) = 300 − 4.9t 2 = 0 to obtain t ≈ 7.8246 s. (We discard the negative time, which took place before the ball was dropped.) The velocity at impact is v(7.8246) = −9.8(7.8246) ≈ −76.68 m/s. This signifies that the ball is falling at 76.68 m/s. 33. A ball is tossed up vertically from ground level and returns to earth 4 s later. What was the initial velocity of the It takes a stone 3 s to hit the ground when dropped from the top of a building. How high is the building and what stone and how high did it go? is the stone’s velocity upon impact? 1 SOLUTION Galileo’s formula gives s(t) = s0 + v0 t − 2 gt 2 = v0 t − 4.9t 2 , where the time t is in seconds (s) and the height s is in meters (m). When the ball hits the ground after 4 seconds its height is 0. Solve 0 = s(4) = 4v0 − 4.9(4)2 to obtain v0 = 19.6 m/s. The stone reaches its maximum height when s (t) = 0, that is, when 19.6 − 9.8t = 0, or t = 2 s. At this time, t = 2 s, 1 s(2) = 0 + 19.6(2) − (9.8)(4) = 19.6 m. 2 35. A man on the tenth floor of a building sees a bucket (dropped by a window washer) pass his window and notes that An object is tossed up vertically from ground level and hits the ground T s later. Show that its maximum height it hits the ground 1.5 s later. Assuming a floor is 16 ft high (and neglecting air friction), from which floor was the bucket was reached after T /2 s. dropped? A falling object moves 16t 2 feet in t seconds. Suppose H is the unknown height from which the bucket fell starting at time zero. The man saw it at some time t (also unknown to us) and it hit the ground, 160 feet down at time t + 1.5. Thus SOLUTION

(H − 16t 2 ) − (H − 16(t + 1.5)2 ) = 160. The H ’s cancel as do the t 2 terms and solving for t gives t = 31/12 s. Thus the bucket fell 16(31/12)2 ≈ 16 · 6.67 feet before the man saw it. Since there are 16 feet in a floor the bucket was dropped from 6.67 floors above the 10th: either the 16th or 17th floor. 37. Show that for an object rising and falling according to Galileo’s formula in Eq. (3), the average velocity over any Which the following statements is true for an object falling under the influence of gravity near the surface time interval [t1 , tof 2 ] is equal to the average of the instantaneous velocities at t1 and t2 . of the earth? Explain. (a) The object covers equal distance in equal time intervals.

110

CHAPTER 3


The simplest way to proceed is to compute both values and show that they are equal. The average velocity

over [t1 , t2 ] is (s0 + v0 t2 − 12 gt22 ) − (s0 + v0 t1 − 12 gt12 ) v0 (t2 − t1 ) + g2 (t2 2 − t1 2 ) s(t2 ) − s(t1 ) = = t2 − t 1 t2 − t1 t2 − t 1 v (t − t1 ) g g = 0 2 − (t2 + t1 ) = v0 − (t2 + t1 ) t2 − t 1 2 2 Whereas the average of the instantaneous velocities at the beginning and end of [t1 , t2 ] is 1 g g s (t1 ) + s (t2 ) 1

= (v0 − gt1 ) + (v0 − gt2 ) = (2v0 ) − (t2 + t1 ) = v0 − (t2 + t1 ). 2 2 2 2 2 The two quantities are the same. In Exercises 39–46, use Eq.up (2)and to estimate A weight oscillates down at the the unit end change. of a spring. Figure 13 shows the height y of the weight through one √ oscillation. √ √ √ sketch of the graph of the velocity as a function of time. cycle of the Make a rough 39. Estimate 2 − 1 and 101 − 100. Compare your estimates with the actual values. √ SOLUTION Let f (x) = x = x 1/2 . Then f (x) = 12 (x −1/2 ). We are using the derivative to estimate the average ROC. That is, √ √ x +h− x ≈ f (x), h so that √

√

x ≈ h f (x). √ √ √ Thus, 2 − 1 ≈ 1 f (1) = 12 (1) = 12 . The actual value, to six decimal places, is 0.414214. Also, 101 − 100 ≈

1 = .05. The actual value, to six decimal places, is 0.0498756. 1 f (100) = 12 10 √

x +h−

as in Example 3. Calculate F(65) and estimate the increase in 41. Let F(s) = 1.1s + 0.03s 2 be the stopping distance −x Suppose that f (x) is a function with f (x) = 2 . Estimate f (7) − f (6). Then estimate f (5), assuming that stopping distance if speed is increased from 65 to 66 mph. Compare your estimate with the actual increase. f (4) = 7. SOLUTION Let F(s) = 1.1s + .03s 2 be as in Example 3. F (s) = 1.1 + 0.06s. • Then F(65) = 198.25 ft and F (65) = 5.00 ft/mph. • F (65) ≈ F(66) − F(65) is approximately equal to the change in stopping distance per 1 mph increase in speed

when traveling at 65 mph. Increasing speed from 65 to 66 therefore increases stopping distance by approximately 5 ft. • The actual increase in stopping distance when speed increases from 65 mph to 66 mph is F(66) − F(65) = 203.28 − 198.25 = 5.03 feet, which differs by less than one percent from the estimate found using the derivative. 3 Determine the cost of producing 43. TheAccording dollar costtoofKleiber’s producing x bagels is C(x)rate = 300 0.25x − 0.5(x/1,000) Law, the metabolic P (in+kilocalories per day) and.body mass m (in kilograms) of an 2,000animal bagelsare and estimate cost of the 2001st your estimate the withincrease the actual cost of therate 2001st bagel. . Estimate in metabolic when body related by athe three-quarter power bagel. law P Compare = 73.3m 3/4 SOLUTION Expanding thetopower mass increases from 60 61 kg.of

3 yields C(x) = 300 + .25x − 5 × 10−10 x 3 .

This allows us to get the derivative C (x) = .25 − 1.5 × 10−9 x 2 . The cost of producing 2000 bagels is C(2000) = 300 + 0.25(2000) − 0.5(2000/1000)3 = 796 dollars. The cost of the 2001st bagel is, by definition, C(2001) − C(2000). By the derivative estimate, C(2001) − C(2000) ≈ C (2000)(1), so the cost of the 2001st bagel is approximately C (2000) = .25 − 1.5 × 10−9 (20002 ) = $.244. C(2001) = 796.244, so the exact cost of the 2001st bagel is indistinguishable from the estimated cost. The function is very nearly linear at this point. 45. The demand for a commodity generally decreases as the price is raised. Suppose that the demand for oil (per capita 2 + 10−8 x 3 . of producing x video cameras is C(x) = 500xFind − 0.003x per year)Suppose is D( p)the = dollar 900/ pcost barrels, where p is the price per barrel in dollars. the demand when p = $40. Estimate (a) Estimate the marginal at production level if x p=is5,000 and to compare the decrease in demand if p risescost to $41 and the increase decreased $39. it with the actual cost C(5,001) − C(5,000). SOLUTION D( p) = 900 p −1 , so D ( p) = −900 p −2 . When the price is $40 a barrel, the per capita demand is (b)=Compare the marginal cost at = 5,000inwith average cost camera, D(40) 22.5 barrels per year. With anxincrease pricethe from $40 to $41per a barrel, thedefined changeasinC(x)/x. demand D(41) − D(40) −2 is approximately D (40) = −900(40 ) = −.5625 barrels a year. With a decrease in price from $40 to $39 a barrel, the change in demand D(39) − D(40) is approximately −D (40) = +.5625. An increase in oil prices of a dollar leads to a decrease in demand of .5625 barrels a year, and a decrease of a dollar leads to an increase in demand of .5625 barrels a year.

S E C T I O N 3.4

Rates of Change

111

of A(s + 1) − melanogaster, A(s) provided grown by Eq.in(2) has error exactly equal to 1. Explain 47. LetThe A =reproduction s 2 . Show that ratethe of estimate the fruit fly Drosophila bottles in a laboratory, decreases as the this result using Figure 14. bottle becomes more crowded. A researcher has found that when a bottle contains p flies, the number of offspring per female per day is f ( p) = (34 − 0.612 p) p −0.658 1 2

(a) Calculate f (15) and f (15). (b) Estimate the decrease in daily offspring per female when p is increased from 15 to 16. Is this estimate larger or s smaller than the actual value f (16) − f (15)? (c) Plot f ( p) for 5 ≤ p ≤ 25 and verify that f ( p) is a decreasing function of p. Do you expect f ( p) to be 14 positive or negative? Plot f ( p) and confirm yourFIGURE expectation. SOLUTION

Let A = s 2 . Then A(s + 1) − A(s) = (s + 1)2 − s 2 = s 2 + 2s + 1 − s 2 = 2s + 1,

while A (s) = 2s. Therefore, regardless of the value of s, A(s + 1) − A(s) − A (s) = (2s + 1) − 2s = 1. To understand this result, consider Figure 14, which illustrates a square of side length s expanded to a square of side length s + 1 by increasing the side length by 1/2 unit in each direction. The difference A(s + 1) − A(s) is the total area between the two squares. On the other hand, A (s) = 2s is the total area of the four rectangles of dimension s by 1/2 bordering the sides of the original square. The error in approximating A(s + 1) − A(s) by A (s) is given by the four 1/2 by 1/2 squares in the corners of the figure. The total area of these four small squares is always exactly 1. 49. Let M(t) be the mass (in kilograms) of a plant as a function of time (in years). Recent studies by Niklas and Enquist According to aSteven’s Law wide in psychology, the perceived magnitude of to a stimulus (howthe strong a person feels have suggested that for remarkably range of plants (from algae and grass palm trees), growth rate during the span stimulus be) is proportional to a power ofpower the actual I of the Although not an C. exact law, some constant the life of thetoorganism satisfies a three-quarter law, intensity that is, d M/dt = stimulus. C2/3 M 3/4 for experiments show that the perceived brightness B of a light satisfies B = k I , where I is the light intensity, (a) If a tree has a growth rate of 6 kg/year when M = 100 kg, what is its growth rate when M = 125 kg? whereas the perceived heaviness H of a weight W satisfies H = kW 3/2 (k is a constant that is different in the (b) If M = 0.5 kg, how much more mass must the plant acquire to double its growth rate? two cases). Compute d B/d I and d H /d W and state whether they are increasing or decreasing functions. Use this to SOLUTION justify the statements: √ (a) Suppose a tree increase has a growth rate d M/dtisoffelt6 more kg/yr strongly when Mwhen = 100, = C(100 = 10C 10, so that (a)√ A one-unit in light intensity I is then small6than when 3/4 I is)large. 3 10 another pound to a load W is felt more strongly when W is large than when W is small. C =(b) . When M = 125, 50Adding √ dM 3 10 = C(1253/4 ) = 25(51/4 ) = 7.09306. dt 50 (b) The growth rate when M = .5 kg is d M/dt = C(.53/4 ). To double the rate, we must find M so that d M/dt = C M 3/4 = 2C(.53/4 ). We solve for M. C M 3/4 = 2C(.53/4 ) M 3/4 = 2(.53/4 ) M = (2(.53/4 ))4/3 = 1.25992. The plant must acquire the difference 1.25992 − .5 = .75992 kg in order to double its growth rate. Note that a doubling of growth rate requires more than a doubling of mass.

Further Insights and Challenges dP 51. TheAs size a certainspreads animalthrough population P(t) at time (in months)p satisfies = 0.2(300at−time P). t (in days) satisfies anof epidemic a population, the tpercentage of infected individuals dt thePequation a differential (a) Is growing(called or shrinking when Pequation) = 250? when P = 350? (b) Sketch the graph of d P/dt as a functiondof p P for 0 ≤ P ≤2300. = 4 p − 0.06 p 0 ≤ p ≤ 100 (c) Which of the graphs in Figure 15 is the graph dt of P(t) if P(0) = 200? (a) (b) (c) (d)

How fast is the epidemic spreading when p = 10% and Pwhen p = 70%? P For which p is the epidemic 300 neither spreading nor diminishing? 300 Plot d p/dt as a function of p. 200 200 What is the maximum possible rate of increase and for which p does this occur? 100

100 t

t 4

8

4

(A)

8 (B)

FIGURE 15

112

CHAPTER 3


Let P (t) = d P/dt = 0.2(300 − P).

(a) Since P (250) = 10, the population is growing when P = 250. Since P (350) = −10, the population is shrinking when P = 350. (b) Here is a graph of d P/dt for 0 ≤ t ≤ 300. dP/dt 60 50 40 30 20 10 50 100 150 200 250 300

P

(c) If P(0) = 200, as in graph (A), then P (0) = 20 > 0 and the population is growing as depicted. Accordingly, graph (A) has the correct shape for P(t). If P(0) = 200, as in graph (B), then P (0) = 20 > 0 and the population is growing, contradicting what is depicted. Thus graph (B) cannot be the correct shape for P(t). In Exercises 53–54, the average cost per unit at production level x is defined as Cavg (x) = C(x)/x, where C(x) is the Studies of internet usage show that website popularity is described quite well by Zipf’s Law, according to cost function. Average cost is a measure of the efficiency of the production process. which the nth most popular website receives roughly the fraction 1/n of all visits. Suppose that on a particular day, the nththat most popular hadtoapproximately (n) = 106 /nthe visitors ≤ 15,000). (x) is site equal the slope of theV line through origin(for andn the point (x, C(x)) on the graph of C(x). 53. Show C avg (a)this Verify that the top 50 websites received nearly of the visits. LetatTpoints (N ) denote sum V (n)16. for Using interpretation, determine whether average cost45% or marginal cost isHint: greater A, B, the C, D in of Figure 1 ≤ n ≤ N . Use a computer algebra system to compute T (45) and T (15,000). Cost (b) Verify, by numerical experimentation, that when Eq. (2) is used to estimate V (n + 1) − V (n), the error in the D estimate decreases as n grows larger. Find (again, by experimentation) an N such that the error is at most 10 for n ≥ N. C received at most 100 more visitors than the (n + 1)st web B (c) Using Eq. (2), show that for n ≥ 100, the nth A website site. Production level

FIGURE 16 Graph of C(x). SOLUTION

By definition, the slope of the line through the origin and (x, C(x)), that is, between (0, 0) and (x, C(x))

is C(x) − 0 C(x) = = Cav . x −0 x At point A, average cost is greater than marginal cost, as the line from the origin to A is steeper than the curve at this point (we see this because the line, tracing from the origin, crosses the curve from below). At point B, the average cost is still greater than the marginal cost. At the point C, the average cost and the marginal cost are nearly the same, since the tangent line and the line from the origin are nearly the same. The line from the origin to D crosses the cost curve from above, and so is less steep than the tangent line to the curve at D; the average cost at this point is less than the marginal cost. The cost in dollars of producing alarm clocks is

3.5 Higher Derivatives

C(x) = 50x 3 − 750x 2 + 3,740x + 3,750

where x is in units of 1,000. (a) Calculate the average cost at x = 4, 6, 8, and 10. 1. An economist who announces that “America’s economic growth is slowing” is making a statement about the gross (b) Use the graphical interpretation of average cost to find the production level x0 at which average cost is lowest. national product (GNP) as a function of time. Is the second derivative of the GNP positive? What about the first derivative? What is the relation between average cost and marginal cost at x 0 (see Figure 17)? SOLUTION If America’s economic growth is slowing, then the GNP is increasing, but the rate of increase is decreasing. Thus, the second derivative of GNP is negative, while the first derivative is positive. FIGURE 17 Cost function C(x) = 50x 3 − 750x 2 + 3,740x + 3,750. 2. On September 4, 2003, the Wall Street Journal printed the headline “Stocks Go Higher, Though the Pace of Their Gains Slows.” Rephrase as a statement about the first and second time derivatives of stock prices and sketch a possible graph.

Preliminary Questions

SOLUTION Because stocks are going higher, stock prices are increasing and the first derivative of stock prices must therefore be positive. On the other hand, because the pace of gains is slowing, the second derivative of stock prices must be negative.

S E C T I O N 3.5

Higher Derivatives

113

Stock price

Time

3. Is the following statement true or false? The third derivative of position with respect to time is zero for an object falling to earth under the influence of gravity. Explain. SOLUTION This statement is true. The acceleration of an object falling to earth under the influence of gravity is constant; hence, the second derivative of position with respect to time is constant. Because the third derivative is just the derivative of the second derivative and the derivative of a constant is zero, it follows that the third derivative is zero.

4. Which type of polynomial satisfies f (x) = 0 for all x? SOLUTION The second derivative of all linear polynomials (polynomials of the form ax + b for some constants a and b) is equal to 0 for all x.

Exercises In Exercises 1–12, calculate the second and third derivatives. 1. y = 14x 2 SOLUTION

Let y = 14x 2 . Then y = 28x, y = 28, and y = 0.

2 25x 3. y =y x=4 − 7− 2x + 2x SOLUTION Let y = x 4 − 25x 2 + 2x. Then y = 4x 3 − 50x + 2, y = 12x 2 − 50, and y = 24x.

4 3 5. y =y =π4t r 3 − 9t 2 + 7 3 SOLUTION

Let y = 43 π r 3 . Then y = 4π r 2 , y = 8π r , and y = 8π .

4/5 − 6t 2/3 7. y = 20t √ y= x 16 4 SOLUTION Let y = 20t 4/5 − 6t 2/3 . Then y = 16t −1/5 − 4t −1/3 , y = − 5 t −6/5 + 3 t −4/3 , and y = 96 t −11/15 − 16 t −7/3 . 25 9

1 9. y =y z=−x −9/5 z SOLUTION

Let y = z − z −1 . Then y = 1 + z −2 , y = −2z −3 , and y = 6z −4 .

11. y = (x 2 + x)(x 3 + 1) y = t 2 (t 2 + t) SOLUTION Since we don’t want to apply the product rule to an ever growing list of products, we multiply through first. Let y = (x 2 + x)(x 3 + 1) = x 5 + x 4 + x 2 + x. Then y = 5x 4 + 4x 3 + 2x + 1, y = 20x 3 + 12x 2 + 2, and y = 60x 2 + 24x. In Exercises 13–24, calculate the derivative indicated. 1 y= x = x4 13. f (4) (1), 1 +f (x) Let f (x) = x 4 . Then f (x) = 4x 3 , f (x) = 12x 2 , f (x) = 24x, and f (4) (x) = 24. Thus f (4) (1) = 24. d 2 y 15. = 4t=−3 + 3t 2 4t −3 g (1),, yg(t) dt 2 t=1 SOLUTION

SOLUTION

2 Let y = 4t −3 + 3t 2 . Then ddty = −12t −4 + 6t and d 2y = 48t −5 + 6. Hence

d 2 y dt 2

dt

= 48(1)−5 + 6 = 54. t=1

√ 17. h (9), h(x) = x 4 d f √ f (t)== 6tx9 − SOLUTION4 Let, h(x) = 2tx51/2 . Then h (x) = 12 x −1/2 , h (x) = − 14 x −3/2 , and h (x) = 38 x −5/2 . Thus dt t=1 1 h (9) = 648 . g (9),

g(x) = x −1/2

114

CHAPTER 3


d 4 x 19. , dt 4 t=16 SOLUTION

Thus

x = t −3/4

2 −11/4 , d 3 x = − 231 t −15/4 , and d 4 x = 3465 t −19/4 . Let x(t) = t −3/4 . Then ddtx = − 34 t −7/4 , d 2x = 21 16 t 64 256 dt dt 3 dt 4

d 4 x dt 4

=

t=16

3465 −19/4 3465 16 . = 256 134217728

x 21. g (1), g(x) = 2 f (4), f (t)x=+2t 1 −t x SOLUTION Let g(x) = . Then x +1 g (x) =

(x + 1)(1) − (x)(1) 1 1 = = 2 (x + 1)2 (x + 1)2 x + 2x + 1

and g (x) =

(x 2 + 2x + 1)(0) − 1(2x + 2) 2x + 2 2 =− =− . (x 2 + 2x + 1)2 (x + 1)3 (x + 1)4

1 Thus, g (1) = − . 4 1 23. h (1), h(x) = √ 1 1 f (1), f (t) =x + t3 + 1 1 SOLUTION Let h(x) = √ . Then x +1 − 1 x −1/2 − 12 x −1/2 h (x) = √2 , √ 2 = x +2 x +1 x +1 and √

3 x −1/2 + x −1 + 1 x −3/2 x + 1) 14 x −3/2 + 12 x −1/2 (1 + x −1/2 ) 4 4 = √ √ (x + 2 x + 1)2 ( x + 1)4

√ 3 x −1 + 1 x −3/2 ( x + 1) 34 x −1 + 14 x −3/2 4 = = . √ √ 4 ( x + 1)4 ( x + 1)3

(x + 2 h (x) =

= 18 . Accordingly, h (1) = (3/4)+(1/4) 23 25. Calculate y (k) (0) for 0 ≤2 k ≤ 5, where y = x 4 + ax 3 + bx 2 + cx + d (with a, b, c, d the constants). x F(x) =the power, constant multiple, and sum rules at each stage, we get (note y (0) is y by convention): F (2),Applying SOLUTION x −3 k

y (k)

0

x 4 + ax 3 + bx 2 + cx + d

1

4x 3 + 3ax 2 + 2bx + c

2

12x 2 + 6ax + 2b

3

24x + 6a

4

24

5

0

from which we get y (0) (0) = d, y (1) (0) = c, y (2) (0) = 2b, y (3) (0) = 6a, y (4) (0) = 24, and y (5) (0) = 0. d 6 −1 (k) 27. UseWhich the result in Example to find satisfy x .f (x) = 0 for all k ≥ 6? of the following2functions dx6 (b) f (x) = x 3 − 2 (a) f (x) = 7x 4 + 4 + x −1 SOLUTION The√equation in Example 2 indicates that (d) f (x) = 1 − x 6 (c) f (x) = x (e) f (x) = x 9/5

d 6 −1 (f) f (x) = 2x 2 + 3x 5 x = (−1)6 6!x −6−1 . 6 dx

S E C T I O N 3.5

Higher Derivatives

115

(−1)6 = 1 and 6! = 6 × 5 × 4 × 3 × 2 × 1 = 720, so d 6 −1 x = 720x −7 . dx6 √ In Exercises 29–32, general formulaoffor f (n) Calculate thefind firstafive derivatives f (x) =(x).x. (n) −n+1/2 . 29. (a) f (x)Show = (x that + 1)f−1 (x) is a multiple of x (n) 1 as (−1)n−1 for n ≥ 1. −2 −3 (b) ShowLet thatf (x) f (x) SOLUTION = (xalternates + 1)−1 in = sign x+1 . By Exercise 12, f (x) = −1(x + 1) , f (x) = 2(x + 1) , f (x) = 2n − 3 −4 (4) −5 −6(x(c) + Find 1) ,a fformula (x) = . . 2. From we conclude that the nth can (x)1)for ,n. ≥ Hint:this Verify that the coefficient ofderivative x −n+1/2 is ±1be· 3written · 5 · · · as fn(n) (x) . = for24(x f (n)+ 2 n −(n+1) . (−1) n!(x + 1) 31. f (x) = x −1/2 f (x) = x −2 −1 −3/2 SOLUTION f (x) = 2 x . We will avoid simplifying numerators and denominators to find the pattern: f (x) =

−3 −1 −5/2 3×1 = (−1)2 2 x −5/2 x 2 2 2

f (x) = −

5 3 × 1 −7/2 5 × 3 × 1 −7/2 x = (−1)3 x 2 22 23

.. . f (n) (x) = (−1)n

(2n − 1) × (2n − 3) × . . . × 1 −(2n+1)/2 x . 2n

33. (a) Find the −3/2 acceleration at time t = 5 min of a helicopter whose height (in feet) is h(t) = −3t 3 + 400t. f (x) = x (b) Plot the acceleration h (t) for 0 ≤ t ≤ 6. How does this graph show that the helicopter is slowing down during this time interval? SOLUTION

(a) Let h(t) = −3t 3 + 400t, with t in minutes and h in feet. The velocity is v(t) = h (t) = −9t 2 + 400 and acceleration is a(t) = h (t) = −18t. Thus a(5) = −90 ft/min2 . (b) The acceleration of the helicopter for 0 ≤ t ≤ 6 is shown in the figure below. As the acceleration of the helicopter is negative, the velocity of the helicopter must be decreasing. Because the velocity is positive for 0 ≤ t ≤ 6, the helicopter is slowing down. y 1

2

3

4

5

6

x

−20 −40 −60 −80 −100

35. Figure 5 shows f , f , and f . Determine which is which. Find an equation of the tangent to the graph of y = f (x) at x = 3, where f (x) = x 4 . y

y

y

1

2

3

x

1

(A)

2

3

x

1

2

3

x

(C)

(B)

FIGURE 5

(a) f (b) f (c) f . The tangent line to (c) is horizontal at x = 1 and x = 3, where (b) has roots. The tangent line to (b) is horizontal at x = 2 and x = 0, where (a) has roots. 37. Figure 7 shows the graph of the position of an object as a function of time. Determine the intervals on which the The second derivative f is shown in Figure 6. Determine which graph, (A) or (B), is f and which is f . acceleration is positive. SOLUTION

Position

10

20

30

40 Time

FIGURE 7

116

CHAPTER 3

D I F F E R E N T I AT I O N SOLUTION Roughly from time 10 to time 20 and from time 30 to time 40. The acceleration is positive over the same intervals over which the graph is bending upward.

+ frespect (x) = xto2 .the length of a side. 39. FindFind a polynomial (x) satisfying equation f (x)with the second fderivative of thethe volume of axcube SOLUTION Since x f (x) + f (x) = x 2 , and x 2 is a polynomial, it seems reasonable to assume that f (x) is a polynomial of some degree, call it n. The degree of f (x) is n − 2, so the degree of x f (x) is n − 1, and the degree of x f (x) + f (x) is n. Hence, n = 2, since the degree of x 2 is 2. Therefore, let f (x) = ax 2 + bx + c. Then f (x) = 2ax + b and f (x) = 2a. Substituting into the equation x f (x) + f (x) = x 2 yields ax 2 + (2a + b)x + c = x 2 , an identity in x. Equating coefficients, we have a = 1, 2a + b = 0, c = 0. Therefore, b = −2 and f (x) = x 2 − 2x. 41. A servomotor controls the vertical movement of a drill bit that will drill a pattern of holes in sheet metal. The the following descriptions could and not apply Figurethe 8? Explain. maximumWhich verticalofspeed of the drill bit is 4 in./s, while to drilling hole, it must move no more than 2.6 in./s (a) Graph of acceleration when velocity to avoid warping the metal. During a cycle, is theconstant bit begins and ends at rest, quickly approaches the sheet metal, and quickly to its initialwhen position after the hole is drilled. Sketch possible graphs of the drill bit’s vertical velocity and (b) returns Graph of velocity acceleration is constant acceleration. Label the point where the bit enters the sheet metal. (c) Graph of position when acceleration is zero SOLUTION There will be multiple cycles, each of which will be more or less identical. Let v(t) be the downward vertical velocity of the drill bit, and let a(t) be the vertical acceleration. From the narrative, we see that v(t) can be no greater than 4 and no greater than 2.6 while drilling is taking place. During each cycle, v(t) = 0 initially, v(t) goes to 4 quickly. When the bit hits the sheet metal, v(t) goes down to 2.6 quickly, at which it stays until the sheet metal is drilled through. As the drill pulls out, it reaches maximum non-drilling upward speed (v(t) = −4) quickly, and maintains this speed until it returns to rest. A possible plot follows: y

4 Metal 2 x

0.5

1

1.5

2

−2 −4

A graph of the acceleration is extracted from this graph: y

40 20 0.5

1

x

1.5

2

−20 − 40 Metal

In Exercises 42–43, refer to the following. In their 1997 study, Boardman and Lave related the traffic speed S on a twolane road to traffic density Q (number of cars per mile of road) by the formula S = 2,882Q −1 − 0.052Q + 31.73 for 60 ≤ Q ≤ 400 (Figure 9). Speed S 70 (mph) 60 50 40 30 20 10 100

200

300

400

Density Q

FIGURE 9 Speed as a function of traffic density.

43. (a) CalculateExplain d S/d Q intuitively and d 2 S/dwhy Q 2 . we should expect that d S/d Q < 0. (b) Show that d 2 S/d Q 2 > 0. Then use the fact that d S/d Q < 0 and d 2 S/d Q 2 > 0 to justify the following statement: A one-unit increase in traffic density slows down traffic more when Q is small than when Q is large. Plot d S/d Q. Which property of this graph shows that d 2 S/d Q 2 > 0? (c) SOLUTION

S E C T I O N 3.5

Higher Derivatives

117

(a) Traffic speed must be reduced when the road gets more crowded so we expect d S/d Q to be negative. This is indeed the case since d S/d Q = −.052 − 2882/Q 2 < 0. (b) The decrease in speed due to a one-unit increase in density is approximately d S/d Q (a negative number). Since d 2 S/d Q 2 = 5764Q −3 > 0 is positive, this tells us that d S/d Q gets larger as Q increases—and a negative number which gets larger is getting closer to zero. So the decrease in speed is smaller when Q is larger, that is, a one-unit increase in traffic density has a smaller effect when Q is large. (c) d S/d Q is plotted below. The fact that this graph is increasing shows that d 2 S/d Q 2 > 0. y 100

200

300

400

x

−0.2 −0.4 −0.6 −0.8 −1 −1.2

Use a computer algebra system to compute f (k) (x) for k = 1, 2, 3 for the functions: According to one model that attempts to account for air resistance, the distance s(t) (in feet) traveled by a 5/3 (a) falling f (x) =raindrop (1 + x 3 )satisfies 1 − x4 (b) f (x) = 0.0005 ds 2 d 2s 1 − 5x − 6x 2 = g − D dt dt 2 45.

SOLUTION

where D= is the and g = algebra 32 ft/s2 system, . Terminal velocity v 1/2 is defined as the velocity at which the (a) Let f (x) (1 +raindrop x 3 )5/3 . diameter Using a computer drop has zero acceleration (one can show that velocity approaches v 1/2 as time proceeds). 2 3 2/3 √ = 5x 2000g D. (1 + x ) ; (a) Show that v 1/2 = f (x) (b) Find v 1/2 for drops (x)diameter = 10x(10.003 + x 3and )2/30.0003 + 10x 4ft.(1 + x 3 )−1/3 ; and f of (c) In this model, do fraindrops accelerate or lower (x) = 10(1 2/3 +rapidly 6 (1 + x 3 )−4/3 . + x 3 )more 60x 3 (1at+higher x 3 )−1/3 − 10xvelocities? (b) Let f (x) =

1 − x4 . Using a computer algebra system, 1 − 5x − 6x 2 f (x) =

12x 3 − 9x 2 + 2x + 5 ; (6x − 1)2

f (x) =

2(36x 3 − 18x 2 + 3x − 31) ; and (6x − 1)3

f (x) =

1110 . (6x − 1)4

x +2 Further Let Insights . Use a CAS to compute the f (x) = and Challenges x −1 47. Find the 100th derivative of f (k) (x)?

f (k) (x) for 1 ≤ k ≤ 4. Can you find a general formula for

p(x) = (x + x 5 + x 7 )10 (1 + x 2 )11 (x 3 + x 5 + x 7 ) SOLUTION

This is a polynomial of degree 70 + 22 + 7 = 99, so its 100th derivative is zero.

49. Use the Product Rule twice to find a formula for ( f g) in terms of the first and second derivative of f and g. What is the p(99) (x) for p(x) as in Exercise 47? SOLUTION Let h = f g. Then h = f g + g f = f g + f g and h = f g + g f + f g + g f = f g + 2 f g + f g . 51. Compute Use the Product Rule to find a formula for ( f g) and compare your result with the expansion of (a + b)3 . Then f (x + h) + f (x − h) − 2 f (x) try to guess the general formula for ( f g)(n) . f (x) = lim h→0 h2 for the following functions: (a) f (x) = x (b) f (x) = x 2 Based on these examples, can you formulate a conjecture about what f is?

(c) f (x) = x 3

118

CHAPTER 3


For f (x) = x, we have f (x + h) + f (x − h) − 2 f (x) = (x + h) + (x − h) − 2x = 0.

Hence, (x) = 0. For f (x) = x 2 , f (x + h) + f (x − h) − 2 f (x) = (x + h)2 + (x − h)2 − 2x 2 = x 2 + 2xh + h 2 + x 2 − 2xh + h 2 − 2x 2 = 2h 2 , so (x 2 ) = 2. Working in a similar fashion, we find (x 3 ) = 6x. One can prove that for twice differentiable functions, f = f . It is an interesting fact of more advanced mathematics that there are functions f for which f exists at all points, but the function is not differentiable.

3.6 Trigonometric Functions Preliminary Questions 1. Determine the sign ± that yields the correct formula for the following: d (a) (sin x + cos x) = ± sin x ± cos x dx d (b) tan x = ± sec2 x dx d (c) sec x = ± sec x tan x dx d (d) cot x = ± csc2 x dx The correct formulas are d (sin x + cos x) = − sin x + cos x dx d tan x = sec2 x dx d sec x = sec x tan x dx d cot x = − csc2 x dx

SOLUTION

(a) (b) (c) (d)

2. Which of the following functions can be differentiated using the rules we have covered so far? (c) y = x 2 cos x (a) y = 3 cos x cot x (b) y = cos(x 2 ) SOLUTION

(a) 3 cos x cot x is a product of functions whose derivatives are known. This function can therefore be differentiated using the Product Rule. (b) cos(x 2 ) is a composition of the functions cos x and x 2 . We have not yet discussed how to differentiate composite functions. (c) x 2 cos x is a product of functions whose derivatives are known. This function can therefore be differentiated using the Product Rule. 3. Compute SOLUTION

d (sin2 x + cos2 x) without using the derivative formulas for sin x and cos x. dx Recall that sin2 x + cos2 x = 1 for all x. Thus, d d (sin2 x + cos2 x) = 1 = 0. dx dx

4. How is the addition formula used in deriving the formula (sin x) = cos x? SOLUTION The difference quotient for the function sin x involves the expression sin(x + h). The addition formula for the sine function is used to expand this expression as sin(x + h) = sin x cos h + sin h cos x.

S E C T I O N 3.6


Exercises In Exercises 1–4, find an equation of the tangent line at the point indicated. 1. y = sin x, SOLUTION

x=

π 4

Let f (x) = sin x. Then f (x) = cos x and the equation of the tangent line is √ √ √

π √2

π

π π 2 2 2 π x− + f = x− + = x+ 1− . y = f 4 4 4 2 4 2 2 2 4

π 3. y = tan x, x = π y = cos x, x 4= 3 SOLUTION Let f (x) = tan x. Then f (x) = sec2 x and the equation of the tangent line is y = f

π

4

x−

π

π π π + f =2 x− + 1 = 2x + 1 − . 4 4 4 2

In Exercises 5–26, use theπProduct and Quotient Rules as necessary to find the derivative of each function. y = sec x, x = 6 5. f (x) = sin x cos x SOLUTION

Let f (x) = sin x cos x. Then f (x) = sin x(− sin x) + cos x(cos x) = − sin2 x + cos2 x.

7. f (x) = sin2 x2 f (x) = x cos x SOLUTION Let f (x) = sin2 x = sin x sin x. Then f (x) = sin x(cos x) + sin x(cos x) = 2 sin x cos x. x x + 12 cot x 9. f (x)f (x) = x=3 sin 9 sec SOLUTION Let f (x) = x 3 sin x. Then f (x) = x 3 cos x + 3x 2 sin x. 11. f (θ ) = tan θ sec θ sin x f (x) = SOLUTION Let xf (θ ) = tan θ sec θ . Then

f (θ ) = tan θ sec θ tan θ + sec θ sec2 θ = sec θ tan2 θ + sec3 θ = tan2 θ + sec2 θ sec θ . 13. h(θ ) = cos2 θθ g(θ ) = SOLUTION Let h(θθ ) = cos2 θ = cos θ cos θ . Then cos h (θ ) = cos θ (− sin θ ) + cos θ (− sin θ ) = −2 cos θ sin θ . 15. f (x) = (x −2x 2 )2cot x k(θ ) = θ sin θ SOLUTION Let f (x) = (x − x 2 ) cot x. Then f (x) = (x − x 2 )(− csc2 x) + cot x(1 − 2x). sec x 17. f (x)f (z) = = 2z tan z x SOLUTION

Let f (x) =

2 sec x (x) = (x ) sec x tan x − 2x sec x = x sec x tan x − 2 sec x . . Then f x2 x4 x3

2 19. g(t) = sin t −x f (x) = cos t tan x SOLUTION

Let g(t) = sin t −

x 21. f (x) = cos y−1 sin x R(y) = + 2 sin y

2 = sin t − 2 sec t. Then g (t) = cos t − 2 sec t tan t. cos t

119

120

CHAPTER 3


Let f (x) =

x . Then 2 + sin x f (x) =

(2 + sin x) (1) − x cos x (2 + sin x)2

=

2 + sin x − x cos x (2 + sin x)2

.

1 + sin x 23. f (x)f (x) = = (x 2 − x − 4) csc x 1 − sin x 1 + sin x SOLUTION Let f (x) = . Then 1 − sin x f (x) =

(1 − sin x) (cos x) − (1 + sin x) (− cos x) (1 − sin x)2

=

2 cos x . (1 − sin x)2

sec x 25. g(x) = 1 + tan x f (x) =x 1 − tan x sec x SOLUTION Let g(x) = . Then x g (x) =

x sec x tan x − (sec x) (1) (x tan x − 1) sec x = . x2 x2

In Exercises 27–30, the second derivative. sincalculate x h(x) = 4 x++ cos x x 27. f (x) = 3 sin 4 cos SOLUTION

Let f (x) = 3 sin x + 4 cos x. Then f (x) = 3 cos x − 4 sin x and f (x) = −3 sin x − 4 cos x.

29. g(θ ) = θ sin θ f (x) = tan x SOLUTION Let g(θ ) = θ sin θ . Then, by the product rule, g (θ ) = θ cos θ + sin θ and g (θ ) = −θ sin θ + cos θ + cos θ = 2 cos θ − θ sin θ . In Exercises h(θ ) 31–36, = csc θfind an equation of the tangent line at the point specified. 31. y = x 2 + sin x, SOLUTION

x =0

Let f (x) = x 2 + sin x. Then f (x) = 2x + cos x and f (0) = 1. The tangent line at x = 0 is y = f (0) (x − 0) + f (0) = 1 (x − 0) + 0 = x.

π 33. y = 2 sin x + 3 cos x, πx = 3 y = θ tan θ , θ = √ 4 π 3 3 SOLUTION Let f (x) = 2 sin x + 3 cos x. Then f (x) = 2 cos x − 3 sin x and f 3 = 1 − 2 . The tangent line at x = π3 is √

π

π 3 3 3

π π √ x− + f = 1− x− + 3+ y= f 3 3 3 2 3 2 √ √ √ 3 3 3 3 π π− . = 1− x + 3+ + 2 2 2 3 π 35. y =y csc sinx,θ , xθ== 40 = θx2−+cot SOLUTION Let f (x) = csc x − cot x. Then f (x) = csc2 x − csc x cot x and f

π 4

=2−

√

2·1=2−

√ 2.

Hence the tangent line is

π

π

√

π π √ x− + f = 2− 2 x − + y = f 2−1 4 4 4 4

√ √ π √ 2−2 . = 2− 2 x + 2−1+ 4

S E C T I O N 3.6


121

In Exercises 37–39, using (sin x) = cos x and (cos x) = − sin x. cos θ verify the formula π y= , θ= 1 + sin θ 3 d 37. cot x = − csc2 x dx cos x . Using the quotient rule and the derivative formulas, we compute: SOLUTION cot x = sin x d d cos x sin x(− sin x) − cos x(cos x) −(sin2 x + cos2 x) −1 cot x = = = = = − csc2 x. dx d x sin x sin2 x sin2 x sin2 x d csc d x = − csc x cot x dx sec x = sec x tan x dx 1 , we can apply the quotient rule and the two known derivatives to get: SOLUTION Since csc x = sin x

39.

d − cos x cos x 1 d 1 sin x(0) − 1(cos x) = =− csc x = = = − cot x csc x. dx d x sin x sin x sin x sin2 x sin2 x 41. Calculate thevalues first five cos x. determine f (8)graph and of f (37) . sin x cos x is horizontal. where theThen tangent line to the y= Find the of xderivatives between 0of andf (x) 2π = SOLUTION Let f (x) = cos x. • Then f (x) = − sin x, f (x) = − cos x, f (x) = sin x, f (4) (x) = cos x, and f (5) (x) = − sin x. • Accordingly, the successive derivatives of f cycle among

{− sin x, − cos x, sin x, cos x} in that order. Since 8 is a multiple of 4, we have f (8) (x) = cos x. • Since 36 is a multiple of 4, we have f (36) (x) = cos x. Therefore, f (37) (x) = − sin x. 43. Calculate f (x) and f (x), where f (x) = tan x. Find y (157) , where y = sin x. SOLUTION Let f (x) = tan x. Then f (x) = sec2 x = sec x sec x f (x) = 2 sec x sec x tan x = 2 sec2 x tan x, and

f (x) = 2 sec2 x sec2 x + (tan x)(2 sec2 x tan x) = 2 sec4 x + 4 sec2 x tan2 x

sin x LetLet g(t)f (x) = t= − sinx t. for x = 0 and f (0) = 1. (a) Plot (a) Plot f (x)the ongraph [−3π ,of3πg].with a graphing utility for 0 ≤ t ≤ 4π . (b) Show that that f (c) 0 ifofc the = tangent tan c. Use numerical root and finder on this a computer (b) Show the=slope linethe is always positive verify on your algebra graph. system to find a good approximation to the smallest valuerange c0 such that f (c0 )line = 0. (c) For which values of t positive in the given is the tangent horizontal? (c) Verify that the horizontal line y = f (c0 ) is tangent to the graph of y = f (x) at x = c0 by plotting them on the same set of axes. 45.

SOLUTION

(a) Here is the graph of f (x) over [−3π , 3π ]. y 1 0.8 0.4 −5

5

−10

(b) Let f (x) =

x 10

sin x . Then x f (x) =

x cos x − sin x . x2

To have f (c) = 0, it follows that c cos c − sin c = 0, or tan c = c. Using a computer algebra system, we find that the smallest positive value c0 such that f (c0 ) = 0 is c0 = 4.493409.

122

CHAPTER 3


(c) The horizontal line y = f (c0 ) = −0.217234 and the function y = f (x) are both plotted below. The horizontal line is clearly tangent to the graph of f (x). y 1 0.8 0.4 −5

5

x

−10

10

47. The height at time t (s) of a weight, oscillating up and down at the end of a spring, is s(t) = 300 + 40 sin t cm. Find π the Show no tangent graph of f (x) = tan x has zero slope. What is the least slope of a tangent line? the velocity and that acceleration at tline = to 3 s. Justify your response by sketching the graph of (tan x) . SOLUTION Let s(t) = 300 + 40 sin t be the height. Then the velocity is v(t) = s (t) = 40 cos t and the acceleration is a(t) = v (t) = −40 sin t. √ At t = π3 , the velocity is v 3 = 20 cm/sec and the acceleration is a π3 = −20 3 cm/sec2 . π

If you stand 1 m Rfrom wall and mark off points on the wall increments δ of angular 49. θ and initial velocityelevation v0 m/s The horizontal range of aa projectile launched from ground levelatatequal an angle (Figure these sin points grow increasingly farθapart. howwill thisthe illustrates the fact or thatdecrease the derivative of tan is θ is R4),=then (v02 /9.8) θ cos θ . Calculate d R/d . If θ Explain = 7π /24, range increase if the angle is increasing. increased slightly? Base your answer on the sign of the derivative. h4

h3 h2 h1 1

FIGURE 4 SOLUTION Let D(θ ) = tan θ represent the distance from the horizontal base to the point on the wall at angular elevation θ . Based on the derivative estimate of the rate of change, for a small change δ in the angle of elevation, the change in the distance can be approximated by:

D(θ + δ ) − D(θ ) d ≈ D (θ ) = tan θ . δ dθ The change in distance resulting from a δ increase in angle is then D(θ + δ ) − D(θ ) ≈ δ

d tan θ . dθ

Knowing that D(θ + δ ) − D(θ ) increases as θ increases, it follows that D (θ ) = ddθ tan θ must increase as θ increases.

Further Insights and Challenges 51. Show that a nonzero polynomial function y = f (x) cannot satisfy the equation y = −y. Use this to prove that Usex the of the derivative and the addition law for the cosine to prove that (cos x) = − sin x. neither sin norlimit cos xdefinition is a polynomial. SOLUTION

• Let p be a nonzero polynomial of degree n and assume that p satisfies the differential equation y + y = 0. Then p + p = 0 for all x. There are exactly three cases.

(a) If n = 0, then p is a constant polynomial and thus p = 0. Hence 0 = p + p = p or p ≡ 0 (i.e., p is equal to 0 for all x or p is identically 0). This is a contradiction, since p is a nonzero polynomial. (b) If n = 1, then p is a linear polynomial and thus p = 0. Once again, we have 0 = p + p = p or p ≡ 0, a contradiction since p is a nonzero polynomial. (c) If n ≥ 2, then p is at least a quadratic polynomial and thus p is a polynomial of degree n − 2 ≥ 0. Thus q = p + p is a polynomial of degree n ≥ 2. By assumption, however, p + p = 0. Thus q ≡ 0, a polynomial of degree 0. This is a contradiction, since the degree of q is n ≥ 2.

S E C T I O N 3.7

The Chain Rule

123

CONCLUSION: In all cases, we have reached a contradiction. Therefore the assumption that p satisfies the differential equation y + y = 0 is false. Accordingly, a nonzero polynomial cannot satisfy the stated differential equation. • Let y = sin x. Then y = cos x and y = − sin x. Therefore, y = −y. Now, let y = cos x. Then y = − sin x and y = − cos x. Therefore, y = −y. Because sin x and cos x are nonzero functions that satisfy y = −y, it follows that neither sin x nor cos x is a polynomial. 53. Let f (x) = x sin x and g(x) = x cos x. Verify the following identity and use it to give another proof of the formula sin x = cos x: (a) Show that f (x) = g(x) + sin x and g (x) = − f (x) + cos x.

x and g (x) =x−g(x) − 2xsin (b) Verify that f (x) = − f (x) + 2 cos sin(x + h) − sin = 2 cos +x.12 h sin 12 h (c) By further experimentation, try to find formulas for all higher derivatives of f and g. Hint: The kth derivative depends on whether k =the 4n,addition 4n + 1,formula 4n + 2,toorprove 4n + that 3. sin(a + b) − sin(a − b) = 2 cos a sin b. Hint: Use Let f (x) = x sin x and g(x) = x cos x. (a) We examine first derivatives: f (x) = x cos x + (sin x) · 1 = g(x) + sin x and g (x) = (x)(− sin x) + (cos x) · 1 = − f (x) + cos x; i.e., f (x) = g(x) + sin x and g (x) = − f (x) + cos x. (b) Now look at second derivatives: f (x) = g (x) + cos x = − f (x) + 2 cos x and g (x) = − f (x) − sin x = −g(x) − 2 sin x; i.e., f (x) = − f (x) + 2 cos x and g (x) = −g(x) − 2 sin x. • The third derivatives are f (x) = − f (x) − 2 sin x = −g(x) − 3 sin x and g (x) = −g (x) − 2 cos x = (c) f (x) − 3 cos x; i.e., f (x) = −g(x) − 3 sin x and g (x) = f (x) − 3 cos x. • The fourth derivatives are f (4) (x) = −g (x) − 3 cos x = f (x) − 4 cos x and g (4) (x) = f (x) + 3 sin x = g(x) + 4 sin x; i.e., f (4) = f (x) − 4 cos x and g (4) (x) = g(x) + 4 sin x. • We can now see the pattern for the derivatives, which are summarized in the following table. Here n = 0, 1, 2, . . . SOLUTION

k

4n

4n + 1

4n + 2

4n + 3

f (k) (x)

f (x) − k cos x

g(x) + k sin x

− f (x) + k cos x

−g(x) − k sin x

g (k) (x)

g(x) + k sin x

− f (x) + k cos x

−g(x) − k sin x

f (x) − k cos x

Show that if π /2 < θ < π , then the distance along the x-axis between θ and the point where the tangent line intersects the x-axis is equal to |tan θ | (Figure 5).

3.7 The Chain Rule

Preliminary Questions 1. Identify the outside and inside functions for each of these composite functions. 4x + 9x 2 (b) y = tan(x 2 + 1) (c) y = sec5 x

(a) y =

SOLUTION

√ (a) The outer function is x, and the inner function is 4x + 9x 2 . (b) The outer function is tan x, and the inner function is x 2 + 1. (c) The outer function is x 5 , and the inner function is sec x. 2. Which of the following can be differentiated easily without using the Chain Rule? y = tan(7x 2 + 2), y=

√

x · sec x,

x , x +1 √ y = x cos x y=

x can be differentiated using the Quotient Rule, and the function √ x · sec x can be differThe function x+1 √ entiated using the Product Rule. The functions tan(7x 2 + 2) and x cos x require the Chain Rule. SOLUTION

3. Which is the derivative of f (5x)? (a) 5 f (x) SOLUTION

(b) 5 f (5x)

(c) f (5x)

The correct answer is (b): 5 f (5x).

4. How many times must the Chain Rule be used to differentiate each function? (b) y = cos((x 2 + 1)4 ) (a) y = cos(x 2 + 1) (c) y = cos((x 2 + 1)4 ) SOLUTION

(a) To differentiate cos(x 2 + 1), the Chain Rule must be used once. (b) To differentiate cos((x 2 + 1)4 ), the Chain Rule must be used twice.

124

CHAPTER 3


(c) To differentiate

cos((x 2 + 1)4 ), the Chain Rule must be used three times.

5. Suppose that f (4) = g(4) = g (4) = 1. Do we have enough information to compute F (4), where F(x) = f (g(x))? If not, what is missing? SOLUTION If F(x) = f (g(x)), then F (x) = f (g(x))g (x) and F (4) = f (g(4))g (4). Thus, we do not have enough information to compute F (4). We are missing the value of f (1).

Exercises In Exercises 1–4, fill in a table of the following type: f (g(x))

1. f (u) = u 3/2 ,

f (u)

f (g(x))

g (x)

( f ◦ g)

g(x) = x 4 + 1

SOLUTION

f (g(x))

f (u)

f (g(x))

g (x)

( f ◦ g)

(x 4 + 1)3/2

3 u 1/2 2

3 (x 4 + 1)1/2 2

4x 3

6x 3 (x 4 + 1)1/2

3. f (u) = tan u,3 g(x) = x 4 f (u) = u , g(x) = 3x + 5 SOLUTION

f (g(x))

f (u)

f (g(x))

g (x)

( f ◦ g)

tan(x 4 )

sec2 u

sec2 (x 4 )

4x 3

4x 3 sec2 (x 4 )

In Exercises 5–6, 4write the function as a composite f (g(x)) and compute the derivative using the Chain Rule. f (u) = u + u, g(x) = cos x 5. y = (x + sin x)4 SOLUTION

Let f (x) = x 4 , g(x) = x + sin x, and y = f (g(x)) = (x + sin x)4 . Then dy = f (g(x))g (x) = 4(x + sin x)3 (1 + cos x). dx

d 3cos u for the following choices of u(x): 7. Calculate y = cos(x dx ) 2 (b) u = x −1 (a) u = 9 − x

(c) u = tan x

SOLUTION

(a) cos(u(x)) = cos(9 − x 2 ). d cos(u(x)) = − sin(u(x))u (x) = − sin(9 − x 2 )(−2x) = 2x sin(9 − x 2 ). dx (b) cos(u(x)) = cos(x −1 ). d 1 sin(x −1 ) . cos(u(x)) = − sin(u(x))u (x) = − sin(x −1 ) − 2 = dx x x2 (c) cos(u(x)) = cos(tan x). d cos(u(x)) = − sin(u(x))u (x) = − sin(tan x)(sec2 x) = − sec2 x sin(tan x). dx In Exercises 9–14, duse the General Power Rule or the Shifting and Scaling Rule to find the derivative. Calculate f (x 2 + 1) for the following choices of f (u): 9. y = (x 2 + 9)4d x (a) f (u) = sin u (b) f (u) = 3u 3/2 SOLUTION Let g(x) = x 2 + 9. We apply the general power rule. 2 (c) f (u) = u − u d 2 d d g(x)4 = (x + 9)4 = 4(x 2 + 9)3 (x 2 + 9) = 4(x 2 + 9)3 (2x) = 8x(x 2 + 9)3 . dx dx dx y = sin5 x

S E C T I O N 3.7

11. y =

√

The Chain Rule

125

11x + 4

SOLUTION

Let g(x) = 11x + 4. We apply the general power rule.

1 d d √ 1 11 d g(x)1/2 = 11x + 4 = (11x + 4)−1/2 (11x + 4) = (11x + 4)−1/2 (11) = √ . dx dx 2 dx 2 2 11x + 4 √ Alternately, let f (x) = x and apply the shifting and scaling rule. Then 1 d √ d 11 11x + 4 = (11) f (11x + 4) = (11x + 4)−1/2 = √ . dx dx 2 2 11x + 4 13. y = sin(1 − 4x) y = (7x − 9)5 SOLUTION Let f (x) = sin x. We apply the shifting and scaling rule. d d f (1 − 4x) = sin(1 − 4x) = −4 cos(1 − 4x). dx dx In Exercises 15–18, find the derivative of f ◦ g. y = (x 4 − x + 2)−3/2 15. f (u) = sin u, g(x) = 2x + 1 SOLUTION

Let h(x) = f (g(x)) = sin(2x + 1). Then, applying the shifting and scaling rule, h (x) = 2 cos(2x + 1).

Alternately, d f (g(x)) = f (g(x))g (x) = cos(2x + 1) · 2 = 2 cos(2x + 1). dx 9 , g(x) = x + x −1 17. f (u)f (u) = u= 2u + 1, g(x) = sin x SOLUTION

Let h(x) = f (g(x)) = (x + x −1 )9 . Then, applying the general power rule, h (x) = 9(x + x −1 )8 (1 − 12 ). x

Alternately, d f (g(x)) = f (g(x))g (x) = 9(x + x −1 )8 (1 − x −2 ). dx In Exercises 19–20,u find the derivatives of f (g(x)) and g( f (x)). f (u) = , g(x) = csc x u−1 19. f (u) = cos u, g(x) = x 2 + 1 SOLUTION

d f (g(x)) = f (g(x))g (x) = − sin(x 2 + 1)(2x) = −2x sin(x 2 + 1). dx d g( f (x)) = g ( f (x)) f (x) = 2(cos x)(− sin x) = −2 sin x cos x. dx In Exercises 21–32, use the Chain 1 Rule to find the derivative. f (u) = u 3 , g(x) = x +1 21. y = sin(x 2 )

SOLUTION Let y = sin x 2 . Then y = cos x 2 · 2x = 2x cos x 2 . 23. y = cot(4t 22 + 9) y = sin x

SOLUTION Let y = cot 4t 2 + 9 . Then

y = − csc2 4t 2 + 9 · 8t = −8t csc2 4t 2 + 9 . 25. y = (t 2 + 3t + 1)−5/2 y = 1 − t2

−5/2 SOLUTION Let y = t 2 + 3t + 1 . Then y = −

y = (x 4 − x 3 − 1)2/3

−7/2 5 2 5 (2t + 3) t + 3t + 1 (2t + 3) = − 7/2 . 2 2 t 2 + 3t + 1

126

CHAPTER 3


27. y =

x +1 4 x −1

SOLUTION

Let y =

x +1 4 . Then x −1

y = 4

x + 1 3 (x − 1) · 1 − (x + 1) · 1 8 (x + 1)3 8(1 + x)3 · =− = . 2 5 x −1 (1 − x)5 (x − 1) (x − 1)

29. y = cos3 (θ 21)

3 y = sec SOLUTION Letx y = cos3 θ 2 = cos θ 2 . Here, we note that calculating the derivative of the inside function, cos(θ 2 ), requires the chain rule. After two applications of the chain rule, we have

2

y = 3 cos θ 2 − sin θ 2 · 2θ = −6θ sin θ 2 cos2 θ 2 . 1 31. y =y = tan(θ + cos θ ) cos(x 2 ) + 1

−1/2 SOLUTION Let y = 1 + cos x 2 . Here, we note that calculating the derivative of the inside function, 1 + cos(x 2 ), requires the chain rule. After two applications of the chain rule, we have

−3/2

x sin x 2 1 y = − 1 + cos x 2 · − sin x 2 · (2x) = 3/2 . 2 1 + cos x 2

√ In Exercises 33–62, find the derivative using the appropriate rule or combination of rules. y = ( x + 1 − 1)3/2 33. y = tan 5x SOLUTION

Let y = tan 5x. By the scaling and shifting rule, dy = 5 sec2 (5x). dx

35. y = x cos(1 − 3x) y = sin(x 2 + 4x) SOLUTION Let y = x cos (1 − 3x). Applying the product rule and then the scaling and shifting rule, y = x (− sin (1 − 3x)) · (−3) + cos (1 − 3x) · 1 = 3x sin (1 − 3x) + cos (1 − 3x) . 37. y = (4t + 9)1/2 y = sin(x 2 ) cos(x 2 ) SOLUTION Let y = (4t + 9)1/2 . By the shifting and scaling rule, dy 1 =4 (4t + 9)−1/2 = 2(4t + 9)−1/2 . dt 2 3 + cos x)−4 39. y =y (x = sin(cos θ) SOLUTION Let y = (x 3 + cos x)−4 . By the general power rule,

y = −4(x 3 + cos x)−5 (3x 2 − sin x) = 4(sin x − 3x 2 )(x 3 + cos x)−5 . √ 41. y =y =sin x cos x x)) sin(cos(sin SOLUTION We start by using a trig identity to rewrite √ 1 1 y = sin x cos x = sin 2x = √ (sin 2x)1/2 . 2 2 Then, after two applications of the chain rule, 1 1 cos 2x . y = √ · (sin 2x)−1/2 · cos 2x · 2 = √ 2 sin 2x 2 2 y = x 2 tan 2x

S E C T I O N 3.7

The Chain Rule

127

43. y = (z + 1)4 (2z − 1)3 SOLUTION

Let y = (z + 1)4 (2z − 1)3 . Applying the product rule and the general power rule,

dy = (z + 1)4 (3(2z − 1)2 )(2) + (2z − 1)3 (4(z + 1)3 )(1) = (z + 1)3 (2z − 1)2 (6(z + 1) + 4(2z − 1)) dz = (z + 1)3 (2z − 1)2 (14z + 2). √ 45. y = (x + x −1 )√ x + 1 y = 3 + 2s s √ SOLUTION Let y = (x + x −1 ) x + 1. Applying the product rule and the shifting and scaling rule, and then factoring, we get: √ 1 1 (x + x −1 + 2(x + 1)(1 − x −2 )) y = (x + x −1 ) (x + 1)−1/2 + x + 1(1 − x −2 ) = √ 2 2 x +1 =

2x 2

1 1 √ √ (x 3 + x + 2x 3 + 2x 2 − 2x − 2) = (3x 3 + 2x 2 − x − 2). 2 x +1 2x x + 1

47. y = (cos 6x2 + sin x 2 )1/2 y = cos (8x) SOLUTION Let y = (cos 6x + sin(x 2 ))1/2 . Applying the general power rule followed by both the scaling and shifting rule and the chain rule, y =

−1/2 x cos(x 2 ) − 3 sin 6x 1 − sin 6x · 6 + cos(x 2 ) · 2x = . cos 6x + sin(x 2 ) 2 cos 6x + sin(x 2 )

3) 49. y = tan3 x + tan(x (x + 1)1/2 y = Let y = tan3 x + tan(x 3 ) = (tan x)3 + tan(x 3 ). Applying the general power rule to the first term and the SOLUTION x +2 chain rule to the second term, y = 3(tan x)2 sec2 x + sec2 (x 3 ) · 3x 2 = 3 x 2 sec2 (x 3 ) + sec2 x tan2 x .

z√ +1 51. y =y = 4 − 3 cos x z−1 z + 1 1/2 SOLUTION Let y = . Applying the general power rule followed by the quotient rule, z−1 dy 1 = dz 2

−1 z + 1 −1/2 (z − 1) · 1 − (z + 1) · 1 · = √ . 2 z−1 z + 1 (z − 1)3/2 (z − 1)

cos(1 + x) 53. y =y = (cos3 x + 3 cos x + 7)9 1 + cos x SOLUTION

Let y=

cos(1 + x) . 1 + cos x

Then, applying the quotient rule and the shifting and scaling rule, dy −(1 + cos x) sin(1 + x) + cos(1 + x) sin x cos(1 + x) sin x − cos x sin(1 + x) − sin(1 + x) = = dx (1 + cos x)2 (1 + cos x)2 =

sin(−1) − sin(1 + x) . (1 + cos x)2

The last line follows from the identity sin( A − B) = sin A cos B − cos A sin B with A = x and B = 1 + x. ) 55. y = cot7 (x 5 y = sec( t 2 − 9)

128

CHAPTER 3


Let y = cot7 x 5 . Applying the general power rule followed by the chain rule,

dy = 7 cot6 x 5 · − csc2 x 5 · 5x 4 = −35x 4 cot6 x 5 csc2 x 5 . dx

57. y = (1 + (x 2 2+ 2)5 )3 cos(x ) y = Let y2 = (1 + (x 2 + 2)5 )3 . Then, applying the general power rule twice, we obtain: SOLUTION 1+x dy = 3(1 + (x 2 + 2)5 )2 (5(x 2 + 2)4 (2x)) = 30x(1 + (x 2 + 2)5 )2 (x 2 + 2)4 . dx

√

59. y =y =1 +1 +1cot +5 (xx4 + 1) 9 1/2 1/2

SOLUTION Let y = 1 + 1 + x 1/2 . Applying the general power rule twice,

1/2 −1/2 1

−1/2 1 dy 1 1 + x 1/2 · · x −1/2 = = 1 + 1 + x 1/2 dx 2 2 2

1

. √ √ √ 8 x 1+ x 1+ 1+ x

√ 61. y = kx + √ b; k and b any constants x +1+1 y= SOLUTION Let y = (kx + b)1/2 , where b and k are constants. By the scaling and shifting rule, y =

1 k . (kx + b)−1/2 · k = √ 2 2 kx + b

df df du 63. Compute 1if = 2 and = 6. du; k, b constants, dx not both zero y = d x kt 4 + b SOLUTION Assuming f is a function of u, which is in turn a function of x, df d f du = · = 2(6) = 12. dx du d x 65. With notation as in Example 5, calculate √ molecular velocity v of a gas in a certain container is given by v = 29 T m/s, where T is the d The average d (a) sin θ (b) θ + tan θ dv dθ d θ (in atmospheres) . temperature in ◦kelvins. The temperature is related to the pressure ◦ T = 200P. Find θ =60 θ =45by d P P=1.5 SOLUTION

(a)

π

π

π d d π 1 π sin θ = sin θ = cos (60) = = . θ =60◦ d θ θ =60◦ dθ 180 180 180 180 2 360

(b)

π

π d

π π d θ + tan θ = θ + tan θ =1+ sec2 =1+ . ◦ ◦ θ =45 θ =45 dθ dθ 180 180 4 90

67. Compute the derivative of h(sin x) at x = π6 , assuming that h (0.5) = 10. Assume that f (0) = 2 and f (0) = 3. Find the derivatives of ( f (x))3 and f (7x) at x = 0. SOLUTION Let u = sin x and suppose that h (0.5) = 10. Then d dh du dh = cos x. (h(u)) = dx du d x du √ When x = π6 , we have u = .5. Accordingly, the derivative of h(sin x) at x = π6 is 10 cos π6 = 5 3. In Exercises 69–72, the table of values to calculate function at the given and f (g(2)) from the Let F(x) = use f (g(x)), where the graphs of f andthe g derivative are shown of in the Figure 1. Estimate g (2) point. graph and compute F (2).

69. f (g(x)),

x =6

x

1

4

6

f (x) f (x) g(x) g (x)

4 5 4 5

0 7 1

6 4 6 3

1 2

S E C T I O N 3.7

SOLUTION

The Chain Rule

129

d = f (g(6))g (6) = f (6)g (6) = 4 × 3 = 12. f (g(x)) dx x=6

√ 71. g( x), x = 16 sin( f (x)), x =4 √ 1 1 d √ 1 1 1 g( x) . SOLUTION = g (4) (1/ 16) = = dx 2 2 2 4 16 x=16 In Exercises compute f (2x73–76, + g(x)), x = the 1 indicated higher derivatives. d2 sin(x 2 ) dx2

73.

SOLUTION

Let f (x) = sin x 2 . Then, by the chain rule, f (x) = 2x cos x 2 and, by the product rule and the chain

rule,

f (x) = 2x − sin x 2 · 2x + 2 cos x 2 = 2 cos x 2 − 4x 2 sin x 2 . d3 2 d(3x + 9)11 d x 3 2 (x 2 + 9)5 dx SOLUTION Let f (x) = (3x + 9)11 . Then, by repeated use of the scaling and shifting rule,

75.

f (x) = 11(3x + 9)10 · 3 = 33(3x + 9)10 f (x) = 330(3x + 9)9 · 3 = 990(3x + 9)9 , f (x) = 8910(3x + 9)8 · 3 = 26730 (3x + 9)8 . Use a computer algebra system to compute f (k) (x) for k = 1, 2, 3 for the following functions: d3 2 sin(2x) (a) f (x) (b) f (x) = x 3 + 1 d x=3 cot(x )

77.

SOLUTION

(a) Let f (x) = cot(x 2 ). Using a computer algebra system, f (x) = −2x csc2 (x 2 ); f (x) = 2 csc2 (x 2 )(4x 2 cot(x 2 ) − 1); and

f (x) = −8x csc2 (x 2 ) 6x 2 cot2 (x 2 ) − 3 cot(x 2 ) + 2x 2 . (b) Let f (x) =

x 3 + 1. Using a computer algebra system, 3x 2 ; f (x) = 2 x3 + 1 f (x) =

3x(x 3 + 4) ; and 4(x 3 + 1)3/2

f (x) = −

3(x 6 + 20x 3 − 8) . 8(x 3 + 1)5/2

79. Compute the second derivative of sin(g(x)) at x = 2, assuming that g(2) = π4 , g (2) = 5, and g (2) = 3. Use the Chain Rule to express the second derivative of f ◦ g in terms of the first and second derivatives of f and g. SOLUTION Let f (x) = sin(g(x)). Then f (x) = cos(g(x))g (x) and f (x) = cos(g(x))g (x) + g (x)(− sin(g(x)))g (x) = cos(g(x))g (x) − (g (x))2 sin(g(x)). Therefore, √ √ 2 f (2) = g (2) cos (g(2)) − g (2) sin (g(2)) = 3 cos π4 − (5)2 sin π4 = −22 · 22 = −11 2

where is at resistance i the current. d P/dt at t volume. = 2 if RFind = 1,000 81. TheAn power P in a sphere circuit is = Ri 2r, = expanding hasPradius 0.4t Rcm time t (inand seconds). Let V Find be the sphere’s d V /dt and iwhen varies according to i = sin(4 π t) (time in seconds). (a) r = 3 and (b) t = 3. di d 2 Ri SOLUTION = 2Ri = 2(1000)4π sin(4π t) cos(4π t)|t=2 = 0. dt dt t=2 t=2 The price (in dollars) of a computer component is P = 2C − 18C −1 , where C is the manufacturer’s cost to produce it. Assume that cost at time t (in years) is C = 9 + 3t −1 and determine the ROC of price with respect to

130

CHAPTER 3


83. The force F (in Newtons) between two charged objects is F = 100/r 2 , where r is the distance (in meters) between them. Find d F/dt at t = 10 if the distance at time t (in seconds) is r = 1 + 0.4t 2 . Let F(r ) = 100/r 2 , and let r (t) = 1 + .4t 2 . We compute F (r ) = −200/r 3 , and r (t) = .8t. This gives us r (10) = 41 and r (10) = 8. SOLUTION

dF d F dr = , dt dr dt so

d F = −200(41)−3 (8) ≈ −0.0232 N/s. dt t=10

85. Conservation of Energy The position at time t (in seconds) of a weight of mass m oscillating at the end of a According to the U.S. standard atmospheric model, developed by the National Oceanic and Atmospheric Adspring is x(t) = L sin(2πω t). Here, L is the maximum length of the spring, and ω the frequency (number of oscillations ministration for use in aircraft and rocket design, atmospheric temperature T (in degrees Celsius), pressure P per second). Let v and a be the velocity and acceleration of the weight. (kPa = 1,000 Pascals), and altitude h (meters) are related by the formulas (valid in the troposphere h ≤ 11,000): (a) By Hooke’s Law, the spring exerts a force of magnitude where k is the spring constant. √ F = −kx onthe weight, Use Newton’s Second Law, F = ma, to show that 2πω = k/m. T + 273.1 5.256 T = 15.04 0.000649h, = 101.29 1 mv 2 and potential Penergy U =+12 kx 2 .288.08 Prove that the total energy E = K + U (b) The weight has kinetic energy K =− 2 is conserved, that is, d E/dt = 0. Calculate d P/dh. Then estimate the change in P (in Pascals, Pa) per additional meter of altitude when h = 3,000. SOLUTION

(a) Let x(t) = L sin(2πω t). Then v(t) = x (t) = 2πω L cos(2πω t) and a(t) = v (t) = −(2πω )2 L sin(2πω t). Equating Hooke’s Law, F = −kx, to Newton’s second law, F = ma, yields −k L sin(2πω t) = −m(2πω )2 L sin(2πω t) or k/m = (2πω )2 . The desired result follows upon taking square roots. (b) We have dx dU = kx = kxv dt dt and dK dv = mv = mva dt dt But F = ma = −kx by Hooke’s Law, so a = (−k/m)x and we get dK = mva = mv(−k/m)x = −kvx. dt This shows that the derivative of U + K is zero.

Further Insights and Challenges 87. if f (−x) = fthen (x) and odd if f (−x) = − f (x). ShowRecall that ifthat f , g,f (x) andish even are differentiable, (a) Sketch a graph of any even function and explain graphically why its derivative is an odd function. [ f (g(h(x)))] = f (g(h(x)))g (h(x))h (x) (b) Show that differentiation reverses parity: If f is even, then f is odd, and if f is odd, then f is even. Hint: Differentiate f (−x). (c) Suppose that f is even. Is f necessarily odd? Hint: Check if this is true for linear functions. A function is even if f (−x) = f (x) and odd if f (−x) = − f (x). (a) The graph of an even function is symmetric with respect to the y-axis. Accordingly, its image in the left half-plane is a mirror reflection of that in the right half-plane through the y-axis. If at x = a ≥ 0, the slope of f exists and is equal to m, then by reflection its slope at x = −a ≤ 0 is −m. That is, f (−a) = − f (a). Note: This means that if f (0) exists, then it equals 0. SOLUTION

y 4 3 2 1 −2

−1

x 1

2

S E C T I O N 3.7

The Chain Rule

131

(b) By the chain rule, ddx f (−x) = − f (−x). Now suppose that f is even. Then f (−x) = f (x) and d d f (−x) = f (x) = f (x). dx dx Hence, when f is even, − f (−x) = f (x) or f (−x) = − f (x) and f is odd. On the other hand, suppose f is odd. Then f (−x) = − f (x) and d d f (−x) = − f (x) = − f (x). dx dx Hence, when f is odd, − f (−x) = − f (x) or f (−x) = f (x) and f is even. (c) Suppose that f is even. Then f is not necessarily odd. Let f (x) = 4x + 7. Then f (x) = 4, an even function. But f is not odd. For example, f (2) = 15, f (−2) = −1, but f (−2) = − f (2). In Exercises 89–91, use the following fact (proved in Chapter 4):q If a differentiable function f satisfies f (x) = 0 for all Power Rule for Fractional Exponents Let f (u) = u and g(x) = x p/q . Show that f (g(x)) = x p (recall the x, then f is a constant function. laws of exponents). Apply the Chain Rule and the Power Rule for integer exponents to show that f (g(x)) g (x) = px p−1 . ThenEquation derive theofPower RuleCosine for x p/qSuppose . 89. Differential Sine and that f (x) satisfies the following equation (called a differential equation): f (x) = − f (x)

3

(a) Show that f (x)2 + f (x)2 = f (0)2 + f (0)2 . Hint: Show that the function on the left has zero derivative. (b) Verify that sin x and cos x satisfy Eq. (3) and deduce that sin2 x + cos2 x = 1. SOLUTION

(a) Let g(x) = f (x)2 + f (x)2 . Then g (x) = 2 f (x) f (x) + 2 f (x) f (x) = 2 f (x) f (x) + 2 f (x)(− f (x)) = 0, where we have used the fact that f (x) = − f (x). Because g (0) = 0 for all x, g(x) = f (x)2 + f (x)2 must be a constant function. In other words, f (x)2 + f (x)2 = C for some constant C. To determine the value of C, we can substitute any number for x. In particular, for this problem, we want to substitute x = 0 and find C = f (0)2 + f (0)2 . Hence, f (x)2 + f (x)2 = f (0)2 + f (0)2 . (b) Let f (x) = sin x. Then f (x) = cos x and f (x) = − sin x, so f (x) = − f (x). Next, let f (x) = cos x. Then f (x) = − sin x, f (x) = − cos x, and we again have f (x) = − f (x). Finally, if we take f (x) = sin x, the result from part (a) guarantees that sin2 x + cos2 x = sin2 0 + cos2 0 = 0 + 1 = 1. 91. Use the result of Exercise 90 to show the following: f (x) = sin x is the unique solution of Eq. (3) such that f (0) = 0 = g (0). Suppose thatg(x) both=f cos andxgissatisfy Eq. (3) and have thethat same initial that=is,0.f This (0) =result g(0)provides and f (0) = 1; and the unique solution such g(0) = 1values, and g (0) a means of and f (0) Show that f (x) = g(x) for all x. Hint: Show that f − g satisfies Eq. (3) and apply Exercise 89(a). defining and developing all the properties of the trigonometric functions without reference to triangles or the unit circle. SOLUTION In part (b) of Exercise 89, it was shown that f (x) = sin x satisfies Eq. (3), and we can directly calculate that f (0) = sin 0 = 0 and f (0) = cos 0 = 1. Suppose there is another function, call it F(x), that satisfies Eq. (3) with the same initial conditions: F(0) = 0 and F (0) = 1. By Exercise 90, it follows that F(x) = sin x for all x. Hence, f (x) = sin x is the unique solution of Eq. (3) satisfying f (0) = 0 and f (0) = 1. The proof that g(x) = cos x is the unique solution of Eq. (3) satisfying g(0) = 1 and g (0) = 0 is carried out in a similar manner.

93. Chain Rule This exercise proves the Chain Rule without the special assumption made in the text. For any number lim g (x), where b, defineAaDiscontinuous new function Derivative Use the limit definition to show that g (0) exists but g (0) = x→0 f (u) ⎧ − f (b) 1 ⎪ F(u) = for all u = b ⎨ 2 u −xb sin x if x = 0 g(x) = ⎪ Observe that F(u) is equal to the slope of the secant line⎩through (b,iff x(b)) 0 = 0and (u, f (u)). (a) Show that if we define F(b) = f (b), then F(u) is continuous at u = b. (b) Take b = g(a). Show that if x = a, then for all u, u − g(a) f (u) − f (g(a)) = F(u) x −a x −a Note that both sides are zero if u = g(a).

4

132

CHAPTER 3


(c) Substitute u = g(x) in Eq. (4) to obtain g(x) − g(a) f (g(x)) − f (g(a)) = F(g(x)) x −a x −a Derive the Chain Rule by computing the limit of both sides as x → a. SOLUTION

For any differentiable function f and any number b, define F(u) =

f (u) − f (b) u−b

for all u = b. (a) Define F(b) = f (b). Then lim F(u) = lim

u→b

u→b

f (u) − f (b) = f (b) = F(b), u−b

i.e., lim F(u) = F(b). Therefore, F is continuous at u = b. u→b

(b) Let g be a differentiable function and take b = g(a). Let x be a number distinct from a. If we substitute u = g(a) into Eq. (4), both sides evaluate to 0, so equality is satisfied. On the other hand, if u = g(a), then f (u) − f (g(a)) f (u) − f (g(a)) u − g(a) f (u) − f (b) u − g(a) u − g(a) = = = F(u) . x −a u − g(a) x −a u−b x −a x −a (c) Hence for all u, we have u − g(a) f (u) − f (g(a)) = F(u) . x −a x −a (d) Substituting u = g(x) in Eq. (4), we have g(x) − g(a) f (g(x)) − f (g(a)) = F(g(x)) . x −a x −a Letting x → a gives f (g(x)) − f (g(a)) g(x) − g(a) = lim F(g(x)) = F(g(a))g (a) = F(b)g (a) = f (b)g (a) x→a x→a x −a x −a lim

= f (g(a))g (a) Therefore ( f ◦ g) (a) = f (g(a))g (a), which is the Chain Rule.

3.8 Implicit Differentiation Preliminary Questions 1. Which differentiation rule is used to show SOLUTION

d dy sin y = cos y ? dx dx

The chain rule is used to show that ddx sin y = cos y dd xy .

2. One of (a)–(c) is incorrect. Find the mistake and correct it. (a)

d sin(y 2 ) = 2y cos(y 2 ) dy

d sin(x 2 ) = 2x cos(x 2 ) dx d sin(y 2 ) = 2y cos(y 2 ) (c) dx

(b)

SOLUTION

(a) This is correct. Note that the differentiation is with respect to the variable y. (b) This is correct. Note that the differentiation is with respect to the variable x. (c) This is incorrect. Because the differentiation is with respect to the variable x, the chain rule is needed to obtain d dy sin(y 2 ) = 2y cos(y 2 ) . dx dx

S E C T I O N 3.8

Implicit Differentiation

133

3. On an exam, Jason was asked to differentiate the equation x 2 + 2x y + y 3 = 7 What are the errors in Jason’s answer? 2x + 2x y + 3y 2 = 0 There are two mistakes in Jason’s answer. First, Jason should have applied the product rule to the second

SOLUTION

term to obtain d dy (2x y) = 2x + 2y. dx dx Second, he should have applied the general power rule to the third term to obtain dy d 3 y = 3y 2 . dx dx 4. Which of (a) or (b) is equal to (a) (x cos t) SOLUTION

d (x sin t)? dx

dt dt (b) (x cos t) + sin t dx dx Using the product rule and the chain rule we see that dt d (x sin t) = x cos t + sin t, dx dx

so the correct answer is (b).

Exercises 1. Show that if you differentiate both sides of x 2 + 2y 3 = 6, the result is 2x + 6y 2 y = 0. Then solve for y and calculate d y/d x at the point (2, 1). SOLUTION

d d 2 (x + 2y 3 ) = 6 dx dx 2x + 6y 2 y = 0 2x + 6y 2 y = 0 6y 2 y = −2x y =

−2x . 6y 2

2 At (2, 1), dd xy = −4 6 = −3.

In Exercises the expression withofrespect to x. x y + 3x + y = 1, the result is (x + 1)y + y + 3 = 0. Show3–8, thatdifferentiate if you differentiate both sides the equation Then solve for y and calculate d y/d x at the point (1, −1). 3. x 2 y 3 SOLUTION

Assuming that y depends on x, then d 2 3 x y = x 2 · 3y 2 y + y 3 · 2x = 3x 2 y 2 y + 2x y 3 . dx

5.

y3 3 x x y

SOLUTION

Assuming that y depends on x, then d −y 3 3y 2 y y3 x(3y 2 y ) − y 3 = 2 + = . 2 dx x x x x

7. z +tan(xt) z2 SOLUTION

Assuming that z depends on x, then

(x 2 + y 2 )3/2

d (z + z 2 ) = z + 2z(z ) dx

134

CHAPTER 3


In Exercises 9–24, calculate the derivative of y (or other variable) with respect to x. 9. 3y 3 + x 2 = 5 SOLUTION

2x Let 3y 3 + x 2 = 5. Then 9y 2 y + 2x = 0, and y = − 2 . 9y

11. x 2 y + 2x y 2 = x + y y4 − y = x 3 + x SOLUTION Let x 2 y + 2x y 2 = x + y. Then x 2 y + 2x y + 2x · 2yy + 2y 2 = 1 + y x 2 y + 4x yy − y = 1 − 2x y − 2y 2 y =

1 − 2x y − 2y 2 . x 2 + 4x y − 1

13. x 2 ysin(xt) + y 4 −=3x t =8 SOLUTION We apply the product rule and the chain rule together: d d 2 (x y + y 4 − 3x) = 8 dx dx x 2 y + 2x y + 4y 3 y − 3 = 0 x 2 y + 4y 3 y = 3 − 2x y y (x 2 + 4y 3 ) = 3 − 2x y 3 − 2x y . y = 2 x + 4y 3 15. x 4 +3z 45= 1 x R =1 SOLUTION Let x 4 + z 4 = 1. Then 4x 3 + 4z 3 z = 0, and z = −x 3 /z 3 . 3/2 = 1 17. y −3/2 y +xx + = 2y x yLet y −3/2 + x 3/2 = 1. Then − 32 y −5/2 y + 32 x 1/2 = 0, and y = x 1/2 y 5/2 = x y 5 . SOLUTION

19.

√

1 1 1/2 x x+ s= +=x+y + yx2/3 s

SOLUTION

Let (x + s)1/2 = x −1 + s −1 . Then 1 (x + s)−1/2 1 + s = −x −2 − s −2 s . 2

√ Multiplying by 2x 2 s 2 x + s and then solving for s gives √ √ x 2 s 2 1 + s = −2s 2 x + s − 2x 2 s x + s √ √ x 2 s 2 s + 2x 2 s x + s = −2s 2 x + s − x 2 s 2

√ √ x 2 s 2 + 2 x + s s = −s 2 x 2 + 2 x + s

√ s2 x 2 + 2 x + s . s = − 2 2 √ x s +2 x +s x 1 √ 21. y +√x = y + y + y = 2x SOLUTION

Let y + x y −1 = 1. Then y + x −y −2 y + y −1 · 1 = 0, and y =

23. sin(x + y) = x + cos y y2 x + =0 y x +1

y −1 y y2 = . · x y −2 − 1 y 2 x − y2



135

Let sin(x + y) = x + cos y. Then (1 + y ) cos(x + y) = 1 − y sin y cos(x + y) + y cos(x + y) = 1 − y sin y (cos(x + y) + sin y) y = 1 − cos(x + y) y =

1 − cos(x + y) . cos(x + y) + sin y

In Exercises 25–26, find d y/d x at the given point. tan(x 2 y) = x + y 25. (x + 2)2 − 6(2y + 3)2 = 3, SOLUTION

(1, −1)

By the scaling and shifting rule, 2(x + 2) − 24(2y + 3)y = 0.

If x = 1 and y = −1, then 2(3) − 24(1)y = 0. so that 24y = 6, or y = 14 . In Exercises 27–30, find anequation 2 − π πof the tangent line at the given point. , sin2 (3y) = x + y; 4 4 27. x y − 2y = 1, (3, 1) SOLUTION

Taking the derivative of both sides of x y − 2y = 1 yields x y + y − 2y = 0. Substituting x = 3, y = 1

yields 3y + 1 − 2y = 0. Solving, we get:

3y + 1 − 2y = 0 y + 1 = 0 y = −1. Hence, the equation of the tangent line at (3, 1) is y − 1 = −(x − 3), or y = 4 − x. 29. x 2/3 2+ 3y 2/3 = 2, (1, 1) x y + 2y = 3x, (2, 1) SOLUTION Taking the derivative of both sides of x 2/3 + y 2/3 = 2 yields 2 −1/3 2 −1/3 + y y = 0. x 3 3 Substituting x = 1, y = 1 yields 23 + 23 y = 0, so that 1 + y = 0, or y = −1. Hence, the equation of the tangent line at (1, 1) is y − 1 = −(x − 1), or y = 2 − x. 31. Find 1/2 the points on the graph of y 2 = x 3 − 3x + 1 (Figure 5) where the tangent line is horizontal. + y −1/2 = 2x y, 2 (1, 1) x (a) First show that 2yy = 3x − 3, where y = d y/d x. (b) Do not solve for y . Rather, set y = 0 and solve for x. This gives two possible values of x where the slope may be zero. (c) Show that the positive value of x does not correspond to a point on the graph. (d) The negative value corresponds to the two points on the graph where the tangent line is horizontal. Find the coordinates of these two points. y 2

x

−2 −1

1

2

−2

FIGURE 5 Graph of y 2 = x 3 − 3x + 1. SOLUTION

136

CHAPTER 3


(a) Applying implicit differentiation to y 2 = x 3 − 3x + 1, we have 2y

dy = 3x 2 − 3. dx

(b) Setting y = 0 we have 0 = 3x 2 − 3, so x = 1 or x = −1. (c) If we return to the equation y 2 = x 3 − 3x + 1 and substitute x = 1, we obtain the equation y 2 = −1, which has no real solutions. (d) Substituting x = −1 into y 2 = x 3 − 3x + 1 yields y 2 = (−1)3 − 3(−1) + 1 = −1 + 3 + 1 = 3, √ √ √ √ so y = 3 or − 3. The tangent is horizontal at the points (−1, 3) and (−1, − 3). 33. Show that no point on the graph of x 22 − 3x y2 + y 2 = 1 has a horizontal tangent line. Find all points on the graph of 3x + 4y + 3x y = 24 where the tangent line is horizontal (Figure 6). SOLUTION Let the implicit curve x 2 − 3x y + y 2 = 1 be given. Then (a) By differentiating the equation of the curve implicitly and setting y = 0, show that if the tangent line is horizontal at (x, y), then y = −2x. 2x − 3x y − 3y + 2yy = 0, (b) Solve for x by substituting y = −2x in the equation of the curve. so y =

2x − 3y . 3x − 2y

Setting y = 0 leads to y = 23 x. Substituting y = 23 x into the equation of the implicit curve gives 2 2 2 = 1, x + x x 2 − 3x 3 3 or − 59 x 2 = 1, which has no real solutions. Accordingly, there are no points on the implicit curve where the tangent line has slope zero. 35. If the derivative d x/d y exists at a point and d x/d y = 0, then the tangent line is vertical. Calculate d x/d y for the 4 + x y = x 3 − x + 2. Find d y/d x at the two points on the graph with x-coordinate thex 2graph of ythe equationFigure y 4 + 11 shows = y2 + and find points on the graph where the tangent line is vertical. 0 and find an equation of 2the tangent line at (1, 1). 4 2 SOLUTION Let y + 1 = y + x . Differentiating this equation with respect to y yields 4y 3 = 2y + 2x

dx , dy

so 4y 3 − 2y y(2y 2 − 1) dx = = . dy 2x x √ 2 dx Thus, = 0 when y = 0 and when y = ± . Substituting y = 0 into the equation y 4 + 1 = y 2 + x 2 gives dy 2 √ √ 2 3 1 = x 2 , so x = ±1. Substituting y = ± , gives x 2 = 3/4, so x = ± . Thus, there are six points on the graph of 2 2 4 2 2 y + 1 = y + x where the tangent line is vertical: √ √ √ √ √ √ √ √ 3 2 3 2 3 2 3 2 (1, 0), (−1, 0), , , − , , ,− , − ,− . 2 2 2 2 2 2 2 2 37. Differentiate the equation x 3 + 3x y 2 = 1 with respect to the variable t and express d y/dtd yin termsy of d xd x/dt, as in =− . the equation x y = 1 with respect to the variable t and derive the relation ExerciseDifferentiate 36. dt x dt 3 2 SOLUTION Let x + 3x y = 1. Then 3x 2

dy dx dx + 6x y + 3y 2 = 0, dt dt dt

and x 2 + y2 d x dy =− . dt 2x y dt In Exercises 38–39, differentiate the equation with respect to t to calculate d y/dt in terms of d x/dt. x 2 − y2 = 1

S E C T I O N 3.8


137

39. y 4 + 2x y + x 2 = 0 SOLUTION

Let y 4 + 2x y + x 2 = 0. Then 4y 3

dy dx dx dy + 2x + 2y + 2x = 0, dt dt dt dt

and x + y dx dy =− . dt x + 2y 3 dt discussed in 1638 by 41. The folium of Descartes is the curve with equation x 3 + y 3 = 3x y (Figure 7). It was first The volume V and pressure P of gas in a piston (which vary in time t) satisfy P V 3/2 = C, where C is a the French philosopher-mathematician René Descartes. The name “folium” means leaf. Descartes’s scientific colleague constant. Prove that Gilles de Roberval called it the “jasmine flower.” Both men believed incorrectly that the leaf shape in the first quadrant was repeated in each giving the appearancedof petals of3a flower. Find an equation of the tangent line to this P P/dt quadrant, =− curve at the point 23 , 43 . d V /dt 2 V The ratio of the derivatives is negative. Could you havey predicted this from the relation P V 3/2 = C? 2

x

−2

2

−2

FIGURE 7 Folium of Descartes: x 3 + y 3 = 3x y. SOLUTION

x2 − y 2 , 4 , we have Let x 3 + y 3 = 3x y. Then 3x 2 + 3y 2 y = 3x y + 3y, and y = . At the point 3 3 x − y2 4 − 4 − 89 4 3 y = 29 16 = 10 = . 5 −9 3 − 9

The tangent line at P is thus y − 43 = 45 x − 23 or y = 45 x + 45 .

43. Plot the equation x 3 + y 3 = 3x y + b for several values of b. Find all points on the folium x 3 + y 3 = 3x y at which the tangent line is horizontal. (a) Describe how the graph changes as b → 0. (b) Compute d y/d x (in terms of b) at the point on the graph where y = 0. How does this value change as b → ∞? Do your plots confirm this conclusion? SOLUTION

(a) Consider the first row of figures below. When b < 0, the graph of x 3 + y 3 = 3x y + b consists of two pieces. As b → 0−, the two pieces move closer to intersecting at the origin. From the second row of figures, we see that the graph of x 3 + y 3 = 3x y + b when b > 0 consists of a single piece that has a “loop” in the first quadrant. As b → 0+, the loop comes closer to “pinching off” at the origin. b = −0.1

b = − 0.001

b = − 0.01

y

y

y

1.5

1.5

1.5

1

1

1

0.5

0.5

0.5

x

− 0.5

0.5 − 0.5

1

1.5

x

− 0.5

0.5 − 0.5

1

1.5

x

− 0.5

0.5 − 0.5

1

1.5

138

CHAPTER 3

D I F F E R E N T I AT I O N b = 0.1

b = 0.001

b = 0.01

y

y

y

1.5

1.5

1.5

1

1

1

0.5

0.5

0.5

x

− 0.5

0.5

1

1.5

x

−0.5

− 0.5

0.5

1

x

−0.5

1.5

0.5

1

1.5

−0.5

−0.5

(b) Differentiating the equation x 3 + y 3 = 3x y + b with respect to x yields 3x 2 + 3y 2 y = 3x y + 3y, so y − x2 y = 2 . y −x Substituting y = 0 into x 3 + y 3 = 3x y + b leads to x 3 = b, or x =

√ 3

b. Thus, at the point on the graph where y = 0,

√ 0 − x2 3 y = 2 = x = b. 0 −x Consequently, as b → ∞, y → ∞ at the point on the graph where y = 0. This conclusion is supported by the figures shown below, which correspond to b = 1, b = 10, and b = 100.

−4

b = 100 y

b = 01 y

b = 10 y

4

4

4

2

2

2

x

−2

2

4

−4

x

−2

2

−4

4

x

−2

2

−2

−2

−2

−4

−4

−4

4

√ a trident curve (Figure 9), named by Isaac Newton in his treatise on 45. The equation x y = x 3 − 5x 2 + 2x − 1 defines Show thatin the1710. tangent lines x = 1where ± 2the to the so-called 8) with equation (x − 1)2 (x 2 + y 2 ) = curves 2published Find theatpoints tangent to theconchoid trident is(Figure horizontal as follows. 2x are vertical. (a) Show that x y + y = 3x 2 − 10x + 2. (b) Set y = 0 in (a), replace y by x −1 (x 3 − 5x 2 + 2x − 1), and solve the resulting equation for x. y 20 4 −2

2

x 6

8

−20

FIGURE 9 Trident curve: x y = x 3 − 5x 2 + 2x − 1. SOLUTION

Consider the equation of a trident curve: x y = x 3 − 5x 2 + 2x − 1

or y = x −1 (x 3 − 5x 2 + 2x − 1). (a) Taking the derivative of x y = x 3 − 5x 2 + 2x − 1 yields x y + y = 3x 2 − 10x + 2.

S E C T I O N 3.8


139

(b) Setting y = 0 in (a) gives y = 3x 2 − 10x + 2. Thus, we have x −1 (x 3 − 5x 2 + 2x − 1) = 3x 2 − 10x + 2. Collecting like terms and setting to zero, we have 0 = 2x 3 − 5x 2 + 1 = (2x − 1)(x 2 − 2x − 1). Hence, x = 12 , 1 ±

√

2.

47. Find equations of the tangent lines at the points where x = 1 on the so-called folium (Figure 11): Find an equation of the tangent line at the four points on the curve (x 2 + y 2 − 4x)2 = 2(x 2 + y 2 ), where x = 1. 25 2 after the father of the French philosopher Blaise This curve (Figure 10) is an example of a limac xy )2 = named (x¸2on+ofy 2Pascal, 4 Pascal, who first described it in 1650. y 2

x 1

−2

FIGURE 11 SOLUTION

2 First, find the points (1, y) on the curve. Setting x = 1 in the equation (x 2 + y 2 )2 = 25 4 x y yields

25 2 y 4 25 2 y 4 + 2y 2 + 1 = y 4 (1 + y 2 )2 =

4y 4 + 8y 2 + 4 = 25y 2 4y 4 − 17y 2 + 4 = 0 (4y 2 − 1)(y 2 − 4) = 0 y2 =

1 or y 2 = 4 4

Hence y = ± 12 or y = ±2. Taking ddx of both sides of the original equation yields 25 2 25 y + x yy 4 2 25 2 25 y + x yy 4(x 2 + y 2 )x + 4(x 2 + y 2 )yy = 4 2 25 25 2 (4(x 2 + y 2 ) − x)yy = y − 4(x 2 + y 2 )x 2 4 2(x 2 + y 2 )(2x + 2yy ) =

25 y 2 − 4(x 2 + y 2 )x y = 4 y(4(x 2 + y 2 ) − 25 2 x) • At (1, 2), x 2 + y 2 = 5, and 25

22 − 4(5)(1) 1 y = 4 = . 25 3 2(4(5) − 2 (1)) Hence, at (1, 2), the equation of the tangent line is y − 2 = 13 (x − 1) or y = 13 x + 53 . • At (1, −2), x 2 + y 2 = 5 as well, and 25

(−2)2 − 4(5)(1) 1 y = 4 =− . 25 3 −2(4(5) − 2 (1)) Hence, at (1, −2), the equation of the tangent line is y + 2 = − 13 (x − 1) or y = − 13 x − 53 .

140

CHAPTER 3

D I F F E R E N T I AT I O N • At (1, 1 ), x 2 + y 2 = 5 , and 2 4

− 4 54 (1) 11

= y = . 1 4 5 − 25 (1) 12 2 4 2 25 4

2 1 2

11 5 Hence, at (1, 12 ), the equation of the tangent line is y − 12 = 11 12 (x − 1) or y = 12 x − 12 . • At (1, − 1 ), x 2 + y 2 = 5 , and 2 4

− 4 54 (1) 11

=− . y = 1 5 25 12 − 2 4 4 − 2 (1) 25 − 1 4 2

2

11 5 Hence, at (1, − 12 ), the equation of the tangent line is y + 12 = − 11 12 (x − 1) or y = − 12 x + 12 . The folium and its tangent lines are plotted below: y 2 1 x 1

0.5

1.5

2

−1 −2

Use a computer algebra system to plot y 2 = x 3 − 4x for −4 ≤ x, y ≤ 4. Show that if d x/d y = 0, then 49. Use a computer algebra system to plot (x 2 + y 2 )2 = 12(x 2 − y 2 ) + 2 for −4 ≤ x, y ≤ 4. How many y = 0. Conclude that the tangent line is vertical at the points where the curve intersects the x-axis. Does your plot confirm horizontal tangent lines does the curve appear to have? Find the points where these occur. this conclusion? SOLUTION

A plot of the curve y 2 = x 3 − 4x is shown below. y 2 1

−2

x

−1

1

2

3

−1 −2

Differentiating the equation y 2 = x 3 − 4x with respect to y yields 2y = 3x 2

dx dx −4 , dy dy

or dx 2y . = 2 dy 3x − 4 From here, it follows that dd xy = 0 when y = 0, so the tangent line to this curve is vertical at the points where the curve intersects the x-axis. This conclusion is confirmed by the plot of the curve shown above. In Exercises 50–53, use implicit differentiation to calculate higher derivatives. 51. Use the method of the previous exercise to show that y = −y −3 if x 2 + y 2 = 1. Consider the equation y 3 − 32 x 2 = 1. x SOLUTION Let x 2 + y 2 =21. Then 2x + 2yy = 0, and y = − . Thus (a) Show that y = x/y and differentiate again to show that y

2 y −yx =− xyy − 2x yy y · 1 − x y y2 + x 2 1 4 y =− =− = = − 3 = −y −3 . y− 2 2 3 y y y y (b) Express y in terms of x and y using part (a). Calculate y at the point (1, 1) on the curve x y 2 + y − 2 = 0 by the following steps:

S E C T I O N 3.9

Related Rates

141

53. Use the method of the previous exercise to compute y at the point 23 , 43 on the folium of Descartes with the equation x 3 + y 3 = 3x y.

x2 − y 2 , 4 , we find SOLUTION Let x 3 + y 3 = 3x y. Then 3x 2 + 3y 2 y = 3x y + 3y, and y = . At y) = (x, 3 3 x − y2 4 − 4 4 − 12 −8 4 3 y = 29 16 = = = . 6 − 16 −10 5 3 − 9

Similarly,

y =

x − y2

2x − y − x 2 − y 1 − 2yy 162 =− 2 125 2 x−y

when (x, y) = 23 , 43 and y = 45 .

Further Insights and Challenges 55. The lemniscate curve (x 2 + y 2 )2 = 4(x 2 − y 2 ) was discovered by Jacob Bernoulli in 1694, who noted that it is if P8, lies the or intersection two curves x 2coordinates − y 2 = c of andthe x yfour = dpoints (c, d atconstants), the “shapedShow like athat figure or aon knot, the bow ofofa the ribbon.” Find the which thethen tangent tangents to the curves at P are perpendicular. line is horizontal (Figure 12). y 1 x

−1

1 −1

FIGURE 12 Lemniscate curve: (x 2 + y 2 )2 = 4(x 2 − y 2 ).

2

Consider the equation of a lemniscate curve: x 2 + y 2 = 4 x 2 − y 2 . Taking the derivative of both sides of this equation, we have

2 x 2 + y 2 2x + 2yy = 4 2x − 2yy . SOLUTION

Therefore,

x 2 + y2 − 2 x 8x − 4x x 2 + y 2 . =− y = 8y + 4y x 2 + y 2 x 2 + y2 + 2 y If y = 0, then either x = 0 or x 2 + y 2 = 2.

• If x = 0 in the lemniscate curve, then y 4 = −4y 2 or y 2 y 2 + 4 = 0. If y is real, then y = 0. The formula for y

in (a) is not defined at the origin (0/0). An alternative parametric analysis shows that the slopes of the tangent lines to the curve at the origin are ±1.

• If x 2 + y 2 = 2 or y 2 = 2 − x 2 , then plugging this into the lemniscate equation gives 4 = 4 2x 2 − 2 which

√ √ yields x = ± 32 = ± 26 . Thus y = ± 12 = ± 22 . Accordingly, the four points at which the tangent lines to the √ √ √ √ √

√ √

√ lemniscate curve are horizontal are − 26 , − 22 , − 26 , 22 , 26 , − 22 , and 26 , 22 . Divide the curve in Figure 13 into five branches, each of which is the graph of a function. Sketch the branches.

3.9 Related Rates Preliminary Questions 1. Assign variables and restate the following problem in terms of known and unknown derivatives (but do not solve it): How fast is the volume of a cube increasing if its side increases at a rate of 0.5 cm/s? SOLUTION

Let s and V denote the length of the side and the corresponding volume of a cube, respectively. Determine

d V if ds = 0.5 cm/s. dt dt

142

CHAPTER 3


2. What is the relation between d V /dt and dr/dt if

4 V = πr 3 3

SOLUTION

Applying the general power rule, we find ddtV = 4π r 2 dr dt .

In Questions 3–4, suppose that water pours into a cylindrical glass of radius 4 cm. The variables V and h denote the volume and water level at time t, respectively. 3. Restate in terms of the derivatives d V /dt and dh/dt: How fast is the water level rising if water pours in at a rate of 2 cm3 /min? dV 3 Determine dh dt if dt = 2 cm /min. 4. Repeat the same for this problem: At what rate is water pouring in if the water level rises at a rate of 1 cm/min?

SOLUTION

SOLUTION

Determine ddtV if dh dt = 1 cm/min.

Exercises In Exercises 1–2, consider a rectangular bathtub whose base is 18 ft2 . 1. How fast is the water level rising if water is filling the tub at a rate of 0.7 ft3 /min? Let h be the height of the water in the tub and V be the volume of the water. Then V = 18h and dh dV = 18 . Thus dt dt

SOLUTION

1 dV 1 dh = = (.7) ≈ .039 ft/min. dt 18 dt 18 3. The radius of a circular oil slick expands at a rate of 2 m/min. At what rate is water pouring into the tub if the water level rises at a rate of 0.8 ft/min? (a) How fast is the area of the oil slick increasing when the radius is 25 m? (b) If the radius is 0 at time t = 0, how fast is the area increasing after 3 min? Let r be the radius of the oil slick and A its area. dA dr = 2π r . Substituting r = 25 and dr (a) Then A = π r 2 and dt = 2, we find dt dt

SOLUTION

dA = 2π (25) (2) = 100π ≈ 314.16 m2 /min. dt (b) Since dr dt = 2 and r (0) = 0, it follows that r (t) = 2t. Thus, r (3) = 6 and dA = 2π (6) (2) = 24π ≈ 75.40 m2 /min. dt In Exercises 5–8,rate assume the radius r ofincreasing a sphere isif expanding at increasing a rate of 14 volume of a sphere is At what is the that diagonal of a cube its edges are at in./min. a rate of The 2 cm/s? V = 43 π r 3 and its surface area is 4π r 2 . 5. Determine the rate at which the volume is changing with respect to time when r = 8 in. d As the radius is expanding at 14 inches per minute, we know that dr dt = 14 in./min. Taking dt of the equation V = 43 π r 3 yields dr dr dV 4 = π 3r 2 = 4π r 2 . dt 3 dt dt SOLUTION

Substituting r = 8 and dr dt = 14 yields dV = 4π (8)2 (14) = 3584π in.3 /min. dt 7. Determine the rate at which the surface area is changing when the radius is r = 8 in. Determine the rate at which the volume is changing with respect to time at t = 2 min, assuming that r = 0 at dA dr dr 2 SOLUTION t = 0. Taking the derivative of both sides of A = 4π r with respect to t yields dt = 8π r dt . dt = 14, so dA = 8π (8)(14) = 896π in.2 /min. dt Determine the rate at which the surface area is changing with respect to time at t = 2 min, assuming that r = 3 at t = 0.

S E C T I O N 3.9

Related Rates

143

9. A road perpendicular to a highway leads to a farmhouse located 1 mile away (Figure 9). An automobile travels past the farmhouse at a speed of 60 mph. How fast is the distance between the automobile and the farmhouse increasing when the automobile is 3 miles past the intersection of the highway and the road?

l 60 mph

Automobile

FIGURE 9 SOLUTION Let l denote the distance between the automobile and the farmhouse, and let s denote the distance past the intersection of the highway and the road. Then l 2 = 1 + s 2 . Taking the derivative of both sides of this equation yields dl = 2s ds , so 2l dt dt

s ds dl = . dt l dt When the auto is 3 miles past the intersection, we have √ 3 · 60 dl 180 = √ = 18 10 ≈ 56.92 mph. = 2 2 dt 10 1 +3 11. Follow the same set-up as Exercise 10, but assume that the water level is rising at a rate of 0.3 m/min when it is 2 m. 3 conical tankflowing has height At whatArate is water in? 3 m and radius 2 m at the top. Water flows in at a rate of 2 m /min. How fast is the water level rising when it is 2 m? SOLUTION Consider the cone of water in the tank at a certain instant. Let r be the radius of its (inverted) base, h its 4 π h 3 . Accordingly, height, and V its volume. By similar triangles, hr = 23 or r = 23 h and thus V = 13 π r 2 h = 27 dV 4 dh = π h2 . dt 9 dt Substituting h = 2 and dh dt = 0.3 yields dV 4 = π (2)2 (.3) ≈ 1.68 m3 /min. dt 9 13. Answer (a) and (b) in Exercise 12 assuming that Sonya begins moving 1 minute after Isaac takes off. Sonya and Isaac are in motorboats located at the center of a lake. At time t = 0, Sonya begins traveling south at SOLUTION IsaacAtx the miles easttime, of the center of the and Sonya miles south its center, let h be the distance a speed ofWith 32 mph. same Isaac takes off, lake heading east at ya speed of 27 of mph. between them. (a) How far have Sonya and Isaac each traveled after 12 min? 12 = 1 hour, Isaac has traveled 1 × 27 = 27 miles. After 11 minutes or 11 hour, Sonya has (a) After minutes (b) At12what rate isorthe between them increasing 60 distance 5 5 at t = 125 min? 60 11 88 traveled 60 × 32 = 15 miles. dh dx dy = 2x + 2y . Thus, (b) We have h 2 = x 2 + y 2 and 2h dt dt dt x d x + y ddty x d x + y ddty dh . = dt = dt dt h x 2 + y2 dy dx 88 Substituting x = 27 5 , dt = 27, y = 15 , and dt = 32 yields

dh = dt

27 (27) + 5

27 5

2

88 15

+ 88 15

(32)

2

5003 = √ ≈ 41.83 mph. 14305

15. At a given moment, a plane passes directly above a radar station at an altitude of 6 miles. A 6-ft man walks away from a 15-ft lamppost at a speed of 3 ft/s (Figure 10). Find the rate at which his shadow (a) If the plane’sinspeed is 500 mph, how fast is the distance between the plane and the station changing half an hour is increasing length. later? (b) How fast is the distance between the plane and the station changing when the plane passes directly above the station? SOLUTION

Let x be the distance of the plane from the station along the ground and h the distance through the air.

144

CHAPTER 3


(a) By the Pythagorean Theorem, we have h 2 = x 2 + 62 = x 2 + 36. dh dx dh x dx = 2x , and = . After an half hour, x = 12 × 500 = 250 miles. With x = 250, h = 2502 + 36, dt dt dt h dt and ddtx = 500, Thus 2h

250 dh = × 500 ≈ 499.86 mph. dt 2502 + 36 (b) When the plane is directly above the station, x = 0, so the distance between the plane and the station is not changing, for at this instant we have 0 dh = × 500 = 0 mph. dt 6 17. A hot air balloon rising vertically is tracked by an observer located 2 miles from the lift-off point. At a certain In the setting of Exercise 15, suppose that the line through the radar station and the plane makes an angle θ with moment, the angle between the observer’s line-of-sight and the horizontal is π5 , and it is changing at a rate of 0.2 rad/min. the horizontal. How fast is θ changing 10 min after the plane passes over the radar station? How fast is the balloon rising at this moment? SOLUTION

Let y be the height of the balloon (in miles) and θ the angle between the line-of-sight and the horizontal. y . Therefore, 2

Via trigonometry, we have tan θ =

sec2 θ ·

1 dy dθ = , dt 2 dt

and dy dθ =2 sec2 θ . dt dt Using ddtθ = 0.2 and θ = π5 yields dy 1 ≈ .61 mi/min. = 2 (.2) 2 dt cos (π /5) In Exercises referaway to a 16-ft down wall, Figuresmoves 1 and twice 2. Theasvariable h isdoes. the height the As a 19–23, man walks from ladder a 12-ftsliding lamppost, thea tip of as hisinshadow fast as he What of is the ladder’s topheight? at time t, and x is the distance from the wall to the ladder’s bottom. man’s 19. Assume the bottom slides away from the wall at a rate of 3 ft/s. Find the velocity of the top of the ladder at t = 2 if the bottom is 5 ft from the wall at t = 0. SOLUTION Let x denote the distance from the base of the ladder to the wall, and h denote the height of the top of the ladder from the floor. The ladder is 16 ft long, so h 2 + x 2 = 162 . At any time t, x = 5 + 3t. Therefore, at time t = 2, the base is x = 5 + 3(2) = 11 ft from the wall. Furthermore, we have

2h Substituting x = 11, h =

dh dx + 2x = 0 so dt dt

162 − 112 and ddtx = 3, we obtain

dh x dx =− . dt h dt

11 dh 11 (3) = − √ ft/s ≈ −2.84 ft/s. = − dt 15 162 − 112 21. Suppose that h(0) = 12 and the top slides down the wall at a rate of 4 ft/s. Calculate x and d x/dt at t = 2 s. Suppose that the top is sliding down the wall at a rate of 4 ft/s. Calculate d x/dt when h = 12. SOLUTION Let h and x be the height of the ladder’s top and the distance from the wall of the ladder’s bottom, respectively. After 2 seconds, h = 12 + 2 (−4) = 4 ft. Since h 2 + x 2 = 162 , √ x = 162 − 42 = 4 15 ft. Furthermore, we have 2h find

√ dh dx dx h dh + 2x = 0, so that =− . Substituting h = 4, x = 4 15, and dh dt = −4, we dt dt dt x dt dx 4 4 = − √ (−4) = √ ≈ 1.03 ft/s. dt 4 15 15

What is the relation between h and x at the moment when the top and bottom of the ladder move at the same speed?

S E C T I O N 3.9

Related Rates

145

23. Show that the velocity dh/dt approaches infinity as the ladder slides down to the ground (assuming d x/dt is constant). This suggests that our mathematical description is unrealistic, at least for small values of h. What would, in fact, happen as the top of the ladder approaches the ground? SOLUTION Let L be the (constant) length of the ladder, let x be the distance from the base of the ladder to the point of contact of the ladder with the ground, and let h be the height of the point of contact of the ladder with the wall. By Pythagoras’ theorem,

x 2 + h2 = L 2. Taking derivatives with respect to t yields: 2x

dh dx + 2h = 0. dt dt

Therefore, dh x dx =− . dt h dt As ddtx is constant and positive and x approaches the length of the ladder as it falls, − hx ddtx gets arbitrarily large as h → 0. In a real situation, the top of the ladder would slide down the wall only part of the way. At some point it would lose contact with the wall and fall down freely with acceleration g. 25. Suppose that both the radius r and height h of a circular cone change at a rate of 2 cm/s. How fast is the volume of The radius r of a right circular cone of fixed height h = 20 cm is increasing at a rate of 2 cm/s. How fast is the the cone increasing when r = 10 and h = 20? volume increasing when r = 10? 1 SOLUTION Let r be the radius, h be the height, and V be the volume of a right circular cone. Then V = 3 π r 2 h, and dV dh 1 dr = π r2 + 2hr . dt 3 dt dt dh When r = 10, h = 20, and dr dt = dt = 2, we find 1000π dV π 2 10 · 2 + 2 · 20 · 10 · 2 = = ≈ 1047.20 cm3 /s. dt 3 3

27. A searchlight rotates at a rate of 3 revolutions per minute. The beam hits a wall located 10 miles away and produces 2 + 16y 2 = 25 (Figure 11). particle the How ellipse 9xis a dot ofAlight that moves moves counterclockwise horizontally alongaround the wall. fast this dot moving when the angle θ between the beam π and the line through the searchlight perpendicular to the wall is 6 ? Note that d θ /dt = 3(2π ) = 6π . SOLUTION Let y be the distance between the dot of light and the point of intersection of the wall and the line through the searchlight perpendicular to the wall.is Let be the angle between the Explain beam ofyour lightanswer. and the line. Using trigonometry, (a) In which of the four quadrants the θderivative d x/dt positive? y we have tan θ = 10 . Therefore, (b) Find a relation between d x/dt and d y/dt. (c) At what rate is the x-coordinate changing when the the point (1, 1) if its y-coordinate is increasing d θparticle 1 passes dy = , sec2 θ · at a rate of 6 ft/s? dt 10 dt and (d) What is d y/dt when the particle is at the top and bottom of the ellipse?

dy dθ = 10 sec2 θ . dt dt With θ = π6 and ddtθ = 6π , we find 1 dy = 80π ≈ 251.33 mi/min ≈ 15,079.64 mph. = 10 (6π ) dt cos2 (π /6) 29. A plane traveling at an altitude of 20,000 ft passes directly overhead at time t = 0. One minute later you observe A rocket travels vertically at a speed of 800 mph. The rocket is tracked through a telescope by an observer located that the angle between the vertical and your line of sight to the plane is 1.14 rad and that this angle is changing at a rate 10 miles from the launching pad. Find the rate at which the angle between the telescope and the ground is increasing of 0.38 rad/min. Calculate the velocity of the airplane. 3 min after lift-off. SOLUTION Let x be the distance of the plane from you along the ground and θ the angle between the vertical and your x line of sight to the plane. Then tan θ = 20000 and dθ dx = 20000 sec2 θ · . dt dt Substituting θ = 1.14 and ddtθ = 0.38, we find dx 20000 = (.38) ≈ 43581.69 ft/min dt cos2 (1.14) or roughly 495.25 mph.

146

CHAPTER 3


31. A jogger runs around a circular track of radius 60 ft. Let (x, y) be her coordinates, where the origin is at the center 2 /s) at which area is swept out by the second hand of a circular clock as a function of Calculate (in cmcoordinates of the track. Whenthe therate jogger’s are (36, 48), her x-coordinate is changing at a rate of 14 ft/s. Find d y/dt. the clock’s radius. dx dy dy x dx SOLUTION We have x 2 + y 2 = 602 . Thus 2x + 2y = 0, and =− . With x = 36, y = 48, and dt dt dt y dt d x = 14, dt dy 36 21 = − (14) = − = −10.5 ft/s. dt 48 2 3 ) of an expanding gas are related In Exercises the pressure P (in volume cmthe A car33–34, travelsassume down a that highway at 55 mph. Ankilopascals) observer is and standing 500Vft(in from highway. b by P(a) V How = C, fast where b and C arebetween constants holds and in adiabatic expansion,atwithout heat gain or passes loss). in front of the is the distance the(this observer the car increasing the moment the car

observer? Can answer without relying 2 , and on 33. Find d P/dt if byou = justify 1.2, P your = 8 kPa, V = 100 cm d Vany /dt calculations? = 20 cm3 /min. (b) How fast is the distance between the observer and the car increasing 1 min later? b SOLUTION Let P V = C. Then PbV b−1

dV dP + Vb = 0, dt dt

and Pb d V dP =− . dt V dt Substituting b = 1.2, P = 8, V = 100, and ddtV = 20, we find dP (8) (1.2) =− (20) = −1.92 kPa/min. dt 100 35. A point moves along the parabola y = x 2 + 1. Let (t) be the2 distance between the point and the origin. Calculate , and /dt = 20 cm3 /min. Find b ifthat P =the25x-coordinate kPa, d P/dt of = the 12 kPa/min, V = 100atcm (t), assuming point is increasing a rate ofd9Vft/s. A point moves along the parabola y = x 2 + 1. Let be the distance between the point and the origin. By the distance formula, we have = (x 2 + (x 2 + 1)2 )1/2 . Hence SOLUTION

−1/2 d x

(2x 2 + 3)x ddtx dx 1 2 d = x + (x 2 + 1)2 + 2 x 2 + 1 · 2x = 2x . dt 2 dt dt x 4 + 3x 2 + 1 With ddtx = 9, we have 9x(2x 2 + 3) (t) = ft/s. x 4 + 3x 2 + 1 37. A water tank in the shape of a right circular cone of radius 300 cm and height 500 cm leaks water from the vertex at The base x of the right triangle in Figure 12 increases at a rate of 5 cm/s, while the height remains constant at a rate of 10 cm3 /min. Find the rate at which the water level is decreasing when it is 200 cm. h = 20. How fast is the angle θ changing when x = 20? SOLUTION Consider the cone of water in the tank at a certain instant. Let r be the radius of its (inverted) base, h its 3 1 2 3 3 height, and V its volume. By similar triangles, hr = 300 500 or r = 5 h and thus V = 3 π r h = 25 π h . Therefore, dV 9 dh π h2 , = dt 25 dt and 25 d V dh = . dt 9π h 2 dt We are given ddtV = −10. When h = 200, it follows that dh 25 1 = ≈ −2.21 × 10−4 cm/min. (−10) = − 2 dt 1440 π 9π (200) Thus, the water level is decreasing at the rate of 2.21 × 10−4 cm/min. Two parallel paths 50 ft apart run through the woods. Shirley jogs east on one path at 6 mph, while Jamail walks west on the other at 4 mph. If they pass each other at time t = 0, how far apart are they 3 s later, and how fast is the distance between them changing at that moment?

S E C T I O N 3.9

Related Rates

147

Further Insights and Challenges 39. Henry is pulling on a rope that passes through a pulley on a 10-ft pole and is attached to a wagon (Figure 13). Assume that the rope is attached to a loop on the wagon 2 ft off the ground. Let x be the distance between the loop and the pole. (a) Find a formula for the speed of the wagon in terms of x and the rate at which Henry pulls the rope. (b) Find the speed of the wagon when it is 12 ft from the pole, assuming that Henry pulls the rope at a rate of 1.5 ft/sec.

2 x

FIGURE 13 SOLUTION Let h be the distance from the pulley to the loop on the wagon. (Note that the rate at which Henry pulls the rope is the rate at which h is decreasing.) Using the Pythagorean Theorem, we have h 2 = x 2 + (10 − 2)2 = x 2 + 82 . dh dx dx h dh (a) Thus 2h = 2x , and = . dt dt dt x dt (b) As Henry pulls the rope at the rate of 1.5 = 32 ft/s, the distance h is decreasing at that rate; i.e., dh/dt = − 32 . When the wagon is 12 feet from the pole, we thus have √ dx 122 + 82 3 = − = − 13/2 ft/s. dt 12 2 √ 13 Thus, the speed of the wagon is ft/s. 2

41. Using a telescope, you track a rocket that was launched 2 miles away, recording the angle θ between the telescope A roller coaster has the shape of the graph in Figure 14. Show that when the roller coaster passes the point and the ground at half-second intervals. Estimate the velocity of the rocket if θ (10) = 0.205 and θ (10.5) = 0.225. (x, f (x)), the vertical velocity of the roller coaster is equal to f (x) times its horizontal velocity. h SOLUTION Let h be the height of the vertically ascending rocket. Using trigonometry, tan θ = , so 2 dθ dh = 2 sec2 θ · . dt dt We are given θ (10) = 0.205, and we can estimate θ (10.5) − θ (10) d θ = 0.04. ≈ dt t=10 0.5 Thus, dh = 2 sec2 (0.205) · (0.04) ≈ 0.083 mi/s, dt or roughly 300 mph. 43. A baseball player runs from home plate toward first base at 20 ft/s. How fast is the player’s distance from second Two trains leave a station at t = 0 and travel with constant velocity v along straight tracks that make an angle θ . base changing when the player is halfway to first base? See Figure 15. √ (a) Show that the trains are separating from each other at a rate of v 2 − 2 cos θ . (b) What does this formula give for θ = π ?

20 ft/s

FIGURE 15 Baseball diamond. SOLUTION In baseball, the distance between bases is 90 feet. Let x be the distance of the player from home plate and h the player’s distance from second base. Using the Pythagorean theorem, we have h 2 = 902 + (90 − x)2 . Therefore, dh dx 2h = 2 (90 − x) − , dt dt

148

CHAPTER 3


and dh 90 − x d x =− . dt h dt We are given ddtx = 20. When the player is halfway to first base, x = 45 and h =

902 + 452 , so

√ 45 dh = − (20) = −4 5 ≈ −8.94 ft/s. 2 2 dt 90 + 45 45. A spectator seated 300 m away from the center of a circular track of radius 100 m watches an athlete run laps at As the wheel of radius r cm in Figure 16 rotates, the rod of length L attached at the point P drives a piston back a speed of 5 m/s. How fast is the distance between the spectator and athlete changing when the runner is approaching and forth in a straight line. Let x be the distance from the origin to the point Q at the end of the rod as in the figure. the spectator and the distance between them is 250 m? Hint: The diagram for this problem is similar to Figure 16, with (a) Use Theorem to show that r = 100 and xthe=Pythagorean 300. SOLUTION

2 θ those of Q are (x, 0). From the diagram, the coordinates cosθθ)2, r+sin ) and L 2 of =P (xare − r(rcos r 2 θsin

• The distance formula gives (b) Differentiate Eq. (8) with respect to t to prove that

d x − r cosdθθ)2 + (−r2 sin θ )2 . d θ L 2(x − r cos θ ) = (x + r sin θ + 2r sin θ cos θ =0 dt dt dt

Thus, (c) Calculate the speed of the piston when θ = π2 , assuming that r = 10 cm, L = 30 cm, and the wheel rotates at 4 L 2 = (x − r cos θ )2 + r 2 sin2 θ . revolutions per minute. Note that x (the distance of the spectator from the center of the track) and r (the radius of the track) are constants. • Differentiating with respect to t gives

2L

dθ dθ dL = 2 (x − r cos θ ) r sin θ + 2r 2 sin θ cos θ . dt dt dt

Thus, dθ rx dL = sin θ . dt L dt

θ , namely s = r θ . Thus

• Recall the relation between arc length s and angle ds = −5, we have dt

dθ 1 ds = . Given r = 100 and dt r dt

dθ 1 1 = rad/s. (−5) = − dt 100 20 (Note: In this scenario, the runner traverses the track in a clockwise fashion and approaches the spectator from Quadrant 1.) • Next, the Law of Cosines gives L 2 = r 2 + x 2 − 2r x cos θ , so cos θ =

r 2 + x2 − L2 1002 + 3002 − 2502 5 = = . 2r x 2 (100) (300) 8

Accordingly, sin θ =

√ 2 5 39 1− = . 8 8

• Finally

dL (300) (100) = dt 250

√

39 8

−

1 20

=−

√ 3 39 ≈ −4.68 m/s. 4

CHAPTER REVIEW EXERCISES In Exercises 1–4, refer to the function f (x) whose graph is shown in Figure 1.


7 6 5 4 3 2 1

149

y

0.5

1.0

1.5

x 2.0

FIGURE 1

1. Compute the average ROC of f (x) over [0, 2]. What is the graphical interpretation of this average ROC? SOLUTION

The average rate of change of f (x) over [0, 2] is f (2) − f (0) 7−1 = = 3. 2−0 2−0

Graphically, this average rate of change represents the slope of the secant line through the points (2, 7) and (0, 1) on the graph of f (x). f (0.7 + h) − f (0.7) for h+= this number larger or smaller than f (0.7)? 3. Estimate f (0.7 h)0.3. − fIs(0.7) equal to the slope of the secant line between the points where For which valuehof h is h SOLUTION Forxh==1.1. 0.3, x = 0.7 and f (0.7 + h) − f (0.7) f (1) − f (0.7) 2.8 − 2 8 = ≈ = . h 0.3 0.3 3 Because the curve is concave up, the slope of the secant line is larger than the slope of the tangent line, so the value of the difference quotient should be larger than the value of the derivative. using the limit definition and find an equation of the tangent line to the graph of f (x) In Exercises 5–8, fcompute f (a) (0.7) and Estimate f (1.1). at x = a. 5. f (x) = x 2 − x, SOLUTION

a=1

Let f (x) = x 2 − x and a = 1. Then f (a + h) − f (a) (1 + h)2 − (1 + h) − (12 − 1) = lim h h h→0 h→0

f (a) = lim

1 + 2h + h 2 − 1 − h = lim (1 + h) = 1 h h→0 h→0

= lim

and the equation of the tangent line to the graph of f (x) at x = a is y = f (a)(x − a) + f (a) = 1(x − 1) + 0 = x − 1. −1 , a = 4 7. f (x)f (x) = x= 5 − 3x, a = 2 SOLUTION Let f (x) = x −1 and a = 4. Then 1 − 1 f (a + h) − f (a) 4 − (4 + h) 4 = lim 4+h = lim h h h→0 h→0 h→0 4h(4 + h)

f (a) = lim

−1 1 1 =− =− 4(4 + 0) 16 h→0 4(4 + h)

= lim

and the equation of the tangent line to the graph of f (x) at x = a is 1 1 1 1 y = f (a)(x − a) + f (a) = − (x − 4) + = − x + . 16 4 16 2 In Exercises 9–12,3 compute d y/d x using the limit definition. f (x) = x , a = −2 9. y = 4 − x 2 SOLUTION

Let y = 4 − x 2 . Then

4 − (x + h)2 − (4 − x 2 ) 4 − x 2 − 2xh − h 2 − 4 + x 2 dy = lim = lim = lim (−2x − h) = −2x − 0 = −2x. dx h h h→0 h→0 h→0 y=

√ 2x + 1

150

CHAPTER 3


11. y =

1 2−x

SOLUTION

Let y =

1 . Then 2−x 1

1

dy 1 (2 − x) − (2 − x − h) 1 2−(x+h) − 2−x . = lim = lim = lim = dx h h→0 h→0 h(2 − x − h)(2 − x) h→0 (2 − x − h)(2 − x) (2 − x)2 In Exercises 13–16, 1 express the limit as a derivative. y√ = 1)21 1(x +− h− 13. lim h h→0 √ SOLUTION Let f (x) = x. Then √ lim

h→0

15. lim

1+h−1 f (1 + h) − f (1) = lim = f (1). h h h→0

sin t cos t

3 t→π limt −xπ + 1 x→−1 x + 1

SOLUTION

Let f (t) = sin t cos t and note that f (π ) = sin π cos π = 0. Then lim

t→π

sin t cos t f (t) − f (π ) = lim = f (π ). t→π t −π t −π

17. Find f (4) and f (4) if the tangent line to the graph of f (x) at x = 4 has equation y = 3x − 14. cos θ − sin θ + 1 lim SOLUTION θ − π of the tangent line to the graph of f (x) at x = 4 is y = f (4)(x − 4) + f (4) = f (4)x + θ →π The equation ( f (4) − 4 f (4)). Matching this to y = 3x − 14, we see that f (4) = 3 and f (4) − 4(3) = −14, so f (4) = −2. 19. Which of (A), (B), or (C) is the graph of the derivative of the function f (x) shown in Figure 3? Each graph in Figure 2 shows the graph of a function f (x) and its derivative f (x). Determine which is the function and which is the derivative. y y = f(x) x

−2 −1 y

2

y x

−2 −1

1

1

2

x

−2 −1

(A)

y

1

2

x

−2 −1

(B)

1

2

(C)

FIGURE 3

The graph of f (x) has four horizontal tangent lines on [−2, 2], so the graph of its derivative must have four x-intercepts on [−2, 2]. This eliminates (B). Moreover, f (x) is increasing at both ends of the interval, so its derivative must be positive at both ends. This eliminates (A) and identifies (C) as the graph of f (x). SOLUTION

21. A girl’s height h(t) (in centimeters) is measured at time t (years) for 0 ≤ t ≤ 14: Let N (t) be the percentage of a state population infected with a flu virus on week t of an epidemic. What = 1.2? percentage is likely to be infected week 4 if 96.7, N (3) 104.5, = 8 and N (3) 52,in75.1, 87.5, 111.8, 118.7, 125.2, 131.5, 137.5, 143.3, 149.2, 155.3, 160.8, 164.7 (a) What is the girl’s average growth rate over the 14-year period? (b) Is the average growth rate larger over the first half or the second half of this period? (c) Estimate h (t) (in centimeters per year) for t = 3, 8. SOLUTION

(a) The average growth rate over the 14-year period is 164.7 − 52 = 8.05 cm/year. 14


151

(b) Over the first half of the 14-year period, the average growth rate is 125.2 − 52 ≈ 10.46 cm/year, 7 which is larger than the average growth rate over the second half of the 14-year period: 164.7 − 125.2 ≈ 5.64 cm/year. 7 (c) For t = 3, h (3) ≈

h(4) − h(3) 104.5 − 96.7 = = 7.8 cm/year; 4−3 1

h (8) ≈

h(9) − h(8) 137.5 − 131.5 = = 6.0 cm/year. 9−8 1

for t = 8,

In Exercises 23–24, use the following table of values for the number A(t) of automobiles (in millions) manufactured3/2in A planet’s period P (time in years to complete one revolution around the sun) is approximately 0.0011 A , the United States in year t. where A is the average distance (in millions of miles) from the planet to the sun. (a) Calculate P and d P/d Mars using 142. 1974 1975 1976 t A for1970 1971the value 1972 A = 1973 (b) Estimate the increase in P if A is increased to 143. A(t) 6.55 8.58 8.83 9.67 7.32 6.72 8.50 23. What is the interpretation of A (t)? Estimate A (1971). Does A (1974) appear to be positive or negative? Because A(t) measures the number of automobiles manufactured in the United States in year t, A (t) measures the rate of change in automobile production in the United States. For t = 1971, SOLUTION

A (1971) ≈

A(1972) − A(1971) 8.83 − 8.58 = = 0.25 million automobiles/year. 1972 − 1971 1

Because A(t) decreases from 1973 to 1974 and from 1974 to 1975, it appears that A (1974) would be negative. In Exercises compute derivative. Given25–50, the data, whichthe of (A)–(C) in Figure 4 could be the graph of the derivative A (t)? Explain. 25. y = 3x 5 − 7x 2 + 4 SOLUTION

Let y = 3x 5 − 7x 2 + 4. Then dy = 15x 4 − 14x. dx

27. y = t −7.3 −3/2 y = 4x SOLUTION Let y = t −7.3 . Then dy = −7.3t −8.3 . dt x +1 29. y =y =2 4x 2 − x −2 x +1 SOLUTION

x +1 . Then Let y = 2 x +1 1 − 2x − x 2 (x 2 + 1)(1) − (x + 1)(2x) dy = . = 2 2 dx (x + 1) (x 2 + 1)2

6 31. y = (x 4 3t −− 9x) 2 y= SOLUTION 4tLet − 9y = (x 4 − 9x)6 . Then

d dy = 6(x 4 − 9x)5 (x 4 − 9x) = 6(4x 3 − 9)(x 4 − 9x)5 . dx dx 33. y = (2 + 9x2 2 )3/2 −3 6 y = (3t + 20t )

152

CHAPTER 3


Let y = (2 + 9x 2 )3/2 . Then 3 d dy = (2 + 9x 2 )1/2 (2 + 9x 2 ) = 27x(2 + 9x 2 )1/2 . dx 2 dx

z 35. y = √ 3 4 y =1(x−+ z 1) (x + 4) SOLUTION

Let y = √

z 1−z

. Then √

dy = dz

1 − z − (− 2z ) √ 1

1−z

1−z

=

1 − z + 2z (1 − z)3/2

=

2−z . 2(1 − z)3/2

√ x4 + x 3 37. y = 1 y = x 12 + x SOLUTION Let y=

√ x4 + x = x 2 + x −3/2 . x2

Then dy 3 = 2x − x −5/2 . dx 2 39. y = tan(t −3 ) 1 y= √ −3 SOLUTION (1Let y = 2tan(t − x) − x ). Then d dy = sec2 (t −3 ) t −3 = −3t −4 sec2 (t −3 ). dt dt 2x 41. y =y sin(2x) cos− = 4 cos(2 3x) SOLUTION Let y = sin(2x) cos2 x = 2 sin x cos3 x. Then

dy = −6 sin2 x cos2 x + 2 cos4 x. dx 43. y = tan3 θ y = sin( x 2 + 1) SOLUTION Let y = tan3 θ . Then d dy = 3 tan2 θ tan θ = 3 tan2 θ sec2 θ . dθ dθ t 45. y = sec t4 y 1=+sin θ SOLUTION Let y =

t . Then 1 + sec t 1 + sec t − t sec t tan t dy . = dt (1 + sec t)2

8 47. y =y = z csc(9z + 1) 1 + cot θ SOLUTION

Let y =

8 = 8(1 + cot θ )−1 . Then 1 + cot θ dy 8 csc2 θ d = −8(1 + cot θ )−2 (1 + cot θ ) = . dθ dθ (1 + cot θ )2

y x)√x + x+ 49. y = =xtan(cos


SOLUTION

Let y =

x+

x+

√

153

x. Then

√ −1/2 d √ 1 dy = x+ x+ x x+ x+ x dx 2 dx −1/2

√ √ −1/2 d √ 1 1 = x+ x+ x x+ x x+ x 1+ 2 2 dx −1/2

√ √ −1/2 1 1 1 = x+ x+ x x+ x 1+ 1 + x −1/2 . 2 2 2 In Exercises 51–56, use the table of values to calculate the derivative of the given function at x = 2. y = cos(cos(cos( θ ))) x

f (x)

g(x)

f (x)

g (x)

2

5

4

−3

9

4

3

2

−2

3

51. S(x) = 3 f (x) − 2g(x) SOLUTION

Let S(x) = 3 f (x) − 2g(x). Then S (x) = 3 f (x) − 2g (x) and S (2) = 3 f (2) − 2g (2) = 3(−3) − 2(9) = −27.

f (x) 53. R(x) = = f (x)g(x) H (x) g(x) SOLUTION

Let R(x) = f (x)/g(x). Then R (x) =

g(x) f (x) − f (x)g (x) g(x)2

and R (2) =

g(2) f (2) − f (2)g (2) 4(−3) − 5(9) 57 = =− . 2 2 16 g(2) 4

55. F(x) = f (g(2x)) G(x) = f (g(x)) SOLUTION Let F(x) = f (g(2x)). Then F (x) = 2 f (g(2x))g (2x) and F (2) = 2 f (g(4))g (4) = 2 f (2)g (4) = 2(−3)(3) = −18. In Exercise 57–60, 2 ) f (x) = x 3 − 3x 2 + x + 4. K (x) = f (xlet 57. Find the points on the graph of f (x) where the tangent line has slope 10. Let f (x) = x 3 − 3x 2 + x + 4. Then f (x) = 3x 2 − 6x + 1. The tangent line to the graph of f (x) will have slope 10 when f (x) = 10. Solving the quadratic equation 3x 2 − 6x + 1 = 10 yields x = −1 and x = 3. Thus, the points on the graph of f (x) where the tangent line has slope 10 are (−1, −1) and (3, 7). SOLUTION

59. Find all values of b such that y = 25x + b is tangent to the graph of f (x). For which values of x are the tangent lines to the graph of f (x) horizontal? SOLUTION Let f (x) = x 3 − 3x 2 + x + 4. The equation y = 25x + b represents a line with slope 25; in order for this line to be tangent to the graph of f (x), we must therefore have f (x) = 3x 2 − 6x + 1 = 25. Solving for x yields x = 4 and x = −2. The equation of the line tangent to the graph of f (x) at x = 4 is y = f (4)(x − 4) + f (4) = 25(x − 4) + 24 = 25x − 76, while the equation of the tangent line at x = −2 is y = f (−2)(x + 2) + f (−2) = 25(x + 2) − 18 = 25x + 32. Thus, when b = −76 and when b = 32, the line y = 25x + b is tangent to the graph of f (x). 61. (a) Find Show there a unique of aonly suchone that f (x) line = x 3of−slope 2x 2 k. + x + 1 has the same slope at both a and allthat values of is k such that value f (x) has tangent a + 1. (b) Plot f (x) together with the tangent lines at x = a and x = a + 1 and confirm your answer to part (a).

154

CHAPTER 3


(a) Let f (x) = x 3 − 2x 2 + x + 1. Then f (x) = 3x 2 − 4x + 1 and the slope of the tangent line at x = a is f (a) = 3a 2 − 4a + 1, while the slope of the tangent line at x = a + 1 is f (a + 1) = 3(a + 1)2 − 4(a + 1) + 1 = 3(a 2 + 2a + 1) − 4a − 4 + 1 = 3a 2 + 2a. In order for the tangent lines at x = a and x = a + 1 to have the same slope, we must have f (a) = f (a + 1), or 3a 2 − 4a + 1 = 3a 2 + 2a. The only solution to this equation is a = 16 . (b) The equation of the tangent line at x = 16 is y = f

1 1 5 1 241 5 113 1 x− + f = x− + = x+ , 6 6 6 12 6 216 12 108

and the equation of the tangent line at x = 76 is 7 7 5 7 223 5 59 7 x− + f = x− + = x+ . y = f 6 6 6 12 6 216 12 108 The graphs of f (x) and the two tangent lines appear below. y 3 2 −1

x −1 −2 −3

0.5

1

1.5

2

In Exercises 63–68, calculate y . The following table gives the percentage of voters supporting each of two candidates in the days before an − 5x 2 +the 3x average ROC of A’s percentage over the intervals from day 20 to 15, day 15 to 10, and day 10 63. election. y = 12x 3Compute to 5. If this trend continues over the last 5 days before the election, will A win? SOLUTION Let y = 12x 3 − 5x 2 + 3x. Then Days before2Election 20 15 10 5 y = 36x − 10x + 3 and y = 72x − 10. Candidate A 44.8% 46.8% 48.3% 49.3%

√ Candidate B 65. y = 2x −2/5 +3 y=x √ SOLUTION Let y = 2x + 3 = (2x + 3)1/2 . Then y =

1 d (2x + 3)−1/2 (2x + 3) = (2x + 3)−1/2 2 dx

55.2% 53.2% 51.7% 50.7%

and

1 d y = − (2x + 3)−3/2 (2x + 3) = −(2x + 3)−3/2 . 2 dx

2) 67. y = tan(x4x y= SOLUTION x Let + 1y = tan(x 2 ). Then

y = 2x sec2 (x 2 ) and d sec(x 2 ) + 2 sec2 (x 2 ) = 8x 2 sec2 (x 2 ) tan(x 2 ) + 2 sec2 (x 2 ). y = 2x 2 sec(x 2 ) dx 69. In Figure 5, label the graphs f , f , and f . y = sin2 (x + 9) y

y

x

FIGURE 5

x


155

SOLUTION First consider the plot on the left. Observe that the green curve is nonnegative whereas the red curve is increasing, suggesting that the green curve is the derivative of the red curve. Moreover, the green curve is linear with negative slope for x < 0 and linear with positive slope for x > 0 while the blue curve is a negative constant for x < 0 and a positive constant for x > 0, suggesting the blue curve is the derivative of the green curve. Thus, the red, green and blue curves, respectively, are the graphs of f , f and f . Now consider the plot on the right. Because the red curve is decreasing when the blue curve is negative and increasing when the blue curve is positive and the green curve is decreasing when the red curve is negative and increasing when the red curve is positive, it follows that the green, red and blue curves, respectively, are the graphs of f , f and f .

q of frozen chocolate cakes a commercial can sell p, per week depends on the price 71. Let qThe be anumber differentiable function of p, say, q =that f ( p). If p changesbakery by an amount then the percentage change p. Ininthis the percentage ROC of q withpercentage respect to change p is called price elasticity of demand Assume p issetting, (100p)/ p, and the corresponding in q the is (100q)/q, where q = f E( ( p p). + p) − f (that p). q = The 50 p(10 − p) forrate 5 0 and consider the limit f (x) − f (0) |x|α = lim . x −0 x→0 x→0 x lim

If 0 < α < 1, then |x|α = −∞ x→0− x lim

while

|x|α =∞ x→0+ x lim

and f (0) does not exist. If α = 1, then lim

|x|

x→0− x

= −1

while

lim

|x|

x→0+ x

=1

and f (0) again does not exist. Finally, if α > 1, then |x|α = 0, x→0 x lim

so f (0) does exist. In summary, f (x) = |x|α is differentiable at x = 0 when α = 0 and when α > 1. is the water level rising when the water level 85. Water pours into the tank−1 in Figure 7 at a rate of 20 m3 /min. How fast 2 sin(x ) for x = 0 and f (0) = 0. Show that f (x) exists for all x (including x = 0) but that Let f (x) = x is h = 4 m? f (x) is not continuous at x = 0 (Figure 6).

36 m 8m 10 m 24 m

FIGURE 7


157

SOLUTION When the water level is at height h, the length of the upper surface of the water is 24 + 3 2 h and the volume of water in the trough is 3 15 1 V = h 24 + 24 + h (10) = 240h + h 2 . 2 2 2

Therefore, dh dV = (240 + 15h) = 20 m3 /min. dt dt When h = 4, we have 20 1 dh = = m/min. dt 240 + 15(4) 15 87. A light moving at 3 ft/s approaches a 6-ft man standing 12 ft from a wall (Figure 8). The light is 3 ft above the The minute hand of a clock is 4 in. long and the hour hand is 3 in. long. How fast is the distance between the tips ground. How fast is the tip P of the man’s shadow moving when the light is 24 ft from the wall? of the hands changing at 3 o’clock?

3 ft 12 ft

3 ft/s

FIGURE 8 SOLUTION

Let x denote the distance between the man and the light. Using similar triangles, we find P −3 3 = x 12 + x

or

P=

36 + 6. x

Therefore, dP 36 d x =− 2 . dt x dt When the light is 24 feet from the wall, x = 12. With ddtx = −3, we have dP 36 = − 2 (−3) = 0.75 ft/s. dt 12 89. (a) Side x of the triangle in Figure 9 is increasing at 2 cm/s and side y is increasing at 3 cm/s. Assume that θ A bead slides down the curve x y = 10. Find the bead’s horizontal velocity if its height at time t seconds is decreases in such a2 way that the area of the triangle has the constant value 4 cm2 . How fast is θ decreasing when x = 4, y = 80 − 16t cm. y = 4? (b) How fast is the distance between P and Q changing when x = 4, y = 4? P x

y

Q

FIGURE 9 SOLUTION

(a) The area of the triangle is A=

1 x y sin θ = 4. 2

Differentiating with respect to t, we obtain dA 1 1 dθ dx dy = x y cos θ + y sin θ + x sin θ = 0. dt 2 dt 2 dt dt When x = y = 4, we have 12 (4)(4) sin θ = 4, so sin θ = 12 . Thus, θ = π6 and √ 1 1 1 1 1 3 dθ (4)(4) + (4) (2) + (4) (3) = 0. 2 2 dt 2 2 2 2

158

CHAPTER 3


Solving for d θ /dt, we find dθ 5 = − √ ≈ −0.72 rad/s. dt 4 3 (b) By the Law of Cosines, the distance D between P and Q satisfies D 2 = x 2 + y 2 − 2x y cos θ , so 2D

dD dθ dy dx dx dy = 2x + 2y + 2x y sin θ − 2x cos θ − 2y cos θ . dt dt dt dt dt dt

With x = y = 4 and θ = π6 , D=

42 + 42 − 2(4)(4)

√

√ 3 = 4 2 − 3. 2

Therefore, √ √ 20 − 12 3 − 8 3 16 + 24 − √ dD 3 = ≈ −1.50 cm/s. √ dt 8 2− 3

APPLICATIONS OF 4 THE DERIVAT IVE 4.1 Linear Approximation and Applications Preliminary Questions 1. Estimate g(1.2) − g(1) if g (1) = 4. SOLUTION

Using the Linear Approximation, g(1.2) − g(1) ≈ g (1)(1.2 − 1) = 4(0.2) = 0.8.

2. Estimate f (2.1), assuming that f (2) = 1 and f (2) = 3. SOLUTION

Using the Linear Approximation, f (2.1) ≈ f (2) + f (2)(2.1 − 2) = 1 + 3(0.1) = 1.3

3. The velocity of a train at a given instant is 110 ft/s. How far does the train travel during the next half-second (use the Linear Approximation)? SOLUTION Using the Linear Approximation, we estimate that the train travels at the constant velocity of 110 ft/sec over the next half-second; hence, we estimate that the train will travel 55 ft over the next half-second.

4. Discuss how the Linear Approximation makes the following statement more precise: The sensitivity of the output to a small change in input depends on the derivative. SOLUTION The Linear Approximation tells us that the change in the output value of a function is approximately equal to the value of the derivative times the change in the input value.

5. Suppose that the linearization of f (x) at a = 2 is L(x) = 2x + 4. What are f (2) and f (2)? SOLUTION

The linearization of f (x) at a = 2 is L(x) = f (2) + f (2)(x − 2) = f (2) · x + ( f (2) − 2 f (2))

Matching this to the given expression, L(x) = 2x + 4, it follows that f (2) = 2. Moreover, f (2) − 2 f (2) = 4, so f (2) = 8.

Exercises In Exercises 1–4, use the Linear Approximation to estimate f = f (3.02) − f (3) for the given function. 1. f (x) = x 2 SOLUTION

Let f (x) = x 2 . Then f (x) = 2x and f ≈ f (3)x = 6(.02) = 0.12.

3. f (x) = x −1 4 f (x) = x 1 SOLUTION Let f (x) = x −1 . Then f (x) = −x −2 and f ≈ f (3)x = − (.02) = −.00222. 9 In Exercises 5–10, 1estimate f using the Linear Approximation and use a calculator to compute both the error and the f (x)error. = percentage x +1 5. f (x) = x − 2x 2 ,

a = 5,

x = −0.4

Let f (x) = x − 2x 2 , a = 5, and x = −0.4. Then f (x) = 1 − 4x, f (a) = f (5) = −19 and f ≈ f (a)x = (−19)(−0.4) = 7.6. The actual change is SOLUTION

f = f (a + x) − f (a) = f (4.6) − f (5) = −37.72 − (−45) = 7.28. The error in the Linear Approximation is therefore |7.28 − 7.6| = 0.32; in percentage terms, the error is 0.32 × 100% ≈ 4.40% 7.28 f (x) =

√

1 + x,

a = 8, x = 1

160

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E

7. f (x) =

1 , 1 + x2

a = 3,

x = 0.5

Let f (x) = 1 2 , a = 3, and x = .5. Then f (x) = − 2x2 2 , f (a) = f (3) = −.06 and f ≈ 1+x (1+x ) f (a)x = −.06(.5) = −0.03. The actual change is

SOLUTION

f = f (a + x) − f (a) = f (3.5) − f (3) ≈ −.0245283. The error in the Linear Approximation is therefore | − .0245283 − (−.03)| = .0054717; in percentage terms, the error is .0054717 −.0245283 × 100% ≈ 22.31% 9. f (x) = tan x, a = π4 , x = 0.013 f (x) = sin x, a = 0, x = 0.02 π π SOLUTION Let f (x) = tan x, a = 4 , and x = .013. Then f (x) = sec2 x, f (a) = f ( 4 ) = 2 and f ≈ f (a)x = 2(.013) = .026. The actual change is

π

π ≈ 1.026344 − 1 = .026344. f = f (a + x) − f (a) = f + .013 − f 4 4 The error in the Linear Approximation is therefore |.026344 − .026| = .000344; in percentage terms, the error is .000344 × 100% ≈ 1.31% .026344 In Exercises estimate using the Linear Approximation and find the error using a calculator. π , quantity x = 0.03 f (x) 11–16, = cos x, a = the 4 √ √ 11. 26 − 25 √ 1 1 SOLUTION Let f (x) = x, a = 25, and x = 1. Then f (x) = 2 x −1/2 and f (a) = f (25) = 10 . • The Linear Approximation is f ≈ f (a)x = 1 (1) = .1. 10 • The actual change is f = f (a + x) − f (a) = f (26) − f (25) ≈ .0990195. • The error in this estimate is |.0990195 − .1| = .000980486.

1 1 1/4 − 161/4 13. √ 16.5− 101 10 SOLUTION

−0.0005.

1 )= Let f (x) = √1 , a = 100, and x = 1. Then f (x) = ddx (x −1/2 ) = − 12 x −3/2 and f (a) = − 12 ( 1000 x

• The Linear Approximation is f ≈ f (a)x = −0.0005(1) = −0.0005. • The actual change is

f = f (a + x) − f (a) = √

1 101

−

1 = −0.000496281. 10

• The error in this estimate is |−0.0005 − (−0.000496281)| = 3.71902 × 10−6 .

15. sin(0.023) Hint: Estimate sin(0.023) − sin(0). 1 1 √ − SOLUTION 98 Let10f (x) = sin x, a = 0, and x = .023. Then f (x) = cos x and f (a) = f (0) = 1. • The Linear Approximation is f ≈ f (a)x = 1(.023) = .023. • The actual change is f = f (a + x) − f (a) = f (0.023) − f (0) = 0.02299797. • The error in this estimate is |.023 − .02299797| ≈ 2.03 × 10−6 .

17. The cube root of 27 is 3. How much larger is the cube root of 27.2? Estimate using the Linear Approximation. (15.8)1/4 1 1 SOLUTION Let f (x) = x 1/3 , a = 27, and x = .2. Then f (x) = 3 x −2/3 and f (a) = f (27) = 27 . The Linear Approximation is 1 f ≈ f (a)x = (.2) = .0074074 27 Approximation. Note: You must express θ in radians. 19. Estimate sin 61◦ − sin √ the√Linear √ √ 60◦ using Which is larger: 2.1 − 2 or 9.1 − 9? Explain using the Linear Approximation.

S E C T I O N 4.1

Linear Approximation and Applications

161

π 1 π SOLUTION Let f (x) = sin x, a = π 3 , and x = 180 . Then f (x) = cos x and f (a) = f ( 3 ) = 2 . Finally, the Linear Approximation is

1 π π f ≈ f (a)x = = ≈ .008727 2 180 360 21. Atmospheric pressure P at altitude h = 40,000 ft is 390 lb/ft2 . Estimate P at h = 41,000 if d P/dh = T = 25◦ C A thin silver wire has length L = 18 cm when the temperature is T = 30◦ C. Estimate the length whenh=40,000 3. −5◦ −1 −0.0188 lb/ft if the coefficient of thermal expansion is k = 1.9 × 10 C (see Example 3). SOLUTION

Let P(h) by the pressure P at altitude h. By the Linear Approximation, P(41000) − P(40000) ≈

d P h = −.0188 lbs/ft3 (1000 ft) = −18.8 lbs/ft2 . dh h=40000

Hence, P(41000) ≈ 390 lbs/ft2 − 18.8 lbs/ft2 = 371.2 lbs/ft2 . 23. The side s of a square carpet is measured at 6 ft. Estimate ◦the maximum error in the area A of the carpet if s◦ is resistance a copper wire at temperature T = 20 C is R = 15 . Estimate the resistance at T = 22 C, accurateThe to within half R anof inch. assuming that d R/d T T =20 = 0.06 /◦ C. SOLUTION Let s be the length in feet of the side of the square carpet. Then A(s) = s 2 is the area of the carpet. With 1 (note that 1 inch equals 1 foot), an estimate of the size of the error in the area is given by the Linear a = 6 and s = 24 12 Approximation: 1 A ≈ A (6)s = 12 = .5 ft2 24 25. A stone tossed vertically in the air with velocitythe v ft/s reaches a maximum height of h= v 2 /64 ft. increases A spherical balloon has a radius of 6initial in. Estimate change in volume and surface area if the radius (a) Estimate by 0.3 in.h if v is increased from 25 to 26 ft/s. (b) Estimate h if v is increased from 30 to 31 ft/s. (c) In general, does a 1 ft/s increase in initial velocity cause a greater change in maximum height at low or high initial velocities? Explain. SOLUTION

A stone tossed vertically with initial velocity v ft/s attains a maximum height of h = v 2 /64 ft.

1 (25)(1) = .78125 ft. (a) If v = 25 and v = 1, then h ≈ h (v)v = 32

1 (30)(1) = .9375 ft. (b) If v = 30 and v = 1, then h ≈ h (v)v = 32 (c) A one foot per second increase in initial velocity v increases the maximum height by approximately v/32 ft. Accordingly, there is a bigger effect at higher velocities. 2 ft, 27. TheIfstopping automobile (aftertoapplying the brakes) F(s) = 1.1s the pricedistance of a bus for passanfrom Albuquerque Los Alamos is set atisx approximately dollars, a bus company takes+in0.054s a monthly where s is the in 1.5x mph.−Use the2Linear Approximation to estimate the change in stopping distance per additional (in thousands of dollars). revenue ofspeed R(x) = 0.01x mph when s = 35 and when s = 55. (a) Estimate the change in revenue if the price rises from $50 to $53. 2. SOLUTION Let that F(s)x==1.1s .054swill (b) Suppose 80.+How revenue be affected by a small increase in price? Explain using the Linear •Approximation. The Linear Approximation at s = 35 mph is

F ≈ F (35)s = (1.1 + .108 × 35)s = 4.88s ft The change in stopping distance per additional mph for s = 35 mph is approximately 4.88 ft. • The Linear Approximation at s = 55 mph is

F ≈ F (55)s = (1.1 + .108 × 55)s = 7.04s ft The change in stopping distance per additional mph for s = 55 mph is approximately 7.04 ft. 29. Estimate the weight loss per mile of altitude gained for a 130-lb pilot. At which altitude would she weigh 129.5 lb? Juan measures the circumference C of a spherical ball at 40 cm and computes the ball’s volume V . Estimate the See Example 4. maximum possible error in V if the error in C is at most 2 cm. Recall that C = 2π r and V = 43 π r 3 , where r is the SOLUTION From the discussion in the text, the weight loss W at altitude h (in miles) for a person weighing W0 at ball’s radius. the surface of the earth is approximately W ≈ −

1 W h 1980 0

If W0 = 130 pounds, then W ≈ −0.066h. Accordingly, the pilot loses approximately 0.066 pounds per mile of altitude gained. She will weigh 129.5 pounds at the altitude h such that −0.066h = −0.5, or h = 0.5/0.066 ≈ 7.6 miles. 31. The volume of a certain gas (in liters) is related to pressure P (in atmospheres) by the formula P V = 24. Suppose How much would a 160-lb astronaut weigh in a satellite orbiting the earth at an altitude of 2,000 miles? Estimate that V = 5 with a possible error of ±0.5 L. the astronaut’s weight loss per additional mile of altitude beyond 2,000.

162

CHAPTER 4


(a) Compute P and estimate the possible error. (b) Estimate the maximum allowable error in V if P must have an error of at most 0.5 atm. SOLUTION

(a) We have P = 24V −1 . For V = 5 L, P = 24/5 = 4.8 atm. If the error in computing the volume is ±0.5 liters, then the error in computing the pressure is approximately 24 12 |P| ≈ |P (V )V | = | − 24V −2 V | = − (±.5) = = .48 atm 25 25 (b) When the volume is V = 5 L, the error in computing the pressure is approximately 24 24 −2 |V | = .5 atm |P| ≈ |P (V )V | = | − 24V V | = − V = 25 25 Thus, the maximum allowable error in V is approximately V = ±.520833 L. In Exercises 33–38, find the linearization at x = a. The dosage D of diphenhydramine for a dog of body mass w kg is D = kw 2/3 mg, where k is a constant. A cocker 33. y = cosspaniel x sin x,hasamass = 0 w = 10 kg according to a veterinarian’s scale. Estimate the maximum allowable error in w if the percentage error in the dosage1D must be less than 5%. SOLUTION Let f (x) = sin x cos x = 2 sin 2x. Then f (x) = cos 2x. The linearization at a = 0 is L(x) = f (a)(x − a) + f (a) = 1(x − 0) + 0 = x. 35. y = (1 + x)−1/2 , a = 0 π y = cos x sin x, a = 1 SOLUTION Let f (x) = (1 +4x)−1/2 . Then f (x) = − 2 (1 + x)−3/2 . The linearization at a = 0 is 1 L(x) = f (a)(x − a) + f (a) = − x + 1. 2 , a=0 37. y = (1 + x 2 )−1/2 y = (1 + x)−1/2 , a = 32 −1/2 SOLUTION Let f (x) = (1 + x ) . Then f (x) = −x(1 + x 2 )−3/2 , f (a) = 1 and f (a) = 0, so the linearization at a is L(x) = f (a)(x − a) + f (a) = 1. √ √ Estimate 16.2 39. π using the linearization L(x) of f (x) = x at a sin x y = determine , a if=the estimate is too large or too small. of axes and x 2 SOLUTION Let f (x) = x 1/2 , a = 16, and x = .2. Then f (x) = linearization to f (x) is

= 16. Plot f (x) and L(x) on the same set 1 x −1/2 and f (a) = f (16) = 1 . The 2 8

1 1 L(x) = f (a)(x − a) + f (a) = (x − 16) + 4 = x + 2. 8 8

√ Thus, we have 16.2 ≈ L(16.2) = 4.025. Graphs of f (x) and L(x) are shown below. Because the graph of L(x) lies above the graph of f (x), we expect that the estimate from the Linear Approximation is too large. y 5 4 3 2 1

L(x)

0

5

f (x) x 10

15

20

25

√ √ to compute the percentage error. In Exercises 41–46, approximate using linearization and use a calculator Estimate 1/ 15 using a suitable linearization of f (x) = 1/ x. Plot f (x) and L(x) on the same set of axes √ 41. and 17determine if the estimate is too large or too small. Use a calculator to compute the percentage error. SOLUTION

to f (x) is

Let f (x) = x 1/2 , a = 16, and x = 1. Then f (x) = 12 x −1/2 , f (a) = f (16) = 18 and the linearization

Thus, we have

√

1 1 L(x) = f (a)(x − a) + f (a) = (x − 16) + 4 = x + 2. 8 8 17 ≈ L(17) = 4.125. The percentage error in this estimate is √ 17 − 4.125 √ × 100% ≈ .046% 17

S E C T I O N 4.1

Linear Approximation and Applications

163

43. (17)1/4 1 √ 1 1 SOLUTION Let f (x) = x 1/4 , a = 16, and x = 1. Then f (x) = 4 x −3/4 , f (a) = f (16) = 32 and the linearization 17 to f (x) is 1 1 3 (x − 16) + 2 = x+ . L(x) = f (a)(x − a) + f (a) = 32 32 2 Thus, we have (17)1/4 ≈ L(17) = 2.03125. The percentage error in this estimate is (17)1/4 − 2.03125 × 100% ≈ .035% (17)1/4 45. (27.001)1/31/3 (27.03) 1 1 SOLUTION Let f (x) = x 1/3 , a = 27, and x = .001. Then f (x) = 3 x −2/3 , f (a) = f (27) = 27 and the linearization to f (x) is 1 1 L(x) = f (a)(x − a) + f (a) = (x − 27) + 3 = x + 2. 27 27 Thus, we have (27.001)1/3 ≈ L(27.001) ≈ 3.0000370370370. The percentage error in this estimate is (27.001)1/3 − 3.0000370370370 × 100% ≈ .000000015% (27.001)1/3

π on the same set of axes. 5/3 f (x) = tan x and its linearization L(x) at a = (1.2)Plot 4 (a) Does the linearization overestimate or underestimate f (x)? (b) Show, by graphing y = f (x) − L(x) and y = 0.1 on the same set of axes, that the error | f (x) − L(x)| is at most 0.1 for 0.55 ≤ x ≤ 0.95. (c) Find an interval of x-values on which the error is at most 0.05. 47.

SOLUTION

Let f (x) = tan x and a = π4 . Then f (x) = sec2 x, f (a) = 2 and

π L(x) = f (a) + f (a)(x − a) = 1 + 2 x − . 4

Graphs of f (x) and L(x) are shown below. y 2 x 1 −2

(a) From the figure above, we see that the graph of the linearization L(x) lies below the graph of f (x); thus, the linearization underestimates f (x). (b) The graph of y = f (x) − L(x) is shown below at the left. Below at the right, portions of the graphs of y = f (x) − L(x) and y = 0.1 are shown. From the graph on the right, we see that | f (x) − L(x)| ≤ 0.1 for 0.55 ≤ x ≤ 0.95.

y 4 3 2 1 x 0

x

1

0.5 0.6 0.7 0.8 0.9

1

(c) Portions of the graphs of y = f (x) − L(x) and y = 0.05 are shown below. From the graph, we see that | f (x) − L(x)| ≤ 0.05 roughly for 0.62 < x < 0.93.

x 0.6

0.7

0.8

0.9

164

CHAPTER 4


In Exercises 49–50, use the following fact derived from2 Newton’s Laws: An object released at an angle θ with initial Compute the linearization L(x) 1of 2f (x) = x − x 3/2 at a = 2. Then plot f (x) − L(x) and find an interval velocity v ft/s travels a total distance s = 32 v sin 2θ ft (Figure 8). around a = 1 such that | f (x) − L(x)| ≤ 0.1. 49. A player located 18.1 ft from a basket launches a successful jump shot from a height of 10 ft (level with the rim of the basket), at an angle θ = 34◦ and initial velocity of v = 25 ft/s. y

q

x

FIGURE 8 Trajectory of an object released at an angle θ .

(a) Show that the distance s of the shot changes by approximately 0.255θ ft if the angle changes by an amount θ . Remember to convert the angles to radians in the Linear Approximation. (b) Is it likely that the shot would have been successful if the angle were off by 2◦ ? 1 v 2 sin 2t = Using Newton’s laws and the given initial velocity of v = 25 ft/s, the shot travels s = 32 625 sin 2t ft, where t is in radians. 32

SOLUTION

(a) If θ = 34◦ (i.e., t = 17 90 π ), then

s ≈ s (t)t =

625 π 17 17 625 π t = π θ · cos cos ≈ 0.255θ . 16 45 16 45 180

(b) If θ = 2◦ , this gives s ≈ 0.51 ft, in which case the shot would not have been successful, having been off half a foot. 51. Compute the linearization of f (x) = 3x − 4 at a = 0 and a = 2. Prove more generally that a linear function Estimate change inatthe distance of the shot if the angle changes from 50◦ to 51◦ for v = 25 ft/s and coincides with itsthe linearization x= a for alls a. v = 30 ft/s. Is the shot more sensitive to the angle when the velocity is large or small? Explain. SOLUTION Let f (x) = 3x − 4. Then f (x) = 3. With a = 0, f (a) = −4 and f (a) = 3, so the linearization of f (x) at a = 0 is L(x) = −4 + 3(x − 0) = 3x − 4 = f (x). With a = 2, f (a) = 2 and f (a) = 3, so the linearization of f (x) at a = 2 is L(x) = 2 + 3(x − 2) = 2 + 3x − 6 = 3x − 4 = f (x). More generally, let g(x) = bx + c be any linear function. The linearization L(x) of g(x) at x = a is L(x) = g (a)(x − a) + g(a) = b(x − a) + ba + c = bx + c = g(x); i.e., L(x) = g(x). 53. Show that the Linear Approximation to f (x) = tan x at x = π4 yields the estimate tan( π4 + h) − 1 ≈ 2h. Compute According to (3), error in the Linear Approximation is of “order two” in h. Show that the Linear Approxi√ √,the the error E for h = 10−n 1 ≤ n ≤ 4, and verify that E satisfies the Error Bound (3) with K = 6.2. mation to f (x) = x at x = 9 yields the estimate 9 + h − 3 ≈ 16 h. Then compute the error E for h = 10−n , SOLUTION Let = tan x. Then fthat ( π4 )E=≤1,0.006h f (x) 2=. sec2 x and f ( π4 ) = 2. Therefore, by the Linear Approxima1 ≤ n ≤ 4, andf (x) verify numerically tion,

π

π

π = tan f +h − f + h − 1 ≈ 2h. 4 4 4 From the following table, we see that for h = 10−n , 1 ≤ n ≤ 4, E ≤ 6.2h 2 . h

E = | tan( π4 + h) − 1 − 2h|

6.2h 2

10−1 10−2 10−3 10−4

2.305 × 10−2 2.027 × 10−4 2.003 × 10−6 2.000 × 10−8

6.20 × 10−2 6.20 × 10−4 6.20 × 10−6 6.20 × 10−8

S E C T I O N 4.2

Extreme Values

165

Further Insights and Challenges 55. (a) Show that f (x) = sin x and g(x) = tan x have the same linearization at a = 0. Show that forisany real numbermore k, (1accurately? + x)k ≈ 1 Explain + kx forusing smallax. Estimate (1.02) π ].0.7 and (1.02)−0.3 . (b) Which function approximated graph over [0, 6

(c) Calculate the error in these linearizations at x = π6 . Does the answer confirm your conclusion in (b)? SOLUTION Let f (x) = sin x and g(x) = tan x. (a) The Linear Approximation of f (x) at a = 0 is L(x) = f (0)(x − 0) + f (0) = cos 0 · (x − 0) + sin 0 = x, whereas the Linear Approximation of g(x) at a = 0 is L(x) = f (0)(x − 0) + f (0) = sec2 0 · (x − 0) + tan 0 = x. (b) The linearization L(x) = x more closely approximates the function f (x) = sin x than g(x) = tan x, since the graph of x lies closer to that of sin x than to tan x. y

tan x x sin x

0.5 0.4 0.3 0.2 0.1 x 0

0.1 0.2 0.3 0.4 0.5

(c) The error in the Linear Approximation for f at x = π6 is sin( π6 ) − π6 ≈ −.02360; the error in the Linear Approximation for g at x = π6 is tan( π6 ) − π6 ≈ .05375. These results confirm what the graphs of the curves told us in (b). error in the Linear Approximation at a = 1. Show directly that 57. Let f (x) = x −1 and let E = | f − f (1)h| be the 2 , and let E In = this | f case, − f (5)h| the error f (5 + h) −that f (5), where 1 . Hint: 1 ≤ 1be 3 . in the Linear ApproxiE = h 2Let /(1 + fh).=Then prove E≤ 2h 2 iff (x) − 12 = ≤ xh ≤ + h ≤ 2 in h). 2 mation. Verify directly that E satisfies (3) with K = 22 (thus E is of order two SOLUTION Let f (x) = x −1 . Then f = f (1 + h) − f (1) =

1 h −1=− 1+h 1+h

and

E = | f − f (1)h| = −

h h2 + h = . 1+h 1+h

1 ≤ 2. Thus, E ≤ 2h 2 for − 1 ≤ h ≤ 1 . If − 12 ≤ h ≤ 12 , then 12 ≤ 1 + h ≤ 32 and 23 ≤ 1+h 2 2

Consider the curve y 4 + x 2 + x y = 13. Calculate d y/d x at P = (3, 1) and find the linearization y = L(x) at P. 4.2 (a)Extreme Values (b) Plot the curve and y = L(x) on the same set of axes. (c) Although y is only defined implicitly as a function of x, we can use y = L(x) to approximate y for x near 3. Use Preliminary Questions this idea to find an approximate solution of y 4 + 3.12 + 3.1y = 13. Based on your plot in (b), is your approximation 1. If f (x) is continuous on (0, 1), then (choose the correct statement): too large or too small? (a) f (x) has a minimum value on (0, 1). (d) Solve y 4 + 3.12 + 3.1y = 13 numerically using a computer algebra system and use the result to find the (b) percentage f (x) has noerror minimum on (0, 1). in yourvalue approximation. (c) f (x) might not have a minimum value on (0, 1). SOLUTION The correct response is (c): f (x) might not have a minimum value on (0, 1). The key is that the interval (0, 1) is not closed. The function might have a minimum value but it does not have to have a minimum value.

2. If f (x) is not continuous on [0, 1], then (choose the correct statement): (a) f (x) has no extreme values on [0, 1]. (b) f (x) might not have any extreme values on [0, 1]. SOLUTION The correct response is (b): f (x) might not have any extreme values on [0, 1]. Although [0, 1] is closed, because f is not continuous, the function is not guaranteed to have any extreme values on [0, 1].

3. What is the definition of a critical point? A critical point is a value of the independent variable x in the domain of a function f at which either f (x) = 0 or f (x) does not exist.

SOLUTION

4. True or false: If f (x) is differentiable and f (x) = 0 has no solutions, then f (x) has no local minima or maxima. True. If f (x) is differentiable but f (x) = 0 has no solutions, then f (x) has no critical points. As critical points are the only candidates for local minima and local maxima, it follows that f (x) has no local extrema. SOLUTION

166

CHAPTER 4


5. Fermat’s Theorem does not claim that if f (c) = 0, then f (c) is a local extreme value (this is false). What does Fermat’s Theorem assert? SOLUTION

Fermat’s Theorem claims: If f (c) is a local extreme value, then either f (c) = 0 or f (c) does not exist.

6. If f (x) is continuous but has no critical points in [0, 1], then (choose the correct statement): (a) f (x) has no min or max on [0, 1]. (b) Either f (0) or f (1) is the minimum value on [0, 1]. SOLUTION The correct response is (b): either f (0) or f (1) is the minimum value on [0, 1]. Remember that extreme values occur either at critical points or endpoints. If a continuous function on a closed interval has no critical points, the extreme values must occur at the endpoints.

Exercises 1. The following questions refer to Figure 15. y 6 5 4 3

f(x)

2 1 x 1

2

3

4

5

6

7

8

FIGURE 15

(a) (b) (c) (d) (e)

How many critical points does f (x) have? What is the maximum value of f (x) on [0, 8]? What are the local maximum values of f (x)? Find a closed interval on which both the minimum and maximum values of f (x) occur at critical points. Find an interval on which the minimum value occurs at an endpoint.

SOLUTION

(a) f (x) has three critical points on the interval [0, 8]: at x = 3, x = 5 and x = 7. Two of these, x = 3 and x = 5, are where the derivative is zero and one, x = 7, is where the derivative does not exist. (b) The maximum value of f (x) on [0, 8] is 6; the function takes this value at x = 0. (c) f (x) achieves a local maximum of 5 at x = 5. (d) Answers may vary. One example is the interval [4, 8]. Another is [2, 6]. (e) Answers may vary. The easiest way to ensure this is to choose an interval on which the graph takes no local minimum. One example is [0, 2]. In Exercises 3–10, find all critical points of the function. State whether f (x) = x −1 (Figure 16) has a minimum or maximum value on the following intervals: (b) (1, 2) (c) [1, 2] 3. (a) f (x)(0, =2) x 2 − 2x + 4 SOLUTION

Let f (x) = x 2 − 2x + 4. Then f (x) = 2x − 2 = 0 implies that x = 1 is the lone critical point of f .

9 3− 5. f (x)f (x) = x= 7x −x 22 − 54x + 2 2 Let f (x) = x 3 − 92 x 2 − 54x + 2. Then f (x) = 3x 2 − 9x − 54 = 3(x + 3)(x − 6) = 0 implies that x = −3 and x = 6 are the critical points of f . x 7. f (x) = 2 3 2 f (t) = x 8t + 1− t

SOLUTION

SOLUTION

x 1 − x2 . Then f (x) = 2 Let f (x) = 2 = 0 implies that x = ±1 are the critical points of f . x +1 (x + 1)2

9. f (x) = x 1/3 f (t) = 4t − t 2 + 1 1 SOLUTION Let f (x) = x 1/3 . Then f (x) = 3 x −2/3 . The derivative is never zero but does not exist at x = 0. Thus, x = 0 is the only critical point of f . 11. (a) (b) (c)

Let f (x) = x 2 2− 4x + 1. x = Findf (x) the critical point c of f (x) and compute f (c). x +1 Compute the value of f (x) at the endpoints of the interval [0, 4]. Determine the min and max of f (x) on [0, 4].

S E C T I O N 4.2

Extreme Values

167

(d) Find the extreme values of f (x) on [0, 1]. SOLUTION

Let f (x) = x 2 − 4x + 1.

(a) Then f (c) = 2c − 4 = 0 implies that c = 2 is the sole critical point of f . We have f (2) = −3. (b) f (0) = f (4) = 1. (c) Using the results from (a) and (b), we find the maximum value of f on [0, 4] is 1 and the minimum value is −3. (d) We have f (1) = −2. Hence the maximum value of f on [0, 1] is 1 and the minimum value is −2. 13. Find the critical points of f (x) = sin x + cos x and determine the extreme values on [0, π2 ]. Find the extreme values of 2x 3 − 9x 2 + 12x on [0, 3] and [0, 2]. SOLUTION

• Let f (x) = sin x + cos x. Then on the interval 0, π , we have f (x) = cos x − sin x = 0 at x = π , the only 2 4 critical point of f in this interval. √ √ 2 and f (0) = f ( π2 ) = 0, the maximum value of f on 0, π2 is 2, while the minimum value is 0.

• Since f ( π ) = 4

√ Plot f (x) = 2 x − x on [0, 4] and determine the maximum value graphically. Then verify your answer 15. Compute the critical points of h(t) = (t 2 − 1)1/3 . Check that your answer is consistent with Figure 17. Then using calculus. find the extreme values of h(t)√on [0, 1] and [0, 2]. SOLUTION The graph of y = 2 x − x over the interval [0, 4] is shown below. From the graph, we see that at x = 1, the function achieves its maximum value of 1. y 1 0.8 0.6 0.4 0.2 x 0

1

2

3

4

√ To verify the information obtained from the plot, let f (x) = 2 x − x. Then f (x) = x −1/2 − 1. Solving f (x) = 0 yields the critical points x = 0 and x = 1. Because f (0) = f (4) = 0 and f (1) = 1, we see that the maximum value of f on [0, 4] is 1. In Exercises 17–45, find the maximum and minimum values of the function on the given interval. Plot f (x) = 2x 3 − 9x 2 + 12x on [0, 3] and locate the extreme values graphically. Then verify your answer 2 − 4x + 2, [0, 3] 17. using y = 2xcalculus. SOLUTION Let f (x) = 2x 2 − 4x + 2. Then f (x) = 4x − 4 = 0 implies that x = 1 is a critical point of f . On the interval [0, 3], the minimum value of f is f (1) = 0, whereas the maximum value of f is f (3) = 8. (Note: f (0) = 2.)

− 1, [−2, 2] 19. y = x 2 − 6x y = −x 2 + 10x + 43, [3, 8] SOLUTION Let f (x) = x 2 − 6x − 1. Then f (x) = 2x − 6 = 0 implies that x = 3 is a critical point of f . The minimum of f on the interval [−2, 2] is f (2) = −9, whereas its maximum is f (−2) = 15. (Note: The critical point x = 3 is outside the interval [−2, 2].) 21. y = −4x 22 + 3x + 4, [−1, 1] y = x − 6x − 1, [−2, 0] 3 SOLUTION Let f (x) = −4x 2 + 3x + 4. Then f (x) = −8x + 3 = 0 implies that x = 8 is a critical point of f . The 3 minimum of f on the interval [−1, 1] is f (−1) = −3, whereas its maximum is f ( 8 ) = 4.5625. (Note: f (1) = 3.) 23. y = x 3 −36x + 1, [−1, 1] y = x − 3x + 1, [0, 2] √ SOLUTION Let f (x) = x 3 − 6x + 1. Then f (x) = 3x 2 − 6 = 0 implies that x = ± 2 are the critical points of f . The minimum of f on the interval [−1, 1] is f (1) = −4, whereas its maximum is f (−1) = 6. (Note: The critical points √ x = ± 2 are not in the interval [−1, 1].) 25. y = x 3 +33x 2 − 9x + 2, [−1, 1] y = x − 12x 2 + 21x, [0, 2] SOLUTION Let f (x) = x 3 + 3x 2 − 9x + 2. Then f (x) = 3x 2 + 6x − 9 = 3(x + 3)(x − 1) = 0 implies that x = 1 and x = −3 are the critical points of f . The minimum of f on the interval [−1, 1] is f (1) = −3, whereas its maximum is f (−1) = 13. (Note: The critical point x = −3 is not in the interval [−1, 1].) 27. y = x 3 +33x 2 −29x + 2, [−4, 4] y = x + 3x − 9x + 2, [0, 2] SOLUTION Let f (x) = x 3 + 3x 2 − 9x + 2. Then f (x) = 3x 2 + 6x − 9 = 3(x + 3)(x − 1) = 0 implies that x = 1 and x = −3 are the critical points of f . The minimum of f on the interval [−4, 4] is f (1) = −3, whereas its maximum is f (4) = 78. (Note: f (−4) = 22 and f (−3) = 29.) 29. y = x 5 −53x 2 , [−1, 5] y = x − x, [0, 2]

168

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E 1/3 ≈ 1.06 are SOLUTION Let f (x) = x 5 − 3x 2 . Then f (x) = 5x 4 − 6x = 0 implies that x = 0 and x = 1 5 (150) critical points of f . Over the interval [−1, 5], the minimum value of f is f (−1) = −4, whereas its maximum value is 9 (150)2/3 ≈ −2.03). f (5) = 3050. (Note: f (0) = 0 and f ( 15 (150)1/3 ) = − 125

x2 + 1 2 6] , 3s [5, 31. y =y = 2s 3 − + 3, x −4 SOLUTION

Let f (x) =

[−4, 4]

x2 + 1 . Then x −4 f (x) =

(x − 4) · 2x − (x 2 + 1) · 1 x 2 − 8x − 1 = =0 2 (x − 4) (x − 4)2

√ implies x = 4 ± 17 are critical points of f . x = 4 is not a critical point because x = 4 is not in the domain of f . On the interval [5, 6], the minimum of f is f (6) = 37 2 = 18.5, whereas the maximum of f is f (5) = 26. (Note: The critical √ points x = 4 ± 17 are not in the interval [5, 6].) 4x 33. y = x − 1 − x , [0, 3] y = x2 + 1 , [1, 4] x + 3x 4x SOLUTION Let f (x) = x − . Then x +1 f (x) = 1 −

4 (x − 1)(x + 3) = =0 (x + 1)2 (x + 1)2

implies that x = 1 and x = −3 are critical points of f . x = −1 is not a critical point because x = −1 is not in the domain of f . The minimum of f on the interval [0, 3] is f (1) = −1, whereas the maximum is f (0) = f (3) = 0. (Note: The critical point x = −3 is not in the interval [0, 3].) 35. y = (2 + x) 2 2 + (2 − x)2 , [0, 2] y = 2 x + 1 − x, [0, 2] SOLUTION Let f (x) = (2 + x) 2 + (2 − x)2 . Then

2(x − 1)2 2 + (2 − x)2 − (2 + x)(2 + (2 − x)2 )−1/2 (2 − x) = =0 2 + (2 − x)2 √ implies that x = 1 is the √ critical point of f . On the interval √ [0, 2], the minimum is f (0) = 2 6 ≈ 4.898979 and the maximum is f (2) = 4 2 ≈ 5.656854. (Note: f (1) = 3 3 ≈ 5.196152.) √ + x 2 − 2 x, [0, 4] 37. y = x y = 1 + x 2 − 2x, [3, 6] √ SOLUTION Let f (x) = x + x 2 − 2 x. Then f (x) =

√

1 1 + 2x − 2 f (x) = (x + x 2 )−1/2 (1 + 2x) − x −1/2 = √ √ 2

√

1+x

2 x 1+x

=0 √

implies that x = 0 and x = 23 are the critical points of f . Neither x = −1 nor x = − 23 is a critical point because

√ neither is in the domain of f . On the interval [0, 4], the minimum of f is f 23 ≈ −.589980 and the maximum is f (4) ≈ .472136. (Note: f (0) = 0.) 39. y = sin x cos x, [0, π2 ] y = (t − t 2 )1/3 , [−1, 2] π 1 π SOLUTION Let f (x) = sin x cos x = 2 sin 2x. On the interval 0, 2 , f (x) = cos 2x = 0 when x = 4 . The π π 1 minimum of f on this interval is f (0) = f ( 2 ) = 0, whereas the maximum is f ( 4 ) = 2 . √ π] 41. y =y =2xθ − −sin secx,θ , [0, [0,2π 3] √ √ π π SOLUTION Let f (θ ) = 2θ − sec θ . On the interval [0, 3 ], f (θ ) = 2 − sec θ tan θ = 0 at θ = 4 . The minimum √ value of f on this interval is f (0) = −1, whereas the maximum value over this interval is f ( π4 ) = 2( π4 − 1) ≈ √ −.303493. (Note: f ( π3 ) = 2 π3 − 2 ≈ −.519039.) 43. y = θ − 2 sin θ , [0, 2π ] y = cos θ + sin θ , [0, 2π ] SOLUTION Let g(θ ) = θ − 2 sin θ . On the interval [0, 2π ], g (θ ) = 1 − 2 cos θ = 0 at √ minimum of g on this interval is g( π3 ) = π3 − 3 ≈ −.685 and the maximum is g( 53 π ) = g(0) = 0 and g(2π ) = 2π ≈ 6.283.) 45. y = tan x − 2x, [0, 1] y = sin3 θ − cos2 θ ,

[0, 2π ]

θ = π3 and θ = 53 π . The √ 5 π + 3 ≈ 6.968. (Note: 3

S E C T I O N 4.2

Extreme Values

169

SOLUTION Let f (x) = tan x − 2x. Then on the interval [0, 1], f (x) = sec2 x − 2 = 0 at x = π 4 . The minimum of f is f ( π4 ) = 1 − π2 ≈ −.570796 and the maximum is f (0) = 0. (Note: f (1) = tan 1 − 2 ≈ −.442592.)

Find the critical points of f (x) = 2 cos 3x + 3 cos 2x. Check your answer against a graph of f (x). Let f (θ ) = 2 sin 2θ + sin 4θ . SOLUTION (x) θisisdifferentiable are2θlooking for points where f (x) = 0 only. Setting f (x) = a critical pointforif all cosx, 4θso = we − cos . (a) Show fthat −6 sin 3x − 6 sin 2x, we get sin 3x = − sin 2x. Looking at a unit circle, we find the relationship between angles y and x (b) Show, using a unit circle, that cos θ1 = − cos θ2 if and only if θ1 = π ± θ2 + 2π k for an integer k. such that sin y = − sin x. This technique is also used in Exercise 46. (c) Show that cos 4θ = − cos 2θ if and only if θ = π /2 + π k or θ = π /6 + (π /3)k. (d) Find the six critical points of f (θ ) on [0, 2π ] and find the extreme values of f (θ ) on this interval. (e) Check your results against a graph of f (θ ). 47.

From the diagram, we see that sin y = − sin x if y is either (i.) the point antipodal to x (y = π + x + 2π k) or (ii.) the point obtained by reflecting x through the horizontal axis (y = −x + 2π k). Since sin 3x = − sin 2x, we get either 3x = π + 2x + 2π k or 3x = −2x + 2π k. Solving each of these equations for x yields x = π + 2π k and x = 25π k, respectively. The values of x between 0 and 2π are 0, 25π , 45π , π , 65π , 85π , and 2π . The graph is shown below. As predicted, it has horizontal tangent lines at 25π k and at x = π2 . Each of these points is a local extremum. y 4 2 x 1

−2

2

3

4

5

6

−4

In Exercises 48–51, find the critical points and the extreme values on [0, 3]. In Exercises 50 and 51, refer to Figure 18. y

y 1

30 20 10 x

−6

2

−π

π 2

2

π

3π 2

x

y = | cos x |

y = | x 2 + 4x − 12 |

FIGURE 18

49. y = |3x − 9| y = |x − 2| SOLUTION Let f (x) = |3x − 9| = 3|x − 3|. For x < 3, we have f (x) = −3. For x > 3, we have f (x) = 3. Now as f (x) − f (3) 3(3 − x) − 0 f (x) − f (3) 3(x − 3) − 0 x → 3−, we have = → −3; whereas as x → 3+, we have = → 3. x −3 x −3 x −3 x −3 f (x) − f (3) does not exist and the lone critical point of f is x = 3. Alternately, we examine Therefore, f (3) = lim x −3 x→3 the graph of f (x) = |3x − 9| shown below. To find the extrema of f (x) on [0, 3], we test the values of f (x) at the critical point and the endpoints. f (0) = 9 and f (3) = 0, so f (x) takes its minimum value of 0 at x = 3, and its maximum value of 9 at x = 0. y 12 8 4 x 2

4

6

51. y = | cos x| y = |x 2 + 4x − 12| SOLUTION Let f (x) = | cos x|. There are two types of critical points: points of the form π n where the derivative is zero and points of the form n π + π /2 where the derivative does not exist.

170

CHAPTER 4


To find points where f (x) does not exist, we look at points where the inside of the absolute value function is equal to zero, and take derivatives in either direction, to find out whether they are the same. cos x = 0 at the points x = − π2 , x = π2 , and x = 32π , etc. Only one of these, x = π2 is in the interval [0, 3], so we check the one critical point and the endpoints. f (0) = 1, f (3) = | cos 3| ≈ .98992, and f ( π2 ) = 0, so f (x) takes its maximum value of 1 at x = 0 and its minimum of 0 at x = π2 . In Exercises 53–56, verify Rolle’s Theorem for the given interval. Let f (x) = 3x − x 3 . Check that f (−2) = f (1). What may we conclude from Rolle’s Theorem? Verify this 1 −1 53. conclusion. f (x) = x + x , [ , 2] 2

SOLUTION

Because f is continuous on [ 12 , 2], differentiable on ( 12 , 2) and 1 5 1 1 1 f = + 1 = = 2 + = f (2), 2 2 2 2 2

we may conclude from Rolle’s Theorem that there exists a c ∈ ( 12 , 2) at which f (c) = 0. Here, f (x) = 1 − x −2 = x 2 −1 , so we may take c = 1. 2 x

x2 π , 5] 3π 55. f (x)f (x) = = sin x, , [[3, 4 4 ] 8x − 15 SOLUTION Because f is continuous on [3, 5], differentiable on (3, 5) and f (3) = f (5) = 1, we may conclude from Rolle’s Theorem that there exists a c ∈ (3, 5) at which f (c) = 0. Here,

f (x) =

(8x − 15)(2x) − 8x 2 2x(4x − 15) = , (8x − 15)2 (8x − 15)2

so we may take c = 15 4 . 57. Use Rolle’s Theorem to prove that f (x) = x 5 + 2x 3 + 4x − 12 has at most one real root. f (x) = sin2 x − cos2 x, [ π4 , 34π ] SOLUTION We use proof by contradiction. Suppose f (x) = x 5 + 2x 3 + 4x − 12 has two real roots, x = a and x = b. Then f (a) = f (b) = 0 and Rolle’s Theorem guarantees that there exists a c ∈ (a, b) at which f (c) = 0. However, f (x) = 5x 4 + 6x 2 + 4 ≥ 4 for all x, so there is no c ∈ (a, b) at which f (c) = 0. Based on this contradiction, we conclude that f (x) = x 5 + 2x 3 + 4x − 12 cannot have more than one real root. bloodstream after t hours is 59. The concentration C(t) (in mg/cm3 ) of a drug in3 a patient’s x x2 Use Rolle’s Theorem to prove that f (x) = + + x + 1 has at most one real root. 6 = 2 0.016t C(t) t 2 + 4t + 4 Find the maximum concentration and the time at which it occurs. SOLUTION

C (t) =

.016(t 2 + 4t + 4) − (.016t (2t + 4)) −t 2 + 4 2−t = .016 2 = .016 . 2 2 (t + 4t + 4) (t + 4t + 4)2 (t + 2)3

C (t) exists for all t ≥ 0, so we are looking for points where C (t) = 0. C (t) = 0 when t = 2, so we check C(2) = .002 mg3 . Since C(0) = 0 and C(t) → 0 as t → ∞, the point (2, .002) is a local maximum of C(t). cm

Hence, the maximum concentration occurs 2 hours after being administered, and that concentration is .002

mg . cm3

61. Bees build honeycomb structures out of cells with a hexagonal base and three rhombus-shaped faces on top as in Maximum of a Resonance Curve size of the of aiscircuit or other oscillatory system to an input Figure 20. Using geometry, we can show thatThe the surface arearesponse of this cell signal of frequency ω (“omega”) is described in suitable units by the function 3 √ A(θ ) = 6hs + s 2 ( 3 csc 1 θ − cot θ ) φ (ω ) = 2 . 2 2 ω02 − ω 2 )2fact + 4D where h, s, and θ are as indicated in the figure. It is a (remarkable thatωbees “know” which angle θ minimizes the surface area (and therefore requires the least amount of wax). Both ω0 (the natural frequency of the system) and D (the damping factor) are positive constants. The graph of φ is (a) Show this anglecurve, is approximately 54.7◦frequency by findingω the>critical point of A(its θ )maximum for 0 < θ value, < π /2if(assume h and s called that a resonance and the positive 0 where φ takes such a positive r √ are constant). frequency exists, is the resonant frequency. Show that a resonant √

frequency exists if and only if 0 < D < ω0 / 2 Confirm, by graphing f (θ ) = 3 csc θ − cot θ , that the critical point indeed minimizes the surface area. (b) (Figure 19). Then show that if this condition is satisfied, ωr = ω02 − 2D 2 .

S E C T I O N 4.2

Extreme Values

171

q

h

s

FIGURE 20 A cell in a honeycomb constructed by bees. SOLUTION

√ (a) Because h and s are constant relative to θ , we have A (θ ) = 32 s 2 (− 3 csc θ cot θ + csc2 θ ) = 0. From this, we get √ 3 csc θ cot θ = csc2 θ , or cos θ = √1 , whence θ = cos−1 √1 = .955317 radians = 54.736◦ . 3 3 √ (b) The plot of 3 csc θ − cot θ , where θ is given in degrees, is given below. We can see that the minimum occurs just below 55◦ . h 1.5 1.48 1.46 1.44 1.42 1.4 40 45 50 55 60 65

Degrees

63. FindMigrating the maximum of yto=swim x a −at x baon [0, 1] where < a < b.the In particular, find theof maximum of According y = x 5 − xto10one on fish tend velocity v that 0minimizes total expenditure energy E. [0, 1]. 3 v model, E is proportional to f (v) = , where vr is the velocity of the river water. SOLUTION v − vr (a) Find the critical points of f (v). • Let f (x) = x a − x b . Then f (x) = ax a−1 − bx b−1 . Since a < b, f (x) = x a−1 (a − bx b−a ) = 0 implies (say, and plot (v).interval Confirm[0,that a minimum the critical (b) Choose of vxr = a < value r = 10) , which is inf the 1] asf (v) a x b on the interval [0, 1], which gives x = (b) f (x) > 0 and thus the maximum value of f on [0, 1] is a 1/(b−a) a a/(b−a) a b/(b−a) − . f = b b b • Let f (x) = x 5 − x 10 . Then by part (a), the maximum value of f on [0, 1] is

1 1 1 1 2 1 1 1/5 f = = − = . − 2 2 2 2 4 4 In Exercises 64–66, plot the function using a graphing utility and find its critical points and extreme values on [−5, 5]. 1 1 y= + 1 y = 1 + |x − 1| 1 + |x − 4| 1 + |x − 1| SOLUTION Let

65.

f (x) =

1 1 + . 1 + |x − 1| 1 + |x − 4|

The plot follows: 1.2 1 0.8 0.6 0.4 0.2 −5 −4 −3 −2 −1

1

2

3

4

5

172

CHAPTER 4


We can see on the plot that the critical points of f (x) lie at the cusps at x = 1 and x = 4 and at the location of the 5 5 4 7 horizontal tangent line at x = 52 . With f (−5) = 17 70 , f (1) = f (4) = 4 , f ( 2 ) = 5 and f (5) = 10 , it follows that the maximum value of f (x) on [−5, 5] is f (1) = f (4) = 54 and the minimum value is f (−5) = 17 70 . 67. (a) Use implicit differentiation to find the critical points on the curve 27x 2 = (x 2 + y 2 )3 . x y = Plot|xthe 2 −curve (b) 1| + and |x 2 the − 4|horizontal tangent lines on the same set of axes. SOLUTION

(a) Differentiating both sides of the equation 27x 2 = (x 2 + y 2 )3 with respect to x yields dy 54x = 3(x 2 + y 2 )2 2x + 2y . dx Solving for d y/d x we obtain x(9 − (x 2 + y 2 )2 ) dy 27x − 3x(x 2 + y 2 )2 = . = 2 2 2 dx 3y(x + y ) y(x 2 + y 2 )2 Thus, the derivative is zero when x 2 + y 2 = 3. Substituting into the equation for the curve, this yields x 2 = 1, or x = ±1. There are therefore four points at which the derivative is zero: √ √ √ √ (−1, − 2), (−1, 2), (1, − 2), (1, 2). There are also critical points where the derivative does not exist. This occurs when y = 0 and gives the following points with vertical tangents: √ 4 (0, 0), (± 27, 0). (b) The curve 27x 2 = (x 2 + y 2 )3 and its horizontal tangents are plotted below. y 1 −2

x

−1

1

2

−1

69. Sketch the graph of a continuous function on (0, 4) having a local minimum but no absolute minimum. Sketch the graph of a continuous function on (0, 4) with a minimum value but no maximum value. SOLUTION Here is the graph of a function f on (0, 4) with a local minimum value [between x = 2 and x = 4] but no absolute minimum [since f (x) → −∞ as x → 0+]. y 10 x 1

2

3

−10

71. Sketch the graph of a function f (x) on [0, 4] with a discontinuity such that f (x) has an absolute minimum but no Sketch the graph of a function on [0, 4] having absolute maximum. (a) Two local maxima and one local minimum. SOLUTION Here is the graph of a function f on [0, 4] that (a) has a discontinuity [at x = 4] and (b) has an absolute (b) An absolute minimum that occurs at an endpoint, and an absolute maximum that occurs at a critical point. minimum [at x = 0] but no absolute maximum [since f (x) → ∞ as x → 4−]. y 4 3 2 1 x 0

1

2

3

4

The Mean Value Theorem and Monotonicity

S E C T I O N 4.3

173

Further Insights and Challenges on only positive values. More generally, find 73. Show, by considering its minimum, that f (x) = x 2 − 2x + 3 takes 2 + b2 . Show that extreme values f (x) = a sinfunction x + b cosfx(x) are=±x 2a+ the conditions on rthe and s under whichofthe quadratic r x + s takes on only positive values. Give examples of r and s for which f takes on both positive and negative values. SOLUTION

• Observe that f (x) = x 2 − 2x + 3 = (x − 1)2 + 2 > 0 for all x. Let f (x) = x 2 + r x + s. Completing the square, we note that f (x) = (x + 12 r )2 + s − 14 r 2 > 0 for all x provided that s > 14 r 2 . • Let f (x) = x 2 − 4x + 3 = (x − 1)(x − 3). Then f takes on both positive and negative values. Here, r = −4 and

s = 3.

75. Generalize Exercise 74: Show that if the horizontal 2line y = c intersects the graph of f (x) = x 2 + r x + s at two Show that if the quadratic polynomial f (x) = x + r x + s takes on both positive and x 1negative + x 2 values, then its points (x1 , f (xvalue (x 2 , at f (x then f (x) takes its (Figure 21). minimum occurs the midpoint between theminimum two roots.value at the midpoint M = 1 )) and 2 )), 2 y f(x) c

y=c

x x1

M

x2

FIGURE 21

Suppose that a horizontal line y = c intersects the graph of a quadratic function f (x) = x 2 + r x + s in two points (x 1 , f (x 1 )) and (x 2 , f (x 2 )). Then of course f (x1 ) = f (x 2 ) = c. Let g(x) = f (x) − c. Then g(x1 ) = g(x 2 ) = 0. By Exercise 74, g takes on its minimum value at x = 12 (x 1 + x2 ). Hence so does f (x) = g(x) + c. SOLUTION

77. Find the minimum and maximum values of f (x) = x p (1 − x)q on [0, 1], where p and q are positive numbers. The graphs in Figure 22 show that a cubic polynomial may have a local min and max, or it may have neither. p (1 − x)q , 0 ≤ x ≤ 1, where p 1and 1 axpositive 2 + bx numbers. SOLUTION Let f (x) q are Then f has neither a local min + c that ensure Find conditions on = thexcoefficients a and b of f (x) = 3 x 3 + 2 nor max. Hint: Apply Exercise 73 to f (x). f (x) = x p q(1 − x)q−1 (−1) + (1 − x)q px p−1 p = x p−1 (1 − x)q−1 ( p(1 − x) − qx) = 0 at x = 0, 1, p+q The minimum value of f on [0, 1] is f (0) = f (1) = 0, whereas its maximum value is p p pqq f . = p+q ( p + q) p+q Prove that if f is continuous and f (a) and f (b) are local minima where a < b, then there exists a value c between a and b such that f (c) is a local maximum. (Hint: Apply Theorem 1 to the interval [a, b].) Show that 4.3continuity The Mean Value Theorem and Monotonicity is a necessary hypothesis by sketching the graph of a function (necessarily discontinuous) with two local minima but no local maximum.

Preliminary Questions 1. Which value of m makes the following statement correct? If f (2) = 3 and f (4) = 9, where f (x) is differentiable, then the graph of f has a tangent line of slope m. SOLUTION

The Mean Value Theorem guarantees that the function must have a tangent line with slope equal to 9−3 f (4) − f (2) = = 3. 4−2 4−2

Hence, m = 3 makes the statement correct. 2. (a) (b) (c) (d)

Which of the following conclusions does not follow from the MVT (assume that f is differentiable)? If f has a secant line of slope 0, then f (c) = 0 for some value of c. If f (5) < f (9), then f (c) > 0 for some c ∈ (5, 9). If f (c) = 0 for some value of c, then there is a secant line whose slope is 0. If f (x) > 0 for all x, then every secant line has positive slope.

Conclusion (c) does not follow from the Mean Value Theorem. As a counterexample, consider the function f (x) = x 3 . Note that f (0) = 0, but no secant line has zero slope.

SOLUTION

174

CHAPTER 4


3. Can a function that takes on only negative values have a positive derivative? Sketch an example or explain why no such functions exist. SOLUTION

Yes. The figure below displays a function that takes on only negative values but has a positive derivative. y x

4. (a) Use the graph of f (x) in Figure 10 to determine whether f (c) is a local minimum or maximum. (b) Can you conclude from Figure 10 that f (x) is a decreasing function? y

x

c

FIGURE 10 Graph of derivative f (x). SOLUTION

(a) To the left of x = c, the derivative is positive, so f is increasing; to the right of x = c, the derivative is negative, so f is decreasing. Consequently, f (c) must be a local maximum. (b) No. The derivative is a decreasing function, but as noted in part (a), f (x) is increasing for x < c and decreasing for x > c.

Exercises In Exercises 1–6, find a point c satisfying the conclusion of the MVT for the given function and interval. 1. y = x −1 , SOLUTION

[1, 4] Let f (x) = x −1 , a = 1, b = 4. Then f (x) = −x −2 , and by the MVT, there exists a c ∈ (1, 4) such that 1

1

− f (b) − f (a) 1 1 = 4 1 =− . − 2 = f (c) = b−a 4−1 4 c Thus c2 = 4 and c = ±2. Choose c = 2 ∈ (1, 4). 3. y = (x − √1)(x − 3), [1, 3] y = x, [4, 9] SOLUTION Let f (x) = (x − 1) (x − 3), a = 1, b = 3. Then f (x) = 2x − 4, and by the MVT, there exists a c ∈ (1, 3) such that f (b) − f (a) 0−0 2c − 4 = f (c) = = = 0. b−a 3−1 Thus 2c − 4 = 0 and c = 2 ∈ (1, 3). x 5. y =y = cos, x − [3,sin 6]x, [0, 2π ] x +1 SOLUTION

Let f (x) = x/ (x + 1), a = 3, b = 6. Then f (x) =

such that

1 , and by the MVT, there exists a c ∈ (3, 6) (x+1)2

6 3 (c) = f (b) − f (a) = 7 − 4 = 1 . = f b−a 6−3 28 (c + 1)2 √ √ Thus (c + 1)2 = 28 and c = −1 ± 2 7. Choose c = 2 7 − 1 ≈ 4.29 ∈ (3, 6).

1

7. Let = x 5 + x 2 . Check that the secant line between x = 0 and x = 1 has slope 2. By the MVT, 3 , f (x) [−3, y = x f (c) = 2 for some c ∈−1] (0, 1). Estimate c graphically as follows. Plot f (x) and the secant line on the same axes. Then plot the lines y = 2x + b for different values of b until you find a value of b for which it is tangent to y = f (x). Zoom in on the point of tangency to find its x-coordinate.



175

Let f (x) = x 5 + x 2 . The slope of the secant line between x = 0 and x = 1 is 2−0 f (1) − f (0) = = 2. 1−0 1

A plot of f (x), the secant line between x = 0 and x = 1, and the line y = 2x − 0.764 is shown below at the left. The line y = 2x − 0.764 appears to be tangent to the graph of y = f (x). Zooming in on the point of tangency (see below at the right), it appears that the x-coordinate of the point of tangency is approximately 0.62. y

0.6

y = x5 + x 2

y

0.5

4

0.4

y = 2x − .764

2

x

0.3 0.52

1

x 0.56

0.6

0.64

9. Determine the intervals on which f (x) is increasing or decreasing, assuming that Figure 11 is the graph of the Determine derivative f (x). the intervals on which f (x) is positive and negative, assuming that Figure 11 is the graph of f (x). y

x 1

2

3

4

5

6

FIGURE 11

f (x) is increasing on every interval (a, b) over which f (x) > 0, and is decreasing on every interval over which f (x) < 0. If the graph of f (x) is given in Figure 11, then f (x) is increasing on the intervals (0, 2) and (4, 6), SOLUTION

and is decreasing on the interval (2, 4).

In Exercises 11–14, sketch the graph of a function f (x) whose derivative f (x) has the given description. Plot the derivative f (x) of f (x) = 3x 5 − 5x 3 and describe the sign changes of f (x). Use this to determine local f (x). > 0extreme for x >values 3 andoff (x) 0 for x > 3 and f (x) < 0 for x < 3. y 10 8 6 4 2 x 0

1

2

3

4

5

13. f (x) is negative on (1, 3) and positive everywhere else. f (x) > 0 for x < 1 and f (x) < 0 for x > 1. SOLUTION Here is the graph of a function f for which f (x) is negative on (1, 3) and positive elsewhere. y 8 6 4 2 x −2

1

2

3

4

In Exercises the First Derivative f (x)15–18, makes use the sign transitions +, −,Test +, to −.determine whether the function attains a local minimum or local maximum (or neither) at the given critical point. 15. y = 7 + 4x − x 2 ,

c=2

176

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION Let f (x) = 7 + 4x − x 2 . Then f (c) = 4 − 2c = 0 implies c = 2 is a critical point of f . Since f makes the sign transition +, − as x increases through c = 2, we conclude that f (2) = 11 is a local maximum of f .

x2 17. y =y = x 3 ,− 27x c =+ 0 2, c = −3 x +1 SOLUTION

x 2 . Then Let f (x) = x+1

f (c) =

2c(c + 1) − c2 c(c + 2) = = 0 at c = 0. (c + 1)2 (c + 1)2

Since f makes the sign transition −, + as x increases through c = 0, we conclude that f (0) = 0 is a local minimum of f. 19. Assuming that Figure 11 πis the graph of the derivative f (x), state whether f (2) and f (4) are local minima or maxima.y = sin x cos x, c = 4 SOLUTION

• f (x) makes a transition from positive to negative at x = 2, so f (2) is a local maximum. • f (x) makes a transition from negative to positive at x = 4, so f (4) is a local minimum.

In Exercises 21–40, find the and the intervals which thef function increasing or decreasing, and deterapply Figure 12 shows the critical graph ofpoints the derivative f (x) ofon a function (x). Findisthe critical points of f (x) and the First Testare to local each critical mineDerivative whether they minima,point. maxima, or neither. SOLUTION

Here is a table legend for Exercises 21–40. SYMBOL

MEANING

−

The entity is negative on the given interval.

0

The entity is zero at the specified point.

+

The entity is positive on the given interval.

U

The entity is undefined at the specified point.

f is increasing on the given interval.

f is decreasing on the given interval.

M

f has a local maximum at the specified point.

m

f has a local minimum at the specified point.

¬

There is no local extremum here.

21. y = −x 2 + 7x − 17 SOLUTION

Let f (x) = −x 2 + 7x − 17. Then f (x) = 7 − 2x = 0 yields the critical point c = 72 . x

−∞, 72

7/2

7,∞ 2

f

+

0

−

f

M

2 23. y = x 3 − 6x y = 5x 2 + 6x − 4 SOLUTION Let f (x) = x 3 − 6x 2 . Then f (x) = 3x 2 − 12x = 3x(x − 4) = 0 yields critical points c = 0, 4.

25. y = 3x 4 + 8x 3 −3 6x 2 − 24x y = x(x + 1)

x

(−∞, 0)

0

(0, 4)

4

(4, ∞)

f

+

0

−

0

+

f

M

m



177

Let f (x) = 3x 4 + 8x 3 − 6x 2 − 24x. Then f (x) = 12x 3 + 24x 2 − 12x − 24 = 12x 2 (x + 2) − 12(x + 2) = 12(x + 2)(x 2 − 1) = 12 (x − 1) (x + 1) (x + 2) = 0

yields critical points c = −2, −1, 1. x

(−∞, −2)

−2

(−2, −1)

−1

(−1, 1)

1

(1, ∞)

f

−

0

+

0

−

0

+

f

m

M

m

27. y = 13 x 3 2+ 32 x 2 + 2x 2+ 4 y = x + (10 − x) 1 3 SOLUTION Let f (x) = 3 x 3 + 2 x 2 + 2x + 4. Then f (x) = x 2 + 3x + 2 = (x + 1) (x + 2) = 0 yields critical points c = −2, −1. x

(−∞, −2)

−2

(−2, −1)

−1

(−1, ∞)

f

+

0

−

0

+

f

M

m

29. y = x 4 +5x 3 3 y = x +x +1 3 SOLUTION Let f (x) = x 4 + x 3 . Then f (x) = 4x 3 + 3x 2 = x 2 (4x + 3) yields critical points c = 0, − 4 . x

−∞, − 34

− 34

− 34 , 0

0

(0, ∞)

f

−

0

+

0

+

f

m

¬

1 31. y =y =2 x 2 − x 4 x +1 SOLUTION

−1

−2 Let f (x) = x 2 + 1 . Then f (x) = −2x x 2 + 1 = 0 yields critical point c = 0. x

(−∞, 0)

0

(0, ∞)

f

+

0

−

f

M

33. y = x +2x x −1 + 1 (x > 0) y= 2 −1 −2 = 0 yields the critical point c = 1. (Note: c = −1 SOLUTION xLet + 1f (x) = x + x for x > 0. Then f (x) = 1 − x is not in the interval under consideration.) x

(0, 1)

1

(1, ∞)

f

−

0

+

f

m

(x > 0) 35. y = x 5/2 4− x 2 3/2 y = x − 4x (x > 0) 5 5 16 SOLUTION Let f (x) = x 5/2 − x 2 . Then f (x) = 2 x 3/2 − 2x = x( 2 x 1/2 − 2) = 0, so the critical point is c = 25 . (Note: c = 0 is not in the interval under consideration.) x

(0, 16 25 )

16 25

( 16 25 , ∞)

f

−

0

+

f

m

178

CHAPTER 4


37. y = sin θ cos θ , [0, 2π ] y = x −2 − 4x −1 (x > 0) SOLUTION Let f (θ ) = sin θ cos θ . Then f (θ ) = − sin2 θ + cos2 θ = 0 implies that tan2 θ = 1. On the interval [0, 2π ], this yields c = π4 , 34π , 54π , 74π . x

(0, π4 )

π 4

( π4 , 34π )

3π 4

( 34π , 54π )

5π 4

( 54π , 74π )

7π 4

( 74π , 2π )

f

+

0

−

0

+

0

−

0

+

f

M

m

M

m

39. y = θ + cos θ , [0, 2π ] y = cos θ + sin θ , [0, 2π ] π SOLUTION Let f (θ ) = θ + cos θ . Then f (θ ) = 1 − sin θ = 0, which yields c = 2 on the interval [0, 2π ]. x

0, π2

π 2

π

2 , 2π

f

+

0

+

f

¬

b b 41. Show =θx, 2 + y =that θ −f (x) 2 cos [0,bx2π+] c is decreasing on (−∞, − 2 ) and increasing on (− 2 , ∞).

Let f (x) = x 2 + bx + c. Then f (x) = 2x + b = 0 yields the critical point c = − b2 .

• For x < − b , we have f (x) < 0, so f is decreasing on −∞, − b . 2 2

• For x > − b , we have f (x) > 0, so f is increasing on − b , ∞ . 2 2

SOLUTION

43. Find conditions on a and3 b that2 ensure that f (x) = x 3 + ax + b is increasing on (−∞, ∞). Show that f (x) = x − 2x + 2x is an increasing function. Hint: Find the minimum value of f (x). SOLUTION Let f (x) = x 3 + ax + b. • If a > 0, then f (x) = 3x 2 + a > 0 and f is increasing for all x. • If a = 0, then

f (x 2 ) − f (x 1 ) = (3x 23 + b) − (3x 13 + b) = 3(x 2 − x 1 )(x 22 + x 2 x 1 + x 12 ) > 0 whenever x2 > x1 . Thus, f is increasing for all x. • If a < 0, then f (x) = 3x 2 + a < 0 and f is decreasing for |x| <

− a3 .

In summary, f (x) = x 3 + ax + b is increasing on (−∞, ∞) whenever a ≥ 0. 45. Sam made two statements that Deborah found dubious. x(x 2 − 1) (a) “Although the average for my is trip 70 mph,ofatano point inf time did my 70describe mph.” the thewas derivative function (x). Plot h(x)speedometer and use the read plot to Suppose that h(x)velocity = x 2 me + 1going 70 mph, my speedometer never read 65 mph.” (b) “Although a policeman clocked local extrema and the increasing/decreasing behavior of f (x). Sketch a plausible graph for f (x) itself. In each case, which theorem did Deborah apply to prove Sam’s statement false: the Intermediate Value Theorem or the Mean Value Theorem? Explain. SOLUTION

(a) Deborah is applying the Mean Value Theorem here. Let s(t) be Sam’s distance, in miles, from his starting point, let a be the start time for Sam’s trip, and let b be the end time of the same trip. Sam is claiming that at no point was s (t) =

s(b) − s(a) . b−a

This violates the MVT. (b) Deborah is applying the Intermediate Value Theorem here. Let v(t) be Sam’s velocity in miles per hour. Sam started out at rest, and reached a velocity of 70 mph. By the IVT, he should have reached a velocity of 65 mph at some point. 47. Show that f (x) = 1 − |x| satisfies the conclusion of the MVT on [a, b] if both a and b are positive or negative, but 2 2 2 2 where not if a Determine < 0 and b > 0. f (x) = (1,000 − x) + x is increasing. Use this to decide which is larger: 1,000 or 998 + 22 . SOLUTION Let f (x) = 1 − |x|. • If a and b (where a < b) are both positive (or both negative), then f is continuous on [a, b] and differentiable on

(a, b). Accordingly, the hypotheses of the MVT are met and the theorem does apply. Indeed, in these cases, any point c ∈ (a, b) satisfies the conclusion of the MVT (since f is constant on [a, b] in these instances).

S E C T I O N 4.3


179

1 f (b) − f (a) 0 − (−1) = . Yet there is no point c ∈ (−2, 1) such that = b−a 1 − (−2) 3 f (c) = 13 . Indeed, f (x) = 1 for x < 0, f (x) = −1 for x > 0, and f (0) is undefined. The MVT does not apply in this case, since f is not differentiable on the open interval (−2, 1).

• For a = −2 and b = 1, we have

a+b 49. Show that if f is aofquadratic then = intervalsatisfies conclusion the MVT on [a, b] Which values c satisfy polynomial, the conclusion ofthe themidpoint MVT on cthe [a, b] ifthe f (x) is a linearoffunction? 2 for any a and b. SOLUTION Let f (x) = px 2 + qx + r with p = 0 and consider the interval [a, b]. Then f (x) = 2 px + q, and by the MVT we have

pb2 + qb + r − pa 2 + qa + r f (b) − f (a) 2 pc + q = f (c) = = b−a b−a

=

(b − a) ( p (b + a) + q) = p (b + a) + q b−a

Thus 2 pc + q = p(a + b) + q, and c =

a+b . 2

51. Suppose that f (2) = −2 and f (x) ≥ 5. Show that f (4) ≥ 8. Suppose that f (0) = 4 and f (x) ≤ 2 for x > 0. Apply the MVT to the interval [0, 3] to prove that f (3) ≤ 10. SOLUTION The MVT, applied to the [2,all 4],xguarantees there exists a c ∈ (2, 4) such that Prove more generally that f (x) ≤ 4interval + 2x for > 0. f (c) =

f (4) − f (2) 4−2

or

f (4) − f (2) = 2 f (c).

Because f (x) ≥ 5, it follows that f (4) − f (2) ≥ 10, or f (4) ≥ f (2) + 10 = 8.

Further Insights and Challenges 53. Prove that if f (0) = g(0) and f (x) ≤ g (x) for x ≥ 0, then f (x) ≤ g(x) for all x ≥ 0. Hint: Show that f (x) − g(x) Show that the cubic function f (x) = x 3 + ax 2 + bx + c is increasing on (−∞, ∞) if b > a 2 /3. is nonincreasing. Let h(x) = f (x) − g(x). By the sum rule, h (x) = f (x) − g (x). Since f (x) ≤ g (x) for all x ≥ 0, h (x) ≤ 0 for all x ≥ 0. This implies that h is nonincreasing. Since h(0) = f (0) − g(0) = 0, h(x) ≤ 0 for all x ≥ 0 (as SOLUTION

h is nonincreasing, it cannot climb above zero). Hence f (x) − g(x) ≤ 0 for all x ≥ 0, and so f (x) ≤ g(x) for x ≥ 0.

55. Use Exercises 53 and 54 to establish the following assertions for all x ≥ 0 (each assertion follows from the previous Use Exercise 53 to prove that sin x ≤ x for x ≥ 0. one): (a) cos x ≥ 1 − 12 x 2 (b) sin x ≥ x − 16 x 3 1 x4 (c) cos x ≤ 1 − 12 x 2 + 24 (d) Can you guess the next inequality in the series? SOLUTION

(a) We prove this using Exercise 53: Let g(x) = cos x and f (x) = 1 − 12 x 2 . Then f (0) = g(0) = 1 and g (x) = − sin x ≥ −x = f (x) for x ≥ 0 by Exercise 54. Now apply Exercise 53 to conclude that cos x ≥ 1 − 12 x 2 for x ≥ 0.

(b) Let g(x) = sin x and f (x) = x − 16 x 3 . Then f (0) = g(0) = 0 and g (x) = cos x ≥ 1 − 12 x 2 = f (x) for x ≥ 0 by part (a). Now apply Exercise 53 to conclude that sin x ≥ x − 16 x 3 for x ≥ 0.

1 x 4 and f (x) = cos x. Then f (0) = g(0) = 1 and g (x) = −x + 1 x 3 ≥ − sin x = f (x) (c) Let g(x) = 1 − 12 x 2 + 24 6 1 x 4 for x ≥ 0. for x ≥ 0 by part (b). Now apply Exercise 53 to conclude that cos x ≤ 1 − 12 x 2 + 24

1 x 5 , valid for x ≥ 0. To construct (d) from (c), we note (d) The next inequality in the series is sin x ≤ x − 16 x 3 + 120 that the derivative of sin x is cos x, and look for a polynomial (which we currently must do by educated guess) whose 1 x 4 . We know the derivative of x is 1, and that a term whose derivative is − 1 x 2 should be of derivative is 1 − 12 x 2 + 24 2 1 x 4 should be of the form Dx 5 . the form C x 3 . ddx C x 3 = 3C x 2 = − 12 x 2 , so C = − 16 . A term whose derivative is 24 1 x 4 , so that 5D = 1 , or D = 1 . From this, ddx Dx 5 = 5Dx 4 = 24 24 120 exists Define = x 3and sin( f1x )(x) for = x = 0 and f (0) = 0.that f (x) = mx + b, where m = f (0) and b = f (0). 57. Assume that ff(x) 0 for all x. Prove (a) Show that f (x) is continuous at x = 0 and x = 0 is a critical point of f . Examine the graphs of f (x) and f (x). Can the First Derivative Test be applied? (b) (c) Show that f (0) is neither a local min nor max. SOLUTION

180

CHAPTER 4


(a) Let f (x) = x 3 sin( 1x ). Then f (x) = 3x 2 sin

1 1 1 1 + x 3 cos (−x −2 ) = x 3x sin − cos . x x x x

This formula is not defined at x = 0, but its limit is. Since −1 ≤ sin x ≤ 1 and −1 ≤ cos x ≤ 1 for all x, 1 1 3x sin 1 + cos 1 ≤ |x|(3|x| + 1) | f (x)| = |x| 3x sin − cos ≤ |x| x x x x so, by the Squeeze Theorem, lim | f (x)| = 0. But does f (0) = 0? We check using the limit definition of the derivative: x→0

1 f (x) − f (0) = lim x 2 sin = 0. x −0 x x→0 x→0

f (0) = lim

Thus f (x) is continuous at x = 0, and x = 0 is a critical point of f . (b) The figure below at the left shows f (x), and the figure below at the right shows f (x). Note how the two functions oscillate near x = 0, which implies that the First Derivative Test cannot be applied. y

−0.2

y

x 0.2

−0.2

x 0.2

(c) As x approaches 0 from either direction, f (x) alternates between positive and negative arbitrarily close to x = 0. This means that f (0) cannot be a local minimum (since f (x) gets lower than f (0) arbitrarily close to 0), nor can f (0) be a local maximum (since f (x) takes values higher than f (0) arbitrarily close to x = 0). Therefore f (0) is neither a local minimum nor a local maximum of f .

4.4 The Shape of a Graph Preliminary Questions 1. Choose the correct answer: If f is concave up, then f is: (a) increasing (b) decreasing The correct response is (a): increasing. If the function is concave up, then f is positive. Since f is the derivative of f , it follows that the derivative of f is positive and f must therefore be increasing. SOLUTION

2. If x 0 is a critical point and f is concave down, then f (x 0 ) is a local: (a) min (b) max (c) undetermined SOLUTION

By the Second Derivative Test, the correct response is (b): maximum.

In Questions 3–8, state whether true or false and explain. Assume that f (x) exists for all x. 3. If f (c) = 0 and f (c) < 0, then f (c) is a local minimum. SOLUTION

False. By the Second Derivative Test, the correct conclusion would be that f (c) is a local maximum.

4. A function that is concave down on (−∞, ∞) can have no minimum value. SOLUTION

True.

5. If f (c) = 0, then f must have a point of inflection at x = c. False. f has an inflection point at x = c provided concavity changes at x = c. Points where f (c) = 0 are simply candidates for inflection points. The function f (x) = x 4 provides a counterexample. Here, f (x) = 12x 2 . Thus, f (0) = 0 but f (x) does not change sign at x = 0. Therefore, f (x) = x 4 does not have an inflection point at x = 0. SOLUTION

6. If f has a point of inflection at x = c, then f (c) = 0. SOLUTION True. In general, if f has an inflection point at x = c, then either f (c) = 0 or f (c) does not exist. Because we are assuming that f (x) exists for all x, we must have f (c) = 0.

7. If f is concave up and f changes sign at x = c, then f changes sign from negative to positive at x = c. True. If f is concave up, then f is positive and f is increasing; therefore, f must go from negative to positive at x = c. SOLUTION

S E C T I O N 4.4

The Shape of a Graph

181

8. If f (c) is a local maximum, then f (c) must be negative. SOLUTION True. In general, it could be that either f (c) does not exist or f (c) is negative. Because we are assuming that f (x) exists for all x, we must have f (c) < 0.

9. Suppose that f (c) = 0 and f (x) changes sign from + to − at x = c. Which of the following statements are correct? (a) f (x) has a local maximum at x = a. (b) f (x) has a local minimum at x = a. (c) f (x) has a local maximum at x = a. (d) f (x) has a point of inflection at x = a. Statements (c) and (d) are correct. Because f (x) goes from positive to negative, it follows that f (x) goes from increasing to decreasing and so must have a local maximum at x = c. Also, because f (c) = 0 and f (x) changes sign at x = c, it follows that f has a point of inflection at x = c. SOLUTION

Exercises 1. Match the graphs in Figure 12 with the description: (a) f (x) < 0 for all x. (c) f (x) > 0 for all x.

(A)

(b) f (x) goes from + to −. (d) f (x) goes from − to +.

(B)

(C)

(D)

FIGURE 12 SOLUTION

(a) (b) (c) (d)

In C, we have f (x) < 0 for all x. In A, f (x) goes from + to −. In B, we have f (x) > 0 for all x. In D, f (x) goes from − to +.

3. Sketch the graph of an increasing function such that f (x) changes from + to − at x = 2 and from − to + at x = 4. Match each statement with a graph in Figure 13 that represents company profits as a function of time. Do the same for a decreasing function. (a) The outlook is great: The growth rate keeps increasing. SOLUTION The graph shown below at the left is an increasing function which changes from concave up to concave (b) We’re losing money, but not as quickly as before. down at x = 2 and from concave down to concave up at x = 4. The graph shown below at the right is a decreasing (c) We’re money, it’s getting worse asdown time at goes function which losing changes from and concave up to concave x =on. 2 and from concave down to concave up at x = 4. (d) We’re doing well, but our growth rate is leveling off. y y (e) Business had been cooling off, but now it’s picking up. 2 6 (f) Business had been picking up, but now it’s cooling off. 4 1 2 x 2

4

x 2

4

5. If Figure 14 is the graph of the derivative f (x), where do the points of inflection of f (x) occur, and on which If Figure 14 is the graph of a function f (x), where do the points of inflection of f (x) occur, and on which interval interval is f (x) concave down? is f (x) concave down? SOLUTION Points of inflection occur when f (x) changes sign. Consequently, points of inflection occur when f (x) changes from increasing to decreasing or from decreasing to increasing. In Figure 14, this occurs at x = b and at x = e; therefore, f (x) has an inflection point at x = b and another at x = e. The function f (x) will be concave down when f (x) < 0 or when f (x) is decreasing. Thus, f (x) is concave down for b < x < e. In Exercises 7–14,14determine theof intervals on which the function concave uppoints or down and find the of inflection. If Figure is the graph the second derivative f (x), is where do the of inflection of points f (x) occur, and on + 7x +is10f (x) concave down? 7. which y = x 2interval

SOLUTION Let f (x) = x 2 + 7x + 10. Then f (x) = 2x + 7 and f (x) = 2 > 0 for all x. Therefore, f is concave up everywhere, and there are no points of inflection.

9. y = x − 2 cos x y = t 3 − 3t 2 + 1

182

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION Let f (x) = x − 2 cos x. Then f (x) = 1 + 2 sin x and f (x) = 2 cos x = 0 at x = 1 2 (2n + 1) π , where n is an integer. Now, f is concave up on the intervals

π π − + 2n π , + 2n π 2 2

where n is any integer since f (x) > 0 there. Moreover, f is concave down on the intervals 3π π + 2n π , + 2n π 2 2 where n is any integer since f (x) < 0 there. Finally, because f (x) changes sign at each x = 12 (2n + 1)π , there is a point of inflection at each of these locations. √ 11. y = x(x − 8 x) 5 y = 4x − 5x 4 √ SOLUTION Let f (x) = x(x − 8 x) = x 2 − 8x 3/2 . Then f (x) = 2x − 12x 1/2 and f (x) = 2 − 6x −1/2 . Now, f is concave down for 0 < x < 9 since f (x) < 0 there. Moreover, f is concave up for x > 9 since f (x) > 0 there. Finally, because f (x) changes sign at x = 9, f (x) has a point of inflection at x = 9. 1 13. y =y =2 (x − 2)(1 − x 3 ) x +3 SOLUTION

1 2x . Then f (x) = − 2 Let f (x) = 2 and x +3 (x + 3)2 f (x) = −

2(x 2 + 3)2 − 8x 2 (x 2 + 3) 6x 2 − 6 = . (x 2 + 3)4 (x 2 + 3)3

Now, f is concave up for |x| > 1 since f (x) > 0 there. Moreover, f is concave down for |x| < 1 since f (x) < 0 there. Finally, because f (x) changes sign at both x = −1 and x = 1, f (x) has a point of inflection at both x = −1 and x = 1. 15. Sketch the7/5 graph of f (x) = x 4 and state whether f has any points of inflection. Verify your conclusion by showing y = x that f (x) does not change sign. SOLUTION

From the plot of f (x) = x 4 below, it appears that f has no points of inflection. Indeed, f (x) = 4x 3 and

f (x) = 12x 2 = 0 at x = 0, but f (x) does not change sign as x increases through 0 because 12x 2 ≥ 0 for all x. y 1

x

−1

1

The her growth of a sunflower its first bean 100 days is modeled by the logistic curve ybelow. = h(t)Inshown in 17. Through website, Leticia hasduring been selling bag chairs with well monthly sales as recorded a report Figure 15. Estimate growth ratereached at the point ofofinflection explain its significance. Then make a rough sketch of to investors, she the states, “Sales a point inflectionand when I started using pay-per-click advertising.” In which the first anddid second derivatives of h(t). month that occur? Explain. Height (cm)

Month Sales

300

1

250 2

2

3

4

5

6

7

8

20

30

35

38

44

60

90

200 150 100 50 t (days) 20

40

60

80

100

FIGURE 15 SOLUTION The point of inflection in Figure 15 appears to occur at t = 40 days. The graph below shows the logistic curve with an approximate tangent line drawn at t = 40. The approximate tangent line passes roughly through the points (20, 20) and (60, 240). The growth rate at the point of inflection is thus

220 240 − 20 = = 5.5 cm/day. 60 − 20 40 Because the logistic curve changes from concave up to concave down at t = 40, the growth rate at this point is the maximum growth rate for the sunflower plant.


S E C T I O N 4.4

183

Height (cm) 300 250 200 150 100 50 t (days) 20

40

60

80

100

Sketches of the first and second derivative of h(t) are shown below at the left and at the right, respectively. h′

h′′

6 5

0.1

4

t

3

20

2

40

60

80

100

−0.1

1 t 20

40

60

80

100

In Exercises 19–28, find the the graph critical of f (x) and Derivative Testof (ifinflection possible)of to fdetermine whether Figure 16 shows ofpoints the derivative f (x)use onthe [0, Second 1.2]. Locate the points (x) and the points eachwhere corresponds to a local minimum or maximum. the local minima and maxima occur. Determine the intervals on which f (x) has the following properties: (b) Decreasing 19. (a) f (x)Increasing = x 3 − 12x 2 + 45x (c) Concave up (d) Concave down SOLUTION Let f (x) = x 3 − 12x 2 + 45x. Then f (x) = 3x 2 − 24x + 45 = 3(x − 3)(x − 5), and the critical points are x = 3 and x = 5. Moreover, f (x) = 6x − 24, so f (3) = −6 < 0 and f (5) = 6 > 0. Therefore, by the Second Derivative Test, f (3) = 54 is a local maximum, and f (5) = 50 is a local minimum.

21. f (x) = 3x 4 − 8x 3 + 6x 2 f (x) = x 4 − 8x 2 + 1 SOLUTION Let f (x) = 3x 4 − 8x 3 + 6x 2 . Then f (x) = 12x 3 − 24x 2 + 12x = 12x(x − 1)2 = 0 at x = 0, 1 and f (x) = 36x 2 − 48x + 12. Thus, f (0) > 0, which implies f (0) is a local minimum; however, f (1) = 0, which is inconclusive. 23. f (x) = x 5 −5x 3 2 f (x) = x − x

3 SOLUTION Let f (x) = x 5 − x 3 . Then f (x) = 5x 4 − 3x 2 = x 2 (5x 2 − 3) = 0 at x = 0, x = ± 5 and f (x) = 3 3 3 2 20x − 6x = x(20x − 6). Thus, f − 35 < 0, 5 > 0, which implies f 5 is a local minimum, and f which implies that f − 35 is a local maximum; however, f (0) = 0, which is inconclusive. 1 25. f (x)f (x) = = sin2 x + cos x, [0, π ] cos x + 2 1 sin x SOLUTION Let f (x) = = 0 at x = 0, ±π , ±2π , ±3π , . . . , and . Then f (x) = cos x + 2 (cos x + 2)2 f (x) =

cos2 x + 2 cos x + 2 sin2 x . (cos x + 2)3

Thus, f (0) > 0, f (±2π ) > 0 and f (±4π ) > 0, which implies that f (x) has local minima when x is an even multiple of π . Similarly, f (±π ) < 0, f (±3π ) < 0, and f (±5π ) < 0, which implies that f (x) has local maxima when x is an odd multiple of π . 27. f (x) = 3x 3/2 − 1x 1/2 f (x) = 3/2 − x 1/2 . Then f (x) = 9 x 1/2 − 1 x −1/2 = 1 x −1/2 (9x − 1), so there are two critical SOLUTION Letx 2f − (x)x = + 23x 2 2 2 1 points: x = 0 and x = 9 . Now, f (x) =

9 −1/2 1 −3/2 1 + x = x −3/2 (9x + 1). x 4 4 4

Thus, f 19 > 0, which implies f 19 is a local minimum. f (x) is undefined at x = 0, so the Second Derivative Test cannot be applied there. In Exercises 29–40, find the intervals on which f is concave up or down, the points of inflection, and the critical points, f (x) = x 7/4 − x and determine whether each critical point corresponds to a local minimum or maximum (or neither).

184

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION

Here is a table legend for Exercises 29–40. SYMBOL

MEANING

−

The entity is negative on the given interval.

0

The entity is zero at the specified point.

+

The entity is positive on the given interval.

U

The entity is undefined at the specified point.

The function ( f , g, etc.) is increasing on the given interval.

The function ( f , g, etc.) is decreasing on the given interval.

The function ( f , g, etc.) is concave up on the given interval.

The function ( f , g, etc.) is concave down on the given interval.

M

The function ( f , g, etc.) has a local maximum at the specified point.

m

The function ( f , g, etc.) has a local minimum at the specified point.

I

The function ( f , g, etc.) has an inflection point here.

¬

There is no local extremum or inflection point here.

29. f (x) = x 3 − 2x 2 + x SOLUTION

Let f (x) = x 3 − 2x 2 + x.

• Then f (x) = 3x 2 − 4x + 1 = (x − 1)(3x − 1) = 0 yields x = 1 and x = 1 as candidates for extrema. 3 • Moreover, f (x) = 6x − 4 = 0 gives a candidate for a point of inflection at x = 2 . 3

x

(−∞, 13 )

1 3

( 13 , 1)

1

(1, ∞)

x

(−∞, 23 )

2 3

( 23 , ∞)

f

+

0

−

0

+

f

−

0

+

f

M

m

f

I

31. f (t) = t 2 − t23 f (x) = x (x − 4) SOLUTION Let f (t) = t 2 − t 3 . • Then f (t) = 2t − 3t 2 = t (2 − 3t) = 0 yields t = 0 and t = 2 as candidates for extrema. 3 • Moreover, f (t) = 2 − 6t = 0 gives a candidate for a point of inflection at t = 1 . 3

t

(−∞, 0)

0

(0, 23 )

2 3

( 23 , ∞)

t

(−∞, 13 )

1 3

( 13 , ∞)

f

−

0

+

0

−

f

+

0

−

f

m

M

f

I

33. f (x) = x 2 − x41/2 2 f (x) = 2x − 3x + 2 SOLUTION Let f (x) = x 2 − x 1/2 . Note that the domain of f is x ≥ 0.

2/3 • Then f (x) = 2x − 1 x −1/2 = 1 x −1/2 4x 3/2 − 1 = 0 yields x = 0 and x = 1 as candidates for extrema. 2 2 4 • Moreover, f (x) = 2 + 1 x −3/2 > 0 for all x ≥ 0, which means there are no inflection points. 4

1 x 35. f (t) = 2 f (x)t =+ 12 x +2

x

0

2/3 0, 14

2/3

f

U

−

0

+

f

M

m

1 4

2/3 1 4

,∞


S E C T I O N 4.4

185

1 . Let f (t) = 2 t +1 2t • Then f (t) = − 2 = 0 yields t = 0 as a candidate for an extremum. 2 t +1 2

2 − 2t 2 − 2 t 2 + 1 (2) − 2t · 2 t 2 + 1 (2t) 2 3t 2 − 1 8t • Moreover, f (t) = − = 4 3 = 3 = 0 gives candidates t2 + 1 t2 + 1 t2 + 1

for a point of inflection at t = ± 13 .

SOLUTION

t

(−∞, 0)

0

(0, ∞)

f

+

0

−

f

M

−∞, − 13

− 13

− 13 , 13

f

+

0

−

0

+

f

I

I

t

1 3

1,∞ 3

37. f (θ ) = θ + sin θ for 0 ≤ θ ≤ 2π 1 f (x) = SOLUTION Let 4f (θ ) = θ + sin θ on [0, 2π ]. x +1 • Then f (θ ) = 1 + cos θ = 0 yields θ = π as a candidate for an extremum. • Moreover, f (θ ) = − sin θ = 0 gives candidates for a point of inflection at θ = 0, at θ = π , and at θ = 2π .

θ

(0, π )

f

+

f

π

(π , 2π )

θ

0

(0, π )

π

(π , 2π )

2π

0

+

f

0

−

0

+

0

¬

f

¬

I

¬

39. f (x) = x − sin x for 0 ≤ x ≤ 2π f (t) = sin2 t for 0 ≤ t ≤ π SOLUTION Let f (x) = x − sin x on [0, 2π ]. • Then f (x) = 1 − cos x > 0 on (0, 2π ). • Moreover, f (x) = sin x = 0 gives a candidate for a point of inflection at x =

π.

(0, 2π )

x

(0, π )

π

(π , 2π )

f

+

f

+

0

−

f

f

I

x

π <x < π 41. infectious slowly at the beginning of an epidemic. The infection process accelerates until a f (x)An = tan x for flu−spreads 2 2 majority of the susceptible individuals are infected, at which point the process slows down. (a) If R(t) is the number of individuals infected at time t, describe the concavity of the graph of R near the beginning and end of the epidemic. (b) Write a one-sentence news bulletin describing the status of the epidemic on the day that R(t) has a point of inflection. SOLUTION

(a) Near the beginning of the epidemic, the graph of R is concave up. Near the epidemic’s end, R is concave down. (b) “Epidemic subsiding: number of new cases declining.” 43. Water is pumped a sphere a variable in such wayh(t) that be thethe water level rises at a constant Water is pumped intointo a sphere at aatconstant raterate (Figure 17).a Let water level at time t. Sketchrate the c (Figure 17). Let V (t) be the volume of water at time t. Sketch the graph of V (t) (approximately, but with graph of h(t) (approximately, but with the correct concavity). Where does the point of inflection occur? the correct concavity). Where does the point of inflection occur?

h

FIGURE 17

186

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION Because water is entering the sphere in such a way that the water level rises at a constant rate, we expect the volume to increase more slowly near the bottom and top of the sphere where the sphere is not as “wide” and to increase more rapidly near the middle of the sphere. The graph of V (t) should therefore start concave up and change to concave down when the sphere is half full; that is, the point of inflection should occur when the water level is equal to the radius of the sphere. A possible graph of V (t) is shown below. V

t

In Exercises 45–47, sketch the graph of a function f (x) satisfying all of the given conditions. (Continuation of Exercise 43) When the water level in a sphere of radius R is h, the volume of water is 2 − 1 3 ). Assume the level rises at a constant rate c = 1 (i.e., h = t). = π> (Rh 0 and < 0 for all x. 45. Vf (x) 3f h(x) (a) Find the inflection point of V (t). Does this agree with your conclusion in Exercise 43? SOLUTION Here is the graph of a function f (x) satisfying f (x) > 0 for all x and f (x) < 0 for all x. (b) Plot V (t) for R = 1. y

0.5 x 1

2

−0.5

47. (i) f (x) < 0 for x < 0 and f (x) > 0 for x > 0, and f (x) for > all2,x,and and f (x) > 0 for |x| < 2. (x) < 0>for0 |x| (ii) f(i) (ii) f (x) < 0 for x < 0 and f (x) > 0 for x > 0. SOLUTION

Interval

(−∞, −2)

(−2, 0)

(0, 2)

(2, ∞)

Direction

Concavity

One potential graph with this shape is the following: y

−10

x 10

Further Insights and Challenges In Exercises 48–50, assume that f (x) is differentiable. (x) > 0. Prove that the graph of f (x) “sits 49. that f Derivative (x) exists and above” its tangent ProofAssume of the Second Testf Let c be a critical point in (a, b) such that f (c) > 0 (the case f lines (c) 0 (b) Show thatpart G(c) (c) that = 0 there and Gexists (x) > for all x. Use(a, thisb)tocontaining conclude that G (x) for< x< (b) Use (a)= to G show an0open interval c such that 0 ifdeduce, c < x <using b. Conclude (c) is > a local SOLUTION

(a) Let c be any number. Then y = f (c)(x − c) + f (c) is the equation of the line tangent to the graph of f (x) at x = c and G(x) = f (x) − f (c)(x − c) − f (c) measures the amount by which the value of the function exceeds the value of the tangent line (see the figure below). Thus, to prove that the graph of f (x) “sits above” its tangent lines, it is sufficient to prove that G(x) ≥ 0 for all c.

S E C T I O N 4.4


187

y

x

(b) Note that G(c) = f (c) − f (c)(c − c) − f (c) = 0, G (x) = f (x) − f (c) and G (c) = f (c) − f (c) = 0. Moreover, G (x) = f (x) > 0 for all x. Now, because G (c) = 0 and G (x) is increasing, it must be true that G (x) < 0 for x < c and that G (x) > 0 for x > c. Therefore, G(x) is decreasing for x < c and increasing for x > c. This implies that G(c) = 0 is a minimum; consequently G(x) > G(c) = 0 for x = c. 51. Let C(x) be the cost of producing x units of a certain good. Assume that the graph of C(x) is concave up. Assume that f (x) exists and let c be a point of inflection of f (x). (a) Show that the average cost A(x) = C(x)/x is minimized at that production level x 0 for which average cost equals (a) Use marginal cost.the method of Exercise 49 to prove that the tangent line at x = c crosses the graph (Figure 18). Hint: Show that G(x) sign at x(0, = 0) c. and (x , C(x )) is tangent to the graph of C(x). (b) Show that changes the line through 0 0 Let C(x) be the cost of producing x units of a commodity. Assume the graph of C is concave up. (a) Let A(x) = C(x)/x be the average cost and let x 0 be the production level at which average cost is minimized. x x C (x 0 ) − C(x 0for ) f (x) = graphing f (x) and the tangent line at each inflection point on ) = 0this conclusion = 0 implies x C 1(xby Then(b) A (x 0Verify 20+ 0 ) − C(x 0 ) = 0, whence C (x 0 ) = C(x 0 )/x 0 = A(x 0 ). In other 3x 2 x the same set of axes. 0 words, A(x0 ) = C (x 0 ) or average cost equals marginal cost at production level x 0 . To confirm that x 0 corresponds to a local minimum of A, we use the Second Derivative Test. We find SOLUTION

x 2 C (x 0 ) − 2(x 0 C (x 0 ) − C(x 0 )) C (x 0 ) A (x 0 ) = 0 = >0 3 x0 x0 because C is concave up. Hence, x0 corresponds to a local minimum. (b) The line between (0, 0) and (x 0 , C(x 0 )) is C(x 0 ) C(x 0 ) − 0 (x − x0 ) + C(x 0 ) = A(x 0 )(x − x 0 ) + C(x 0 ) (x − x 0 ) + C(x 0 ) = x0 − 0 x0 = C (x0 )(x − x 0 ) + C(x 0 ) which is the tangent line to C at x0 . 53. Critical Points and Inflection Points If f (c) = 0 and f (c) is neither a local min or max, must x = c be a point Let f (x) be a polynomial of degree n. of inflection? This is true of most “reasonable” examples (including the examples in this text), but it is not true in general. (a) Show that if n is odd and n > 1, then f (x) has at least one point of inflection. Let have a point of inflection if n is even. (b) Show by giving an example that f (x) need not x 2 sin 1x for x = 0 f (x) = 0 for x = 0 (a) Use the limit definition of the derivative to show that f (0) exists and f (0) = 0. (b) Show that f (0) is neither a local min nor max. (c) Show that f (x) changes sign infinitely often near x = 0 and conclude that f (x) does not have a point of inflection at x = 0. x 2 sin (1/x) for x = 0 . SOLUTION Let f (x) = 0 for x = 0 1 f (x) − f (0) x 2 sin (1/x) (a) Now f (0) = lim = lim = lim x sin = 0 by the Squeeze Theorem: as x → 0 x −0 x x x→0 x→0 x→0 we have x sin 1 − 0 = |x| sin 1 → 0, x x since | sin u| ≤ 1. (b) Since sin( 1x ) oscillates through every value between −1 and 1 with increasing frequency as x → 0, in any open interval (−δ , δ ) there are points a and b such that f (a) = a 2 sin( a1 ) < 0 and f (b) = b2 sin( b1 ) > 0. Accordingly, f (0) = 0 can neither be a local minimum value nor a local maximum value of f . (c) In part (a) it was shown that f (0) = 0. For x = 0, we have 1 1 1 1 1 f (x) = x 2 cos − 2 + 2x sin = 2x sin − cos . x x x x x As x → 0, f (x) oscillates increasingly rapidly; consequently, f (x) changes sign infinitely often near x = 0. From this we conclude that f (x) does not have a point of inflection at x = 0.

188

CHAPTER 4


4.5 Graph Sketching and Asymptotes Preliminary Questions 1. Sketch an arc where f and f have the sign combination ++. Do the same for −+. SOLUTION An arc with the sign combination ++ (increasing, concave up) is shown below at the left. An arc with the sign combination −+ (decreasing, concave up) is shown below at the right. y

y

x

x

2. If the sign combination of f and f changes from ++ to +− at x = c, then (choose the correct answer): (a) f (c) is a local min (b) f (c) is a local max (c) c is a point of inflection SOLUTION

Because the sign of the second derivative changes at x = c, the correct response is (c): c is a point of

inflection. 3. What are the following limits? (a) lim x 3 x→∞

(b)

lim x 3

(c)

x→−∞

lim x 4

x→−∞

SOLUTION

(a) limx→∞ x 3 = ∞ (b) limx→−∞ x 3 = −∞ (c) limx→−∞ x 4 = ∞ 4. What is the sign of a2 if f (x) = a2 x 2 + a1 x + a0 satisfies

lim

x→−∞

f (x) = −∞?

The behavior of f (x) = a2 x 2 + a1 x + a0 as x → −∞ is controlled by the leading term; that is, limx→−∞ f (x) = limx→−∞ a2 x 2 . Because x 2 → +∞ as x → −∞, a2 must be negative to have limx→−∞ f (x) = −∞. SOLUTION

5. What is the sign of the leading coefficient a7 if f (x) is a polynomial of degree 7 such that

lim

x→−∞

f (x) = ∞?

SOLUTION The behavior of f (x) as x → −∞ is controlled by the leading term; that is, lim x→−∞ f (x) = limx→−∞ a7 x 7 . Because x 7 → −∞ as x → −∞, a7 must be negative to have limx→−∞ f (x) = ∞.

6. The second derivative of the function f (x) = (x − 4)−1 is f (x) = 2(x − 4)−3 . Although f (x) changes sign at x = 4, f (x) does not have a point of inflection at x = 4. Why not? SOLUTION

The function f does not have a point of inflection at x = 4 because x = 4 is not in the domain of f.

Exercises 1. Determine the sign combinations of f and f for each interval A–G in Figure 18. y

y = f(x)

A

B C

D

E F

G

FIGURE 18 SOLUTION

• In A, f is decreasing and concave up, so f < 0 and f > 0. • In B, f is increasing and concave up, so f > 0 and f > 0.

• In C, f is increasing and concave down, so f > 0 and f < 0. • In D, f is decreasing and concave down, so f < 0 and f < 0. • In E, f is decreasing and concave up, so f < 0 and f > 0.

x

S E C T I O N 4.5

Graph Sketching and Asymptotes

189

• In F, f is increasing and concave up, so f > 0 and f > 0. • In G, f is increasing and concave down, so f > 0 and f < 0. f 19. takeExample: on the given sign combinations. In Exercises drawchange the graph of atransition function for which State 3–6, the sign at each point A–Gfinand Figure f (x) goes from + to − at A.

3. ++,

+−,

SOLUTION

x = 0.

−−

This function changes from concave up to concave down at x = −1 and from increasing to decreasing at −1

0

1

x

−1

y

5. −+, −−, −+ +−, −−, −+ SOLUTION The function is decreasing everywhere and changes from concave up to concave down at x = −1 and from concave down to concave up at x = − 12 . y

0.05

x

−1

0

7. Sketch graph +− of y = x 2 − 2x + 3. −+,the++, SOLUTION Let f (x) = x 2 − 2x + 3. Then f (x) = 2x − 2 and f (x) = 2. Hence f is decreasing for x < 1, is increasing for x > 1, has a local minimum at x = 1 and is concave up everywhere. y 5 4 3 2 1 x

−1

1

2

3

9. Sketch graph of the cubic f (x) = x 3 2− 3x 2 + 2. For extra accuracy, plot the zeros of f (x), which are x = 1 and √ the the Sketch of y2.73. = 3 + 5x − 2x . x = 1 ± 3 or x ≈graph −0.73, Let f (x) = x 3 − 3x 2 + 2. Then f (x) = 3x 2 − 6x = 3x(x − 2) = 0 yields x = 0, 2 and f (x) = 6x − 6. Thus f is concave down for x < 1, is concave up for x > 1, has an inflection point at x = 1, is increasing for x < 0 and for x > 2, is decreasing for 0 < x < 2, has a local maximum at x = 0, and has a local minimum at x = 2. SOLUTION

y 2 1 −1

x −1

1

2

3

−2

11. Extend the sketch of the graph of f (x) = cos x + 1 x over [0, π ] in Example 4 to the interval [0, 5π ]. Show that the cubic x 3 − 3x 2 + 6x has a point 2of inflection but no local extreme values. Sketch the graph. 1 1 π 5π 13π 17π SOLUTION Let f (x) = cos x + 2 x. Then f (x) = − sin x + 2 = 0 yields critical points at x = 6 , 6 , 6 , 6 , 25π , and 29π . Moreover, f (x) = − cos x so there are points of inflection at x = π , 3π , 5π , 7π , and 9π . 6 6 2 2 2 2 2

190

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E y 6 4 2 x 0

2

4

6

8 10 12 14

In Exercises 13–30, sketch the graph of the function. Indicate the transition points (local extrema and points of inflection). Sketch the graphs of y = x 2/3 and y = x 4/3 . 13. y = x 3 + 32 x 2 Let f (x) = x 3 + 32 x 2 . Then f (x) = 3x 2 + 3x = 3x (x + 1) and f (x) = 6x + 3. This shows that f has critical points at x = 0 and x = −1 and a candidate for an inflection point at x = − 12 . SOLUTION

Interval

(−∞, −1)

(−1, − 12 )

(− 12 , 0)

(0, ∞)

+−

−−

−+

++

Signs of f and f

Thus, there is a local maximum at x = −1, a local minimum at x = 0, and an inflection point at x = − 12 . Here is a graph of f with these transition points highlighted as in the graphs in the textbook. y 2 1 −2

x

−1

−1

15. y = x 2 −34x 3 y = x − 3x + 5 SOLUTION Let f (x) = x 2 − 4x 3 . Then f (x) = 2x − 12x 2 = 2x(1 − 6x) and f (x) = 2 − 24x. Critical points are 1 . at x = 0 and x = 16 , and the sole candidate point of inflection is at x = 12 Interval Signs of f and f

(−∞, 0)

1 ) (0, 12

1 , 1) ( 12 6

( 16 , ∞)

−+

++

+−

−−

1 . Here is the graph Thus, f (0) is a local minimum, f ( 16 ) is a local maximum, and there is a point of inflection at x = 12 of f with transition points highlighted as in the textbook: y 0.04 x

−0.2

0.2 −0.04

17. y = 4 −12x32 + 162x 4 y = 3 x + x + 3x

1 2 2 SOLUTION Let f (x) = 6 x 4 − 2x 2 + 4. Then f (x) = 3 x 3 − 4x = 3 x x 2 − 6 and f (x) = 2x 2 − 4. This shows √ √ that f has critical points at x = 0 and x = ± 6 and has candidates for points of inflection at x = ± 2. Interval Signs of f and f

√ (−∞, − 6)

√ √ (− 6, − 2)

√ (− 2, 0)

−+

++

+−

(0,

√

2)

−−

√ √ ( 2, 6)

√ ( 6, ∞)

−+

++

√ √ Thus, f has local minima at x = ± 6, a local maximum at x = 0, and inflection points at x = ± 2. Here is a graph of f with transition points highlighted.

S E C T I O N 4.5


191

y 10 5 −2

2

x

19. y = x 5 + 5x y = 7x 4 − 6x 2 + 1 SOLUTION Let f (x) = x 5 + 5x. Then f (x) = 5x 4 + 5 = 5(x 4 + 1) and f (x) = 20x 3 . f (x) > 0 for all x, so the graph has no critical points and is always increasing. f (x) = 0 at x = 0. Sign analyses reveal that f (x) changes from negative to positive at x = 0, so that the graph of f (x) has an inflection point at (0, 0). Here is a graph of f with transition points highlighted. y 40 20 −2

−1

x 1

−20

2

−40

21. y = x 4 −53x 3 +34x y = x − 5x SOLUTION Let f (x) = x 4 − 3x 3 + 4x. Then f (x) = 4x 3 − 9x 2 + 4 = (4x 2 − x − 2)(x − 2) and f (x) = √ 1 ± 33 2 and candidate points of 12x − 18x = 6x(2x − 3). This shows that f has critical points at x = 2 and x = 8 √ 3 inflection at x = 0 and x = 2 . Sign analyses reveal that f (x) changes from negative to positive at x = 1−8 33 , from √

√

positive to negative at x = 1+8 33 , and again from negative to positive at x = 2. Therefore, f ( 1−8 33 ) and f (2) are √ local minima of f (x), and f ( 1+ 8 33 ) is a local maximum. Further sign analyses reveal that f (x) changes from positive to negative at x = 0 and from negative to positive at x = 32 , so that there are points of inflection both at x = 0 and x = 32 . Here is a graph of f (x) with transition points highlighted. y 6 4 2 −1

x 1

−2

2

23. y = 6x 7 − 7x 6 y = x 2 (x − 4)2 SOLUTION Let f (x) = 6x 7 − 7x 6 . Then f (x) = 42x 6 − 42x 5 = 42x 5 (x − 1) and f (x) = 252x 5 − 210x 4 = 42x 4 (6x − 5). Critical points are at x = 0 and x = 1, and candidate inflection points are at x = 0 and x = 56 . Sign analyses reveal that f (x) changes from positive to negative at x = 0 and from negative to positive at x = 1. Therefore f (0) is a local maximum and f (1) is a local minimum. Also, f (x) changes from negative to positive at x = 56 . Therefore, there is a point of inflection at x = 56 . Here is a graph of f with transition points highlighted. y 2 1 −0.5

x −1

0.5

1

√ 25. y = x − x y = x 6 − 3x 4 √ 1 SOLUTION Let f (x) = x − x = x − x 1/2 . Then f (x) = 1 − 2 x −1/2 . This shows that f has critical points at x = 0 1 (where the derivative does not exist) and at x = 4 (where the derivative is zero). Because f (x) < 0 for 0 < x < 14

and f (x) > 0 for x > 14 , f 14 is a local minimum. Now f (x) = 14 x −3/2 > 0 for all x > 0, so the graph is always concave up. Here is a graph of f with transition points highlighted.

192

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E y 0.8 0.6 0.4 0.2 x 0.5

−0.2

1

1.5

2

√ √ + 9−x 27. y = x√ y = x +2 √ √ SOLUTION Let f (x) = x + 9 − x = x 1/2 + (9 − x)1/2 . Note that the domain of f is [0, 9]. Now, f (x) = 1 x −1/2 − 1 (9 − x)−1/2 and f (x) = − 1 x −3/2 − 1 (9 − x)−3/2 . Thus, the critical points are x = 0, x = 9 and 2 2 4 4 2 x = 9. Sign analysis reveals that f (x) > 0 for 0 < x < 92 and f (x) < 0 for 92 < x < 9, so f has a local maximum at x = 92 . Further, f (x) < 0 on (0, 9), so the graph is always concave down. Here is a graph of f with the transition point highlighted. y 4 3 2 1 x 0

1 2 3 4 5 6 7 8 9

29. y = (x 2 − x)1/3 1/3 y = x(1 − x) SOLUTION Let f (x) = (x 2 − x)1/3 . Then f (x) = and 1 f (x) =

3

1 (2x − 1)(x 2 − x)−2/3 3

2 2(x 2 − x)−2/3 − (2x − 1)(2x − 1)(x 2 − x)−5/3

3

2 2 = (x 2 − x)−5/3 3(x 2 − x) − (2x − 1)2 = − (x 2 − x)−5/3 (x 2 − x + 1). 9 9

Critical points of f (x) are points where the numerator 2x − 1 = 0 or where f (x) doesn’t exist, that is, at x = 12 and points where x 2 − x = 0 so that x = 1 or x = 0. Candidate points of inflection lie at points where f (x) = 0 (of which there are none), and points where f (x) does not exist (at x = 0 and x = 1). Sign analyses reveal that f (x) < 0 for x < 0 and for x > 1, while f (x) > 0 for 0 < x < 1. Therefore, the graph of f (x) has points of inflection at x = 0 and x = 1. Since (x 2 − x)−2/3 is positive wherever it is defined, the sign of f (x) depends solely on the sign of 2x − 1. Hence, f (x) does not change sign at x = 0 or x = 1, and goes from negative to positive at x = 12 . f ( 12 ) is, in that case, a local minimum. Here is a graph of f (x) with the transition points indicated. y 1.5 1 0.5 −1

x −0.5

1

2

31. Sketch the3graph of f (x) = 18(x − 3)(x − 1)2/3 using the following formulas: y = (x − 4x)1/3 3 30(x − 95 ) (x) = 20(x − 5 ) , f f (x) = (x − 1)1/3 (x − 1)4/3 SOLUTION

f (x) =

30(x − 95 ) (x − 1)1/3

S E C T I O N 4.5


193

yields critical points at x = 95 , x = 1. f (x) =

20(x − 35 ) (x − 1)4/3

yields potential inflection points at x = 35 , x = 1. Interval

signs of f and f

(−∞, 35 )

+−

( 35 , 1) (1, 95 ) ( 95 , ∞)

++ −+ ++

The graph has an inflection point at x = 35 , a local maximum at x = 1 (at which the graph has a cusp), and a local minimum at x = 95 . The sketch looks something like this. y −2

40 −1 20

1

2

3

−20 −40 −60 −80

x

In Exercises 32–37, sketch the graph over the given interval. Indicate the transition points. 33. y = sin x + cos x, [0, 2π ] y = x + sin x, [0, 2π ] SOLUTION Let f (x) = sin x + cos x. Setting f (x) = cos x − sin x = 0 yields sin x = cos x, so that tan x = 1, and x = π4 , 54π . Setting f (x) = − sin x − cos x = 0 yields sin x = − cos x, so that − tan x = 1, and x = 34π , x = 74π . Interval

signs of f and f

(0, π4 )

+−

( π4 , 34π )

−−

( 34π , 54π )

−+

( 54π , 74π )

++

( 74π , 2π )

+−

The graph has a local maximum at x = π4 , a local minimum at x = 54π , and inflection points at x = 34π and x = 74π . Here is a sketch of the graph of f (x): y 1 x 1

2

3

4

5

6

−1

35. y = sin x + 12 x, [0,2 2π ] y = 2 sin x − cos x, [0, 2π ] 1 1 2π 4π SOLUTION Let f (x) = sin x + 2 x. Setting f (x) = cos x + 2 = 0 yields x = 3 or 3 . Setting f (x) = − sin x = 0 yields potential points of inflection at x = 0, π , 2π .

194

CHAPTER 4


Interval

signs of f and f

(0, 23π )

+−

( 23π , π )

−−

(π , 43π )

−+

( 43π , 2π )

++

The graph has a local maximum at x = 23π , a local minimum at x = 43π , and an inflection point at x = π . Here is a graph of f without transition points highlighted. y 3 2 1 x 0

1

2

3

4

5

6

1 √ ] π] [0, π[0, 37. y =y sin x −x +sin 32x, = sin cos x, 2 Hint for Exercise 37: We find numerically that there is one point of inflection in the interval, occurring at x ≈ 1.3182. Let f (x) = sin x − 12 sin 2x. Setting f (x) = cos x − cos 2x = 0 yields cos 2x = cos x. Using the double angle formula for cosine, this gives 2 cos2 x − 1 = cos x or (2 cos x + 1)(cos x − 1) = 0. Solving for x ∈ [0, π ], we find x = 0 or 23π . Setting f (x) = − sin x + 2 sin 2x = 0 yields 4 sin x cos x = sin x, so sin x = 0 or cos x = 14 . Hence, there are potential points of inflection at x = 0, x = π and x = cos−1 41 ≈ 1.31812. SOLUTION

Interval

Sign of f and f

(0, cos−1 41 )

++

(cos−1 14 , 23π )

+−

( 23π , π )

−−

The graph of f (x) has a local maximum at x = 23π and a point of inflection at x = cos−1 41 . y

1

x 0

1

2

3

39. Suppose that f is twice differentiable satisfying (i) f (0) = 1, (ii) f (x) > 0 for all x = 0, and (iii) f (x) < 0 for all > sign transitions possible? x < 0 andAre f (x) 0 for x > 0. Let g(x) =Explain f (x 2 ). with a sketch why the transitions ++ → −+ and −− → +− do not occur if the function is differentiable. (See Exercise 83 for a proof.) (a) Sketch a possible graph of f (x). (b) Prove that g(x) has no points of inflection and a unique local extreme value at x = 0. Sketch a possible graph of g(x). SOLUTION

(a) To produce a possible sketch, we give the direction and concavity of the graph over every interval. Interval

(−∞, 0)

(0, ∞)

Direction

Concavity

A sketch of one possible such function appears here:


S E C T I O N 4.5

195

y

2 −2

x

−1

1

2

(b) Let g(x) = f (x 2 ). Then g (x) = 2x f (x 2 ). If g (x) = 0, either x = 0 or f (x 2 ) = 0, which implies that x = 0 as well. Since f (x 2 ) > 0 for all x = 0, g (x) < 0 for x < 0 and g (x) > 0 for x > 0. This gives g(x) a unique local extreme value at x = 0, a minimum. g (x) = 2 f (x 2 ) + 4x 2 f (x 2 ). For all x = 0, x 2 > 0, and so f (x 2 ) > 0 and f (x 2 ) > 0. Thus g (x) > 0, and so g (x) does not change sign, and can have no inflection points. A sketch of g(x) based on the sketch we made for f (x) follows: indeed, this sketch shows a unique local minimum at x = 0. y

2 −1

x 1

In Exercises following limits theofnumerator and Explain. denominator by the highest power of x Which41–50, of the calculate graphs in the Figure 20 cannot be (divide the graph a polynomial? appearing in the denominator). 41. lim

x

x→∞ x + 9

SOLUTION

1 x x −1 (x) 1 = = lim −1 = 1. = lim x→∞ x + 9 x→∞ x (x + 9) x→∞ 1 + 9 1+0 x lim

3x 22 + 20x 43. lim 3x + 20x 4 x→∞ lim 2x + 3x 3 − 29 x→∞ 4x 2 + 9 SOLUTION

3 + 20 0 3x 2 + 20x x −4 (3x 2 + 20x) x2 x3 = = 0. = lim = lim x→∞ 2x 4 + 3x 3 − 29 x→∞ x −4 (2x 4 + 3x 3 − 29) x→∞ 2 + 3 − 29 2 4 x

lim

x

7x − 9 45. lim 4 x→∞ 4x + 3 lim x→∞ x + 5 SOLUTION

7− 7x − 9 x −1 (7x − 9) = lim −1 = lim x→∞ 4x + 3 x→∞ x (4x + 3) x→∞ 4 + lim

47.

9 7 x = . 3 4 x

7x 2 2− 9 lim 7x − 9 x→−∞ lim 4x + 3 x→∞ 4x + 3

SOLUTION

7x − 9x 7x 2 − 9 x −1 (7x 2 − 9) = −∞. = lim = lim x→−∞ 4x + 3 x→−∞ x −1 (4x + 3) x→−∞ 4 + 3 x lim

49.

x2 − 1 limlim 5x − 9 x→−∞ +34 x→∞x4x +1

SOLUTION

x− x2 − 1 x −1 (x 2 − 1) = lim = lim x→−∞ x + 4 x→−∞ x −1 (x + 4) x→−∞ 1 + lim

1 x = −∞. 4 x

196

CHAPTER 4


In Exercises 51–56, calculate the limit. 2x 5 + 3x 4 − 31x lim 41 x→∞x 8x 2+ − 31x 2 + 12 51. lim x→∞ x + 1 SOLUTION

√

√

Looking at only the highest powers in the top and bottom of

x 2 +1 x+1 , it seems this quotient behaves like

x2 x −1 yields x = x . Multiplying top and bottom by x

x2 + 1 = x +1

1 + 12 x

. 1 + 1x

From this we get: lim 1 + 1/x 2 1 x2 + 1 x→∞ lim = = = 1. x→∞ x + 1 lim 1 + 1/x 1

x→∞

x +1 53. lim 4 x→∞ 3 x 2 +x1 + 1 lim x→−∞ x 3 + 1 SOLUTION We can see that the denominator behaves like x 2/3 . Dividing top and bottom by x 2/3 yields: lim x 1/3 + 1/x 2/3 x +1 ∞ x→∞ = = = ∞. lim 3 x→∞ 3 x 2 + 1 2 1 lim 1 + 1/x x→∞

55.

x

lim 4/3 +1/3 4x 1/3 x→−∞ (xx6 + 1) lim x→∞ x 5/4 − 2x

SOLUTION

Looking at highest exponents of x, we see the denominator behaves like x 2 . Dividing top and bottom by

x 2 yields: x = x→−∞ (x 6 + 1)1/3

lim 1/x

x→−∞

lim

lim (1 + 1/x 6 )1/3

=

x→−∞

0 = 0. 1

4 4xFigure − 9 21 is the graph of f (x) = 2x − 1 ? Explain on the basis of horizontal asymptotes. 57. Which curve in lim 1 + x4 x→−∞ (3x 4 − 2)1/3 y

y

2

−4

2

x

−2

2

4

−4

−1.5

x

−2

2

4

−1.5

(A)

(B)

FIGURE 21 SOLUTION

Since 2 2x 4 − 1 = · lim 1 = 2 x→±∞ 1 + x 4 1 x→±∞ lim

the graph has left and right horizontal asymptotes at y = 2, so the left curve is the graph of f (x) =

2x 4 − 1 . 1 + x4

59. Match the functions with their graphs in Figure 23. 3x 2 3x 2 1 the graphs in Figure 22 with the two functions y = x and y = Match . Explain. (a) y = 2 (b) yx 2=− 12 x2 − 1 x −1 x +1 x 1 (d) y = 2 (c) y = 2 x +1 x −1

S E C T I O N 4.5 y


197

y

x

x (A)

(B)

y

y x x

(C)

(D)

FIGURE 23 SOLUTION

1 should have a horizontal asymptote at y = 0 and vertical asymptotes at x = ±1. Further, the x 2 −1 graph should consist of positive values for |x| > 1 and negative values for |x| < 1. Hence, the graph of 21 is (D). x −1 2 (b) The graph of 2x should have a horizontal asymptote at y = 1 and no vertical asymptotes. Hence, the graph of x +1 x 2 is (A). 2 x +1 (c) The graph of 21 should have a horizontal asymptote at y = 0 and no vertical asymptotes. Hence, the graph of x +1 1 is (B). 2 x +1 (d) The graph of 2x should have a horizontal asymptote at y = 0 and vertical asymptotes at x = ±1. Further, the x −1

(a) The graph of

graph should consist of positive values for −1 < x < 0 and x > 1 and negative values for x < 1 and 0 < x < 1. Hence, the graph of 2x is (C). x −1

x 61. Sketch the graph of f (x) = 2 2x −using 1 the formulas x Sketch the graph of f (x) = + 1 . x +1 1 − x2 f (x) = , (1 + x 2 )2 SOLUTION

• Because

f (x) =

2x(x 2 − 3) . (x 2 + 1)3

x . Let f (x) = 2 x +1 lim

x→±∞

f (x) = 11 ·

lim x −1 = 0, y = 0 is a horizontal asymptote for f .

x→±∞

1 − x2 2 is negative for x < −1 and x > 1, positive for −1 < x < 1, and 0 at x = ±1. x2 + 1 Accordingly, f is decreasing for x < −1 and x > 1, is increasing for −1 < x < 1, has a local minimum value at x = −1 and a local maximum value at x = 1. • Moreover,

2x x 2 − 3 f (x) = 3 . x2 + 1 • Now f (x) =

Here is a sign chart for the second derivative, similar to those constructed in various exercises in Section 4.4. (The legend is on page 184.)

√

√ √

√ √ √ −∞, − 3 − 3, 0 0, 3 3 3, ∞ x − 3 0 f

−

0

+

0

−

0

+

f

I

I

I

• Here is a graph of f (x) =

x x2 + 1

.

198

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E y 0.5 −10

−5

x 5

10

−0.5

In Exercises 62–77, sketch the graph of the function. Indicate the asymptotes, local extrema, and points of inflection. x 63. y = 1 y 2x = −1 2x − 1 SOLUTION Let f (x) =

x −1 , so that f is decreasing for all x = 12 . Moreover, f (x) = . Then f (x) = 2x − 1 (2x − 1)2 4 x 1 , so that f is concave up for x > 12 and concave down for x < 12 . Because lim = , f has a x→±∞ 2x − 1 2 (2x − 1)3 horizontal asymptote at y = 12 . Finally, f has a vertical asymptote at x = 12 with lim

x→ 12 −

x = −∞ 2x − 1

and

lim

x→ 12 +

x = ∞. 2x − 1

y 4 2 −1

x 1

−2

2

3

−4

x +3 65. y = x +1 y x=− 2 2x − 1 x +3 −5 SOLUTION Let f (x) = , so that f is decreasing for all x = 2. Moreover, f (x) = . Then f (x) = x −2 (x − 2)2 x +3 10 , so that f is concave up for x > 2 and concave down for x < 2. Because lim = 1, f has a horizontal x→±∞ x − 2 (x − 2)3 asymptote at y = 1. Finally, f has a vertical asymptote at x = 2 with lim

x +3

x→2− x − 2

= −∞

and

lim

x +3

x→2+ x − 2

= ∞.

y 10 5 −10

−5

x 5

−5

10

−10

1 1 67. y = + 1 x −1 y x= x + x 1 2x 2 − 2x + 1 1 SOLUTION Let f (x) = , so that f is decreasing for all x = 0, 1. Moreover, + . Then f (x) = − x x− x 2 (x − 1)2 1

2 2x 3 − 3x 2 + 3x − 1 , so that f is concave up for 0 < x < 12 and x > 1 and concave down for x < 0 f (x) = x 3 (x − 1)3 1 1 and 12 < x < 1. Because lim + = 0, f has a horizontal asymptote at y = 0. Finally, f has vertical x→±∞ x x −1 asymptotes at x = 0 and x = 1 with 1 1 1 1 lim + = −∞ and lim + =∞ x −1 x −1 x→0− x x→0+ x and

1 1 + x −1 x→1− x lim

= −∞

and

1 1 + x −1 x→1+ x lim

= ∞.

S E C T I O N 4.5


199

y 5 x

−1

1

−5

2

4 3 69. y = 1 −1 + 13 y = x− x x x −1 3 4 SOLUTION Let f (x) = 1 − + 3 . Then x x 3 12 3(x − 2)(x + 2) , f (x) = 2 − 4 = x x x4 so that f is increasing for |x| > 2 and decreasing for −2 < x < 0 and for 0 < x < 2. Moreover, 6 48 6(8 − x 2 ) , f (x) = − 3 + 5 = x x x5 √ √ √ so that f is √ concave down for −2 2 < x < 0 and for x > 2 2, while f is concave up for x < −2 2 and for 0 < x < 2 2. Because 4 3 lim 1 − + 3 = 1, x→±∞ x x f has a horizontal asymptote at y = 1. Finally, f has a vertical asymptote at x = 0 with 4 4 3 3 and lim 1 − + 3 = ∞. lim 1 − + 3 = −∞ x x x→0− x→0+ x x y 6 4 2 x

−6 −4 −2

−2

2

4

6

−4 −6

1 1 71. y = 2 + 1 2 y x= 2 (x − 2) x − 6x + 8 1 1 SOLUTION Let f (x) = + . Then 2 x (x − 2)2 f (x) = −2x −3 − 2 (x − 2)−3 = −

4(x − 1)(x 2 − 2x + 4) , x 3 (x − 2)3

so that f is increasing for x < 0 and for 1 < x < 2, is decreasing for 0 < x < 1 and for x > 2, and has a local minimum at x = 1. Moreover, f (x) = 6x −4 + 6 (x − 2)−4 , so that f is concave up for all x = 0, 2. Because 1 1 + lim = 0, f has a horizontal asymptote at y = 0. Finally, f has vertical asymptotes at x = 0 and x→±∞ x 2 (x − 2)2 x = 2 with 1 1 1 1 + + lim = ∞ and lim =∞ x→0− x 2 x→0+ x 2 (x − 2)2 (x − 2)2 and

1 1 + x→2− x 2 (x − 2)2 lim

=∞

1 1 + x→2+ x 2 (x − 2)2

and

lim

y

2 −2 −1

x 1

2

3

4

= ∞.

200

CHAPTER 4


4 73. y = 2 1 1 y x= −29− x (x − 2)2 4 8x . Then f (x) = − SOLUTION Let f (x) = 2 , so that f is increasing for x < −3 and for −3 < x < 0, 2 x2 − 9 x −9

24 x 2 + 3 is decreasing for 0 < x < 3 and for x > 3, and has a local maximum at x = 0. Moreover, f (x) = 3 , so x2 − 9 4 = 0, f that f is concave up for x < −3 and for x > 3 and is concave down for −3 < x < 3. Because lim 2 x→±∞ x − 9 has a horizontal asymptote at y = 0. Finally, f has vertical asymptotes at x = −3 and x = 3, with 4 4 lim = ∞ and lim = −∞ x→−3− x 2 − 9 x→−3+ x 2 − 9 and

4 x→3− x 2 − 9

lim

= −∞

4 x→3+ x 2 − 9

and

lim

= ∞.

y 2 1 x

−5

5

−1 −2

x12 75. y =y =2 2 (x (x −21)(x + 1)2+ 1) SOLUTION Let f (x) =

x2 (x 2 − 1)(x 2 + 1)

.

Then f (x) = −

2x(1 + x 4 ) (x − 1)2 (x + 1)2 (x 2 + 1)2

,

so that f is increasing for x < −1 and for −1 < x < 0, is decreasing for 0 < x < 1 and for x > 1, and has a local maximum at x = 0. Moreover, f (x) =

2 + 24x 4 + 6x 8 , (x − 1)3 (x + 1)3 (x 2 + 1)3

so that f is concave up for |x| > 1 and concave down for |x| < 1. Because

lim

x2

x→±∞ (x 2 − 1)(x 2 + 1)

horizontal asymptote at y = 0. Finally, f has vertical asymptotes at x = −1 and x = 1, with lim

x2

x→−1− (x 2 − 1)(x 2 + 1)

=∞

and

x2

lim

x→−1+ (x 2 − 1)(x 2 + 1)

= −∞

and lim

x2

x→1− (x 2 − 1)(x 2 + 1)

= −∞

and

lim

y 2 1 −2

x

−1

1 −1 −2

x2

x→1+ (x 2 − 1)(x 2 + 1)

2

= ∞.

= 0, f has a

S E C T I O N 4.5


201

x 77. y = 1 2 y =x + 1 2 x +1 SOLUTION Let f (x) =

x x2 + 1

.

Then f (x) = (x 2 + 1)−3/2

and

f (x) =

−3x . (x 2 + 1)5/2

Thus, f is increasing for all x, is concave up for x < 0, is concave down for x > 0, and has a point of inflection at x = 0. Because x x lim =1 and lim = −1, x→∞ x 2 + 1 x→−∞ x 2 + 1 f has horizontal asymptotes of y = −1 and y = 1. There are no vertical asymptotes. y 1

x

−5

5 −1

Further Insights and Challenges In Exercises 78–82, we explore functions whose graphs approach a nonhorizontal line as x → ∞. A line y = ax + b is called a slant asymptote if lim ( f (x) − (ax + b)) = 0

x→∞

or lim ( f (x) − (ax + b)) = 0.

x→−∞

If f (x) =x 2P(x)/Q(x), where P and Q are polynomials of degrees m + 1 and m, then by long division, we 79. . Verify the following: Let f (x) = can write x −1 (a) f is concave down on (−∞, 1) andf concave up+onb)(1, as in Figure 24. (x) = (ax + ∞) P (x)/Q(x) 1

where P1 is a polynomial of degree < m. Show that y = ax + b is the slant asymptote of f (x). Use this procedure to find the slant asymptotes of the functions: (b) f (0)2 is a local max and f (2) a local min. x x3 + x (a) y(c)= lim f (x) = −∞ and lim f (x) = ∞. (b) y = 2 x +2 x +x +1 x→1− x→1+ (d) y = x + 1 is a slant asymptote of f (x) as x → ±∞. SOLUTION Since deg(P1 ) < deg(Q), (e) The slant asymptote lies above the graph of f (x) for x < 1 and below the graph for x > 1. P1 (x) lim = 0. x→±∞ Q(x) Thus lim ( f (x) − (ax + b)) = 0

x→±∞

and y = ax + b is a slant asymptote of f . x2 4 x2 (a) = x −2+ ; hence y = x − 2 is a slant asymptote of . x +2 x +2 x +2 3 x +x x +1 x3 + x (b) 2 = (x − 1) + 2 ; hence, y = x − 1 is a slant asymptote of 2 . x +x +1 x −1 x +x +1 81. Show that y = 3x is a slant asymptote for f (x) = 3x + x −2 . Determine whether f (x) approaches the slant asympx2 tote from abovethe or graph belowof andf (x) make of the graph. . Proceed as in the previous exercise to find the slant asymptote. Sketch = a sketch x +1

202

CHAPTER 4


Let f (x) = 3x + x −2 . Then lim ( f (x) − 3x) =

x→±∞

lim (3x + x −2 − 3x) =

x→±∞

lim x −2 = 0

x→±∞

which implies that 3x is the slant asymptote of f (x). Since f (x) − 3x = x −2 > 0 as x → ±∞, f (x) approaches the slant asymptote from above in both directions. Moreover, f (x) = 3 − 2x −3 and f (x) = 6x −4 . Sign analyses reveal

−1/3 a local minimum at x = 32 ≈ .87358 and that f is concave up for all x = 0. Limit analyses give a vertical asymptote at x = 0. y 5 −4

x

−2

2

4

−5

83. Assume that f (x) and f (x) exist for all x and let c be a critical point of f (x). Show that f (x) cannot make a − x 2Apply the MVT to f (x). transition from the ++graph to −+ x ==c. 1Hint: Sketch ofatf (x) . 2−x SOLUTION Let f (x) be a function such that f (x) > 0 for all x and such that it transitions from ++ to −+ at a critical point c where f (c) is defined. That is, f (c) = 0, f (x) > 0 for x < c and f (x) < 0 for x > c. Let g(x) = f (x). The previous statements indicate that g(c) = 0, g(x 0 ) > 0 for some x 0 < c, and g(x1 ) < 0 for some x 1 > c. By the Mean Value Theorem, g(x 1 ) − g(x 0 ) = g (c0 ), x1 − x0 for some c0 between x 0 and x1 . Because x1 > c > x 0 and g(x 1 ) < 0 < g(x 0 ), g(x 1 ) − g(x 0 ) < 0. x1 − x0 But, on the other hand g (c0 ) = f (c0 ) > 0, so there is a contradiction. This means that our assumption of the existence of such a function f (x) must be in error, so no function can transition from ++ to −+. If we drop the requirement that f (c) exist, such a function can be found. The following is a graph of f (x) = −x 2/3 . f (x) > 0 wherever f (x) is defined, and f (x) transitions from positive to negative at x = 0. y −1

−0.5

x 0.5

1

−0.8

Assume that f (x) exists and f (x) > 0 for all x. Show that f (x) cannot be negative for all x. Hint: Show f (b) = 0 for some b and use the result of Exercise 49 in Section 4.4.

4.6 Applied Optimization Preliminary Questions

1. The problem is to find the right triangle of perimeter 10 whose area is as large as possible. What is the constraint equation relating the base b and height h of the triangle? The perimeter of a right triangle is the sum of the lengths of the base, the height and the hypotenuse. If the 2 2 base has length b and the height is h, then the length of the hypotenuse is b + h and the perimeter of the triangle is 2 2 P = b + h + b + h . The requirement that the perimeter be 10 translates to the constraint equation b + h + b2 + h 2 = 10. SOLUTION

2. What are the relevant variables if the problem is to find a right circular cone of surface area 20 and maximum volume?


Applied Optimization

203

The relevant variables are the radius r and the height h of the cone. We are asked to maximize the volume V =

1 2 πr h 3

subject to the constraint S = π r r 2 + h 2 + π r 2 = 20. 3. Does a continuous function on an open interval always have a maximum value? SOLUTION No, it is possible for a continuous function on an open interval to have no maximum value. As an example, consider the function f (x) = x −1 on the open interval (0, 1). Because f is a decreasing function on (0, 1) and

lim f (x) = lim

1

x→0+ x

x→0+

= ∞,

it follows that f does not have a maximum value on (0, 1). 4. Describe a way of showing that a continuous function on an open interval (a, b) has a minimum value. If the function tends to infinity at the endpoints of the interval, then the function must take on a minimum value at a critical point. SOLUTION

Exercises 1. (a) (b) (c) (d)

Find the dimensions of the rectangle of maximum area that can be formed from a 50-in. piece of wire. What is the constraint equation relating the lengths x and y of the sides? Find a formula for the area in terms of x alone. Does this problem require optimization over an open interval or a closed interval? Solve the optimization problem.

SOLUTION

(a) The perimeter of the rectangle is 50 inches, so 50 = 2x + 2y, which is equivalent to y = 25 − x. (b) Using part (a), A = x y = x(25 − x) = 25x − x 2 . (c) This problem requires optimization over the closed interval [0, 25], since both x and y must be non-negative. 25 (d) A (x) = 25 − 2x = 0, which yields x = 25 2 and consequently, y = 2 . Because A(0) = A(25) = 0 and 25 25 2 A( 2 ) = 156.25, the maximum area 156.25 in is achieved with x = y = 2 inches. 3. Find the positive number x such that the sum of x and its reciprocal is as small as possible. Does this problem require A 100-in. piece of wire is divided into two pieces and each piece is bent into a square. How should this be done optimization over an open interval or a closed interval? in order to minimize the sum of the areas of the two squares? SOLUTION Let x > 0 and f (x) = x + x −1 . Here we require optimization over the open interval (0, ∞). Solve (a) Express the sum of the areas of the squares in terms of the lengths x and y of the two pieces. f (x) = 1 − x −2 = 0 for x > 0 to obtain x = 1. Since f (x) → ∞ as x → 0+ and as x → ∞, we conclude that f has (b) What is the constraint equation relating x and y? an absolute minimum of f (1) = 2 at x = 1. (c) Does this problem require optimization over an open or closed interval? 5. Find positive numbers x, y such that x y = 16 and x + y is as small as possible. legs a right triangle have lengths a and b satisfying a + b = 10. Which values of a and b maximize the (d) The Solve theofoptimization problem. −1 SOLUTION x, y > 0. Now x y = 16 implies y = 16 area of theLet triangle? x . Let f (x) = x + y = x + 16x . Solve f (x) = −2 = 0 for x > 0 to obtain x = 4 and, consequently, y = 4. Since f (x) → ∞ as x → 0+ and as x → ∞, we 1 − 16x conclude that f has an absolute minimum of f (4) = 8 at x = y = 4. 7. (a) (b)

Let S be the set of all rectangles with area 100. A 20-in. piece of wire is bent into an L-shape. Where should the bend be made to minimize the distance between What areends? the dimensions of the rectangle in S with the least perimeter? the two Is there a rectangle in S with the greatest perimeter? Explain.

SOLUTION

Consider the set of all rectangles with area 10.

(a) Let x, y > 0 be the lengths of the sides. Now x y = 100, so that y = 100/x. Let p(x) = 2x + 2y = 2x + 200x −1 be the perimeter. Solve p (x) = 2 − 200x −2 = 0 for x > 0 to obtain x = 10. Since p(x) → ∞ as x → 0+ and as x → ∞, the least perimeter is p (10) = 40 when x = 10 and y = 10. (b) There is no rectangle in this set with greatest perimeter. For as x → 0+ or as x → ∞, we have p(x) = 2x + 200x −1 → ∞. 9. Suppose that 600 ft of fencing are used to enclose a corral in the shape of a rectangle with a semicircle whose A box has a square base of side x and height y. diameter is a side of the rectangle as in Figure 10. Find the dimensions of the corral with maximum area. (a) Find the dimensions x, y for which the volume is 12 and the surface area is as small as possible. (b) Find the dimensions for which the surface area is 20 and the volume is as large as possible.

FIGURE 10

204

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION Let x be the width of the corral and therefore the diameter of the semicircle, and let y be the height of the rectangular section. Then the perimeter of the corral can be expressed by the equation 2y + x + π2 x = 2y + (1 + π )x = 600 ft or equivalently, y = 1 600 − (1 + π )x . Since x and y must both be nonnegative, it follows that x must 2 2 2 600 ]. The area of the corral is the sum of the area of the rectangle and semicircle, be restricted to the interval [0, 1+ π /2

A = x y + π8 x 2 . Making the substitution for y from the constraint equation, A(x) =

1

1

π π π 2 π 2 x 600 − (1 + )x + x 2 = 300x − 1+ x + x . 2 2 8 2 2 8

Now, A (x) = 300 − 1 + π2 x + π4 x = 0 implies x = 300π ≈ 168.029746 feet. With A(0) = 0 ft2 , 1+ 4 600 300 and A ≈ 25204.5 ft2 ≈ 21390.8 ft2 , A 1 + π /4 1 + π /2 it follows that the corral of maximum area has dimensions x=

300 ft 1 + π /4

y=

and

150 ft. 1 + π /4

11. A landscape architect wishes to enclose a rectangular garden on one side by a brick wall costing $30/ft and on the Find the rectangle of maximum area that can be inscribed in a right triangle with legs of length 3 and 4 if the other three sides by a metal fence costing $10/ft. If the area of the garden is 1,000 ft2 , find the dimensions of the garden sides of the rectangle are parallel to the legs of the triangle, as in Figure 11. that minimize the cost. SOLUTION

Let x be the length of the brick wall and y the length of an adjacent side with x, y > 0. With x y = 1000

or y = 1000 x , the total cost is C(x) = 30x + 10 (x + 2y) = 40x + 20000x −1 . √ −2 = 0 for x > 0 to obtain x = 10 5. Since C(x) → ∞ as x → 0+ and as x → ∞, the Solve C (x) = 40 − 20000x √ √ √ √ minimum cost is C(10 5) = 800 5 ≈ $1788.85 when x = 10 5 ≈ 22.36 ft and y = 20 5 ≈ 44.72 ft. 13. FindFind the the point P on x 2 closest the point point onthe theparabola line y = yx = closest to thetopoint (1, 0).(3, 0) (Figure 12). y

y = x2 P x 3

FIGURE 12 SOLUTION

With y = x 2 , let’s equivalently minimize the square of the distance, f (x) = (x − 3)2 + y 2 = x 4 + x 2 − 6x + 9.

Then f (x) = 4x 3 + 2x − 6 = 2(x − 1)(2x 2 + 2x + 3), so that f (x) = 0 when x = 1 (plus two complex solutions, which we discard). Since f (x) → ∞ as x → ±∞, P = (1, 1) is the point on y = x 2 closest to (3, 0). 15. Find the dimensions of the rectangle of maximum area that can be inscribed in a circle of radius r (Figure 13). A box is constructed out of two different types of metal. The metal for the top and bottom, which are both square, costs $1/ft2 and the metal for the sides costs $2/ft2 . Find the dimensions that minimize cost if the box has a volume of 20 ft3 . r

FIGURE 13

Place the center of the circle at the origin with the sides of the rectangle (of lengths 2x > 0 and 2y > 0) parallel to the coordinate axes. By the Pythagorean Theorem, x 2 + y 2 = r 2 , so that y = r 2 − x 2 . Thus the area of the rectangle is A(x) = 2x · 2y = 4x r 2 − x 2 . To guarantee both x and y are real and nonnegative, we must restrict x to the interval [0, r ]. Solve SOLUTION

4x 2 A (x) = 4 r 2 − x 2 − =0 r2 − x2

S E C T I O N 4.6


205

√ for x > 0 to obtain x = √r . Since A(0) = A(r ) = 0 and A(r/ 2) = 2r 2 , the rectangle of maximum area has dimensions x = y = √r .

2

2

17. Find the angle θ that maximizes the area of the isosceles triangle whose legs have length (Figure 14). Problem of Tartaglia (1500–1557) Among all positive numbers a, b whose sum is 8, find those for which the product of the two numbers and their difference is largest. Hint: Let x = a − b and express abx in terms of x alone. q

FIGURE 14 SOLUTION

The area of the triangle is A(θ ) =

1 2 sin θ , 2

where 0 ≤ θ ≤ π . Setting A (θ ) =

1 2 cos θ = 0 2

yields θ = π2 . Since A(0) = A(π ) = 0 and A( π2 ) = 12 2 , the angle that maximizes the area of the isosceles triangle is θ = π2 . 19. Rice production requires both labor and capital investment in equipment and land. Suppose that if x dollars per acre 2 + h 2 . Find the dimensions of the r 2 h andinitsequipment surface area S = πthen r rthe The volume a right circular is π3invested are invested in laborof and y√dollars per cone acre are andis land, yield P of rice per acre is given √ cone with surface area 1x and maximal by the formula P = 100 + 150 y. If volume a farmer(Figure invests15). $40/acre, how should he divide the $40 between labor and capital investment in order to maximize the amount of rice produced? √ √ √ 50 √ SOLUTION Since x + y = 40, we have P(x) = 100 x + 150 y = 100 x + 150 40 − x. Solve P (x) = √ − x √ √ 75 160 = 0 to obtain x = 13 ≈ $12.31. Since P(0) = 300 10 ≈ 948.68 and P(40) = 200 10 ≈ 632.46, we have √ 40 − x

√ √ 160 360 that the maximum yield is P 160 13 = 100 13 10 ≈ 1140.18 when x = 13 ≈ $12.31 and y = 13 ≈ $27.69. 21. Find the dimensions x and y of the rectangle inscribed in a circle of radius r that maximizes the quantity x y 2 . Kepler’s Wine Barrel Problem The following problem was stated and solved in the work Nova stereometria SOLUTION the center the circle of radius r at the originpublished with the sides of the rectangle (of lengths x >Kepler 0 and doliorum Place vinariorum (NewofSolid Geometry of a Wine Barrel), in 1615 by the astronomer Johannes y inscribed x )2can 2 = r 2 , whence 2 = 4r 2 radius What are the dimensions the cylinderTheorem, of largestwe volume be in the sphere of y > (1571–1630). 0) parallel to the coordinate axes. By theofPythagorean have (that + ( ) y − x 2. 2 2 − x22 ), where 2 2 3 2 2 π x(R x is one-half the height of the R? Hint: Show that the volume of an inscribed cylinder is 2 for Let f (x) = x y = 4xr − x . Allowing for degenerate rectangles, we have 0 ≤ x ≤ 2r . Solve √ f 3(x) = 4r − 3x 2r 2r 16 2r cylinder. x ≥ 0 to obtain x = √ . Since f (0) = f (2r ) = 0, the maximal value of f is f ( √ ) = 9 3r when x = √ and 3 3 3

2 y = 2 3r. 23. Consider a rectangular industrial warehouse consisting of three separate spaces of equal size as in Figure 17. Assume Find the angle θ that maximizes the area of the trapezoid with a base of length 4 and sides of length 2, as in that the wall materials cost $200 per linear ft and the company allocates $2,400,000 for the project. Figure 16. (a) Which dimensions maximize the total area of the warehouse? (b) What is the area of each compartment in this case?

FIGURE 17

Let the dimensions of one compartment be height x and width y. Then the perimeter of the warehouse is given by P = 4x + 6y and the constraint equation is cost = 2,400,000 = 200(4x + 6y), which gives y = 2000 − 23 x.

(a) Area is given by A = 3x y = 3x 2000 − 23 x = 6000x − 2x 2 , where 0 ≤ x ≤ 3000. Then A (x) = 6000 − 4x = 0 yields x = 1500 and consequently y = 1000. Since A(0) = A(3000) = 0 and A(1500) = 4, 500, 000, the area of the warehouse is maximized with a height of 1500 feet and a total width of 3000 feet. (b) The area of one compartment is 1500 · 1000 = 1, 500, 000 square feet. SOLUTION

2 . Suppose that 25. TheSuppose, amount of lightprevious reachingexercise, a point that at a the distance r fromconsists a lightof source A of intensity A is Isize. A /r Find in the warehouse n separate spaces of Iequal a formula a second lightofsource B of intensitypossible I B = 4Iarea located 10 ft from A. Find the point on the segment joining A and B A isof in terms n for the maximum the warehouse. where the total amount of light is at a minimum.

206

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION Place the segment in the x y-plane with A at the origin and B at (10, 0). Let x be the distance from A. Then 10 − x is the distance from B. The total amount of light is IA IB 4 1 f (x) = 2 + = I + . A x x2 (10 − x)2 (10 − x)2

Solve f (x) = I A

2 − 3 3 x (10 − x) 8

=0

for 0 ≤ x ≤ 10 to obtain (10 − x)3 = 4= x3

3 10 −1 x

or

x=

10 √ ≈ 3.86 ft. 1+ 3 4

Since f (x) → ∞ as x → 0+ and x → 10− we conclude that the minimal amount of light occurs 3.86 feet from A. 27. According to postal regulations, a carton is classified as “oversized” if the sum of its height and girth (the perimeter 4−x of its base) in.largest Find the dimensions of abecarton withinsquare base bounded that is not Findexceeds the area108 of the rectangle that can inscribed the region byoversized the graphand of yhas = maximum and 2 +x volume. the coordinate axes (Figure 18). SOLUTION Let h denote the height of the carton and s denote the side length of the square base. Clearly the volume will be maximized when the sum of the height and girth equals 108; i.e., 4s + h = 108, whence h = 108 − 4s. Allowing for degenerate cartons, the carton’s volume is V (s) = s 2 h = s 2 (108 − 4s), where 0 ≤ s ≤ 27. Solve V (s) = 216s − 12s 3 = 0 for s to obtain s = 0 or s = 18. Since V (0) = V (27) = 0, the maximum volume is V (18) = 11664 in3 when s = 18 in and h = 36 in. 29. What is the area of the largest rectangle that can be circumscribed around a rectangle of sides L and H ? Hint: Find the maximum area of a triangle formed in the first quadrant by the x-axis, y-axis, and a tangent line to the Express the area of the circumscribed rectangle in terms of the angle θ (Figure 19). graph of y = (x + 1)−2 .

q H L

FIGURE 19 SOLUTION Position the L × H rectangle in the first quadrant of the x y-plane with its “northwest” corner at the origin. Let θ be the angle the base of the circumscribed rectangle makes with the positive x-axis, where 0 ≤ θ ≤ π2 . Then the area of the circumscribed rectangle is A = L H + 2 · 12 (H sin θ )(H cos θ ) + 2 · 12 (L sin θ )(L cos θ ) = L H + 12 (L 2 + H 2 ) sin 2θ , which has a maximum value of L H + 12 (L 2 + H 2 ) when θ = π4 because sin 2θ achieves its maximum when θ = π4 .

31. An 8-billion-bushel corn crop brings a price of $2.40/bushel. A commodity broker uses the following rule of thumb: Optimal Priceby Let r be thethen monthly rent increases per unit inby an10x apartment building with 100results units. in A maximum survey reveals that If the crop is reduced x percent, the price cents. Which crop size revenue all units canprice be rented when rHint: = $900 and that one unit becomes with each $10 increase in rent. Suppose that and what is the per bushel? Revenue is equal to price timesvacant crop size. the average monthly maintenance per occupied unit is $100/month. SOLUTION Let x denote the percentage reduction in crop size. Then the price for corn is 2.40 + 0.10x, the crop size is (a) Show that the number of units rented is n = 190 − r/10 for 900 ≤ r ≤ 1,900. 8(1 − 0.01x) and the revenue (in billions of dollars) is (b) Find a formula for the net cash intake (revenue minus maintenance) and determine the rent r that maximizes intake. R(x) = (2.4 + .1x)8(1 − .01x) = 8(−.001x 2 + .076x + 2.4), where 0 ≤ x ≤ 100. Solve R (x) = −.002x + .076 = 0 to obtain x = 38 percent. Since R(0) = 19.2, R(38) = 30.752, and R(100) = 0, revenue is maximized when x = 38. So we reduce the crop size to 8(1 − .38) = 4.96 billion bushels. The price would be $2.40 + .10(38) = 2.40 + 3.80 = $6.20. 33. Let P = (a, b) be a point in the first quadrant. Given n numbers x1 , . . . , x n , find the value of x minimizing the sum of the squares: (a) Find the slope of the line through P such that the triangle bounded by this line and the axes in the first quadrant has minimal area. (x − x 1 )2 + (x − x 2 )2 + · · · + (x − xn )2 First solve the problem for n = 2, 3 and then solve it for arbitrary n.


S E C T I O N 4.6

207

(b) Show that P is the midpoint of the hypotenuse of this triangle. SOLUTION Let P(a, b) be a point in the first quadrant (thus a, b > 0) and y − b = m(x − a), −∞ < m < 0, be a line through P that cuts the positive x- and y-axes. Then y = L(x) = m(x − a) + b.

b , 0 . Hence the area of the triangle (a) The line L(x) intersects the y-axis at H (0, b − am) and the x-axis at W a − m is 1 b 1 1 = ab − a 2 m − b2 m −1 . A(m) = (b − am) a − 2 m 2 2

Solve A (m) = 12 b2 m −2 − 12 a 2 = 0 for m < 0 to obtain m = − ab . Since A → ∞ as m → −∞ or m → 0−, we conclude that the minimal triangular area is obtained when m = − ab . (b) For m = −b/a, we have H (0, 2b) and W (2a, 0). The midpoint of the line segment connecting H and W is thus P(a, b). 35. Figure 20 shows a rectangular plot of size 100 × 200 feet. Pipe is to be laid from A to a point P on side BC and A truck gets 10 mpg (miles per gallon) traveling along an interstate highway at 50 mph, and this is reduced by from there to C. The cost of laying pipe through the lot is $30/ft (since it must be underground) and the cost along the 0.15 mpg for each mile per hour increase above 50 mph. side of the plot is $15/ft. (a) If the truck driver is paid $30/hour and diesel fuel costs P = $3/gal, which speed v between 50 and 70 mph (a) Let f (x) be the total cost, where x is the distance from P to B. Determine f (x), but note that f is discontinuous at will minimize the cost of a trip along the highway? Notice that the actual cost depends on the length of the trip but x = 0 (when x = 0, the cost of the entire pipe is $15/ft). the optimal speed does not. (b) What is the most economical way to lay the pipe? What if the cost along the sides is $24/ft? (b) Plot cost as a function of v (choose the length arbitrarily) and verify your answer to part (a). (c) Do you expect the optimal speed v to Bincrease or decrease P C if fuel costs go down to P = $2/gal? Plot the 200same − x axis and verify your conclusion. graphs of cost as a function of v for P = 2 and P =x 3 on the 100 A

200

FIGURE 20 SOLUTION

(a) Let x be the distance from P to B. If x > 0, then the length of the underground pipe is of the pipe along the side of the plot is 200 − x. The total cost is f (x) = 30 1002 + x 2 + 15(200 − x).

1002 + x 2 and the length

If x = 0, all of the pipe is along the side of the plot and f (0) = 15(200 + 100) = $4500. (b) To locate the critical points of f , solve 30x f (x) = − 15 = 0. 1002 + x 2

√ We find√x = ±100/ 3. Note that only the positive value is in the domain of the problem. Because f (0) = $4500, f (100/ 3) = $5598.08 and f (200) = $6708.20, the most economical way to lay the pipe is to place the pipe along the side of the plot. If the cost of laying the pipe along the side of the plot is $24 per foot, then f (x) = 30 1002 + x 2 + 24(200 − x) and 30x f (x) = − 24. 1002 + x 2 The only critical point in the domain of the problem is x = 400/3. Because f (0) = $7200, f (400/3) = $6600 and f (200) = $6708.20, the most economical way to lay the pipe is place the underground pipe from A to a point 133.3 feet to the right of B and continuing to C along the side of the plot. 37. In Example 6 in this section, find the x-coordinate of the point P where the light beam strikes the mirror if h 1 = 10, the=dimensions of a cylinder of volume 1 m3 of minimal cost if the top and bottom are made of material that h 2 = 5,Find and L 20. costs twice as much as the material for the side. SOLUTION Substitute h 1 = 10 feet, h 2 = 5 feet, and L = 20 feet into

L−x = x 2 + h 21 (L − x)2 + h 22 x

Solving for x gives x = 40 3 feet.

to obtain

x x 2 + 102

=

20 − x (20 − x)2 + 52

.

208

CHAPTER 4


In Exercises 38–40, a box (with no top) is to be constructed from a piece of cardboard of sides A and B by cutting out squares of length h from the corners and folding up the sides (Figure 21).

B h A

FIGURE 21

39. Which value of h maximizes the volume if A = B? Find the value of h that maximizes the volume of the box if A = 15 and B = 24. What are the dimensions of the SOLUTION When A = B, the volume of the box is resulting box? V (h) = hx y = h ( A − 2h)2 = 4h 3 − 4 Ah 2 + A2 h, A 2 2 where 0 ≤ h ≤ A 2 (allowing for degenerate boxes). Solve V (h) = 12h − 8 Ah + A = 0 for h to obtain h = 2 or A A A 2 A 3 h = 6 . Because V (0) = V ( 2 ) = 0 and V ( 6 ) = 27 A , volume is maximized when h = 6 .

41. The monthly output P of a light bulb factory is given by the formula P = 350L K , where L is the amount invested cardboard ABto=produce 144). Which that a box invested of heightinh equipment = 3 in. is constructed using 144 in.2 Ifofthe in laborSuppose and K the amount (in thousands of dollars). company(i.e., needs 10,000values units A and Bhow maximize the investment volume? be divided among labor and equipment to minimize the cost of production? The per month, should the cost of production is L + K . 200 . Accordingly, the cost of production is SOLUTION Since P = 10000 and P = 350L K , we have L = 7K C(K ) = L + K = K + 200 for K ≥ 0 to obtain K = 7K 2 minimum cost of production is achieved for K = L = Solve C (K ) = 1 −

200 . 7K

√

10 14. Since C(K ) → ∞ as K → 0+ and as K → ∞, the 7 √ 10 14 ≈ 5.345 or $5435 invested in both labor and equipment. 7

43. Janice can swim 3 mph and run 8 mph. She is standing at one bank of a river that is 300 ft wide and wants to reach Use calculus to show that among right trianglesaswith hypotenuse ofswim length 1, the isosceles triangle has a point located 200 ft downstream on the otherallside as quickly possible. She will diagonally across the river and maximum area. Can you see more directly why this must be true by reasoning from Figure 22? then jog along the river bank. Find the best route for Janice to take. Let lengths be in feet, times in seconds, and speeds in ft/s. Let x be the distance from the point directly opposite Janice on the other shore to the point where she starts to run after having swum. Then x 2 + 3002 is the 176 distance she swims and 200 − x is the distance she runs. Janice swims at 22 5 ft/s (i.e., 3 mph) and runs at 15 ft/s (i.e., 8 mph). The total time she travels is 375 200 − x 5 2 15 x 2 + 3002 f (x) = x + 90000 + + = − x, 0 ≤ x ≤ 200. 22/5 176/15 22 22 176 SOLUTION

Solve 5 x 15 =0 − 22 x 2 + 90000 176 √ √ 4875 750 to obtain x = 180 11 55. Since f (0) = 22 ≈ 221.59 and f (200) = 11 13 ≈ 245.83, weconclude that the minimum √ √ √ 375 375 480 2 2 amount of time f 180 11 55 = 44 55 + 22 ≈ 80.35 s occurs when Janice swims x + 300 = 11 55 ≈ √ 323.62 ft and runs 200 − x = 200 − 180 11 55 ≈ 78.64 ft. f (x) =

45. (a) Find the radius and height of a cylindrical can of total surface area A whose volume is as large as possible. Delivery Schedule gas station gallons total of gasoline (b) CanOptimal you design a cylinder with totalAsurface area sells A andQminimal volume?per year, which is delivered N times per year in equal shipments of Q/N gallons. The cost of each delivery is d dollars and the yearly storage costs are SOLUTION Let a closed cylindrical can be of radius r and height h. s QT , where T is the length shipments and s is a constant. Show that costs √ of time (a fraction of a year) between A .) Find the optimal number of deliveries are minimized for N = s Q/d. (Hint: Express T in terms of N 2 2 (a) Its total surface area is S = 2π r + 2π r h = A, whence h = − r. Its volume is thus V (r ) = π r 2 h =if 12QAr= − r million gal, d = $8,000, and s = 30 cents/gal-yr. Your answer2πshould be a whole number, so compare costs for the

A integer nearest the) = optimal A .NSolve 1 A −value. A (r 2 for r > 0 to obtain r = π r 3 ,two where 0 < values r ≤ of V 3 π r . Since V (0) = V ( 2π 2 2π ) = 0 and 6π √ A 6 A3/2 = V √ , 6π 18 π the maximum volume is achieved when

r=

A 6π

and

1 h= 3

6A . π


S E C T I O N 4.6

209

(b) For a can of total surface area A, there are cans of arbitrarily small volume since lim V (r ) = 0. r →0+

47. A billboard of height b is mounted on the side of a building with its bottom edge at a distance h from the street. At Find the area of the largest isosceles triangle that can be inscribed in a circle of radius r . what distance x should an observer stand from the wall to maximize the angle of observation θ (Figure 23)? (a) Find x using calculus. You may wish to use the addition formula cot(a − b) =

1 + cot a cot b . cot b − cot a

(b) Solve the problem again using geometry (without any calculation!). There is a unique circle passing through points B and C which is tangent to the street. Let R be the point of tangency. Show that θ is maximized at the point R. Hint: The two angles labeled ψ are, in fact, equal because they subtend equal arcs on the circle. Let A be the intersection of the circle with PC and show that ψ = θ + P B A > θ . (c) Prove that the two answers in (a) and (b) agree.

C b

P

A

h

q x

y

B y

q

P

R

FIGURE 23 SOLUTION

(a) From the leftmost diagram in Figure 23 and the addition formula for the cotangent function, we see that 1+ x x x 2 + h(b + h) cot θ = x b+hx h = , bx h − b+h where b and h are constant. Now, differentiate with respect to x and solve − csc2 θ

x 2 − h(b + h) dθ = =0 dx bx 2

to obtain x = bh + h 2 . Since this is the only critical point, and since θ → 0 as x → 0+ and θ → 0 as x → ∞, θ (x) reaches its maximum at x = bh + h 2 . (b) Following the directions, and mindful of the diagram in Figure 23, let C be the point at the top of the painting along the wall. For every radius r , there is a unique circle of radius r through B and C with center to the front of the billboard. Only one of these is tangent to the street (all the others are either too big or too small). Draw such a circle (call it D) , as shown in the right half of Figure 23, and let R be the point where the circle touches the street. Let P be any other such point along the street. We will prove that the angle of observation θ at P is less than the angle of observation ψ at R, thus proving that R is the point with maximum angle of observation along the floor. P, C, and B form a triangle. Let A be the topmost point of intersection of the triangle and circle D as shown. Because the two angles both subtend the arc BC, the measure of angle C AB is equal to the measure of angle C R B. By supplementarity, ψ + P AB = 180◦ . Because they form a triangle, θ + P AB + P B A = 180◦ . Thus, ψ = θ + P B A > θ . (c) To show that the two answers agree, let O be the center of the circle. One observes that if d is the distance from R to the wall, then O has coordinates (−d, b2 + h). This is because the height of the center is equidistant from points B and C and because the center must lie directly above R if the circle is tangent to the floor. Now we can solve for d. The radius of the circle is clearly b2 + h, by the distance formula:

2

O B = d2 +

2 2 b b +h−h = +h 2 2

This gives d2 = or d =

2 2 b b = bh + h 2 +h − 2 2

bh + h 2 as claimed.

49. Snell’s Law, derived in Exercise 48, explains why it is impossible to see above the water if the angle of vision θ2 is 24 represents the surface of a swimming pool. A light beam travels from point A located such thatSnell’s sin θ2 Law > v2 /vFigure 1. above the pool to point B located underneath the water. Let v1 be the velocity of light in air and let v2 be the velocity (a) Show that if this inequality holds, then there is no angle θ1 for which Snell’s Law holds. of light in water (it is a fact that v1 > v2 ). Prove Snell’s Law of Refraction according to which the path from A to B that takes the least time satisfies the relation sin θ

sin θ

210

CHAPTER 4


(b) What will you see if you look through the water with an angle of vision θ2 such that sin θ2 > v2 /v1 ? SOLUTION

v sin θ1 sin θ2 v v v (a) Suppose sin θ2 > 2 . By Snell’s Law, = , whence sin θ1 = 1 sin θ2 > 1 · 2 = 1, a contradiction, v1 v1 v2 v2 v2 v1 since sin θ ≤ 1 for all real values of θ . Accordingly, there is no real value θ1 for which Snell’s Law holds. (b) The underwater observer sees what appears to be a silvery mirror due to total internal reflection. 51. Vascular Branching A small blood vessel of radius r branches off at an angle θ from a larger vessel of radius R to blank margins oftowidth 6 in. onLaw, the top and bottom andto4 blood in. onflow the sides. Find the A poster of aarea ft2 has supply blood along path6 from A to B. According Poiseuille’s the total resistance is proportional to dimensions that maximize the printed area. a − b cot θ b csc θ T = + R4 r4 where a and b are as in Figure 25. Show that the total resistance is minimized when cos θ = (r/R)4 . B r b R

q

A a

FIGURE 25

SOLUTION

a − b cot θ b csc θ + . Set R4 r4 b csc2 θ b csc θ cot θ = 0. − T (θ ) = R4 r4

With a, b, r, R > 0 and R > r , let T (θ ) =

Then

b r 4 − R 4 cos θ R 4 r 4 sin2 θ

so that cos θ =

r 4 cos θ = . R

r 4 R

. Since

lim T (θ ) = ∞ and

θ →0+

= 0,

lim T (θ ) = ∞, the minimum value of T (θ ) occurs when

θ →π −

53. Let (a, b) be a fixed point in the first quadrant and let S(d) be the sum of the distances from (d, 0) to the points Find the minimum length of a beam that can clear a fence of height h and touch a wall located b ft behind the (0, 0), (a, b), and (a, −b). √ √ fence (Figure 26). (a) Find the value of √ d for which S(d) is minimal. The answer depends on whether b < 3a or b ≥ 3a. Hint: Show that d = 0 when b ≥ 3a. √ Let a = 1. Plot S(d) for b = 0.5, 3, 3 and describe the position of the minimum. (b) SOLUTION

(a) If d < 0, then the distance from (d, 0) to the other three points can all be reduced by increasing the value of d. Similarly, if d > a, then the distance from (d, 0) to the other three points can all be reduced by decreasing the value of d. It follows that the minimum of S(d) must occur for 0 ≤ d ≤ a. Restricting attention to this interval, we find

S(d) = d + 2 (d − a)2 + b2 . Solving 2(d − a) S (d) = 1 + =0 (d − a)2 + b2 √ √ √ yields the critical point d = √ a − b/ 3. If b < 3a, then d = a − b/ 3 > 0 and the minimum occurs at this value of d. On the other hand, if b ≥ 3a, then the minimum occurs at the endpoint d = 0. √ (b) Let a = 1. Plots of S(d) for b = 0.5,√ b = 3 and b = 3 are shown below. For b = 0.5, the results√of (a) indicate the minimum should occur for d = 1 − 0.5/ 3 ≈ 0.711, and this is confirmed in the plot. For both b = 3 and b = 3, the results of (a) indicate that the minimum should occur at d = 0, and both of these conclusions are confirmed in the plots.

S E C T I O N 4.6 y


211

y b = 0.5

2.1 2 1.9 1.8 1.7 1.6 1.5

x 0

0.2 0.4 0.6 0.8

b = √3

4.4 4.3 4.2 4.1 4

1

x 0

0.2 0.4 0.6 0.8

1

y b=3 6.8 6.6 6.4 x 0

0.2 0.4 0.6 0.8

1

55. In the of Exercise 54, show any f the minimal force required is proportional to 1/ 1 to + f 2. Thesetting minimum force required to that drivefor a wedge of angle α into a block (Figure 27) is proportional k SOLUTION Let F(α ) = , where k > 0 is a proportionality constant and 0 ≤ α ≤ π2 . Solve 1 sin α + f cos α F(α ) = sin α + f cos α k f sin α − cos α ) ( (α ) = = 0 is required, assuming f = 0.4. where f is a positive constant. Find theFangle α for which the least2 force (sin α + f cos α ) for 0 ≤ α ≤ π2 to obtain tan α = 1f . From the diagram below, we note that when tan α = 1f , 1 sin α = 1+ f2

f cos α = . 1+ f2

and

Therefore, at the critical point the force is k 2 √1 +√f 2 1+ f 2 1+ f

k = . 1+ f2

Since F(0) = 1f k > √ k 2 and F( π2 ) = k > √ k 2 , we conclude that the minimum force is proportional to 1+ f 1+ f 1/ 1 + f 2 .

√1 + f 2 1

α f

57. Find the maximum length of a pole that can be carried horizontally around a corner joining corridors of widths 8 ft The problem is to put a “roof” of side s on an attic room of height h and width b. Find the smallest length s for and 4 ft (Figure 29). which this is possible. See Figure 28. 4

8

FIGURE 29 SOLUTION In order to find the length of the longest pole that can be carried around the corridor, we have to find the shortest length from the left wall to the top wall touching the corner of the inside wall. Any pole that does not fit in this shortest space cannot be carried around the corner, so an exact fit represents the longest possible pole. Let θ be the angle between the pole and a horizontal line to the right. Let c1 be the length of pole in the 8 ft corridor and let c2 be the length of pole in the 4 foot corridor. By the definitions of sine and cosine,

4 = sin θ c2

and

8 = cos θ , c1

212

CHAPTER 4


so that c1 = cos8 θ , c2 = sin4 θ . What must be minimized is the total length, given by f (θ ) =

8 4 + . cos θ sin θ

Setting f (θ ) = 0 yields 4 cos θ 8 sin θ − =0 cos2 θ sin2 θ 8 sin θ 4 cos θ = cos2 θ sin2 θ 8 sin3 θ = 4 cos3 θ As θ < π2 (the pole is being turned around a corner, after all), we can divide both sides by cos3 θ , getting tan3 θ = 12 .

This implies that tan θ = 12

1/3

(tan θ > 0 as the angle is acute).

1/3 Since f (θ ) → ∞ as θ → 0+ and as θ → π2 −, we can tell that the minimum is attained at θ0 where tan θ0 = 12 . Because tan θ0 =

opposite 1 = 1/3 , adjacent 2

we draw a triangle with opposite side 1 and adjacent side 21/3 .

c 1 q 21/3

By Pythagoras, c =

1 + 22/3 , so 1 sin θ0 = 1 + 22/3

and

21/3 cos θ0 = . 1 + 22/3

From this, we get f (θ0 ) =

8 4 8 + = 1/3 1 + 22/3 + 4 1 + 22/3 = 4 1 + 22/3 22/3 + 1 . cos θ0 sin θ0 2

59. Find the isosceles triangle of smallest area that circumscribes a circle of radius 1 (from Thomas Simpson’s The Redo Exercise 57 for corridors of arbitrary widths a and b. Doctrine and Application of Fluxions, a calculus text that appeared in 1750). See Figure 30.

q

1

FIGURE 30 SOLUTION From the diagram, we see that the height h and base b of the triangle are h = 1 + csc θ and b = 2h tan θ = 2(1 + csc θ ) tan θ . Thus, the area of the triangle is

A(θ ) =

1 hb = (1 + csc θ )2 tan θ , 2

where 0 < θ < π . We now set the derivative equal to zero: A (θ ) = (1 + csc θ )(−2 csc θ + sec2 θ (1 + csc θ )) = 0. The first factor gives θ = 3π /2 which is not in the domain of the problem. To find the roots of the second factor, multiply through by cos2 θ sin θ to obtain −2 cos2 θ + sin θ + 1 = 0,

S E C T I O N 4.6


213

or 2 sin2 θ + sin θ − 1 = 0. This is a quadratic equation in sin θ with roots sin θ = −1 and sin θ = 1/2. Only the second solution is relevant and gives us θ = π /6. Since A(θ ) → ∞ as θ → 0+ and as θ → π −, we see that the minimum area occurs when the triangle is an equilateral triangle. basketballand player stands d feet from the basket. Let h and α be as in Figure 31. Using physics, one can show Further AInsights Challenges

that if the player releases the ball at an angle θ , then the initial velocity required to make the ball go through the 61. Bird Migration Power P is the rate at which energy E is consumed per unit time. Ornithologists have found that basket satisfies the power consumed by a certain pigeon flying at velocity v m/s is described well by the function P(v) = 17v −1 + 4 16denergy as body fat. 2 10 10−3 v 3 J/s. Assume that the pigeon can store 5v× usable = J of . 2 θ (tan θ − tan α ) cos (a) Find the velocity vpmin that minimizes power consumption. (b) Show that a why pigeon at velocity v and usingfor allαof D, so c = D/(2 3) = 3/3 (0) = L + D = 6, f (L) =

∈ [0, L]. f √ √ √ 1 2 2 2 L + D /4 = 2 17 ≈ 8.24621, and f (c) = 4 − ( 3/3) + 2 3 + 1 = 4 + 3 ≈ 5.73204. Therefore, the √ total length is minimized where √ x = c = 3/3. √ √ • D = 8, L = 2; 2 3L = 4 3 < D, so c does not lie in the interval [0, L]. f (0) = 2 + 2 64/4 = 10, and √ √ √ f (L) = 0 + 2 4 + 64/4 = 2 20 = 4 5 ≈ 8.94427. Therefore, the total length is minimized were x = L, or where P = C. A jewelry designer plans to incorporate a component made of gold in the shape of a frustum of a cone of height 1 cm and fixed lower radius r (Figure 35). The upper radius x can take on any value between 0 and r . Note that

216

CHAPTER 4


4.7 Newton’s Method Preliminary Questions 1. How many iterations of Newton’s Method are required to compute a root if f (x) is a linear function? SOLUTION Remember that Newton’s Method uses the linear approximation of a function to estimate the location of a root. If the original function is linear, then only one iteration of Newton’s Method will be required to compute the root.

2. What happens in Newton’s Method if your initial guess happens to be a zero of f ? SOLUTION

If x0 happens to be a zero of f, then f (x ) x1 = x0 − 0 = x0 − 0 = x0 ; f (x 0 )

in other words, every term in the Newton’s Method sequence will remain x0 . 3. What happens in Newton’s Method if your initial guess happens to be a local min or max of f ? SOLUTION Assuming that the function is differentiable, then the derivative is zero at a local maximum or a local minimum. If Newton’s Method is started with an initial guess such that f (x 0 ) = 0, then Newton’s Method will fail in the sense that x 1 will not be defined. That is, the tangent line will be parallel to the x-axis and will never intersect it.

4. Is the following a reasonable description of Newton’s Method: “A root of the equation of the tangent line to f (x) is used as an approximation to a root of f (x) itself”? Explain. SOLUTION Yes, that is a reasonable description. The iteration formula for Newton’s Method was derived by solving the equation of the tangent line to y = f (x) at x 0 for its x-intercept.

Exercises In Exercises 1–4, use Newton’s Method with the given function and initial value x 0 to calculate x1 , x 2 , x3 . 1. f (x) = x 2 − 2, SOLUTION

x0 = 1

Let f (x) = x 2 − 2 and define x n+1 = x n −

f (xn ) x n2 − 2 − . = x n f (x n ) 2xn

With x 0 = 1, we compute n

1

2

3

xn

1.5

1.416666667

1.414215686

3. f (x) = x 3 −25, x 0 = 1.6 f (x) = x − 7, x 0 = 2.5 SOLUTION Let f (x) = x 3 − 5 and define x n+1 = x n −

f (xn ) x3 − 5 = xn − n 2 . f (x n ) 3x n

With x 0 = 1.6 we compute n

1

2

3

xn

1.717708333

1.710010702

1.709975947

3 5. UsefFigure 6 toxchoose (x) = cos − x, an x0 initial = 0.8guess x0 to the unique real root of x + 2x + 5 = 0. Then compute the first three iterates of Newton’s Method. y

−2

−1

x 1

2

FIGURE 6 Graph of y = x 3 + 2x + 5.


Newton’s Method

217

Let f (x) = x 3 + 2x + 5 and define f (x n ) x 3 + 2x n + 5 . = xn − n 2 f (x n ) 3xn + 2

x n+1 = x n −

We take x 0 = −1.4, based on the figure, and then calculate n

1

2

3

xn

−1.330964467

−1.328272820

−1.328268856

In Exercises 7–10, useMethod Newton’s to approximate thecos root decimal[0, places compare the value π ] toand Use Newton’s to Method find a solution of sin x = 2xtointhree the interval three decimalwith places. Then 2 obtained from a calculator. guess the exact solution and compare with your approximation. √ 7. 10 SOLUTION

Let f (x) = x 2 − 10, and let x 0 = 3. Newton’s Method yields: n

1

2

3

xn

3.16667

3.162280702

3.16227766

A calculator yields 3.16227766. 9. 51/3 1/4 7 SOLUTION

Let f (x) = x 3 − 5, and let x0 = 2. Here are approximations to the root of f (x), which is 51/3 . n

1

2

3

4

xn

1.75

1.710884354

1.709976429

1.709975947

A calculator yields 1.709975947. 11. Use Newton’s Method to approximate the largest positive root of f (x) = x 4 − 6x 2 + x + 5 to within an error of at −1/2 2 −4 most 10 . Refer to Figure 5. SOLUTION

Figure 5 from the text suggests the largest positive root of f (x) = x 4 − 6x 2 + x + 5 is near 2. So let

f (x) = x 4 − 6x 2 + x + 5 and take x 0 = 2. n

1

2

3

4

xn

2.111111111

2.093568458

2.093064768

2.093064358

The largest positive root of x 4 − 6x 2 + x + 5 is approximately 2.093064358. Use a graphing calculator to3 choose an initial guess for the unique positive root of x 4 + x 2 − 2x − 1 = 0. 13. + 1 and use Newton’s Method to approximate the largest positive root to Sketch graph of f of (x)Newton’s = x − 4x Calculate the firstthe three iterates Method. −3 . within an error of at most 10 SOLUTION Let f (x) = x 4 + x 2 − 2x − 1. The graph of f (x) shown below suggests taking x 0 = 1. Starting from x 0 = 1, the first three iterates of Newton’s Method are: n

1

2

3

xn

1.25

1.189379699

1.184171279

y 30 20 10 −1

x 1

2

3

sin θ the smallest positive solution of π . Use = 0.9 to three decimal places. π Use graphing calculator The Estimate first positive solution of sin x = 0 is x = Newton’s Method to calculate to afour decimal places. to θ choose the initial guess.

15.

218

CHAPTER 4


Let f (θ ) =

sin θ − .9. θ

We use the following plot to obtain an initial guess: y 1

x

0.5

−0.1 −0.2 −0.3 −0.4

1.5

θ = 0.8 seems a good first guess for the location of a zero of the function. A series of approximations based on θ0 = 0.8 follows: n

1

2

3

θn

0.7867796879

0.7866830771

0.7866830718

A three digit approximation for the point where sinθ θ = .9 is 0.787. Plugging sin(0.787)/0.787 into a calculator yields 0.900 to three decimal places. 17. Let x 1 , x 2 be the estimates to a root obtained by applying Newton’s Method with x0 = 1 to the function graphed in In 1535, the mathematician Antonio Fior challenged his rival Niccolo Tartaglia to solve this problem: A tree Figure 7. Estimate the numerical values of x1 and x2 , and draw the tangent lines used to obtain them. stands 12 braccia high; it is broken into two parts at such a point that the height of the part left standing is the cube root of the length of the part cut away. What is ythe height of the part left standing? Show that this is equivalent to solving x 3 + x = 12 and find the height to three decimal places. Tartaglia, who had discovered the secrets of cubic equations, was able to determine the exact answer: x −1 1 3

2 √ 3 3 3 2,919 + 54 − 2,919 − 54 9 x= FIGURE 7 SOLUTION The graph with tangent lines drawn on it appears below. The tangent line to the curve at (x 0 , f (x 0 )) has an x-intercept at approximately x 1 = 3.0. The tangent line to the curve at (x 1 , f (x 1 )) has an x-intercept at approximately x 2 = 2.2. y

x

−1

1

2

3

19. Find the x-coordinate to two decimal places of the first point in the region x > 0 where y = x intersects y = tan x Find the coordinates to two decimal places of the point P in Figure 8 where the tangent line to y = cos x passes (draw a graph). through the origin. SOLUTION Here is a plot of tan x and x on the same axes: y 5 x 1

2

3

4

−5

The first intersection with x > 0 lies on the second “branch” of y = tan x, between x = 54π and x = 32π . Let f (x) = tan x − x. The graph suggests an initial guess x 0 = 54π , from which we get the following table: n

1

2

3

4

xn

6.85398

21.921

4480.8

7456.27

This is clearly leading nowhere, so we need to try a better initial guess. Note: This happens with Newton’s Method—it is sometimes difficult to choose an initial guess. We try the point directly between 54π and 32π , x 0 = 118π :

Newton’s Method

S E C T I O N 4.7

n

1

2

3

4

5

6

7

xn

4.64662

4.60091

4.54662

4.50658

4.49422

4.49341

4.49341

219

The first point where y = x and y = tan x cross is at approximately x = 4.49341, which is approximately 1.4303π . Newton’s Method is often used to determine interest rates in financial calculations. In Exercises 20–22, r denotes a yearly interest rate expressed as a decimal (rather than as a percent). 21. If you borrow L dollars for N years at a yearly interest rate r , your monthly payment of P dollars is calculated using If P dollars are deposited every month in an account earning interest at the yearly rate r , then the value S of the the equation account after N years is 1 − b−12N r L=P where b = 1 + .r b12N +1 − b b−1 12 . S=P where b = 1 + b−1 12 (a) What is the monthly payment if L = $5,000, N = 3, and r = 0.08 (8%)? decideda to deposit 100 dollars month. (b) You have are offered loan of L P==$5,000 to beper paid back over 3 years with monthly payments of P = $200. Use Newton’s Method tovalue compute findif the interestrate rateisrrof=this (a) What is the afterb5and years the implied yearly interest 0.07loan. (i.e.,Hint: 7%)?Show that (L/P)b12N +1 − (1 + 12N L/P)b + 1that = 0.to save $10,000 after 5 years, you must earn interest at a rate r determined by the equation b61 − (b) Show SOLUTION 101b +

100 = 0. Use Newton’s Method to solve for b (note that b = 1 is a root, but you want the root satisfying > (1 1).+ Then find r=. 1.00667 (a) b = .08/12) b−1 1.00667 − 1 P=L = 5000 ≈ $156.69 1 − b−12N 1 − 1.00667−36

(b) Starting from

L=P

1 − b−12N b−1

,

divide by P, multiply by b − 1, multiply by b12N and collect like terms to arrive at (L/P)b12N +1 − (1 + L/P)b12N + 1 = 0. Since L/P = 5000/200 = 25, we must solve 25b37 − 26b36 + 1 = 0. Newton’s Method gives b ≈ 1.02121 and r = 12(b − 1) = 12(.02121) ≈ .25452 So the interest rate is around 25.45%. 23. Kepler’s Problem Although planetary motion is beautifully described by Kepler’s three laws (see Section 14.6), If you deposit P dollars in a retirement fund every year for N years with the intention of then withdrawing there is no simple formula for the position of a planet P along its elliptical orbit as a function of time. Kepler developed a Q dollars per year for M years, you must earn interest at a rate r satisfying P(b N − 1) = Q(1 − b−M ), where method for locating P at time t by drawing the auxiliary dashed circle in Figure 9 and introducing the angle θ (note that b = 1 + r . Assume that $2,000 is deposited each year for 30 years and the goal is to withdraw $8,000 per year for P determines θ , which is the central angle of the point B on the circle). Let a = O A and e = O S/O A (the eccentricity 20 years. Use Newton’s Method to compute b and then find r . of the orbit). (a) Show that sector BSA has area (a 2 /2)(θ − e sin θ ). (b) It follows from Kepler’s Second Law that the area of sector BSA is proportional to the time t elapsed since the planet passed point A. More precisely, since the circle has area π a 2 , BSA has area (π a 2 )(t/T ), where T is the period of the orbit. Deduce that 2π t = θ − e sin θ . T (c) The eccentricity of Mercury’s orbit is approximately e = 0.2. Use Newton’s Method to find θ after a quarter of Mercury’s year has elapsed (t = T /4). Convert θ to degrees. Has Mercury covered more than a quarter of its orbit at t = T /4? Auxiliary circle

B P

q O

Elliptical orbit

FIGURE 9

S Sun

A

220

CHAPTER 4


(a) The sector S AB is the slice O AB with the triangle O P S removed. O AB is a central sector with arc θ and radius 2 O A = a, and therefore has area a 2θ . O P S is a triangle with height a sin θ and base length O S = ea. Hence, the area of the sector is 1 a2 a2 θ − ea 2 sin θ = (θ − e sin θ ). 2 2 2 (b) Since Kepler’s second law indicates that the area of the sector is proportional to the time t since the planet passed point A, we get

π a 2 (t/T ) = a 2 /2 (θ − e sin θ ) 2π

t = θ − e sin θ . T

(c) If t = T /4, the last equation in (b) gives:

π = θ − e sin θ = θ − .2 sin θ . 2 Let f (θ ) = θ − .2 sin θ − π2 . We will use Newton’s Method to find the point where f (θ ) = 0. Since a quarter of the year on Mercury has passed, a good first estimate θ0 would be π2 . n

1

2

3

4

xn

1.7708

1.76696

1.76696

1.76696

From the point of view of the Sun, Mercury has traversed an angle of approximately 1.76696 radians = 101.24◦ . Mercury has therefore traveled more than one fourth of the way around (from the point of view of central angle) during this time. 25. What happens when you apply Newton’s Method to the equation x 3 − 20x = 1/3 0 with the unlucky initial guess x 0 = 2?What happens when you apply Newton’s Method to find a zero of f (x) = x ? Note that x = 0 is the only zero. SOLUTION Let f (x) = x 3 − 20x. Define x n+1 = x n −

f (x n ) x n3 − 20x n . − = x n f (x n ) 3x n2 − 20

Take x 0 = 2. Then the sequence of iterates is −2, 2, −2, 2, . . . , which diverges by oscillation.

Further Insights and Challenges 27. The roots of f (x) = 13 x 3 − 4x + 1 to three decimal places are −3.583, 0.251, and 3.332 (Figure 10). Determine Let c be a positive number and let f (x) = x −1 − c. the root to which Newton’s Method converges for the initial choices x0 = 1.85, 1.7, and 1.55. The answer shows that a 2x − on cx 2the . Thus, Newton’s MethodMethod. provides a way of computing reciprocals Showinthat x −have ( f (x)/ f (x)) = effect small(a) change x 0 can a significant outcome of Newton’s without performing division. y with c = 10.324 and the two initial guesses x = 0.1 and (b) Calculate the first three iterates of Newton’s Method 0 x0 = 0.5. 4 (c) Explain graphically why x0 = 0.5 does not yield a sequence of approximations to the reciprocal 1/10.324. 0.2513 3.332 −3.583

x

−4

FIGURE 10 Graph of f (x) = 13 x 3 − 4x + 1. SOLUTION

Let f (x) = 13 x 3 − 4x + 1, and define xn+1 = x n −

1 x 3 − 4x + 1 f (x n ) n 3 n − = x . n 2 f (xn ) xn − 4

• Taking x 0 = 1.85, we have

n

1

2

3

4

5

6

7

xn

−5.58

−4.31

−3.73

−3.59

−3.58294362

−3.582918671

−3.58291867


Antiderivatives

S E C T I O N 4.8

n

1

2

3

4

5

6

7

8

9

xn

−2.05

−33.40

−22.35

−15.02

−10.20

−7.08

−5.15

−4.09

−3.66

n

10

11

12

13

xn

−3.585312288

−3.582920989

−3.58291867

−3.58291867

221


n

1

2

3

4

5

6

xn

−0.928

0.488

0.245

0.251320515

0.251322863

0.251322863

In Exercises 28–29, consider a metal rod of length L inches that is fastened at both ends. If you cut the rod and weld on an additional m inches of rod, leaving the ends fixed, the rod will bow up into a circular arc of radius R (unknown), as indicated in Figure 11. 29. Let L = 3 and m = 1. Apply Newton’s Method to Eq. (2) to estimate θ and use this to estimate h. Let h be the maximum vertical displacement of the rod. SOLUTION andθmand = 1. We want the solution of: conclude that (a) ShowWe thatletL L==2R3 sin h=

sin L(1θ− cos θL) = . θ 2 sin θL + m

L sin θ − =0 θ L +m 3 sin θ − = 0. θ 4 sin θ L = . θ L +m

(b) Show L + m = 2R θ and then prove Let f (θ ) = sinθ θ − 34 . y 0.2 0.1

x 0.5

−0.2 −0.2

1

1.5

The figure above suggests that θ0 = 1.5 would be a good initial guess. The Newton’s Method approximations for the solution follow: n

1

2

3

4

θn

1.2854388

1.2757223

1.2756981

1.2756981

L is approximately 1.2757. Hence The angle where sinθ θ = L+m

h=L

Quadratic Convergence to Square Roots

1 − cos θ ≈ 1.11181. 2 sin θ √ Let f (x) = x 2 − c and let en = x n − c be the error in x n .

1

Show that xn+1 = 2 (x n + c/xn ) and that en+1 = en2 /2xn . 4.8 (a)Antiderivatives √ √

(b) Show that if x0 > c, then xn > c for all n. Explain this graphically. √ √ (c) Assuming that x0 > c, show that en+1 ≤ en2 /2 c. Preliminary Questions 1. Find an antiderivative of the function f (x) = 0. SOLUTION

Since the derivative of any constant is zero, any constant function is an antiderivative for the function

f (x) = 0. ! 2. What is the difference, if any, between finding the general antiderivative of a function f (x) and evaluating f (x) d x? SOLUTION

No difference. The indefinite integral is the symbol for denoting the general antiderivative.

3. Jacques happens to know that f (x) and g(x) have the same derivative, and he would like to know if f (x) = g(x). Does Jacques have sufficient information to answer his question?

222

CHAPTER 4

A P P L I C AT I O N S O F T H E D E R I VATI V E SOLUTION No. Knowing that the two functions have the same derivative is only good enough to tell Jacques that the functions may differ by at most an additive constant. To determine whether the functions are equal for all x, Jacques needs to know the value of each function for a single value of x. If the two functions produce the same output value for a single input value, they must take the same value for all input values.

4. Write any two antiderivatives of cos x. Which initial conditions do they satisfy at x = 0? SOLUTION

Antiderivatives for cos x all have the form sin x + C for some constant C. At x = 0, sin x + C has the

value C. 5. (a) (b) (c)

Suppose that F (x) = f (x) and G (x) = g(x). Are the following statements true or false? Explain. If f = g, then F = G. If F and G differ by a constant, then f = g. If f and g differ by a constant, then F = G.

SOLUTION

(a) False. Even if f (x) = g(x), the antiderivatives F and G may differ by an additive constant. (b) True. This follows from the fact that the derivative of any constant is 0. (c) False. If the functions f and g are different, then the antiderivatives F and G differ by a linear function: F(x) − G(x) = ax + b for some constants a and b. 6. Determine if y = x 2 is a solution to the differential equation with initial condition dy = 2x, dx

y(0) = 1

d SOLUTION Although d x x 2 = 2x, x 2 takes the value 0 when x = 0, so y = x 2 is not a solution of the indicated initial value problem.

Exercises In Exercises 1–6, find the general antiderivative of f (x) and check your answer by differentiating. 1. f (x) = 12x SOLUTION

"

" 12x d x = 12

x d x = 12 ·

1 2 x + C = 6x 2 + C. 2

As a check, we have d (6x 2 + C) = 12x dx as needed. 3. f (x) = x 2 +23x + 2 f (x) = x

SOLUTION

"

" (x 2 + 3x + 2) d x = =

" x 2d x + 3

" x1 dx + 2

x0 dx =

1 3 1 1 x + 3 · x2 + 2 · x1 + C 3 2 1

1 3 3 2 x + x + 2x + C. 3 2

As a check, we have d dx

1 3 3 2 x + x + 2x + C 3 2

= x 2 + 3x + 2

as needed. 5. f (x) = 8x −42 f (x) = x + 1

SOLUTION

"

8x −4 d x = 8

"

x −4 d x = 8 ·

As a check, we have d dx as needed.

1 −3 8 x + C = − x −3 + C. −3 3

8 − x −3 + C 3

= 8x −4

S E C T I O N 4.8

Antiderivatives

223

In Exercises f (x)7–10, = cosmatch x + 3 the sin xfunction with its antiderivative (a)–(d). (a) F(x) = cos(1 − x) (b) F(x) = − cos x (c) F(x) = − 12 cos(x 2 ) (d) F(x) = sin x − x cos x 7. f (x) = sin x SOLUTION

An antiderivative of sin x is − cos x, which is (b). As a check, we have ddx (− cos x) = − (− sin x) = sin x.

9. f (x) = sin(1 − x) f (x) = x sin(x 2 ) An antiderivative of sin (1 − x) is cos (1 − x) or (a). As a check, we have ddx cos(1 − x) = − sin(1 − x) · (−1) = sin(1 − x). SOLUTION

In Exercises the indefinite integral. f (x)11–36, = x sinevaluate x " 11. (x + 1) d x " 1 (x + 1) d x = x 2 + x + C. SOLUTION 2 " "5 13. (t + 3t + 2) dt (9 − 5x) d x " 1 3 SOLUTION (t 5 + 3t + 2) dt = t 6 + t 2 + 2t + C. 6 2 " "−9/5 15. t dt 8s −4 ds " 5 SOLUTION t −9/5 dt = − t −4/5 + C. 4 " " 17. 2 dx 3 (5x − x −2 − x 3/5 ) d x " 2 d x = 2x + C. SOLUTION " " 19. (5t −19) dt √ dx x" 5 (5t − 9) dt = t 2 − 9t + C. SOLUTION 2 " "−2 21. x d3x (x + 4x −2 ) d x " 1 −1 1 SOLUTION x +C =− +C x −2 d x = −1 x " " 23. (x +√3)−2 d x x dx " 1 1 SOLUTION (x + 3)−1 + C = − + C. (x + 3)−2 d x = −1 x +3 " "3 dz 25. z 5 (4t − 9)−3 dt SOLUTION

" "

3 dz = z5

"

1 3z −5 dz = 3 − z −4 4

3 + C = − z −4 + C. 4

√ "

x(x3− 1) d x dx x 3/2 √ SOLUTION To perform this indefinite integral, we need to distribute x = x 1/2 over (x − 1): √ x(x − 1) = x 1/2 (x − 1) = x 3/2 − x 1/2 .

27.

Written in this form, we can take the indefinite integral: "

" √ 1 1 2 2 x 3/2 − x 1/2 d x = x(x − 1) d x = x 5/2 − x 3/2 + C = x 5/2 − x 3/2 + C. 5/2 3/2 5 3

224

CHAPTER 4


" 29.

t"− 7 √ dt −1 (x t + x )(3x 2 − 5x) d x

SOLUTION

To proceed, we first distribute √1 = t −1/2 over (t − 7): t

t −7 √ = (t − 7)t −1/2 = t 1/2 − 7t −1/2 . t From this, we get: " " 1 1/2 2 t −7 1 3/2 1/2 −1/2 t t − 7t ) dt = −7 + C = t 3/2 − 14t 1/2 + C. √ dt = (t 3/2 1/2 3 t "

" sin2x − 3 cos x) d x (4 x + 2x − 3 dx " x4 (4 sin x − 3 cos x) d x = −4 cos x − 3 sin x + C. SOLUTION " " 33. cos(6t + 4) dt sin 9x d x SOLUTION Using the integral formula for cos(kt + b), we get: " 1 cos(6t + 4) dt = sin(6t + 4) + C. 6

31.

"

" cos(3 − 4t) dt (4θ + cos 8θ ) d θ SOLUTION From the integral formula for cos(kt + b) with k = −4, b = 3, we get: " 1 1 cos(3 − 4t) dt = sin(3 − 4t) + C = − sin(3 − 4t) + C. −4 4

35.

37. In Figure " 2, which of (A) or (B) is the graph of an antiderivative of f (x)? 18 sin(3z + 8) dz y y y x x x f(x)

(A)

(B)

FIGURE 2

Let F(x) be an antiderivative of f (x). By definition, this means F (x) = f (x). In other words, f (x) provides information as to the increasing/decreasing behavior of F(x). Since, moving left to right, f (x) transitions from − to + to − to + to − to +, it follows that F(x) must transition from decreasing to increasing to decreasing to increasing to decreasing to increasing. This describes the graph in (A)! SOLUTION

39. Use the formulas for the derivatives of f (x) = tan x and f (x) = sec x to evaluate the integrals. " In Figure 3, which of (A), (B), (C) is not the graph of an antiderivative " of f (x)? Explain. (a) sec2 (3x) d x (b) sec(x + 3) tan(x + 3) d x Recall that ddx (tan x) = sec2 x and ddx (sec x) = sec x tan x. ! (a) Accordingly, we have sec2 (3x) d x = 13 tan(3x) + C. ! (b) Moreover, we see that sec(x + 3) tan(x + 3) d x = sec(x + 3) + C. SOLUTION

In Exercises 41–52, solvefor thethe differential equation initial condition. Use the formulas derivatives of f (x)with = cot x and f (x) = csc x to evaluate the integrals. " " d y 2 (a) csc x d x (b) csc x cot x d x = cos 2x, y(0) = 3 41. dx SOLUTION

Since dd xy = cos 2x, we have

" y=

cos 2x d x =

1 sin 2x + C. 2

Thus 3 = y(0) = so that C = 3. Therefore, y = 12 sin 2x + 3.

1 sin (2 · 0) + C = C, 2

S E C T I O N 4.8

dy =5 d=y x, y(0) dx = x 3 , y(0) = 2 dx dy SOLUTION Since d x = x, we have

Antiderivatives

225

43.

" y=

x dx =

1 2 x + C. 2

Thus 5 = y(0) = 0 + C, so that C = 5. Therefore, y = 12 x 2 + 5. dy 45. d=y 5 − 2t 2 , y(1) = 2 dt = 0, y(3) = 5 dt dy SOLUTION Since dt = 5 − 2t 2 , we have " 2 y = (5 − 2t 2 ) dt = 5t − t 3 + C. 3 Thus, 2 2 = y(1) = 5(1) − (13 ) + C, 3 so that C = −3 + 23 = − 73 . Therefore y = 5t − 23 t 3 − 73 . dy d=y 4t + 9,3 y(0)2 = 1 dt = 8x + 3x − 3, y(1) = 1 dx dy SOLUTION Since dt = 4t + 9, we have

47.

" (4t + 9) dt = 2t 2 + 9t + C.

y=

Thus 1 = y(0) = 0 + C, so that C = 1. Therefore, y = 2t 2 + 9t + 1.

π dy =1 49. x, y d=y sin √ dx 2 =1 = t, y(1) dt dy SOLUTION Since d x = sin x, we have " y = sin x d x = − cos x + C. Thus 1=y

π 2

= 0 + C,

so that C = 1. Therefore, y = 1 − cos x. dy π 3 d=y cos 5x, y(π ) = dx =4 = sin 2z, y dz 4 dy SOLUTION Since d x = cos 5x, we have

51.

" y=

cos 5x d x =

1 sin 5x + C. 5

Thus 3 = y(π ) = 0 + C, so that C = 3. Therefore, y = 3 + 15 sin 5x.

πf and then find f . In Exercises first find d y 53–58, = sec2 3x, y =2 d x = x, f (0) = 1,4 f (0) = 0 53. f (x) Let g(x) = f (x). g (x) = x and g(0) = 1, so g(x) = 12 x 2 + C with g(0) = 1 = C. Hence f (x) = g(x) = 12 x 2 + 1. f (x) = 12 x 2 + 1 and f (0) = 0, so f (x) = 12 ( 13 x 3 ) + x + C = 16 x 3 + x + C, and f (0) = 0 = C. Therefore f (x) = 16 x 3 + x. SOLUTION

55. f (x) = x 3 −32x + 1, f (1) = 0, f (1) = 4 f (x) = x − 2x + 1, f (0) = 1, f (0) = 0

226

CHAPTER 4


Let g(x) = f (x). The problem statement gives us g (x) = x 3 − 2x + 1, g(0) = 0. From g (x), we get

g(x) = 14 x 4 − x 2 + x + C, and from g(1) = 0, we get 0 = 14 − 1 + 1 + C, so that C = − 14 . This gives f (x) = g(x) = 1 x 4 − x 2 + x − 1 . From f (x), we get f (x) = 1 ( 1 x 5 ) − 1 x 3 + 1 x 2 − 1 x + C = 1 x 5 − 1 x 3 + 1 x 2 − 1 x + C. 4 4 4 5 3 2 4 20 3 2 4

From f (1) = 4, we get

1 1 1 1 − + − + C = 4, 20 3 2 4 so that C = 121 30 . Hence, 1 5 1 3 1 2 1 121 x − x + x − x+ . 20 3 2 4 30

f (x) =

π = 1, f π = 6 = cos θ , f 57. f (fθ)(t) −3/2 =t , f 2(4) = 1, f (4) 2 =4 SOLUTION

Let g(θ ) = f (θ ). The problem statement gives g (θ ) = cos θ ,

g

π 2

= 1.

From g (θ ) we get g(θ ) = sin θ + C. From g( π2 ) = 1 we get 1 + C = 1, so C = 0. Hence f (θ ) = g(θ ) = sin θ . From f (θ ) we get f (θ ) = − cos θ + C. From f ( π2 ) = 6 we get C = 6, so f (θ ) = − cos θ + 6. 59. Show that f (x) = tan2 x and g(x) = sec2 x have the same derivative. What can you conclude about the relation f (t) = t − cos t, f (0) = 2, f (0) = −2 between f and g? Verify this conclusion directly. SOLUTION

Let f (x) = tan2 x and g(x) = sec2 x. Then f (x) = 2 tan x sec2 x and g (x) = 2 sec x · sec x tan x =

2 tan x sec2 x; hence f (x) = g (x). Accordingly, f (x) and g(x) must differ by a constant; i.e., f (x) − g(x) = tan2 x − sec2 x = C for some constant C. To see that this is true directly, divide the identity sin2 x + cos2 x = 1 by cos2 x. This yields tan2 x + 1 = sec2 x, so that tan2 x − sec2 x = −1. 61. A particle located at the origin at t = 0 begins along the x-axis with velocity v(t) = 12 t 2 − t ft/s. Let s(t) 1 cos 2 x moving =− 2x + C for some constant C. Find C bys(t). setting x = 0. Show, by computing that sin 2 initial be its position at time t. Statederivatives, the differential equation with condition satisfied by s(t) and find SOLUTION

Given that v(t) = ds dt and that the particle starts at the origin, the differential equation is 1 ds = t 2 − t, dt 2

s(0) = 0.

Accordingly, we have " s(t) =

1 2 t −t 2

dt =

1 2

"

" t 2 dt −

t dt =

1 1 3 1 2 1 1 · t − t + C = t 3 − t 2 + C. 2 3 2 6 2

Thus 0 = s(0) = 16 (0)3 − 12 (0)2 + C, so that C = 0. Therefore, the position is s(t) = 16 t 3 − 12 t 2 . 2 63. A particle along velocity v(t) = 25t be the position at time t. Repeat moves Exercise 61, the but x-axis replacewith the initial condition s(0)−=t 0 ft/s. with Let s(2)s(t) = 3. (a) Find s(t), assuming that the particle is located at x = 5 at time t = 0. (b) Find s(t), assuming that the particle is located at x = 5 at time t = 2. 2 The differential equation is v(t) = ds dt = 25t − t . (a) Suppose s(0) = 5. Then " " " s(t) = (25t − t 2 ) dt = 25 t dt − t 2 dt

SOLUTION

1 1 25 2 1 3 t − t + C. = 25 · t 2 − t 3 + C = 2 3 2 3 1 2 3 Thus 5 = s(0) = 25 2 (0) − 3 (0) + C, so that C = 5. Therefore, the position is

s(t) =

25 2 1 3 t − t + 5. 2 3

1 3 25 1 142 2 2 3 (b) From (a), we originally ascertained that s(t) = 25 2 t − 3 t + C. Thus 5 = s(2) = 2 (2) − 3 (2) + C = 3 + C, 127 so that C = 5 − 142 3 = − 3 . Therefore, the position is

s(t) =

25 2 1 3 127 t − t − . 2 3 3

S E C T I O N 4.8

Antiderivatives

227

the car come 65. A car traveling 84 ft/s begins to decelerate at a constant rate of 14 ft/s2 . After how many seconds does particle located at the at t = 0before movesstopping? in a straight line with acceleration a(t) = 4 − 12 t ft/s2 . Let v(t) to a stopAand how far will the carorigin have traveled be the velocity and s(t) the position at time t. SOLUTION Since the acceleration of the car is a constant −14ft/s2 , v is given by the differential equation: (a) State and solve the differential equation for v(t) assuming that the particle is at rest at t = 0. (b) Find s(t). dv = −14, v(0) = 84. dt ! ft From dv dt , we get v(t) = −14 dt = −14t + C. Since v(0) = 84, C = 84. From this, v(t) = −14t + 84 s . To find the time until the car stops, we must solve v(t) = 0: −14t + 84 = 0 14t = 84 t = 84/14 = 6 s. Now we have a differential equation for s(t). Since we want to know how far the car has traveled from the beginning of its deceleration at time t = 0, we have s(0) = 0 by definition, so: ds = v(t) = −14t + 84, dt

s(0) = 0.

! From this, s(t) = (−14t + 84) dt = −7t 2 + 84t + C. Since s(0) = 0, we have C = 0, and s(t) = −7t 2 + 84t. At stopping time t = 6 s, the car has traveled s(6) = −7(36) + 84(6) = 252 ft. 67. A 900-kg rocket is released from a spacecraft. As the rocket burns fuel, its mass decreases and its velocity increases. Beginning at rest, an object moves in a straight line with constant acceleration a, covering 100 ft in 5 s. Find a. Let v(m) be the velocity (in meters per second) as a function of mass m. Find the velocity when m = 500 if dv/dm = −50m −1/2 . Assume that v(900) = 0. ! dv SOLUTION Since dm = −50m −1/2 , we have v(m) = −50m −1/2 dm = −100m 1/2 + C. Thus 0 = v(900) = √ √ −100 900 + C = −3000 + C, and C = 3000. Therefore, v(m) = 3000 − 100 m. Accordingly, √ √ v(500) = 3000 − 100 500 = 3000 − 1000 5 ≈ 764 meters/sec. 69. Find constants c1 and c2 such that F(x) = c1 x sin x + c2 cos x is an antiderivative of f (x) = x cos x. As water flows through a tube of radius R = 10 cm, the velocity of an individual water particle depends on dv SOLUTION Let F(x) = c1 x sin x + c2 cos x. If F(x) is to be an antiderivative of f (x) = x cos x, we must have = −0.06r . Determine v(r ), assuming that its distance r from the center of the tube according to the formula F (x) = f (x) for all x. Hence c1 (x cos x + sin x) − c2 sin x = x cos x for dr all x. Equating coefficients on the left- and particles at the velocity. right-hand sides, wewalls have of c1 the = 1tube (i.e.,have the zero coefficients of x cos x are equal) and c1 − c2 = 0 (i.e., the coefficients of sin x are equal). Thus c1 = c2 = 1 and hence F(x) = x sin x + cos x. As a check, we have F (x) = x cos x + sin x − sin x = x cos x = f (x), as required. 71. Verify the linearity properties of the indefinite integral stated in Theorem 3. Find the general antiderivative of (2x + 9)10 . SOLUTION To verify the Sum Rule, let F(x) and G(x) be any antiderivatives of f (x) and g(x), respectively. Because d d d (F(x) + G(x)) = F(x) + G(x) = f (x) + g(x), dx dx dx it follows that F(x) + G(x) is an antiderivative of f (x) + g(x); i.e., " " " ( f (x) + g(x)) d x = f (x) d x + g(x) d x. To verify the Multiples Rule, again let F(x) be any antiderivative of f (x) and let c be a constant. Because d d (cF(x)) = c F(x) = c f (x), dx dx it follows that cF(x) is and antiderivative of c f (x); i.e., " " (c f (x)) d x = c f (x) d x.

228

CHAPTER 4


Further Insights and Challenges 73. (a) (b)

Suppose that F (x) = f (x). Suppose that F (x) = f (x) and G (x) = g(x). Is it true that F(x)G(x) is an antiderivative of f (x)g(x)? Confirm Show that 12 F(2x) is an antiderivative of f (2x). or provide a counterexample. Find the general antiderivative of f (kx) for any constant k.

Let F (x) = f (x). (a) By the Chain Rule, we have

SOLUTION

d dx

1 1 F(2x) = F (2x) · 2 = F (2x) = f (2x). 2 2

Thus 12 F(2x) is an antiderivative of f (2x). (b) For nonzero constant k, the Chain Rules gives d 1 1 F (kx) = F (kx) · k = F (kx) = f (kx). dx k k Thus k1 F(kx) is an antiderivative of f (kx). Hence the general antiderivative of f (kx) is 1k F(kx) + C, where C is a constant. −1 75. TheFind Power for antiderivatives does an Rule antiderivative for f (x) = |x|.not apply to f (x) = x . Which of the graphs in Figure 4 could plausibly −1 represent an antiderivative of f (x) = x ? y

y

y x

x 1

3

5

1

3

5

x 1 (A)

3 (B)

5 (C)

FIGURE 4 SOLUTION Let F(x) be an antiderivative of f (x) = x −1 . Then for x > 0, we have F (x) = f (x) = x −1 > 0. In other words, the graph of F(x) is increasing for x > 0. The only graph for which this is true is (A). Accordingly, neither graph (B) nor graph (C) is the graph of F(x).

CHAPTER REVIEW EXERCISES In Exercises 1–6, estimate using the Linear Approximation or linearization and use a calculator to compute the error. 1. 8.11/3 − 2 1 1 SOLUTION Let f (x) = x 1/3 , a = 8 and x = 0.1. Then f (x) = 3 x −2/3 , f (a) = 12 and, by the Linear Approximation,

1 f = 8.11/3 − 2 ≈ f (a)x = (0.1) = 0.00833333. 12 Using a calculator, 8.11/3 − 2 = 0.00829885. The error in the Linear Approximation is therefore |0.00829885 − 0.00833333| = 3.445 × 10−5 . 3. 6251/41 − 62411/4 √ − 1 −3/4 , f (a) = 1 and, by the Linear 1/4 SOLUTION 4.1 Let2 f (x) = x , a = 625 and x = −1. Then f (x) = 4 x 500 Approximation, 1 f = 6241/4 − 6251/4 ≈ f (a)x = (−1) = −0.002. 500 Thus 6251/4 − 6241/4 ≈ 0.002. Using a calculator, 6251/4 − 6241/4 = 0.00200120. The error in the Linear Approximation is therefore |0.00200120 − (0.002)| = 1.201 × 10−6 .


5.

229

1√ 1.02 101

Let f (x) = x −1 and a = 1. Then f (a) = 1, f (x) = −x −2 and f (a) = −1. The linearization of f (x) at a = 1 is therefore SOLUTION

L(x) = f (a) + f (a)(x − a) = 1 − (x − 1) = 2 − x, 1 ≈ L(1.02) = 0.98. Using a calculator, 1 = 0.980392, so the error in the Linear Approximation is and 1.02 1.02

|0.980392 − 0.98| = 3.922 × 10−4 . √ In Exercises 7–10, find the linearization at the point indicated. 5 33 √ 7. y = x, a = 25 √ 1 1 SOLUTION Let y = x and a = 25. Then y(a) = 5, y = 2 x −1/2 and y (a) = 10 . The linearization of y at a = 25 is therefore L(x) = y(a) + y (a)(x − 25) = 5 +

1 (x − 25). 10

9. A(r ) = 43 π r 3 , a = 3 v(t) = 32t − 4t 2 , a = 2 4 SOLUTION Let A(r ) = 3 π r 3 and a = 3. Then A(a) = 36π , A (r ) = 4π r 2 and A (a) = 36π . The linearization of A(r ) at a = 3 is therefore L(r ) = A(a) + A (a)(r − a) = 36π + 36π (r − 3) = 36π (r − 2). In Exercises the Linear V (h)11–15, = 4h(2use − h)(4 − 2h),Approximation. a=1 11. The position of an object in linear motion at time t is s(t) = 0.4t 2 + (t + 1)−1 . Estimate the distance traveled over the time interval [4, 4.2]. Let s(t) = 0.4t 2 + (t + 1)−1 , a = 4 and t = 0.2. Then s (t) = 0.8t − (t + 1)−2 and s (a) = 3.16. Using the Linear Approximation, the distance traveled over the time interval [4, 4.2] is approximately SOLUTION

s = s(4.2) − s(4) ≈ s (a)t = 3.16(0.2) = 0.632. 13. A store sells 80 MP3 players per week when the players are priced at P = $75. Estimate the number N sold if P is bondassuming that paysthat $10,000 offered Nforifsale at a price P. Thetopercentage yield Y of the bond is raised toA$80, d N /dinP 6=years −4. is Estimate the price is lowered $69. SOLUTION If P is raised to $80, then P = 5. With the assumption 10,000 1/6that d N /d P = −4, we estimate, using the Linear Y = 100 −1 . Approximation, that P dN Verify that if P = $7,500, then Y = 4.91%. the(−4)(5) drop in = yield if the price rises to $7,700. P = −20; N ≈ Estimate dP therefore, we estimate that only 60 MP3 players will be sold per week when the price is $80. On the other hand, if the price is lowered to $69, then P = −6 and N ≈ (−4)(−6) = 24. We therefore estimate that 104 MP3 players will be sold per week when the price is $69. √ b 15. Show a 2 + b ≈ of a +a sphere is measured small. Useatthis to 100 estimate 26 and the findmaximum the error using a calculator. Thethat circumference C = cm. Estimate percentage error in V if the 2a if b is error in C is at most 3 cm. 2 SOLUTION Let a > 0 and let f (b) = a + b. Then 1 f (b) = . 2 a2 + b By the Linear Approximation, f (b) ≈ f (0) + f (0)b, so

To estimate

√ 26, let a = 5 and b = 1. Then √

b a2 + b ≈ a + . 2a

26 =

1 52 + 1 ≈ 5 + = 5.1. 10

√ The error in this estimate is | 26 − 5.1| = 9.80 × 10−4 .

Use the Intermediate Value Theorem to prove that sin x − cos x = 3x has a solution and use Rolle’s Theorem to show that this solution is unique.

230

CHAPTER 4


x 17. Show that f (x) = x + 2 has precisely one real root. x +1 Let f (x) = x + 2x and note that f (0) = 0. Thus, f (x) has at least one real root. Now, suppose f (x) x +1 has two real roots, say a and b. Because f is continuous on [a, b], differentiable on (a, b) and f (a) = f (b) = 0, Rolle’s Theorem guarantees there exists c ∈ (a, b) such that f (c) = 0. However, SOLUTION

f (x) = 1 +

x 2 + 1 − 2x 2 x4 + x2 + 2 = >0 2 2 (x + 1) (x 2 + 1)2

for all x. We have reached a contradiction. Consequently, f (x) must have precisely one real root. 19. Suppose that f (1) = 5 and f (x) ≥ 2 for x ≥ 1. Use the MVT to show that f (8) ≥ 19. Verify the MVT for f (x) = x −1/2 on [1, 4]. SOLUTION Because f is continuous on [1, 8] and differentiable on (1, 8), the Mean Value Theorem guarantees there exists a c ∈ (1, 8) such that f (8) − f (1) 8−1

f (c) =

f (8) = f (1) + 7 f (c).

or

Now, we are given that f (1) = 5 and that f (x) ≥ 2 for x ≥ 1. Therefore, f (8) ≥ 5 + 7(2) = 19. In Exercises 21–24, the local they arethen minima, neither. Use the MVTfind to prove thatextrema if f (x)and ≤ 2determine for x > 0whether and f (0) = 4, f (x) maxima, ≤ 2x + 4orfor all x ≥ 0. 21. f (x) = x 3 − 4x 2 + 4x Let f (x) = x 3 − 4x 2 + 4x. Then f (x) = 3x 2 − 8x + 4 = (3x − 2)(x − 2), so that x = 23 and x = 2 are critical points. Next, f (x) = 6x − 8, so f ( 23 ) = −4 < 0 and f (2) = 4 > 0. Therefore, by the Second Derivative Test, f ( 23 ) is a local maximum while f (2) is a local minimum. SOLUTION

23. f (x) = x 2 (x + 2)3 s(t) = t 4 − 8t 2 SOLUTION Let f (x) = x 2 (x + 2)3 . Then f (x) = 3x 2 (x + 2)2 + 2x(x + 2)3 = x(x + 2)2 (3x + 2x + 4) = x(x + 2)2 (5x + 4), so that x = 0, x = −2 and x = − 45 are critical points. The sign of the first derivative on the intervals surrounding the critical points is indicated in the table below. Based on this information, f (−2) is neither a local maximum nor a local minimum, f (− 45 ) is a local maximum and f (0) is a local minimum. Interval

(−∞, −2)

(−2, − 45 )

(− 45 , 0)

(0, ∞)

+

+

−

+

Sign of f

In Exercises 25–30, find the extreme values on the interval. f (x) = x 2/3 (1 − x) 25. f (x) = x(10 − x),

[−1, 3]

Let f (x) = x(10 − x) = 10x − x 2 . Then f (x) = 10 − 2x, so that x = 5 is the only critical point. As this critical point is not in the interval [−1, 3], we only need to check the value of f at the endpoints to determine the extreme values. Because f (−1) = −11 and f (3) = 21, the maximum value of f (x) = x(10 − x) on the interval [−1, 3] is 21 while the minimum value is −11. SOLUTION

27. g(θ ) = sin2 θ 4− cos θ6 , [0, 2π ] f (x) = 6x − 4x , [−2, 2] SOLUTION Let g(θ ) = sin2 θ − cos θ . Then g (θ ) = 2 sin θ cos θ + sin θ = sin θ (2 cos θ + 1) = 0 when θ = 0, 23π , π , 43π , 2π . The table below lists the value of g at each of the critical points and the endpoints of the interval [0, 2π ]. Based on this information, the minimum value of g(θ ) on the interval [0, 2π ] is −1 and the maximum value is 54 .

θ g(θ ) 1/3 , [−1, 3] 29. f (x) = x 2/3 − 2x t , [0, 3] R(t) = 2 t +t +1

0

2π /3

π

4π /3

2π

−1

5/4

1

5/4

−1


231

Let f (x) = x 2/3 − 2x 1/3 . Then f (x) = 23 x −1/3 − 23 x −2/3 = 23 x −2/3 (x 1/3 − 1), so that the critical √ √ points are x = 0 and x = 1. With f (−1) = 3, f (0) = 0, f (1) = −1 and f (3) = 3 9 − 2 3 3 ≈ −0.804, it follows that the minimum value of f (x) on the interval [−1, 3] is −1 and the maximum value is 3. SOLUTION

31. Find the critical points and extreme values of f (x) = |x − 1| + |2x − 6| in [0, 8]. f (x) = x − tan x, [−1, 1] SOLUTION Let ⎧ ⎪ ⎨7 − 3x, f (x) = |x − 1| + |2x − 6| = 5 − x, ⎪ ⎩ 3x − 7,

x 43 , while y < 0 and y is concave down for x < 43 . Hence, there is a point of inflection at x = 43 . x2 35. y =y =2 x − 2 cos x x +4 SOLUTION

4 x2 8x =1− 2 . Then y = 2 Let y = 2 and x +4 x +4 (x + 4)2 y =

(x 2 + 4)2 (8) − 8x(2)(2x)(x 2 + 4) 8(4 − 3x 2 ) = . (x 2 + 4)4 (x 2 + 4)3

Thus, y > 0 and y is concave up for 2 2 −√ < x < √ , 3 3 while y < 0 and y is concave down for 2 |x| ≥ √ . 3 Hence, there are points of inflection at 2 x = ±√ . 3 37. (a) (b) (c)

Match the description of f (x) with the graph of its derivative f (x) in Figure 1. x y = f (x) is increasing and concave up. (x 2 − 4)1/3 f (x) is decreasing and concave up. f (x) is increasing and concave down. y

y

y

x

x

x (i)

(ii)

(iii)

FIGURE 1 Graphs of the derivative. SOLUTION

(a) If f (x) is increasing and concave up, then f (x) is positive and increasing. This matches the graph in (ii). (b) If f (x) is decreasing and concave up, then f (x) is negative and increasing. This matches the graph in (i). (c) If f (x) is increasing and concave down, then f (x) is positive and decreasing. This matches the graph in (iii). In Exercises thefor limit. Draw39–52, a curveevaluate y = f (x) which f and f have signs as indicated in Figure 2. 39. lim (9x 4 − 12x 3 ) x→∞

232

CHAPTER 4


Because the leading term in the polynomial has a positive coefficient, lim (9x 4 − 12x 3 ) = ∞.

x→∞

x 3 + 2x 41. lim lim 2 (7x 2 − 9x 5 ) x→∞ x→−∞ 4x − 9 SOLUTION

x 3 + 2x (x 3 + 2x)x −2 x + 2x −1 = lim = lim = ∞. 2 2 −2 x→∞ 4x − 9 x→∞ (4x − 9)x x→∞ 4 − 9x −2 lim

x 3 + 2x x 3 + 2x x→∞lim 4x 3 − 92 x→−∞ 4x − 9

43. lim

SOLUTION

lim

x 3 + 2x

x→∞ 4x 3 − 9

= lim

(x 3 + 2x)x −3

x→∞ (4x 3 − 9)x −3

= lim

1 + 2x −2

x→∞ 4 − 9x −3

=

1 . 4

x 2 − 9x3 45. lim x + 2x x→∞limx 2 + 4x x→−∞ 4x 5 − 40x 3 √ SOLUTION For x > 0, x −2 = |x|−1 = x −1 . Then, x 2 − 9x (x 2 − 9x)x −1 x −9 = lim lim = lim = ∞. √ x→∞ x→∞ x 2 + 4x x 2 + 4x x −2 1 + 4x −1

x→∞

x 1/212x + 1 47. lim lim √ x→∞ 4x 92 x→−∞ −4x + 4x SOLUTION

lim √

x→∞

49.

x 1/2 4x − 9

= lim √ x→∞

1 x 1/2 x −1/2 1 = lim = . √ x→∞ −1 −1 2 4x − 9 x 4 − 9x

1 − 2x

lim x 3/2 x→3+ limx − 3 6 x→∞ (16x − 9x 4 )1/4

SOLUTION

As x → 3+, 1 − 2x → −5 and x − 3 → 0+, so 1 − 2x = −∞. x→3+ x − 3 lim

x −5 lim 1 limx − 2 x→2+ x→2− x 2 − 4 SOLUTION As x → 2+, x − 5 → −3 and x − 2 → 0+, so

51.

lim

x −5

x→2+ x − 2

= −∞.

In Exercises 53–62, 1sketch the graph, noting the transition points and asymptotic behavior. lim (x 2+ 3)3 53. y =x→−3+ 12x − 3x Let y = 12x − 3x 2 . Then y = 12 − 6x and y = −6. It follows that the graph of y = 12x − 3x 2 is increasing for x < 2, decreasing for x > 2, has a local maximum at x = 2 and is concave down for all x. Because SOLUTION

lim (12x − 3x 2 ) = −∞,

x→±∞

the graph has no horizontal asymptotes. There are also no vertical asymptotes. The graph is shown below.


233

y 10 5 x

−1 −5

1

2

3

4

5

−10

2+3 55. y = x 3 − 2x y = 8x 2 − x 4 SOLUTION Let y = x 3 − 2x 2 + 3. Then y = 3x 2 − 4x and y = 6x − 4. It follows that the graph of y = x 3 − 2x 2 + 3 is increasing for x < 0 and x > 43 , is decreasing for 0 < x < 43 , has a local maximum at x = 0, has a local minimum at x = 43 , is concave up for x > 23 , is concave down for x < 23 and has a point of inflection at x = 23 . Because

lim (x 3 − 2x 2 + 3) = −∞

lim (x 3 − 2x 2 + 3) = ∞,

and

x→−∞

x→∞

the graph has no horizontal asymptotes. There are also no vertical asymptotes. The graph is shown below. y 10 5 −1

x 1

−5

2

−10

x 57. y = 3 3/2 y x= 4x + 1− x SOLUTION

x Let y = 3 . Then x +1 y =

x 3 + 1 − x(3x 2 ) 1 − 2x 3 = 3 3 2 (x + 1) (x + 1)2

and (x 3 + 1)2 (−6x 2 ) − (1 − 2x 3 )(2)(x 3 + 1)(3x 2 ) 6x 2 (2 − x 3 ) =− . 3 4 (x + 1) (x 3 + 1)3

x is increasing for x < −1 and −1 < x < 3 12 , is decreasing for x > 3 12 , has a It follows that the graph of y = 3 x +1

√ √ 3 3 3 1 local maximum at x = 2 , is concave up for x < −1 and x > 2, is concave down for −1 < x < 0 and 0 < x < 2 √ 3 and has a point of inflection at x = 2. Note that x = −1 is not an inflection point because x = −1 is not in the domain of the function. Now, y =

x

lim

x→±∞ x 3 + 1

= 0,

so y = 0 is a horizontal asymptote. Moreover, lim

x

x→−1− x 3 + 1

=∞

and

x

lim

x→−1+ x 3 + 1

so x = −1 is a vertical asymptote. The graph is shown below. y 4 2 −3 −2 −1

x −2 −4

1 x 59. y = y |x = + 2| + 1 2/3 2 (x − 4)

1

2

3

= −∞,

234

CHAPTER 4


SOLUTION

Let y =

1 . Because |x + 2| + 1 1

lim

= 0,

x→±∞ |x + 2| + 1

the graph of this function has a horizontal asymptote of y = 0. The graph has no vertical asymptotes as |x + 2| + 1 ≥ 1 for all x. The graph is shown below. From this graph we see there is a local maximum at x = −2. y 1 0.8 0.6 0.4 0.2 x

−8 −6 − 4 −2

2

4

61. y = 2 sin x − cos x on [0, 2π ] y = 2 −yx 3= √3 sin x − cos x. Then y = √3 cos x + sin x and y = −√3 sin x + cos x. It follows that the SOLUTION Let √ graph of y = 3 sin x − cos x is increasing for 0 < x < 5π /6 and 11π /6 < x < 2π , is decreasing for 5π /6 < x < 11π /6, has a local maximum at x = 5π /6, has a local minimum at x = 11π /6, is concave up for 0 < x < π /3 and 4π /3 < x < 2π , is concave down for π /3 < x < 4π /3 and has points of inflection at x = π /3 and x = 4π /3. The graph is shown below. y 1 −1

4 1

2

x

3

5

6

63. Find the maximum volume of a right-circular cone placed upside-down in a right-circular cone of radius R and y = 2x − tan x on [0, 2π ] height H (Figure 3). The volume of a cone of radius r and height h is 43 π r 2 h.

h

r

FIGURE 3 SOLUTION

Let r denote the radius and h the height of the upside down cone. By similar triangles, we obtain the

relation H −h H = r R

so

r h = H 1− R

and the volume of the upside down cone is

4 2 4 r3 2 V (r ) = π r h = π H r − 3 3 R

for 0 ≤ r ≤ R. Thus,

3r 2 4 dV = π H 2r − , dr 3 R

and the critical points are r = 0 and r = 2R/3. Because V (0) = V (R) = 0 and 2R 4R 2 4 8R 2 16 2 V π R H, = πH − = 3 3 9 27 81 the maximum volume of a right-circular cone placed upside down in a right-circular cone of radius R and height H is 16 2 π R H. 81


235

65. Show that the maximum area of a parallelogram ADEF inscribed in a triangle ABC, as in Figure 4, is equal to On a certain farm, the corn yield is one-half the area of ABC. Y = −0.118x 2 + 8.5x + B12.9 (bushels per acre) where x is the number of corn plants per acre (in thousands). Assume that corn seed costs $1.25 (per thousand seeds) and that corn can be sold for $1.50/bushel. D E (a) Find the value x 0 that maximizes yield Y . Then compute the profit (revenue minus the cost of seeds) at planting level x 0 . (b) Compute the profit P(x) as a function ofA x and findF the valueCx 1 that maximizes profit. 4 yield lead to maximum profit? maximum (c) Compare the profit at levels x0 and x 1 . Does aFIGURE SOLUTION Let θ denote the measure of angle B AC. Then the area of the parallelogram is given by AD · AF sin θ . Now, suppose that

B E/BC = x. Then, by similar triangles, AD = (1 − x) AB, AF = D E = x AC, and the area of the parallelogram becomes AB · AC x(1 − x) sin θ . The function x(1 − x) achieves its maximum value of 14 when x = 12 . Thus, the maximum area of a parallelogram inscribed in a triangle ABC is 1 1 1 1 AB · AC sin θ = AB · AC sin θ = (area of ABC) . 4 2 2 2 67. Let f (x) be a function whose graph does not pass through the x-axis and let Q = (a, 0). Let P = (x0 , f (x 0 )) be the 1 r 2 (θ − sin θ ). What is the maximum possible area of this region if of closest the shaded in 6). Figure 5 isthat point onThe the area graph to Qregion (Figure Prove 2 P Q is perpendicular to the tangent line to the graph of x 0 . Hint: the length the circular Let q(x) be theofdistance fromarc (x,isf 1? (x)) to (a, 0) and observe that x 0 is a critical point of q(x). y

y = f(x) P = (x 0 , f(x 0))

Q = (a, 0)

x

FIGURE 6 SOLUTION Let P = (a, 0) and let Q = (x 0 , f (x 0 )) be the point on the graph of y = f (x) closest to P. The slope of the segment joining P and Q is then

f (x 0 ) . x0 − a Now, let q(x) =

(x − a)2 + ( f (x))2 ,

the distance from the arbitrary point (x, f (x)) on the graph of y = f (x) to the point P. As (x 0 , f (x 0 )) is the point closest to P, we must have q (x 0 ) =

2(x 0 − a) + 2 f (x 0 ) f (x 0 ) = 0. (x 0 − a)2 + ( f (x 0 ))2

Thus, x −a f (x 0 ) = − 0 =− f (x 0 )

f (x 0 ) −1 . x0 − a

In other words, the slope of the segment joining P and Q is the negative reciprocal of the slope of the line tangent to the graph of y = f (x) at x = x 0 ; hence; the two lines are perpendicular. √ 3 to four decimalaplaces. 69. UseTake Newton’s Method to of estimate a circular piece paper of 25 radius R, remove sector of angle θ (Figure 7), and fold the remaining piece 3 SOLUTION Let f (x) = x − 25 and define into a cone-shaped cup. Which angle θ produces the cup of largest volume? x n+1 = x n − With x 0 = 3, we find

f (x n ) x 3 − 25 = xn − n 2 . f (x n ) 3xn

236

CHAPTER 4


Thus, to four decimal places

√ 3

n

1

2

3

xn

2.925925926

2.924018982

2.924017738

25 = 2.9240.

In Exercises 71–80, calculate the indefinite integral. Use Newton’s Method to find a root of f (x) = x 2 − x − 1 to four decimal places. " 3 4x − 2x 2 d x 71. " 2 SOLUTION (4x 3 − 2x 2 ) d x = x 4 − x 3 + C. 3 " " − 8) d θ 73. sin(θ9/4 dx x " SOLUTION sin(θ − 8) d θ = − cos(θ − 8) + C. "

" (4t −3 − 12t −4 ) dt cos(5 − 7θ ) d θ " SOLUTION (4t −3 − 12t −4 ) dt = −2t −2 + 4t −3 + C.

75.

"

" dx sec2 x−2/3 + 4t 7/3 ) dt (9t " SOLUTION sec2 x d x = tan x + C.

77.

"

" (y + 2)4 d y tan 3θ sec 3θ d θ " 1 (y + 2)4 d y = (y + 2)5 + C. SOLUTION 5

79.

" In Exercises 81–84, solve the differential equation with initial condition. 3x 3 − 9 dx 2 dy 81. = 4x x3 , y(1) = 4 dx SOLUTION

Let dd xy = 4x 3 . Then

" y(x) =

4x 3 d x = x 4 + C.

Using the initial condition y(1) = 4, we find y(1) = 14 + C = 4, so C = 3. Thus, y(x) = x 4 + 3. dy d=y x −1/22, y(1) = 1 dx = 3t + cos t, y(0) = 12 dt dy SOLUTION Let d x = x −1/2 . Then

83.

" y(x) =

x −1/2 d x = 2x 1/2 + C.

√ Using the initial condition y(1) = 1, we find y(1) = 2 1 + C = 1, so C = −1. Thus, y(x) = 2x 1/2 − 1. 85. Find f (t), assuming that f (t) = 1 − 2t, f (0) = 2, and f (0) = −1. dy π4 ) = = 12 − 2t. Then = sec2 x, y( SOLUTION d x Suppose f (t) " " f (t) dt = (1 − 2t) dt = t − t 2 + C. f (t) = Using the initial condition f (0) = −1, we find f (0) = 0 − 02 + C = −1, so C = −1. Thus, f (t) = t − t 2 − 1. Now, " " 1 1 f (t) = f (t) dt = (t − t 2 − 1) dt = t 2 − t 3 − t + C. 2 3 Using the initial condition f (0) = 2, we find f (0) = 12 02 − 13 03 − 0 + C = 2, so C = 2. Thus, f (t) =

1 2 1 3 t − t − t + 2. 2 3

The driver of an automobile applies the brakes at time t = 0 and comes to a halt after traveling 500 ft. Find the automobile’s velocity at t = 0 assuming that the rate of deceleration was a constant −10 ft/s2 .

5 THE INTEGRAL 5.1 Approximating and Computing Area Preliminary Questions 1. Suppose that [2, 5] is divided into six subintervals. What are the right and left endpoints of the subintervals? 1 If the interval [2, 5] is divided into six subintervals, the length of each subinterval is 5−2 6 = 2 . The right 5 7 9 5 7 9 endpoints of the subintervals are then 2 , 3, 2 , 4, 2 , 5, while the left endpoints are 2, 2 , 3, 2 , 4, 2 . SOLUTION

2. If f (x) = x −2 on [3, 7], which is larger: R2 or L 2 ? SOLUTION On [3, 7], the function f (x) = x −2 is a decreasing function; hence, for any subinterval of [3, 7], the function value at the left endpoint is larger than the function value at the right endpoint. Consequently, L 2 must be larger than R2 . 3. Which of the following pairs of sums are not equal? 4 4 # # (a) i, (c)

i=1 4 #

j,

j=1

=1 5 #

(i − 1)

(b)

(d)

i=2

4 # j=1 4 #

j 2,

5 #

k2

k=2

i(i + 1),

i=1

5 #

( j − 1) j

j=2

SOLUTION

(a) Only the name of the index variable has been changed, so these two sums are the same. (b) These two sums are not the same; the second squares the numbers two through five while the first squares the numbers one through four. (c) These two sums are the same. Note that when i ranges from two through five, the expression i − 1 ranges from one through four. (d) These two sums are the same. Both sums are 1 · 2 + 2 · 3 + 3 · 4 + 4 · 5. 4. Explain why

100 $

j is equal to

j=1 SOLUTION

100 $ j=0

The first term in the sum

j but $100

100 $

1 is not equal to

j=1

1.

j=0

j=0 j is equal to zero, so it may be dropped. More specifically, 100 #

j =0+

j=0

On the other hand, the first term in

100 $

100 #

j=

j=1

100 #

j.

j=1

$100

j=0 1 is not zero, so this term cannot be dropped. In particular, 100 # j=0

1=1+

100 #

1 =

j=1

100 #

1.

j=1

5. We divide the interval [1, 5] into 16 subintervals. (a) What are the left endpoints of the first and last subintervals? (b) What are the right endpoints of the first two subintervals? 1 Note that each of the 16 subintervals has length 5−1 16 = 4 . (a) The left endpoint of the first subinterval is 1, and the left endpoint of the last subinterval is 5 − 14 = 19 4 .

(b) The right endpoints of the first two subintervals are 1 + 14 = 54 and 1 + 2 14 = 32 . SOLUTION

6. (a) (b) (c)

Are the following statements true or false? The right-endpoint rectangles lie below the graph if f (x) is increasing. If f (x) is monotonic, then the area under the graph lies between R N and L N . If f (x) is constant, then the right-endpoint rectangles all have the same height.

SOLUTION

(a) False. If f is increasing, then the right-endpoint rectangles lie above the graph. (b) True. If f (x) is increasing, then the area under the graph is larger than L N but smaller than R N ; on the other hand, if f (x) is decreasing, then the area under the graph is larger than R N but smaller than L N . (c) True. The height of the right-endpoint rectangles is given by the value of the function, which, for a constant function, is always the same.

THE INTEGRAL

Exercises 1. An athlete runs with velocity 4 mph for half an hour, 6 mph for the next hour, and 5 mph for another half-hour. Compute the total distance traveled and indicate on a graph how this quantity can be interpreted as an area. SOLUTION The figure below displays the velocity of the runner as a function of time. The area of the shaded region equals the total distance traveled. Thus, the total distance traveled is (4) (0.5) + (6) (1) + (5) (0.5) = 10.5 miles. 6 5 4 3 2 1

Velocity (mph)

CHAPTER 5

0.5 1 1.5 Time (hours)

2

3. A rainstorm hit Portland, Maine, in October 1996, resulting in record rainfall. The rainfall rate R(t) on October 21 Figure 14 shows the velocity of an object over a 3-min interval. Determine the distance traveled over the intervals is recorded, in inches per hour, in the following table, where t is the number of hours since midnight. Compute the total [0, 3] and [1, 2.5] (remember to convert from miles per hour to miles per minute). rainfall during this 24-hour period and indicate on a graph how this quantity can be interpreted as an area.

SOLUTION

t

0–2

2–4

4–9

9–12

12–20

20–24

R(t)

0.2

0.1

0.4

1.0

0.6

0.25

Over each interval, the total rainfall is the time interval in hours times the rainfall in inches per hour. Thus R = 2(.2) + 2(.1) + 5(.4) + 3(1.0) + 8(.6) + 4(.25) = 11.4 inches.

The figure below is a graph of the rainfall as a function of time. The area of the shaded region represents the total rainfall. 1 Rainfall (inches per hour)

238

0.8 0.6 0.4 0.2 5

10 15 Time (hours)

20

25

5. Compute R6 , L 6 , and M3 to estimate the distance traveled over [0, 3] if the velocity at half-second intervals is as The velocity of an object is v(t) = 32t ft/s. Use Eq. (2) and geometry to find the distance traveled over the time follows: intervals [0, 2] and [2, 5].

SOLUTION

t (s)

0

0.5

1

1.5

2

2.5

3

v (ft/s)

0

12

18

25

20

14

20

3−0 For R6 and L 6 , t = 3−0 6 = .5. For M3 , t = 3 = 1. Then

R6 = 0.5 sec (12 + 18 + 25 + 20 + 14 + 20) ft/sec = .5(109) ft = 54.5 ft, L 6 = 0.5 sec (0 + 12 + 18 + 25 + 20 + 14) ft/sec = .5(89) ft = 44.5 ft, and M3 = 1 sec (12 + 25 + 14) ft/sec = 51 ft. 7. (a) (b)

Consider f (x) = 2x + 3 on [0, 3]. Use the following table of values to estimate the area under the graph of f (x) over [0, 1] by computing the Compute L 6 over [0, 3]. 6 and average ofRR 5 and L 5 . Find the error in these approximations by computing the area exactly using geometry.

SOLUTION

Let f (x) = 2x + 3 on [0, 3]. x

0 0.2 0.4 0.6 0.8 1 % & (a) We partition [0, 3] into 6 equally-spaced subintervals. The left 44 endpoints of40 the subintervals are 0, 12 , 1, 32 , 2, 52 f (x) 50 48 46 42 & % whereas the right endpoints are 12 , 1, 32 , 2, 52 , 3 .

S E C T I O N 5.1

Approximating and Computing Area

239

• Let a = 0, b = 3, n = 6, x = (b − a) /n = 1 , and x k = a + kx, k = 0, 1, . . . , 5 (left endpoints). Then 2

L6 =

5 #

f (x k )x = x

k=0

5 #

f (x k ) =

k=0

1 (3 + 4 + 5 + 6 + 7 + 8) = 16.5. 2

• With x k = a + kx, k = 1, 2, . . . , 6 (right endpoints), we have

R6 =

6 #

f (x k )x = x

k=1

6 #

f (x k ) =

k=1

1 (4 + 5 + 6 + 7 + 8 + 9) = 19.5. 2

(b) Via geometry (see figure below), the exact area is A = 12 (3) (6) + 32 = 18. Thus, L 6 underestimates the true area (L 6 − A = −1.5), while R6 overestimates the true area (R6 − A = +1.5). y 9 6 3 x 0.5

1

1.5

2

2.5

3

9. Estimate R6 , L 6 , and M6 over [0, 1.5] for the function in Figure 15. Let f (x) = x 2 + x − 2. y (a) Calculate R3 and L 3 over [2, 5]. 3 (b) Sketch the graph of f and the rectangles that make up each approximation. Is the area under the graph larger or smaller than R3 ? Than L 3 ? 2 1 x 0.5

1

1.5

FIGURE 15 SOLUTION

& % Let f (x) on [0, 32 ] be given by Figure 15. For n = 6, x = ( 32 − 0)/6 = 14 , {x k }6k=0 = 0, 14 , 12 , 34 , 1, 54 , 32 .

Therefore L6 =

5 1# 1 f (x k ) = (2.4 + 2.35 + 2.25 + 2 + 1.65 + 1.05) = 2.925, 4 k=0 4

R6 =

6 1# 1 f (x k ) = (2.35 + 2.25 + 2 + 1.65 + 1.05 + 0.65) = 2.4875, 4 k=1 4

6 1# 1 1 f x k − x = (2.4 + 2.3 + 2.2 + 1.85 + 1.45 + 0.8) = 2.75. M6 = 4 k=1 2 4 1 . Sketch the graph of f (x) and draw the rectangles whose area is represented by 11. LetEstimate f (x) = R x,2M + 1, and and Lxfor = the 3 graph in Figure 16. 2 3 6 $6 the sum i=1 f (1 + ix)x. SOLUTION Because the summation index runs from i = 1 through i = 6, we will treat this as a right-endpoint approximation to the area under the graph of y = x 2 + 1. With x = 13 , it follows that the right endpoints of the subintervals are x1 = 43 , x2 = 53 , x3 = 2, x 4 = 73 , x5 = 83 and x 6 = 3. The sketch of the graph with the rectangles $6 represented by the sum i=1 f (1 + ix)x is given below. y 3 2.5 2 1.5 1 0.5 x 0.5

1

1.5

2

2.5

3

In Exercises 13–24, theshaded approximation forinthe given17. function interval. do these rectangles represent? Calculate thecalculate area of the rectangles Figure Whichand approximation

240

CHAPTER 5

THE INTEGRAL

13. R8 ,

f (x) = 7 − x,

SOLUTION

[3, 5]

% & Let f (x) = 7 − x on [3, 5]. For n = 8, x = (5 − 3)/8 = 14 , and {x k }8k=0 = 3, 3 14 , 3 12 , 3 34 , 4, 4 14 , 4 12 , 4 34 , 5 .

Therefore R8 =

8 1# 1 1 (7 − x k ) = (3.75 + 3.5 + 3.25 + 3 + 2.75 + 2.5 + 2.25 + 2) = (23) = 5.75. 4 k=1 4 4

= x=2 , 7 − [0,x,1] [3, 5] 15. M4 ,M ,f (x)f (x) 4 ' (3 1 SOLUTION Let f (x) = x 2 on [0, 1]. For n = 4, x = (1 − 0)/4 = 4 and x k∗ = {.125, .375, .625, .875}. k=0 Therefore M4 =

3 1# 1 (x ∗ )2 = (0.1252 + 0.3752 + 0.6252 + 0.8752 ) = .328125. 4 k=0 k 4

17. R6 , f (x) = 2x 2 √ − x + 2, [1, 4] M6 , f (x) = x, [2, 5] % & 1 1 1 1 SOLUTION Let f (x) = 2x 2 − x + 2 on [1, 4]. For n = 6, x = (4 − 1)/6 = 2 , {x k }6 k=0 = 1, 1 2 , 2, 2 2 , 3, 3 2 , 4 . Therefore R6 =

6 1# 1 (2x 2 − x k + 2) = (5 + 8 + 12 + 17 + 23 + 30) = 47.5. 2 k=1 k 2

19. L 5 , f (x) = x −1 , 2 [1, 2] L 6 , f (x) = 2x − x + 2, [1, 4] % & (2−1) 1 6 7 8 9 5 SOLUTION Let f (x) = x −1 on [1, 2]. For n = 5, x = 5 = 5 , {x k }k=0 = 1, 5 , 5 , 5 , 5 , 2 . Therefore L5 =

4 1# 1 5 5 5 5 (x k )−1 = 1+ + + + ≈ .745635. 5 k=0 5 6 7 8 9

21. L 4 , f (x) = cos x, [ π4 , π2 ] M4 , f (x) = x −2 , [1, 3] π π SOLUTION Let f (x) = cos x on [ 4 , 2 ]. For n = 4, x =

π (π /2 − π /4) = 4 16

and

{x k }4k=0 =

π 5π 3π 7π π , , , , . 4 16 8 16 2

Therefore L4 =

3 π # cos x k ≈ .361372. 16 k=0

1 π ,5]3π ] = =2sin x,, [[1, 23. M4 ,R , f (x) f (x) 5 4 4 x +1 SOLUTION

Let f (x) =

1 . For n = 4, x = 5−1 = 1, and {x ∗ }3 k k=0 = {1.5, 2.5, 3.5, 4.5}. Therefore 4 x 2 +1

M4 = 1

3 #

1 = ∗ (x )2 + 1 k=0 k

4 4 4 4 + + + 13 29 53 85

≈ .568154.

In Exercises 25–28, use2the Graphical Insight on page 254 to obtain bounds on the area. L 5 , f (x) = x + 3|x|, [−3, 2] √ 25. Let A be the area under the graph of f (x) = x over [0, 1]. Prove that 0.51 ≤ A ≤ 0.77 by computing R4 and L 4 . Explain your reasoning. SOLUTION

1 1 1 3 4 For n = 4, x = 1−0 4 = 4 and {xi }i=0 = {0 + ix} = {0, 4 , 2 , 4 , 1}. Therefore, √ √ 4 # 1 1 2 3 f (xi ) = + + + 1 ≈ .768 R4 = x 4 2 2 2 i=1

√ √ 1 2 3 1 f (xi ) = L 4 = x 0+ + + ≈ .518. 4 2 2 2 i=0 3 #


S E C T I O N 5.1

241

In the plot below, you can see the rectangles whose area is represented by L 4 under the graph and the top of those whose area is represented by R4 above the graph. The area A under the curve is somewhere between L 4 and R4 , so .518 ≤ A ≤ .768.

L 4 , R4 and the graph of f (x). 27. Use R4 and L 4 to show that the area A under the graph of y = sin x over [0, π /2] satisfies 0.79 ≤ A ≤ 1.19. Use R6 and L 6 to show that the area A under y = x −2 over [10, 12] satisfies 0.0161 ≤ A ≤ 0.0172. SOLUTION Let f (x) = sin x. f (x) is increasing over the interval [0, π /2], so the Insight on page 254 applies, which

4 4 = π8 and {xi }i=0 = {0 + ix}i=0 = {0, π8 , π4 , 38π , π2 }. From indicates that L 4 ≤ A ≤ R4 . For n = 4, x = π /2−0 4 this,

L4 =

3 π# f (xi ) ≈ .79, 8 i=0

R4 =

4 π# f (xi ) ≈ 1.18. 8 i=1

Hence A is between .79 and 1.19.

Left and Right endpoint approximations to A. 29. Show that the area A in Exercise 25 satisfies L−1 N ≤ A ≤ R N for all N . Then use a computer algebra system ShowL that the area A under the graph of f (x) = x over [1, 8] satisfies to calculate N and R N for N = 100 and 150. Which of these calculations allows you to conclude that A ≈ 0.66 to two decimal places? 1 1 1 1 1 1 1 1 1 1 1 1 1 + √+ + + + + ≤ A ≤ 1 + + + + + + 2 =3 x 4is an 5increasing 6 7function; 8 7 . Now, SOLUTION On [0, 1], f (x) therefore,2 L N3≤ A4≤ R5N for6 all N L 100 = .6614629

R100 = .6714629

L 150 = .6632220

R150 = .6698887

Using the values obtained with N = 150, it follows that .6632220 ≤ A ≤ .6698887. Thus, to two decimal places, A ≈ .66. 31. Calculate the following sums: Show that the area A in Exercise 265 satisfies R N ≤ A ≤ L N for all N . Use 5 4 a computer algebra system to # # # L N and R N for N sufficiently (a) calculate 3 (b) large3to determine A to within an error(c)of at most k 3 10−4 . i=1 4 #

π (d) sin j 2 j=3

(e)

i=0 4 #

1 k − 1 k=2

(f)

k=2 3 #

3j

j=0

SOLUTION

(a)

(b)

(c)

5 # i=1 5 # i=0 4 #

3 = 3 + 3 + 3 + 3 + 3 = 15. Alternatively,

5 #

3=3

i=1

3 = 3 + 3 + 3 + 3 + 3 + 3 = 18. Alternatively,

5 #

1 = (3)(5) = 15.

i=1 5 # i=0

3=3

5 #

= (3)(6) = 18.

i=0

k 3 = 23 + 33 + 43 = 99. Alternatively,

k=2 4 # k=2

k3 =

4 #

k=1

k3 −

1 #

k=1

k3

=

44 43 42 + + 4 2 4

−

14 13 12 + + 4 2 4

= 99.

242

CHAPTER 5

THE INTEGRAL

(d)

4 #

sin

j=3 4 #

jπ 2

= sin

3π 2

+ sin

4π 2

= −1 + 0 = −1.

1 1 11 1 =1+ + = . k − 1 2 3 6 k=2 3 # 3 j = 1 + 3 + 32 + 33 = 40. (f)

(e)

j=0 200 # = 1, b3 = it17, b4 = −17. Let b1 = 3, b2j by 33. Calculate writing asand a difference of Calculate two sums the andsums. using formula (3). 4 2 3 # # # j=101 bk b j (a) bi (b) (b j + 2 ) (c) b SOLUTION i=2 j=1 k=1 k+1 200 #

j=

j=101

200 #

j−

j=1

100 #

j=

j=1

2002 200 + 2 2

−

1002 100 + 2 2

= 20100 − 5050 = 15050.

In Exercises 34–39, write the sum in summation notation. 35. (22 + 2) + (32 + 3) + (42 + 4) + (52 + 5) 47 + 57 + 67 + 77 + 87 SOLUTION The first term is 22 + 2, and the last term is 52 + 5, so it seems that the sum limits are 2 and 5, and the kth term is k 2 + k. Therefore, the sum is: 5 #

(k 2 + k).

k=2

1 +213 + 2 +323 + · · · +4 n + n 3 5 (2 + 2) + (2 + 2) + (2 + 2) + (2 + 2) SOLUTION The first term is 1 + 13 and the last term is n + n 3 , so it seems the summation limits are 1 through n, and the k-th term is k + k 3 . Therefore, the sum is 37.

n #

k + k3.

k=1

π

π π 39. sin(π1) + sin 2 + sin + ·n· · + sin + 2 + ··· + 3 n+1 2·3 3·4 (n + 1)(n + 2)

π π SOLUTION The first summand is sin 1 = sin 0+1 hence, sin(π ) + sin

π 2

+ sin

π 3

+ · · · + sin

π n+1

=

n # i=0

sin

π . i +1

In Exercises 40–47, use linearity and formulas (3)–(5) to rewrite and evaluate the sums. 41.

20 # 15 + 1) # (2k 12 j 3 k=1 j=1

20 #

SOLUTION

(2k + 1) = 2

k=1

43.

20 # k=1

k+

20 # k=1

1=2

202 20 + 2 2

+ 20 = 440.

200 # 150k 3 # (2k + 1) k=100 k=51

SOLUTION

By rewriting the sum as a difference of two power sums,

200 #

k3 =

k=100

30 2 # 4 j 10 #6 j + 45. (3 − 322 ) j=2 =1

200 # k=1

k3 −

99 # k=1

k3 =

2004 2003 2002 + + 4 2 4

−

994 993 992 + + 4 2 4

= 379507500.



30 # j=2

4 j2 6j + 3

⎛ ⎞ ⎛ ⎞ 30 30 1 30 1 # # # # 4# 4 2 2 2 j+ j = 6⎝ j− j⎠ + ⎝ j − j ⎠ =6 3 j=2 3 j=1 j=2 j=1 j=1 j=1 30 4 303 302 30 302 + −1 + + + −1 =6 2 2 3 3 2 6 30 #

= 6 (464) +

47.

37816 46168 4 = . (9454) = 2784 + 3 3 3

30 # 50 2 − 4s − 1) # (3s j ( j − 1) s=1 j=0

SOLUTION

30 #

(3s 2 − 4s − 1) = 3

s=1

30 #

s2 − 4

s=1

30 #

s−

s=1

30 #

1=3

s=1

303 302 30 + + 3 2 6

In Exercises 48–51, calculate the sum, assuming that a1 = −1,

10 #

ai = 10, and

i=1

49.

10 #

302 30 + 2 2

− 30 = 26475.

bi = 7.

i=1

10 # 10 − b ) # (a i i 2ai i=1 i=1

10 #

SOLUTION

(ai − bi ) =

i=1

51.

−4

10 #

ai −

i=1

10 #

bi = 10 − 7 = 3.

i=1

10 # # a10i

(3a + 4b ) 10 10 # # SOLUTION ai = ai − a1 = 10 − (−1) = 11. i=2 =1

i=2

i=1

In Exercises 52–55, use formulas (3)–(5) to evaluate the limit. 53.

N 3 # Nj # i N →∞ limj=1 N 4 2 N →∞ N i=1

lim

SOLUTION

Let s N =

N # j3 . Then N4 j=1 N 1 # 1 j3 = 4 sN = 4 N j=1 N

Therefore, lim s N = N →∞

N4 N3 N2 + + 4 2 4

=

1 1 1 . + + 4 2N 4N 2

1 . 4

N # i 32 20 N # 55. lim i −i +1 N N →∞ lim N4 i=1 N →∞ N3 i=1 N # 20 i3 SOLUTION Let s N = − . Then N N4 i=1 N N 1 # 20 # 1 i3 − 1= 4 sN = 4 N i=1 N i=1 N

Therefore, lim s N = N →∞

1 79 − 20 = − . 4 4

N4 N3 N2 + + 4 2 4

− 20 =

1 1 1 − 20. + + 4 2N 4N 2

243

244

CHAPTER 5

THE INTEGRAL

In Exercises 56–59, calculate the limit for the given function and interval. Verify your answer by using geometry. 57.

lim L N , f (x) = 5x, [1, 3] lim R N , f (x) = 5x, [0, 3] N →∞ N →∞

SOLUTION Let f (x) = 5x on [1, 3]. Let N > 0 be an integer, and set a = 1, b = 3, and x = (b − a)/N = 2/N . Also, let x k = a + kx = 1 + 2k N , k = 0, 1, . . . N − 1 be the left endpoints of the N subintervals of [1, 3]. Then

−1 2 N# 2k 10 5 1+ = N N N k=0 k=0 10 20 (N − 1)2 N −1 = N+ 2 + = 20 − N 2 2 N

L N = x

N −1 #

f (xk ) =

N −1 #

1+

k=0

−1 20 N# k N k=0

30 20 + 2. N N

The area under the graph is lim L N = 20.

N →∞

The region under the curve is a trapezoid with base width 2 and heights 5 and 15. Therefore the area is 12 (2)(5 + 15) = 20, which agrees with the value obtained from the limit of the left-endpoint approximations. 59.

lim M N , f (x) = x, [0, 1] lim L N , f (x) = 6 − 2x, N →∞ N →∞

[0, 2]

Let f (x) = x on [0, 1]. Let N > 0 be an integer and set a = 0, b = 1, and x = (b − a)/N = N1 . Also, let x k∗ = 0 + (k − 12 )x = 2k−1 2N , k = 1, 2, . . . N be the midpoints of the N subintervals of [0, 1]. Then SOLUTION

N N 1 # 2k − 1 1 # = (2k − 1) N k=1 2N 2N 2 k=1 k=1 N # 1 1 N 1 1 N2 = k−N = 2 + − = . 2 2 2 2 2N 2 2N N k=1

M N = x

N #

f (x k∗ ) =

The area under the curve over [0, 1] is lim M N =

N →∞

1 . 2

The region under the curve over [0, 1] is a triangle with base and height 1, and thus area 12 , which agrees with the answer obtained from the limit of the midpoint approximations. In Exercises 60–69, find a formula for R N for the given function and interval. Then compute the area under the graph as a limit. 61. f (x) = x 3 , 2[0, 1] f (x) = x , [0, 1] SOLUTION

1−0 1 Let f (x) = x 3 on the interval [0, 1]. Then x = = and a = 0. Hence, N N N N 1 # 1 1 # 3 f (0 + jx) = j3 j = R N = x N j=1 N3 N 4 j=1 j=1 N4 1 N3 N2 1 1 1 = 4 + + = + + 4 2 4 4 2N N 4N 2 N #

and

1 1 1 + + 2N N →∞ 4 4N 2

lim R N = lim

N →∞

1] 63. f (x) = 1 − x33 , [0, f (x) = x + 2x 2 , [0, 3] SOLUTION

Let f (x) = 1 − x 3 on the interval [0, 1]. Then x = R N = x

N # j=1

f (0 + jx) =

N 1 1 # 1 − j3 3 N j=1 N

=

1 . 4

1−0 1 = and a = 0. Hence, N N


S E C T I O N 5.1

N N 1 # 1 # 1 1 = 1− 4 j3 = N − 4 N j=1 N N j=1 N

and

N4 N3 N2 + + 4 2 4

1 1 3 − − 2N N →∞ 4 4N 2

lim R N = lim

N →∞

=

=1−

1 1 1 − − 4 2N 4N 2

3 . 4

65. f (x) = 3x 2 −2x + 4, [1, 5] f (x) = 3x − x + 4, [0, 1] 5−1 4 SOLUTION Let f (x) = 3x 2 − x + 4 on the interval [1, 5]. Then x = = and a = 1. Hence, N N N N N N N # 4 # 20 80 # 24 # 48 192 # f (1 + jx) = j2 + 2 j+ 1 R N = x +6 = 3 j2 2 + j N j=1 N N j=1 N N j=1 N j=1 j=1 192 = 3 N

N3 N2 N + + 3 2 6

80 + 2 N

and

N2 N + 2 2

+

40 96 32 24 N = 64 + + 2 + 40 + + 24 N N N N

136 32 128 + + 2 = 128. N N →∞ N

lim R N = lim

N →∞

2 , [2, 4] 67. f (x)f (x) = x= 2x + 7, [3, 6] 4−2 2 SOLUTION Let f (x) = x 2 on the interval [2, 4]. Then x = N = N and a = 2. Hence, N N N N N # 2 # 16 # 8 # 4 8 8 # 2 R N = x f (2 + jx) = 1 + j + j2 4+ j + j = N j=1 N N j=1 N2 N 2 j=1 N 3 j=1 j=1

8 16 = N+ 2 N N

N2 N + 2 2

8 + 3 N

N3 N2 N + + 3 2 6

and

56 12 4 + + N N →∞ 3 3N 2

lim R N = lim

N →∞

8 8 4 4 + + + N 3 N 3N 2

=8+8+

=

56 . 3

2 , [a, b] (a, b constants with a < b) 69. f (x)f (x) = x= 2x + 1, [a, b] (a, b constants with a < b)

b−a Let f (x) = x 2 on the interval [a, b]. Then x = . Hence, N N N # (b − a) # (b − a) (b − a)2 2 2 a + 2a j f (a + jx) = R N = x +j N N N2 j=1 j=1

SOLUTION

=

N N N a 2 (b − a) # 2a(b − a)2 # (b − a)3 # 1+ j + j2 2 3 N N N j=1 j=1 j=1

a 2 (b − a) 2a(b − a)2 = N+ N N2 = a 2 (b − a) + a(b − a)2 + and

lim R N = lim

N →∞

N →∞

N N2 + 2 2

N2 N N3 + + 3 2 6

a(b − a)2 (b − a)3 (b − a)3 (b − a)3 + + + N 3 2N 6N 2

(b − a)3 1 1 = b3 − a 3 . 3 3 3

In Exercises 70–73, describe the area represented by the limits.

lim

(b − a)3 + N3

a(b − a)2 (b − a)3 (b − a)3 (b − a)3 a 2 (b − a) + a(b − a)2 + + + + N 3 2N 6N 2

= a 2 (b − a) + a(b − a)2 +

N 4 1 # j

245

246

CHAPTER 5

THE INTEGRAL

71.

N 3 # 3j 4 2+ N N →∞ N j=1 lim

SOLUTION

The limit N 3 # 3 4 2+ j · N N →∞ N j=1

lim R N = lim

N →∞

represents the area between the graph of f (x) = x 4 and the x-axis over the interval [2, 5]. N jπ π # π −1 73. lim sin + j 4 5 N# 3 + 2N N →∞ lim2N −2 5 N N →∞ Nj=1 j=0 SOLUTION The limit N π # π jπ sin + 3 2N N →∞ 2N j=1 lim

represents the area between the graph of f (x) = sin x and the x axis over the interval [ π3 , 56π ]. In Exercises 75–80, use the approximation 2indicated (in summation notation) to express the area under the graph as a N j 1 # limit butEvaluate do not evaluate. lim 1− by interpreting it as the area of part of a familiar geometric figure. N N →∞ N j=1 [0, π ] 75. R N , f (x) = sin x over SOLUTION

Let f (x) = sin x over [0, π ] and set a = 0, b = π , and x = (b − a) /N = π /N . Then R N = x

N #

f (x k ) =

k=1

N kπ π # sin . N k=1 N

Hence N kπ π # sin N N →∞ N k=1

lim R N = lim

N →∞

is the area between the graph of f (x) = sin x and the x-axis over [0, π ]. 77. M N , f (x) = tan x−1 over [ 1 , 1] R N , f (x) = x over2[1, 7] SOLUTION

1− 1

1 and a = 1 . Hence Let f (x) = tan x over the interval [ 12 , 1]. Then x = N 2 = 2N 2 N N # 1 1 1 1 # 1 1 M N = x f tan + j− x = + j− 2 2 2N j=1 2 2N 2 j=1

and so N 1 1 1 1 # tan + j− 2 2N 2 N →∞ 2N j=1

lim M N = lim

N →∞

is the area between the graph of f (x) = tan x and the x-axis over [ 12 , 1]. 79. L N , f (x) = cos x over [ π8 , π ] M N , f (x) = x −2 over [3, 5] SOLUTION

π − π8 7π Let f (x) = cos x over the interval π8 , π . Then x = = and a = π8 . Hence, N 8N N −1

−1 # π π 7π N# 7π f cos + jx = +j L N = x 8 8N j=0 8 8N j=0

and −1 7π N# π 7π cos +j 8 8N N →∞ 8N j=0

lim L N = lim

N →∞

is the area between the graph of f (x) = cos x and the x-axis over [ π8 , π ]. LN,

f (x) = cos x over [ π8 , π4 ]

S E C T I O N 5.1


247

In Exercises 81–83, let f (x) = x 2 and let R N , L N , and M N be the approximations for the interval [0, 1]. 1 1 1 1 1 . Interpret the quantity as the area of a region. + + + 2 3 2N 2N 6N 6N 2 1 Let f (x) = x 2 on [0, 1]. Let N > 0 be an integer and set a = 0, b = 1 and x = 1−0 N = N . Then N N # 1 # 1 1 N2 N 1 1 1 N3 2 R N = x f (0 + jx) = j = 3 . + + = + + 2 N 3 2 6 3 2N N N 6N 2 j=1 j=1

Show that R N =

81. SOLUTION

The quantity 6 1 + 2 2N N

in

RN =

1 1 1 + + 3 2N 6N 2

represents the collective area of the parts of the rectangles that lie above the graph of f (x). It is the error between R N and the true area A = 13 . y 1 0.8 0.6 0.4 0.2 x 0.2

0.4

0.6

0.8

1

83. For each of R N , L N , and M N , find the smallest integer N for which the error is less than 0.001. Show that SOLUTION

1

1

1

L Nwhen: = − , + • For R N , the error is less than .001 3 2N 6N 2

MN =

1 1 − 3 12N 2

1 1 Then rank the three approximations R N , L N , and M+N in order of increasing accuracy (use the formula for R N in < .001. 2N 6N 2 Exercise 81). We find an adequate solution in N : 1 1 < .001 + 2N 6N 2 3N + 1 < .006(N 2 ) 0 < .006N 2 − 3N − 1,

√ 9.024 = 500.333. Hence R in particular, if N > 3+ .012 501 is within .001 of A. • For L N , the error is less than .001 if

1 1 − + < .001. 2N 6N 2

We solve this equation for N :

1 1 < .001 − 2N 6N 2 3N − 1 6N 2 < .001 3N − 1 < .006N 2 0 < .006N 2 − 3N + 1,

√

9−.024 = 499.666. Therefore, L 500 is within .001 units of A. which is satisfied if N > 3+ .012 1 • For M N , the error is given by − 2 , so the error is less than .001 if 12N

1 < .001 12N 2 1000 < 12N 2 9.13 < N Therefore, M10 is within .001 units of the correct answer.

248

CHAPTER 5

THE INTEGRAL

Further Insights and Challenges 85. Draw the graph of a positive continuous function on an interval such that R2 and L 2 are both smaller than the exact Although the accuracy of R N generally improves as N increases, this need not be true for small values of N . area under the graph. Can such a function be monotonic? Draw the graph of a positive continuous function f (x) on an interval such that R1 is closer than R2 to the exact area SOLUTION the plot area under the saw-tooth function f (x) is 3, whereas L 2 = R2 = 2. Thus L 2 and R2 under the In graph. Canbelow, such athe function be monotonic? are both smaller than the exact area. Such a function cannot be monotonic; if f (x) is increasing, then L N underestimates and R N overestimates the area for all N , and, if f (x) is decreasing, then L N overestimates and R N underestimates the area for all N . Left/right-endpoint approximation, n = 2 2

1

1

2

Assume f (x) isstatement monotonic. Prove thatThe M Nendpoint lies between R N and Lare M N iswhen closerf to 87. N and Explain thethat following graphically: approximations lessthat accurate (x)the is actual area under the graph than both R N and L N . Hint: Argue from Figure 18; the part of the error in R N due to the ith large. rectangle is the sum of the areas A + B + D, and for M N it is |B − E|. A B C

D E F

x i − 1 midpoint x i

x

FIGURE 18 SOLUTION

Suppose f (x) is monotonic increasing on the interval [a, b], x =

b−a , N

N = {a, a + x, a + 2x, . . . , a + (N − 1)x, b} {x k }k=0

and ' ∗ ( N −1 xk k=0 =

(a + (N − 1)x) + b a + (a + x) (a + x) + (a + 2x) , ,..., . 2 2 2

Note that xi < xi∗ < xi+1 implies f (xi ) < f (xi∗ ) < f (xi+1 ) for all 0 ≤ i < N because f (x) is monotone increasing. Then −1 −1 N b − a N# b − a N# b−a # ∗ f (x k ) < M N = f (xk ) < R N = f (xk ) LN = N k=0 N k=0 N k=1 Similarly, if f (x) is monotone decreasing, −1 −1 N b − a N# b − a N# b−a # ∗ LN = f (x k ) > M N = f (xk ) > R N = f (xk ) N k=0 N k=0 N k=1 Thus, if f (x) is monotonic, then M N always lies in between R N and L N . Now, as in Figure 18, consider the typical subinterval [xi−1 , xi ] and its midpoint xi∗ . We let A, B, C, D, E, and F be the areas as shown in Figure 18. Note that, by the fact that xi∗ is the midpoint of the interval, A = D + E and F = B + C. Let E R represent the right endpoint approximation error ( = A + B + D), let E L represent the left endpoint approximation error ( = C + F + E) and let E M represent the midpoint approximation error ( = |B − E|). • If B > E, then E M = B − E. In this case,

E R − E M = A + B + D − (B − E) = A + D + E > 0, so E R > E M , while E L − E M = C + F + E − (B − E) = C + (B + C) + E − (B − E) = 2C + 2E > 0, so E L > E M . Therefore, the midpoint approximation is more accurate than either the left or the right endpoint approximation.


S E C T I O N 5.1

249

• If B < E, then E M = E − B. In this case,

E R − E M = A + B + D − (E − B) = D + E + D − (E − B) = 2D + B > 0, so that E R > E M while E L − E M = C + F + E − (E − B) = C + F + B > 0, so E L > E M . Therefore, the midpoint approximation is more accurate than either the right or the left endpoint approximation. • If B = E, the midpoint approximation is exactly equal to the area. Hence, for B < E, B > E, or B = E, the midpoint approximation is more accurate than either the left endpoint or the right endpoint approximation. thisfor exercise, we prove that lim R N and lim L N exist and are equal if f (x) is positive and ProveInthat any function f (x) onthe [a,limits b], N →∞ N →∞ increasing [the case of f (x) decreasing is similar]. We usebthe − aconcept of a least upper bound discussed in Appendix B. = 1. ( f (b) − f (a)). N − N ≥ (a) Explain with a graph why L N ≤ R M forRall N ,LM N (b) By part (a), the sequence {L N } is bounded by R M for any M, so it has a least upper bound L. By definition, L is the smallest number such that L N ≤ L for all N . Show that L ≤ R M for all M. (c) According to part (b), L N ≤ L ≤ R N for all N . Use Eq. (8) to show that lim L N = L and lim R N = L. 89.

N →∞

N →∞

SOLUTION

(a) Let f (x) be positive and increasing, and let N and M be positive integers. From the figure below at the left, we see that L N underestimates the area under the graph of y = f (x), while from the figure below at the right, we see that R M overestimates the area under the graph. Thus, for all N , M ≥ 1, L N ≤ R M . y

y

x

x

(b) Because the sequence {L N } is bounded above by R M for any M, each R M is an upper bound for the sequence. Furthermore, the sequence {L N } must have a least upper bound, call it L. By definition, the least upper bound must be no greater than any other upper bound; consequently, L ≤ R M for all M. (c) Since L N ≤ L ≤ R N , R N − L ≤ R N − L N , so |R N − L| ≤ |R N − L N |. From this, lim |R N − L| ≤ lim |R N − L N |.

N →∞

N →∞

By Eq. (8), lim |R N − L N | = lim

N →∞

1

N →∞ N

|(b − a)( f (b) − f (a))| = 0,

so lim |R N − L| ≤ |R N − L N | = 0, hence lim R N = L. N →∞

N →∞

Similarly, |L N − L| = L − L N ≤ R N − L N , so

|L N − L| ≤ |R N − L N | =

(b − a) ( f (b) − f (a)). N

This gives us that lim |L N − L| ≤ lim

N →∞

1

N →∞ N

|(b − a)( f (b) − f (a))| = 0,

so lim L N = L. N →∞

This proves lim L N = lim R N = L . N →∞

N →∞

− the A| <area 10−4 for its thegraph given over function andUse interval. In Exercises 91–92,that use fEq. to find aand value of N suchand thatlet |RA N be Assume (x)(9) is positive monotonic, under [a, b]. Eq. (8) to √ show that 91. f (x) = x, [1, 4] √ b and −a SOLUTION Let f (x) = x on [1, 4]. Then b = 4, a = 1, |R N − A| ≤ | f (b) − f (a)| N 4−1 3 3 |R N − A| ≤ ( f (4) − f (1)) = (2 − 1) = . N N N √ We need N3 < 10−4 , which gives N > 30000. Thus |R30001 − A| < 10−4 for f (x) = x on [1, 4]. f (x) =

9 − x 2,

[0, 3]

250

CHAPTER 5

THE INTEGRAL

5.2 The Definite Integral Preliminary Questions " b

1. What is a

d x [here the function is f (x) = 1]?

" b

SOLUTION

a

dx =

" b a

1 · d x = 1(b − a) = b − a.

2. Are the following statements true or false [assume that f (x) is continuous]? " b f (x) d x is the area between the graph and the x-axis over [a, b]. (a) a " b (b) f (x) d x is the area between the graph and the x-axis over [a, b] if f (x) ≥ 0. a " b f (x) d x is the area between the graph of f (x) and the x-axis over [a, b]. (c) If f (x) ≤ 0, then − a

SOLUTION

! (a) False. ab f (x) d x is the signed area between the graph and the x-axis. (b) True. (c) True. " π cos x d x = 0. 3. Explain graphically why 0

Because cos(π − x) = − cos x, the “negative” area between the graph of y = cos x and the x-axis over [ π2 , π ] exactly cancels the “positive” area between the graph and the x-axis over [0, π2 ]. " −1 8 d x negative? 4. Is SOLUTION

−5

No, the integrand is the positive constant 8, so the value of the integral is 8 times the length of the integration interval (−1 − (−5) = 4), or 32. " 6 5. What is the largest possible value of f (x) d x if f (x) ≤ 13 ? SOLUTION

SOLUTION

Because f (x) ≤ 13 ,

" 6 0

0

f (x) d x ≤

1 (6 − 0) = 2. 3

Exercises In Exercises 1–10, draw a graph of the signed area represented by the integral and compute it using geometry. " 3 1.

−3

2x d x

The region bounded by the graph of y = 2x and the x-axis over the interval [−3, 3] consists of two right triangles. One has area 12 (3)(6) = 9 below the axis, and the other has area 12 (3)(6) = 9 above the axis. Hence, SOLUTION

" 3 −3

2x d x = 9 − 9 = 0. y 6 4 2

−3

3.

" 1 " 3 (3x + 4) d x (2x + 4) d x −2 −2

−2

−1 −2 −4 −6

x 1

2

3

S E C T I O N 5.2

The Definite Integral

251

SOLUTION The region bounded by the graph of y = 3x + 4 and the x-axis over the interval [−2, 1] consists of two right triangles. One has area 12 ( 23 )(2) = 23 below the axis, and the other has area 12 ( 73 )(7) = 49 6 above the axis. Hence,

" 1 −2

49 2 15 − = . 6 3 2

(3x + 4) d x =

y 8 6 4 2 −2

x

−1

1

−2

" 8 " 1 5. (7 − x) d x 4 dx 6 −2

The region bounded by the graph of y = 7 − x and the x-axis over the interval [6, 8] consists of two right triangles. One triangle has area 12 (1)(1) = 12 above the axis, and the other has area 12 (1)(1) = 12 below the axis. Hence, SOLUTION

" 8 6

(7 − x) d x =

1 1 − = 0. 2 2

y 1 0.5 x 2

−0.5

4

6

8

−1

7.

" 5 " 3π /2 2 25 − x d x sin x d x 0

The region bounded by the graph of y = 25 − x 2 and the x-axis over the interval [0, 5] is one-quarter of a circle of radius 5. Hence, " 5 1 25π 25 − x 2 d x = π (5)2 = . 4 4 0 π /2

SOLUTION

y 5 4 3 2 1 x 1

9.

2

3

4

5

" 2 " 3 (2 − |x|) d x |x| d x −2 −2

The region bounded by the graph of y = 2 − |x| and the x-axis over the interval [−2, 2] is a triangle above the axis with base 4 and height 2. Consequently, " 2 1 (2 − |x|) d x = (2)(4) = 4. 2 −2 SOLUTION

y 2 1

−2

" 1

(2x − |x|) d x

−1

x 1

2

252

CHAPTER 5

THE INTEGRAL

" 6 11. Calculate 0

(4 − x) d x in two ways:

(a) As the limit lim R N N →∞

(b) By sketching the relevant signed area and using geometry ! ! Let f (x) = 4 − x over [0, 6]. Consider the integral 06 f (x) d x = 06 (4 − x) d x. (a) Let N be a positive integer and set a = 0, b = 6, x = (b − a) /N = 6/N . Also, let x k = a + kx = 6k/N , k = 1, 2, . . . , N be the right endpoints of the N subintervals of [0, 6]. Then N N N N # # 6 # 6 # 6 6k f (x k ) = 1 − k = 4 4− R N = x N k=1 N N N k=1 k=1 k=1 6 6 N2 N 18 = 4N − + =6− . N N 2 2 N SOLUTION

18 = 6. N N →∞ N →∞ (b) The region bounded by the graph of y = 4 − x and the x-axis over the interval [0, 6] consists of two right triangles. One triangle has area 12 (4)(4) = 8 above the axis, and the other has area 12 (2)(2) = 2 below the axis. Hence, Hence lim R N = lim

6−

" 6 0

(4 − x) d x = 8 − 2 = 6.

y 4 2 x 1

2

3

4

5

6

−2

13. Evaluate the integrals for f (x) shown in Figure 13. " 5 " " 2 Calculate (2x + 1) d x in two ways: As the limit lim R 6N and using geometry. f (x) d x 2 (b) f (x) d x (a) N →∞ 0 0 " 6 " 4 f (x) d x (d) | f (x)| d x (c) 1

1

y

y = f(x)

x 2

4

6

FIGURE 13 The two parts of the graph are semicircles.

Let f (x) be given by Figure 13. ! (a) The definite integral 02 f (x) d x is the signed area of a semicircle of radius 1 which lies below the x-axis. Therefore,

SOLUTION

" 2 0

1 π f (x) d x = − π (1)2 = − . 2 2

! (b) The definite integral 06 f (x) d x is the signed area of a semicircle of radius 1 which lies below the x-axis and a semicircle of radius 2 which lies above the x-axis. Therefore, " 6 1 1 3π f (x) d x = π (2)2 − π (1)2 = . 2 2 2 0 ! (c) The definite integral 14 f (x) d x is the signed area of one-quarter of a circle of radius 1 which lies below the x-axis and one-quarter of a circle of radius 2 which lies above the x-axis. Therefore, " 4 1 1 3 f (x) d x = π (2)2 − π (1)2 = π . 4 4 4 1

S E C T I O N 5.2


253

! (d) The definite integral 16 | f (x)| d x is the signed area of one-quarter of a circle of radius 1 and a semicircle of radius 2, both of which lie above the x-axis. Therefore, " 6 1 9π 1 | f (x)| d x = π (2)2 + π (1)2 = . 2 4 4 1 In Exercises 14–15, refer to Figure 14. y

y = g (t)

2 1

t 1

−1

2

3

4

5

−2

FIGURE 14

" a " c " c3 such that " g(t) 5 dt and 15. Find a, b, and g(t) dt are as large as possible. Evaluate g(t) dt and0 g(t) dt. b 0 3" a SOLUTION To make the value of g(t) dt as large as possible, we want to include as much positive area as possible. 0 " c g(t) dt as large as possibe, we want to make sure to This happens when we take a = 4. Now, to make the value of b

include all of the positive area and only the positive area. This happens when we take b = 1 and c = 4. In Exercises 17–20, the Psigned area by the integral. the regions of positive and negative area. Describe thesketch partition and the setrepresented of intermediate points C forIndicate the Riemann sum shown in Figure 15. Compute " the2 value of the Riemann sum. 17. (x − x 2 ) d x 0

SOLUTION

! Here is a sketch of the signed area represented by the integral 02 (x − x 2 ) d x. y 0.5 +

x 1

−0.5

−

2

−1 −1.5 −2

19.

" 2π " 3 sin x d x 2 (2x − x ) d x π 0

SOLUTION

! Here is a sketch of the signed area represented by the integral π2π sin x d x. y 0.4 x −0.4 −0.8

1

2

3

4

5

6

7

−

−1.2

" 3π 21–24, determine the sign of the integral without calculating it. Draw a graph if necessary. In Exercises sin x d x " 1 21.

04

−2

x dx

SOLUTION The integrand is always positive. The integral must therefore be positive, since the signed area has only positive part. " " 1 2π x sin x d x 23. x03 d x

−2

As you can see from the graph below, the area below the axis is greater than the area above the axis. Thus, the definite integral is negative. SOLUTION

254

CHAPTER 5

THE INTEGRAL y 0.2

− 0.2

+ 1

x

2

3

4

5

6

7

−

− 0.4 − 0.6

" 225–28, In Exercises π sin xcalculate the Riemann sum R( f, P, C) for the given function, partition, and choice of intermediate points. Also, sketch the d xgraph of f and the rectangles corresponding to R( f, P, C). x 0 25. f (x) = x, P = {1, 1.2, 1.5, 2}, C = {1.1, 1.4, 1.9} SOLUTION

Let f (x) = x. With P = {x 0 = 1, x 1 = 1.2, x3 = 1.5, x 4 = 2}

C = {c1 = 1.1, c2 = 1.4, c3 = 1.9},

and

we get R( f, P, C) = x1 f (c1 ) + x 2 f (c2 ) + x 3 f (c3 ) = (1.2 − 1)(1.1) + (1.5 − 1.2)(1.4) + (2 − 1.5)(1.9) = 1.59. Here is a sketch of the graph of f and the rectangles. y 2 1.5 1 0.5 x 0.5

1

1.5

2

2.5

27. f (x) = x + 1, P = {−2, −1.6, −1.2, −0.8, −0.4, 0}, (x) = −1.3, x 2 + x, P −0.5, = {2, 3, C = f{−1.7, −0.9, 0} 4.5, 5}, C = {2, 3.5, 5} SOLUTION

Let f (x) = x + 1. With P = {x 0 = −2, x1 = −1.6, x3 = −1.2, x4 = −.8, x5 = −.4, x 6 = 0}

and C = {c1 = −1.7, c2 = −1.3, c3 = −.9, c4 = −.5, c5 = 0}, we get R( f, P, C) = x 1 f (c1 ) + x 2 f (c2 ) + x 3 f (c3 ) + x 4 f (c4 ) + x 5 f (c5 ) = (−1.6 − (−2))(−.7) + (−1.2 − (−1.6))(−.3) + (−0.8 − (−1.2))(.1) + (−0.4 − (−0.8))(.5) + (0 − (−0.4))(1) = .24. Here is a sketch of the graph of f and the rectangles. y 1 0.5 −2

−1

x − 0.5 −1

In Exercises the=basic integral f (x) 29–36, = sin x,use P {0, π6properties , π3 , π2 }, of C the = {0.4, 0.7,and 1.2}the formulas in the summary to calculate the integrals. " 4 x2 dx 29. 0

" 4

SOLUTION

By formula (4), 0

" 4

x2 dx

x2 dx =

1 3 64 (4) = . 3 3


S E C T I O N 5.2

" 3 31. 0

255

(3t + 4) dt " 3

SOLUTION

0

(3t + 4) dt = 3

" 3 0

" 1 " 32 (u − 2u) du 33. (3x + 4) d x 0

t dt + 4

" 3 0

1 51 . (3)2 + 4(3 − 0) = 2 2

1 dt = 3 ·

−2

SOLUTION

" 1 0

(u 2 − 2u) du =

" 1 0

u 2 du − 2

" 1 0

u du =

1 2 1 3 1 (1) − 2 (1)2 = − 1 = − . 3 2 3 3

" 1 " 32 (x + 2x) d x −a (6y + 7y + 1) d y 0 !b !b !b 1 1 SOLUTION First, 0 (x 2 + x) d x = 0 x 2 d x + 0 x d x = 3 b3 + 2 b2 . Therefore " 1 " 0 " 1 " 1 " −a (x 2 + x) d x = (x 2 + x) d x + (x 2 + x) d x = (x 2 + x) d x − (x 2 + x) d x

35.

−a

−a

=

0

1 3 1 2 ·1 + ·1 − 3 2

0

1 1 (−a)3 + (−a)2 3 2

0

=

1 3 1 2 5 a − a + . 3 2 6

37. Prove computing the limit of right-endpoint approximations: " aby 2 " b x2 dx b4 a x3 dx = . 4 0

7

Let f (x) = x 3 , a = 0 and x = (b − a)/N = b/N . Then

SOLUTION

N N N3 N2 b4 b4 b # b3 b4 # b4 N 4 b4 3 3 + + + + R N = x f (xk ) = k . k · 3 = 4 = 4 = N k=1 4 2 4 4 2N N N N 4N 2 k=1 k=1 " b b4 b4 b4 b4 Hence x 3 d x = lim R N = lim + + . = 2 2N 4 N →∞ N →∞ 4 4N 0 N #

In Exercises 38–45, use the formulas in the summary and Eq. (7) to evaluate the integral. 39.

" 2 " 32 (x +2 2x) d x x dx 0 0

Applying the linearity of the definite integral and the formulas from Examples 4 and 5,

SOLUTION

" 2 0

41.

(x 2 + 2x) d x =

0

x2 dx + 2

" 2 0

x dx =

1 3 1 20 (2) + 2 · (2)2 = . 3 2 3

" 2 " 3 3 (x − 3x ) d x x dx 0 0

Applying the linearity of the definite integral, the formula from Example 5 and Eq. (7):

SOLUTION

" 2 0

43.

" 2

(x − x 3 ) d x =

" 2 0

x dx −

" 2 0

x3 dx =

1 2 1 4 (2) − (2) = −2. 2 4

" 0 " 1 (2x − 35) d x −3 (2x − x + 4) d x 0

Applying the linearity of the definite integral, reversing the limits of integration, and using the formulas for the integral of x and of a constant: SOLUTION

" 0 −3

" 3 1

(2x − 5) d x = 2

x3 dx

" 0 −3

x dx −

" 0 −3

5 d x = −2

" −3 0

x dx −

" 0

1 5 d x = −2 · (−3)2 − 15 = −24. 2 −3

256

CHAPTER 5

THE INTEGRAL

" 2 45. 1

(x − x 3 ) d x

SOLUTION

Applying the linearity and the additivity of the definite integral: " 2 1

(x − x 3 ) d x =

" 2

=

1

x dx −

" 2 1

x3 dx =

1 2 1 (2 ) − (12 ) − 2 2

" 2 0

x dx −

1 4 1 4 (2) − (1) 4 4

In Exercises 46–50, calculate the integral, assuming that " 5 f (x) d x = 5,

" 5

0

0

" 1 0

=

" x dx −

0

2

x3 dx −

" 1

x3 dx

0

9 3 15 − =− . 2 4 4

g(x) d x = 12

" 5 " 5 ( f (x) + 4g(x)) d x ( f (x) + g(x)) d x 0 0 " 5 " 5 " 5 SOLUTION ( f (x) + 4g(x)) d x = f (x) d x + 4 g(x) d x = 5 + 4(12) = 53.

47.

0

0

0

" 5 " 0 49. (3 f (x) − 5g(x)) d x g(x) d x 0 5 " 5 " 5 " 5 SOLUTION (3 f (x) − 5g(x)) d x = 3 f (x) d x − 5 g(x) d x = 3(5) − 5(12) = −45. 0

0

0

In Exercises 51–54, calculate the"integral, assuming that 5 Is it possible to calculate g(x) f (x) d x from " " the information given? " 1

0

" 4 51. 0

2

0

f (x) d x = 1,

0

f (x) d x = 4,

4

1

f (x) d x = 7

f (x) d x " 4

SOLUTION

0

f (x) d x =

" 1 0

f (x) d x +

" 4 1

f (x) d x = 1 + 7 = 8.

" 1 " 2 f (x) d x f (x) d x 4 1 " 1 " 4 SOLUTION f (x) d x = − f (x) d x = −7.

53.

4

1

" 4 55–58, express each integral as a single integral. In Exercises f (x) d x" 7 " 3 2 f (x) d x + f (x) d x 55. 0

" 3

SOLUTION

0

3

f (x) d x +

" 7 3

f (x) d x =

" 7 0

f (x) d x.

" 5 " 9 " 9 " 9 f (x) d x − f (x) d x f (x) d x − f (x) d x 2 2 " " " " 5 " 9 " 9 2 4 9 5 5 SOLUTION f (x) d x − f (x) d x = f (x) d x + f (x) d x − f (x) d x = f (x) d x.

57.

2

2

2

5

2

5

" b " 3 " 9 In Exercises f59–62, calculate the integral, assuming that f is an integrable function such that f (x) d x = 1 − b−1 (x) d x + f (x) d x 7 0. for all b > " 3 59. f (x) d x 1

" 3

SOLUTION

1

61.

f (x) d x = 1 − 3−1 =

" 4 " 4 (4 f (x) − 2) d x f (x) d x 1 2

1

3

2 . 3


S E C T I O N 5.2

" 4 SOLUTION

1

(4 f (x) − 2) d x = 4

" 4

f (x) d x − 2

1

" 4 1

257

1 d x = 4(1 − 4−1 ) − 2(4 − 1) = −3.

" b " b " 1 Explain the difference in graphical interpretation between f (x) d x and | f (x)| d x. f (x) d x a a 1/2 !b SOLUTION When f (x) takes on both positive and negative values on [a, b], a f (x) d x represents the signed area !b between f (x) and the x-axis, whereas a | f (x)| d x represents the total (unsigned) area between f (x) and the x-axis. ! ! Any negatively signed areas that were part of ab f (x) d x are regarded as positive areas in ab | f (x)| d x. Here is a graphical example of this phenomenon. 63.

Graph of | f (x)|

Graph of f (x) 10 −4

−2

30 x 2

−10

20

4

10

−20 −30

−4

x

−2

2

4

In Exercises 65–68," calculate the integral. " 2π 2π sin2 x d x and J = cos2 x d x. Use the following trick to prove that I = J = π : First show " 6 Let I = 0 0 " 2π 65. |3 − x| d x 0 a graph that I = J and then prove I + J = with d x. 0 SOLUTION Over the interval, the region between the curve and the interval [0, 6] consists of two triangles above the x axis, each of which has height 3 and width 3, and so area 92 . The total area, hence the definite integral, is 9. y 3 2 1 x 1

2

3

4

5

6

Alternately, " 6 0

|3 − x| d x =

" 3

=3

0

(3 − x) d x +

" 3 0

dx −

" 3 0

" 6 3

(x − 3) d x

x dx +

" 6 0

x dx −

" 3 x dx 0

−3

" 6 dx 3

1 1 1 = 9 − 32 + 62 − 32 − 9 = 9. 2 2 2 67.

" 1 " 33 |x | d x |2x − 4| d x −1 1

SOLUTION

|x 3 | =

x3 −x 3

x ≥0 x < 0.

Therefore, " 1 −1

|x 3 | d x =

" 0 −1

−x 3 d x +

" 1 0

x3 dx =

" −1 0

69. Use"the Comparison Theorem to show that 2 " 1 " 1 |x 2 − 1| d x 5 0 x dx ≤ x 4 d x, 0

0

x3 dx +

" 2 1

" 1 0

x3 dx =

x4 dx ≤

" 2 1

1 1 1 (−1)4 + (1)4 = . 4 4 2

x5 dx

258

CHAPTER 5

THE INTEGRAL SOLUTION

On the interval [0, 1], x 5 ≤ x 4 , so, by Theorem 5, " 1 0

x5 dx ≤

" 1

x 4 d x.

0

On the other hand, x 4 ≤ x 5 for x ∈ [1, 2], so, by the same Theorem, " 2 " 2 x4 dx ≤ x 5 d x. 1

1

" " 0.3 71. Prove that 0.0198 1 ≤ 6 1 sin x d1x ≤ 0.0296. Hint: Show that 0.198 ≤ sin x ≤ 0.296 for x in [0.2, 0.3]. Prove that ≤ 0.2 d x ≤ . 3 2 4 x π d SOLUTION For 0 ≤ x ≤ 6 ≈ 0.52, we have d x (sin x) = cos x > 0. Hence sin x is increasing on [0.2, 0.3]. Accordingly, for 0.2 ≤ x ≤ 0.3, we have m = 0.198 ≤ 0.19867 ≈ sin 0.2 ≤ sin x ≤ sin 0.3 ≈ 0.29552 ≤ 0.296 = M Therefore, by the Comparison Theorem, we have " 0.3 " 0.3 " 0.3 m dx ≤ sin x d x ≤ M d x = M(0.3 − 0.2) = 0.0296. 0.0198 = m(0.3 − 0.2) = 0.2

Prove that

73.

Prove that 0.277 ≤

Hint: Graph y = SOLUTION

" π /4 π /8

0.2

0.2

√ sin x 2 dx ≤ x 2 π /4

cos x d x ≤ 0.363. " π /2

sin x and observe that it is decreasing on [ π4 , π2 ]. x

Let f (x) =

sin x . x

As we can see in the sketch √ below, f (x) is decreasing on the interval [π /4, π /2]. Therefore f (x) ≤ f (π /4) for all x in [π /4, π /2]. f (π /4) = 2 π 2 , so:

√ √ " π /2 " π /2 √ sin x 2 2 π2 2 2 dx = = dx ≤ . x π 4 π 2 π /4 π /4 y y = sin x

2 2/p

x

2/p

p /4

p/2

x

" b " b " Suppose that f (x) ≤ g(x) on 1[a, b]. f (x) d x ≤ g(x) d x. Is it also d xBy the Comparison Theorem, Find upper and lower bounds for . a a 3 0 give x a+counterexample. 4 true that f (x) ≤ g (x) for x ∈ [a, b]? If not,

75.

The assertion f (x) ≤ g (x) is false. Consider a = 0, b = 1, f (x) = x, g(x) = 2. f (x) ≤ g(x) for all x in the interval [0, 1], but f (x) = 1 while g (x) = 0 for all x. SOLUTION

whether trueChallenges or false. If false, sketch the graph of a counterexample. Further State Insights and "" ab

(x)ddxx=>00.if f (x) is an odd function. (a) If f (x) > 0, then 77. Explain graphically: f f(x) a −a " b SOLUTION is danx odd then f (−x) = − f (x) for all x. Accordingly, for every positively signed area in the (b) If Iff f(x) > 0,function, then f (x) > 0. a where f is above the x-axis, there is a corresponding negatively signed area in the left half-plane where right half-plane f is below the x-axis. Similarly, for every negatively signed area in the right half-plane where f is below the x-axis, there is a corresponding positively signed area in the left half-plane where f is above the x-axis. We conclude that the net area between the graph of f and the x-axis over [−a, a] is 0, since the positively signed areas and negatively signed areas cancel each other out exactly.

The Fundamental Theorem of Calculus, Part I

S E C T I O N 5.3

259

y 4 2 −1 −2

x 1

−2

2

−4

79. Let k and b be" positive. Show, by comparing the right-endpoint approximations, that 1

Compute

" 1 sin(sin(x))(sin2 (x) + 1)" dbx. x k d x = bk+1 xk dx

−1

0

0

! Let k and b be any positive numbers. Let f (x) = x k on [0, b]. Since f is continuous, both 0b f (x) d x ! and 01 f (x) d x exist. Let N be a positive integer and set x = (b − 0) /N = b/N . Let x j = a + jx = bj/N , j = 1, 2, . . . , N be the right endpoints of the N subintervals of [0, b]. Then the right-endpoint approximation to !b k !b 0 f (x) d x = 0 x d x is ⎛ ⎞ N N k N # # 1 b # bj f (x j ) = = bk+1 ⎝ k+1 jk⎠ . R N = x N N N j=1 j=1 j=1 SOLUTION

! ! In particular, if b = 1 above, then the right-endpoint approximation to 01 f (x) d x = 01 x k d x is N N N k # j 1 # 1 # 1 f (x j ) = = k+1 j k = k+1 R N S N = x N N N b j=1 j=1 j=1 In other words, R N = bk+1 S N . Therefore, " b " 1 x k d x = lim R N = lim bk+1 S N = bk+1 lim S N = bk+1 x k d x. N →∞

0

N →∞

N →∞

0

81. Show that Eq. (4) holds for b ≤ 0. Verify by interpreting the integral as an area: SOLUTION Let c = −b. Since b < 0, c > 0, so by Eq. (4), " b 1 1 1 "− cx 2 d x = b1 1 − b2 + θ 2 3 2 2 0 x dx = c . 3 0 Here, 0 ≤ b ≤ 1 and θ is the angle between 0 and π2 such that sin θ = b. Furthermore, x 2 is an even function, so symmetry of the areas gives " 0 −c

x2 dx =

" c

x 2 d x.

0

Finally, " b 0

x2 dx =

" −c 0

x2 dx = −

" 0 −c

x2 dx = −

" c 0

1 1 x 2 d x = − c3 = b3 . 3 3

Theorem 4 remains true without the assumption a ≤ b ≤ c. Verify this for the cases b < a < c and c < a < b.

5.3 The Fundamental Theorem of Calculus, Part I Preliminary Questions 1. Assume that f (x) ≥ 0. What is the area under the graph of f (x) over [0, 2] if f (x) has an antiderivative F(x) such that F(0) = 3 and F(2) = 7? SOLUTION

Because f (x) ≥ 0, the area under the graph of y = f (x) over the interval [0, 2] is " 2 0

f (x) d x = F(2) − F(0) = 7 − 3 = 4.

2. Suppose that F(x) is an antiderivative of f (x). What is the graphical interpretation of F(4) − F(1) if f (x) takes on both positive and negative values?

260

CHAPTER 5

THE INTEGRAL

! Because F(x) is an antiderivative of f (x), it follows that F(4) − F(1) = 14 f (x) d x. Hence, F(4) − F(1) represents the signed area between the graph of y = f (x) and the x-axis over the interval [1, 4]. " 7 " 7 f (x) d x and f (x) d x, assuming that f (x) has an antiderivative F(x) with values from the follow3. Evaluate SOLUTION

0

2

ing table:

SOLUTION

0

2

7

F(x)

3

7

9

Because F(x) is an antiderivative of f (x), " 7 0

and

" 7 2

4. (a) (b) (c)

x

f (x) d x = F(7) − F(0) = 9 − 3 = 6

f (x) d x = F(7) − F(2) = 9 − 7 = 2.

Are the following statements true or false? Explain. The FTC I is only valid for positive functions. To use the FTC I, you have to choose the right antiderivative. If you cannot find an antiderivative of f (x), then the definite integral does not exist.

SOLUTION

(a) False. The FTC I is valid for continuous functions. (b) False. The FTC I works for any antiderivative of the integrand. (c) False. If you cannot find an antiderivative of the integrand, you cannot use the FTC I to evaluate the definite integral, but the definite integral may still exist. " 9 5. What is the value of f (x) d x if f (x) is differentiable and f (2) = f (9) = 4? 2

SOLUTION

" 9

Because f is differentiable, 2

f (x) d x = f (9) − f (2) = 4 − 4 = 0.

Exercises In Exercises 1–4, sketch the region under the graph of the function and find its area using the FTC I. 1. f (x) = x 2 ,

[0, 1]

SOLUTION y 1 0.8 0.6 0.4 0.2 x 0.2

0.4

0.6

0.8

1

We have the area A=

" 1 0

x2 dx =

1 3 1 1 x = . 3 0 3

3. f (x) = sin x, [0, π /2] f (x) = 2x − x 2 , [0, 2]

SOLUTION

y 1 0.8 0.6 0.4 0.2 x 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

S E C T I O N 5.3


Let A be the area indicated. Then A=

" π /2 0

π /2 sin x d x = − cos x = 0 − (−1) = 1. 0

In Exercises f (x) 5–34, = cosevaluate x, [0, πthe /2]integral using the FTC I. " 6 x dx 5. 3

" 6

SOLUTION

3

x dx =

" 2 " 92 u du 7. 2 dx −3 0 " 2 SOLUTION

−3

1 2 6 1 1 27 x = (6)2 − (3)2 = . 2 3 2 2 2

u 2 du =

1 3 2 1 1 35 u = (2)3 − (−3)3 = . 3 −3 3 3 3

" 3 " 31 (t − t 2 ) dt 9. (x − x 2 ) d x 1 0 " 3 1 4 1 3 3 1 4 1 3 34 1 1 SOLUTION (t 3 − t 2 ) dt = t − t = (3) − (3) − − = . 4 3 4 3 4 3 3 1 1 " 4 " 12 (x + 2) d x4 11. −3 (4 − 5u ) du 0

SOLUTION

" 4 −3

4 1 3 1 3 1 133 x + 2x = (4) + 2(4) − (−3)3 + 2(−3) = . 3 3 3 3 −3

(x 2 + 2) d x =

" 2 " 4 9 (10x 5+ 3x 52) d x −2 (3x + x − 2x) d x 0 " 2 1 6 2 1 6 1 6 9 5 10 10 10 SOLUTION (10x + 3x ) d x = x + x = 2 + 2 − 2 + 2 = 0. 2 2 2 −2 −2 " 3 " 13/2 (4t + t 7/2 ) dt 15. (5u 4 − 6u 2 ) du 1 13.

−1

SOLUTION

" 3 1

(4t 3/2 + t 7/2 ) dt =

√ √ √ 8 5/2 2 9/2 3 72 3 8 2 162 3 82 + t t + 18 3 − + = − . = 5 9 5 5 9 5 45 1

" 4 " 12 dt2 −2 1 t 2 (x − x ) d x 1 " 4 " 4 4

1 −2 dt = −t −1 = −(4)−1 − −(1)−1 = 3 . SOLUTION dt = t 4 1 t2 1 1 " 27 " 41/3 19. x √ dx y dy 1 0 27 " 27 3 3 3 SOLUTION x 1/3 d x = x 4/3 = (81) − = 60. 4 4 4 1 1 " 9 " −1/2 t 4 −4dt 21. x dx 1 1 9 " 9 SOLUTION t −1/2 dt = 2t 1/2 = 2(9)1/2 − 2(1)1/2 = 4. 17.

1

" −1 " 91 23. 8d x −2 x 3 3 d x x 4

1

261

262

CHAPTER 5

THE INTEGRAL

" −1 1 −2 −1 1 1 3 1 d x = − = − (−1)−2 + (−2)−2 = − . x 3 2 2 2 8 −2 x −2

SOLUTION

25.

" 27 " t4 + 1 √ 2 dt πt d x 1 2

SOLUTION

27 " 27 " 27 2 3/2 t +1 (t 1/2 + t −1/2 ) dt = + 2t 1/2 t √ dt = 3 t 1 1 1 √ √ √ 2 2 8 = (81 3) + 6 3 − + 2 = 60 3 − . 3 3 3 " π /2 " π /2 cos x d x cos θ d θ −π /2 0 π /2 " π /2 SOLUTION cos x d x = sin x = 1 − (−1) = 2.

27.

−π /2

" 3π /4 " 2π sin θ d θ 29. cos t dt π /4 0 " 3π /4 SOLUTION

π /4

−π /2

√ √ 3π /4 2 2 √ sin θ d θ = − cos θ = + = 2. 2 2 π /4

" π /4 " 4π 2 31. sec t dt sin x d x 0 2π " π /4 π /4 π 2 SOLUTION sec t dt = tan t = tan − tan 0 = 1. 4 0 0 " π /3 " π /4 33. csc x cot x d x sec θ tan θ d θ π /6 0 π /3

" π /3 π

π 2√ − − csc =2− SOLUTION csc x cot x d x = (− csc x) = − csc 3. 3 6 3 π /6 π /6 " π /235–40, write the integral as a sum of integrals without absolute values and evaluate. In Exercises csc2 y d y " 1 35.

π /6

−2

|x| d x

SOLUTION

1 2 0 1 2 1 1 1 5 |x| d x = (−x) d x + x d x = − x + x = 0 − − (4) + = . 2 2 2 2 2 −2 −2 0 −2 0

" 1

37.

" 0

" 1

" 3 " 53 |x | d x |3 − x| d x −2 0

SOLUTION

" 3 −2

|x 3 | d x =

" 0 −2

(−x 3 ) d x +

1 0 1 3 x 3 d x = − x 4 + x 4 4 −2 4 0 0

" 3

1 1 97 = 0 + (−2)4 + 34 − 0 = . 4 4 4 39.

" π "|cos 3 x| d x |x 2 − 1| d x 0 0

SOLUTION

" π 0

" 5

|cos x| d x =

" π /2 0

|x 2 − 4x + 3| d x

cos x d x +

" π π /2

π /2 π (− cos x) d x = sin x − sin x 0

π /2

= 1 − 0 − (−1 − 0) = 2.


S E C T I O N 5.3

263

In Exercises 41–44, evaluate the integral in terms of the constants. " b 41. x3 dx 1

" b

SOLUTION

1

" b " 5a 43. x d x4 x dx 1 b " b

x3 dx =

1 4 b 1 1 1 4 b − 1 for any number b. x = b4 − (1)4 = 4 1 4 4 4

1 6 b 1 1 1 x = b6 − (1)6 = (b6 − 1) for any number b. 6 1 6 6 6 1 " 1 " x x n d x = 0 if n is an odd whole number. Explain graphically. 45. Use the FTC I to show that (t 3 + t) dt −1 SOLUTION

x5 dx =

−x

SOLUTION

We have " 1 −1

xn dx =

x n+1 1 (1)n+1 (−1)n+1 = − . n + 1 −1 n+1 n+1

Because n is odd, n + 1 is even, which means that (−1)n+1 = (1)n+1 = 1. Hence (−1)n+1 1 1 (1)n+1 − = − = 0. n+1 n+1 n+1 n+1 Graphically speaking, for an odd function such as x 3 shown here, the positively signed area from x = 0 to x = 1 cancels the negatively signed area from x = −1 to x = 0. y 1 0.5 − 0.5 −1

x − 0.5

0.5

1

−1

47. Show that the area of a parabolic arch (the shaded region in Figure 5) is equal to four-thirds the area of the triangle What is the area (a positive number) between the x-axis and the graph of f (x) on [1, 3] if f (x) is a negative shown. function whose antiderivative F has the values F(1) = 7 and F(3) = 4? y

a

a+b 2

x b

FIGURE 5 Graph of y = (x − a)(b − x). SOLUTION

We first calculate the area of the parabolic arch: " b a

(x − a)(b − x) d x = −

" b a

(x − a)(x − b) d x = −

" b a

(x 2 − ax − bx + ab) d x

b 1 3 a 2 b 2 x − x − x + abx 3 2 2 a

b 1 2x 3 − 3ax 2 − 3bx 2 + 6abx =− a 6 1 3 = − (2b − 3ab2 − 3b3 + 6ab2 ) − (2a 3 − 3a 3 − 3ba 2 + 6a 2 b) 6 1

= − (−b3 + 3ab2 ) − (−a 3 + 3a 2 b) 6 1 1

= − a 3 + 3ab2 − 3a 2 b − b3 = (b − a)3 . 6 6

=−

264

CHAPTER 5

THE INTEGRAL

The indicated triangle has a base of length b − a and a height of a+b b−a 2 a+b . −a b− = 2 2 2 Thus, the area of the triangle is 1 b−a 2 1 = (b − a)3 . (b − a) 2 2 8 Finally, we note that 4 1 1 (b − a)3 = · (b − a)3 , 6 3 8 as required.

" 3 " 1 f (x) d x, where 49. Calculate Does −2 x n d x get larger or smaller as n increases? Explain graphically. 0 12 − x 2 for x ≤ 2 f (x) = for x > 2 x3 SOLUTION

" 3 −2

f (x) d x =

" 2 −2

f (x) d x +

" 3 2

f (x) d x =

" 2 −2

(12 − x 2 ) d x +

" 3

x3 dx

2

2 1 1 3 12x − x 3 + x 4 3 4 2 −2 1 1 1 1 = 12(2) − (2)3 − 12(−2) − (−2)3 + 34 − 24 3 3 4 4 =

=

128 65 707 + = . 3 4 12

the function f (x) = sin 3x − x. Find the positive root of Further Plot Insights and Challenges

f (x) to three places and use it to find the area under the graph of f (x) in the first quadrant. 51. In this exercise, we generalize the result of Exercise 47 by proving the famous result of Archimedes: For r < s, the area of the shaded region in Figure 6 is equal to four-thirds the area of triangle AC E, where C is the point on the parabola at which the tangent line is parallel to secant line AE. (a) Show that C has x-coordinate (r + s)/2. (b) Show that AB D E has area (s − r )3 /4 by viewing it as a parallelogram of height s − r and base of length C F. (c) Show that AC E has area (s − r )3 /8 by observing that it has the same base and height as the parallelogram. (d) Compute the shaded area as the area under the graph minus the area of a trapezoid and prove Archimedes’s result. y B

C

D

A

F

E

r

r+s 2

x

s

FIGURE 6 Graph of f (x) = (x − a)(b − x). SOLUTION

(a) The slope of the secant line AE is (s − a)(b − s) − (r − a)(b − r ) f (s) − f (r ) = = a + b − (r + s) s −r s −r and the slope of the tangent line along the parabola is f (x) = a + b − 2x. If C is the point on the parabola at which the tangent line is parallel to the secant line AE, then its x-coordinate must satisfy a + b − 2x = a + b − (r + s)

or

x=

r +s . 2

S E C T I O N 5.3


265

(b) Parallelogram AB D E has height s − r and base of length C F. Since the equation of the secant line AE is y = [a + b − (r + s)] (x − r ) + (r − a)(b − r ), the length of the segment C F is r +s r +s r +s (s − r )2 −a b− − [a + b − (r + s)] − r − (r − a)(b − r ) = . 2 2 2 4 )3 Thus, the area of AB D E is (s−r 4 .

(c) Triangle AC E is comprised of AC F and C E F. Each of these smaller triangles has height s−r 2 and base of

)2 length (s−r 4 . Thus, the area of AC E is

1 s − r (s − r )2 1 s − r (s − r )2 (s − r )3 · + · = . 2 2 4 2 2 4 8 (d) The area under the graph of the parabola between x = r and x = s is " s 1 1 3 s 2 (x − a)(b − x) d x = −abx + (a + b)x − x 2 3 r r 1 1 1 1 = −abs + (a + b)s 2 − s 3 + abr − (a + b)r 2 + r 3 2 3 2 3 1 1 = ab(r − s) + (a + b)(s − r )(s + r ) + (r − s)(r 2 + r s + s 2 ), 2 3 while the area of the trapezoid under the shaded region is 1 (s − r ) [(s − a)(b − s) + (r − a)(b − r )] 2 1 = (s − r ) −2ab + (a + b)(r + s) − r 2 − s 2 2 1 1 = ab(r − s) + (a + b)(s − r )(r + s) + (r − s)(r 2 + s 2 ). 2 2 Thus, the area of the shaded region is 1 1 1 2 1 1 1 1 2 1 1 r + r s + s 2 − r 2 − s 2 = (s − r ) r − r s + s 2 = (s − r )3 , (r − s) 3 3 3 2 2 6 3 6 6 which is four-thirds the area of the triangle AC E. 53. Use the method of Exercise 52 to prove that (a) Apply the Comparison Theorem (Theorem 5 in Section 5.2) to the inequality sin x ≤ x (valid for x ≥ 0) to x2 x4 x2 prove ≤ cos x ≤ 1 − + 1− 2 2 24 x2 3 3 1 − ≤ cos x x x x 5≤ 1. x− ≤ sin x ≤ x 2− + (for x ≥ 0) 6 6 120 (b) Apply it again to prove Verify these inequalities for x = 0.1. Why have we specified x ≥ 0 for sin x but not cos x? 3 3 SOLUTION By Exercise 52, t − 1 ≤ tx for≤t sin > 0. Integrating inequality over the interval [0, x], and then 6 t ≤ sin xt − x≤ x (for xthis ≥ 0). 6 solving for cos x, yields: (c) Verify these inequalities for x 1= 20.3. 1 4 x − x ≤ 1 − cos x ≤ 2 24 1 1 − x 2 ≤ cos x ≤ 1 − 2

1 2 x 2 1 2 1 4 x + x . 2 24

2 2 x 4 are all even functions, they also apply for These inequalities apply for x ≥ 0. Since cos x, 1 − x2 , and 1 − x2 + 24 x ≤ 0. Having established that

1−

t2 t2 t4 ≤ cos t ≤ 1 − + , 2 2 24

for all t ≥ 0, we integrate over the interval [0, x], to obtain: x−

x3 x3 x5 ≤ sin x ≤ x − + . 6 6 120

266

CHAPTER 5

THE INTEGRAL 1 x 5 are all odd functions, so the inequalities are reversed for x < 0. The functions sin x, x − 16 x 3 and x − 16 x 3 + 120 Evaluating these inequalities at x = .1 yields

0.995000000 ≤ 0.995004165 ≤ 0.995004167 0.0998333333 ≤ 0.0998334166 ≤ 0.0998334167, both of which are true. 55. Assume that | f (x)| ≤ K for x ∈ [a, b]. Use FTC I to prove that | f (x) − f (a)| ≤ K |x − a| for x ∈ [a, b]. Calculate the next pair of inequalities for sin x and cos x by integrating the results of Exercise 53. Can you guess SOLUTION Let a > b be real numbers, and let f (x) be such that | f (x)| ≤ K for x ∈ [a, b]. By FTC, the general pattern? " x f (t) dt = f (x) − f (a). a

Since f (x) ≥ −K for all x ∈ [a, b], we get: f (x) − f (a) =

" x a

f (t) dt ≥ −K (x − a).

Since f (x) ≤ K for all x ∈ [a, b], we get: f (x) − f (a) =

" x a

f (t) dt ≤ K (x − a).

Combining these two inequalities yields −K (x − a) ≤ f (x) − f (a) ≤ K (x − a), so that, by definition, | f (x) − f (a)| ≤ K |x − a|. (a) Prove that | sin a − sin b| ≤ |a − b| for all a, b (use Exercise 55). (b) Let f (x) = sin(x + a) − sin x. Use part (a) to show that the graph of f (x) lies between the horizontal lines 5.4 y The Fundamental Theorem of Calculus, Part II = ±a. (c) Produce a graph of f (x) and verify part (b) for a = 0.5 and a = 0.2. Preliminary Questions " x 1. What is A(−2), where A(x) = f (t) dt? −2

SOLUTION

By definition, A(−2) =

2. Let G(x) =

" −2

" x t 3 + 1 dt.

−2

f (t) dt = 0.

4

(a) Is the FTC needed to calculate G(4)? (b) Is the FTC needed to calculate G (4)? SOLUTION

! (a) No. G(4) = 44 t 3 + 1 dt = 0. √ (b) Yes. By the FTC II, G (x) = x 3 + 1, so G (4) = 65. 3. Which of the following defines an antiderivative F(x) of f (x) = x 2 satisfying F(2) = 0? " x " 2 " x (a) 2t dt (b) t 2 dt (c) t 2 dt 2 SOLUTION

" x The correct answer is (c):

0

2

t 2 dt.

2

4. True or false? Some continuous functions do not have antiderivatives. Explain. " x SOLUTION False. All continuous functions have an antiderivative, namely f (t) dt. 5. Let G(x) =

a

" x3 sin t dt. Which of the following statements are correct? 4

(a) G(x) is the composite function sin(x 3 ).

S E C T I O N 5.4

(b) G(x) is the composite function A(x 3 ), where A(x) = (c) (d) (e) (f)

The Fundamental Theorem of Calculus, Part II

267

" x sin(t) dt. 4

G(x) is too complicated to differentiate. The Product Rule is used to differentiate G(x). The Chain Rule is used to differentiate G(x). G (x) = 3x 2 sin(x 3 ).

Statements (b), (e), and (f) are correct. " 3 t 3 dt at x = 2. 6. Trick question: Find the derivative of

SOLUTION

1

! Note that the definite integral 13 t 3 dt does not depend on x; hence the derivative with respect to x is 0 for any value of x. SOLUTION

Exercises 1. Write the area function of f (x) = 2x + 4 with lower limit a = −2 as an integral and find a formula for it. SOLUTION

Let f (x) = 2x + 4. The area function with lower limit a = −2 is " x " x A(x) = f (t) dt = (2t + 4) dt. −2

a

Carrying out the integration, we find x " x (2t + 4) dt = (t 2 + 4t) −2

−2

= (x 2 + 4x) − ((−2)2 + 4(−2)) = x 2 + 4x + 4

or (x + 2)2 . Therefore, A(x) = (x + 2)2 . " x (t 2for − the 2) dt. 3. LetFind G(x)a = formula area function of f (x) = 2x + 4 with lower limit a = 0. 1

(a) What is G(1)? (b) Use FTC II to find G (1) and G (2). (c) Find a formula for G(x) and use it to verify your answers to (a) and (b). !x SOLUTION Let G(x) = 1 (t 2 − 2) dt. ! (a) Then G(1) = 11 (t 2 − 2) dt = 0.

(b) Now G (x) = x 2 − 2, so that G (1) = −1 and G (2) = 2. (c) We have x " x 1 3 1 3 1 3 1 5 (t 2 − 2) dt = t − 2t = x − 2x − (1) − 2(1) = x 3 − 2x + . 3 3 3 3 3 1 1 Thus G(x) = 13 x 3 − 2x + 53 and G (x) = x 2 − 2. Moreover, G(1) = 13 (1)3 − 2(1) + 53 = 0, as in (a), and G (1) = −1 and G (2) = 2, as in (b). " x (π /4), where G(x) = " x tan t dt. 5. Find G(1), G (0), and G Find F(0), F (0), and F (3), where F(x) = 1 t 2 + t dt. 0 !1 π SOLUTION By definition, G(1) = 1 tan t dt = 0. By FTC, G (x) = tan x, so that G (0) = tan 0 = 0 and G ( 4 ) = π tan 4 = 1. " x In Exercises 7–14, find formulas for the functions represented du by the integrals. (−2), where H (x) = . Find H (−2) and H " x 2 −2 u + 1 3 u du 7. 2 SOLUTION

F(x) =

" x 2

u 3 du =

1 4 x 1 u = x 4 − 4. 4 2 4

" x"2 x 9. t dtsin u du 1

0

SOLUTION

" x

F(x) =

(t 2 − t) dt

" x2 1

t dt =

2 1 2 x 1 1 t = x4 − . 2 1 2 2

268

CHAPTER 5

THE INTEGRAL

" 5 11. x

(4t − 1) dt

SOLUTION

F(x) =

" x " x 2 sec θ d θ 13. −π /4 cos u du π /4

SOLUTION

F(x) =

" 5 x

5 (4t − 1) dt = (2t 2 − t) = 45 + x − 2x 2 . x

" x −π /4

x sec2 θ d θ = tan θ

−π /4

= tan x − tan(−π /4) = tan x + 1.

In Exercises " √x15–18, express the antiderivative F(x) of f (x) satisfying the given initial condition as an integral. t 3 dt 15. f (x)2= x 4 + 1, F(3) = 0 " x SOLUTION The antiderivative F(x) of f (x) = x 4 + 1 satisfying F(3) = 0 is F(x) = t 4 + 1 dt. 3

17. f (x) = sec x, F(0) = 0 x +1 " x , F(7) = 0 f (x) = 2 x antiderivative +9 SOLUTION The F(x) of f (x) = sec x satisfying F(0) = 0 is F(x) = sec t dt. 0

In Exercises 19–22, calculate the derivative. f (x) = sin(x 3 ), F(−π ) = 0 " x d 19. (t 3 − t) dt dx 0 " x d SOLUTION By FTC II, (t 3 − t) dt = x 3 − x. dx 0 " t d " x d cos 21. 5x d x dt 100 sin(t 2 ) dt dx 1 " t d SOLUTION By FTC II, cos(5x) d x = cos 5t. dt 100 " x " 23. Sketch f (t) dt for each of the functions shown in Figure 10. d thes graph of 1A(x) = du 0 tan ds −2 1 + u2 y

y 2 1 0 −1

2 1 0 −1

x 1

2

3

4

x 1

2

3

4

(B)

(A)

FIGURE 10 SOLUTION

• Remember that A (x) = f (x). It follows from Figure 10(A) that A (x) is constant and consequently A(x) is linear

on the intervals [0, 1], [1, 2], [2, 3] and [3, 4]. With A(0) = 0, A(1) = 2, A(2) = 3, A(3) = 2 and A(4) = 2, we obtain the graph shown below at the left. • Since the graph of y = f (x) in Figure 10(B) lies above the x-axis for x ∈ [0, 4], it follows that A(x) is increasing over [0, 4]. For x ∈ [0, 2], area accumulates more rapidly with increasing x, while for x ∈ [2, 4], area accumulates more slowly. This suggests A(x) should be concave up over [0, 2] and concave down over [2, 4]. A sketch of A(x) is shown below at the right. y

y

3 2.5 2 1.5 1 0.5

4 3 2 1 x 1

Let A(x) =

" x 0

2

3

4

x 1

2

3

4

f (t) dt for f (x) shown in Figure 11. Calculate A(2), A(3), A (2), and A (3). Then find a

formula for A(x) (actually two formulas, one for 0 ≤ x ≤ 2 and one for 2 ≤ x ≤ 4) and sketch the graph of A(x).


S E C T I O N 5.4

269

25. Make a rough sketch of the graph of the area function of g(x) shown in Figure 12. y y = g(x)

x 1

2

3

4

FIGURE 12 SOLUTION The graph of y = g(x) lies above the x-axis over the interval [0, 1], below the x-axis over [1, 3], and above the x-axis over [3, 4]. The corresponding area function should therefore be increasing on (0, 1), decreasing on (1, 3) and increasing on (3, 4). Further, it appears from Figure 12 that the local minimum of the area function at x = 3 should be negative. One possible graph of the area function is the following. y 4 3 2 1 x 1

−1 −2 −3

2

3

4

" x3 " x 27. FindShow G (x), where G(x) = dt. Hint: Consider x ≥ 0 and x ≤ 0 separately. that |t| dt is equal totan12 tx|x|. 3

0

SOLUTION

By combining the FTC and the chain rule, we have G (x) = tan(x 3 ) · 3x 2 = 3x 2 tan(x 3 ).

In Exercises 29–36, calculate the derivative. " x2 t 3 + 3 dt. Find " x 2 G (1), where G(x) = d 0 2 29. sin t dt dx 0 " x2 SOLUTION Let G(x) = sin2 t dt. By applying the Chain Rule and FTC, we have 0

G (x) = sin2 (x 2 ) · 2x = 2x sin2 (x 2 ). " cos s d " 1/x (u 4 − 3u) du ds d−6 sin(t 2 ) dt dx 1 " s SOLUTION Let G(s) = (u 4 − 3u) du. Then, by the chain rule,

31.

−6

" cos s d d (u 4 − 3u) du = G(cos s) = − sin s(cos4 s − 3 cos s). ds −6 ds 33.

" 0 d " d sin0 2 t dt d x x3 sin2 t dt dx x

SOLUTION

Let F(x) =

" x 0

sin2 t dt. Then

" 0 x3

sin2 t dt = −

" x3 0

sin2 t dt = −F(x 3 ). From this,

" 0 d d sin2 t dt = (−F(x 3 )) = −3x 2 F (x 3 ) = −3x 2 sin2 x 3 . d x x3 dx 35.

" x2 d " tan4 t dt d x d√x x √ t dt d x x2 Hint for Exercise 34:F(x) = A(x 4 ) − A(x 2 ), where A(x) =

" x √

t dt

270

CHAPTER 5

THE INTEGRAL SOLUTION

Let " x2

G(x) = √ tan t dt = x

" x2 0

" √x

tan t dt −

tan t dt. 0

Applying the Chain Rule combined with FTC twice, we have √ √ 1 tan( x) G (x) = tan(x 2 ) · 2x − tan( x) · x −1/2 = 2x tan(x 2 ) − √ . 2 2 x " x " x " 3u+9 d 37–38, In Exercises let 2A(x) = f (t) dt and B(x) = f (t) dt, with f (x) as in Figure 13. x + 1 dx 0 2 du −u y 2 1 0 −1 −2

y = f(x) x 1

2

3

4

5

6

FIGURE 13

37. Find the min and max of A(x) on [0, 6]. SOLUTION The minimum values of A(x) on [0, 6] occur where A (x) = f (x) goes from negative to positive. This occurs at one place, where x = 1.5. The minimum value of A(x) is therefore A(1.5) = −1.25. The maximum values of A(x) on [0, 6] occur where A (x) = f (x) goes from positive to negative. This occurs at one place, where x = 4.5. The maximum value of A(x) is therefore A(4.5) = 1.25. " x ffor (t)dt, with (x) asvalid in Figure 39. LetFind A(x)formulas = A(x) andf B(x) on [2,14. 4].

0

(a) (b) (c) (d)

Does A(x) have a local maximum at P? Where does A(x) have a local minimum? Where does A(x) have a local maximum? True or false? A(x) < 0 for all x in the interval shown. y R P

S

x

y = f(x) Q

FIGURE 14 Graph of f (x). SOLUTION

(a) In order for A(x) to have a local maximum, A (x) = f (x) must transition from positive to negative. As this does not happen at P, A(x) does not have a local maximum at P. (b) A(x) will have a local minimum when A (x) = f (x) transitions from negative to positive. This happens at R, so A(x) has a local minimum at R. (c) A(x) will have a local maximum when A (x) = f (x) transitions from positive to negative. This happens at S, so A(x) has a local maximum at S. (d) It is true that A(x) < 0 on I since the signed area from 0 to x is clearly always negative from the figure. " x Find 41–42, the smallest positive point of f (x) is continuous. In Exercises let A(x) = critical f (t) dt, where a " x cos(t 3/2 ) dt F(x) = 41. Area Functions and Concavity Explain why0 the following statements are true. Assume f (x) is differentiable. and determine whether it is a local min or max. (a) If c is an inflection point of A(x), then f (c) = 0. (b) A(x) is concave up if f (x) is increasing. (c) A(x) is concave down if f (x) is decreasing. SOLUTION

(a) If x = c is an inflection point of A(x), then A (c) = f (c) = 0. (b) If A(x) is concave up, then A (x) > 0. Since A(x) is the area function associated with f (x), A (x) = f (x) by FTC II, so A (x) = f (x). Therefore f (x) > 0, so f (x) is increasing.

S E C T I O N 5.4


271

(c) If A(x) is concave down, then A (x) < 0. Since A(x) is the area function associated with f (x), A (x) = f (x) by FTC II, so A (x) = f (x). Therefore, f (x) < 0 and so f (x) is decreasing. " x f (t) dt, (x) as Figure 15. Determine: 43. LetMatch A(x) = the property of with A(x) fwith theincorresponding property of the graph of f (x). Assume f (x) is differentiable. 0

(a) (b) (c) (d)

The intervals on which A(x) is increasing and decreasing Areavalues function A(x) A(x) has a local min or max The x where (a) A(x) is decreasing. The inflection points of A(x) (b) intervals A(x) has where a localA(x) maximum. The is concave up or concave down (c) A(x) is concave up. y (d) A(x) goes from concave up to concave down. y = f(x) Graph of f(x) (i) Lies below the x-axis. (ii) Crosses the x-axis from positive to negative. 2 4 6 8 (iii) Has a local maximum. FIGURE 15 (iv) f (x) is increasing.

x 10

12

SOLUTION

(a) A(x) is increasing when A (x) = f (x) > 0, which corresponds to the intervals (0, 4) and (8, 12). A(x) is decreasing when A (x) = f (x) < 0, which corresponds to the intervals (4, 8) and (12, ∞). (b) A(x) has a local minimum when A (x) = f (x) changes from − to +, corresponding to x = 8. A(x) has a local maximum when A (x) = f (x) changes from + to −, corresponding to x = 4 and x = 12. (c) Inflection points of A(x) occur where A (x) = f (x) changes sign, or where f changes from increasing to decreasing or vice versa. Consequently, A(x) has inflection points at x = 2, x = 6, and x = 10. (d) A(x) is concave up when A (x) = f (x) is positive or f (x) is increasing, which corresponds to the intervals (0, 2) and (6, 10). Similarly, A(x) is concave down when f (x) is decreasing, which corresponds to the intervals (2, 6) and (10, ∞). " x " f (t) dt are decreasing. 45. Sketch the graph 2of an increasing function fx(x) such that both f (x) and A(x) = f (t) dt. Let f (x) = x − 5x − 6 and F(x) = 0 " x 0 (x) is points (x) must whether (a) Find the of F(x) andf determine they are local minima or maxima. SOLUTION If fcritical decreasing, then be negative. Furthermore, if A(x) = f (t) dt is decreasing, 0 (b) Find the points of inflection of F(x) and determine whether the concavity changes from up to down or vice then A (x) = f (x) must also be negative. Thus, we need a function which is negative but increasing and concave down. versa. The graph of one such function is shown below. (c) Plot f (x) and F(x) on the same set of axes and confirm your answers to (a) and (b). y x

47.

Find the smallest positive inflection point of " x Figure 16 shows the graph of f (x) = x sin x. "Letx F(x) = t sin t dt. 0 cos(t 3/2 ) dt F(x) = 0

3/2 ) to determine whether the concavity changes from up to down or vice versa at this point of Use a(a)graph of ythe = local cos(xmaxima Locate and absolute maximum of F(x) on [0, 3π ]. inflection. (b) Justify graphically that F(x) has precisely one zero in the interval [π , 2π ]. SOLUTION points F(x) where F (x) = 0. By FTC, F (x) = cos(x 3/2 ), so F (x) = ]? (c) How Candidate many zerosinflection does F(x) have of in [0, 3πoccur 1/2 3/2 −(3/2)x ). Finding theof smallest positive F (x) 0, we get: the concavity changes from up to and, forofeach one,=state whether (d) Findsin(x the inflection points F(x) on [0, 3π ]solution down or vice versa. −(3/2)x 1/2 sin(x 3/2 ) = 0

sin(x 3/2 ) = 0 x 3/2 =

(since x > 0)

π

x = π 2/3 ≈ 2.14503. From the plot below, we see that F (x) = cos(x 3/2 ) changes from decreasing to increasing at π 2/3 , so F(x) changes from concave down to concave up at that point.

272

CHAPTER 5

THE INTEGRAL y 1 0.5 x 1

− 0.5

2

3

−1

49. Determine the function g(x) and all"values of c such that x 2 " Determine f (x), assuming that f (t) dt x is equal to x + x. 2 0 g(t) dt = x + x − 6 c

SOLUTION

By the FTC II we have g(x) =

d 2 (x + x − 6) = 2x + 1 dx

and therefore, " x c

g(t) dt = x 2 + x − (c2 + c)

We must choose c so that c2 + c = 6. We can take c = 2 or c = −3.

Further Insights and Challenges

" b Proof of FTC proof that in the textf (t) assumes that f−(x) is increasing. it for allFTC continuous functions, 51. Proof of FTC I II FTCThe I asserts dt = F(b) F(a) if F (x) =Tof prove (x). Assume II and give a new a " x and maximum of f (x) on [x, x + h] (Figure 17). The continuity of f (x) let m(h) and M(h) denote the minimum m(h)Set = lim (x). dt.Show that for h > 0, proofimplies of FTCthat I aslim follows. A(x)M(h) = = f f(t) h→0

h→0

a

(a) Show that F(x) = A(x) + C for some constant. hm(h) " ≤ b A(x + h) − A(x) ≤ h M(h) f (t) dt. (b) Show that F(b) − F(a) = A(b) − A(a) = For h < 0, the inequalities are reversed. Provea that A (x) = f (x). ! x SOLUTION Let F (x) = f (x) and A(x) = a f (t) dt. (a) Then by the FTC, Part II, A (x) = f (x) and thus A(x) and F(x) are both antiderivatives of f (x). Hence F(x) = A(x) + C for some constant C. (b) F(b) − F(a) = ( A(b) + C) − ( A(a) + C) = A(b) − A(a) " b " a " b " b = f (t) dt − f (t) dt = f (t) dt − 0 = f (t) dt a

a

a

a

which proves the FTC, Part I. " b " x 53. Find the values a ≤ b such that (x 2 − 9) d x has minimal value. f (t) dt is an antiderivative Can Every Antiderivative Bea Expressed as an Integral? The area function a !x 2 2 of f (x) for every value of a. However, not all antiderivatives are obtained in this way. The general antiderivative of SOLUTION Let a be given, and let Fa (x) = a (t − 9) dt. Then Fa (x) = x − 9, and the critical points are x = ±3. 1 x 2 +C. Show that F(x) is an area function if C ≤ 0 but not if C > 0. f (x) = x is F(x) = Because Fa (−3) = −6 2and Fa (3) = 6, we see that Fa (x) has a minimum at x = 3. Now, we find a minimizing !3 2 !3 2 2 a (x − 9) d x. Let G(x) = x (x − 9) d x. Then G (x) = −(x − 9), yielding critical points x = 3 or x = −3. With x = −3, 3 " 3 1 3 G(−3) = (x 2 − 9) d x = x − 9x = −36. 3 −3 −3 With x = 3, G(3) = Hence a = −3 and b = 3 are the values minimizing

" 3 3

" b a

(x 2 − 9) d x = 0.

(x 2 − 9) d x.

S E C T I O N 5.5

Net or Total Change as the Integral of a Rate

273

5.5 Net or Total Change as the Integral of a Rate Preliminary Questions 1. An airplane makes the 350-mile trip from Los Angeles to San Francisco in 1 hour. Assuming that the plane’s velocity " 1 at time t is v(t) mph, what is the value of the integral v(t) dt? 0

!1

SOLUTION The definite integral 0 v(t) dt represents the total distance traveled by the airplane during the one hour ! flight from Los Angeles to San Francisco. Therefore the value of 01 v(t) dt is 350 miles.

2. A hot metal object is submerged in cold water. The rate at which the object cools (in degrees per minute) is a function " T f (t) dt? f (t) of time. Which quantity is represented by the integral 0

! The definite integral 0T f (t) dt represents the total drop in temperature of the metal object in the first T minutes after being submerged in the cold water. SOLUTION

3. (a) (b) (c) (d)

Which of the following quantities would be naturally represented as derivatives and which as integrals? Velocity of a train Rainfall during a 6-month period Mileage per gallon of an automobile Increase in the population of Los Angeles from 1970 to 1990

SOLUTION Quantities (a) and (c) involve rates of change, so these would naturally be represented as derivatives. Quantities (b) and (d) involve an accumulation, so these would naturally be represented as integrals.

4. Two airplanes take off at t = 0 from the same place and in the same direction. Their velocities are v1 (t) and v2 (t), respectively. What is the physical interpretation of the area between the graphs of v1 (t) and v2 (t) over an interval [0, T ]? SOLUTION The area between the graphs of v1 (t) and v2 (t) over an interval [0, T ] represents the difference in distance traveled by the two airplanes in the first T hours after take off.

Exercises 1. Water flows into an empty reservoir at a rate of 3,000 + 5t gal/hour. What is the quantity of water in the reservoir after 5 hours? SOLUTION

The quantity of water in the reservoir after five hours is 5 2 5 30125 (3000 + 5t) dt = 3000t + t = = 15,062.5 gallons. 2 2 0 0

" 5

2 perv(t) day.=Find insect after 3 3. A population of insects increases at a moving rate of 200 10t +line 0.25t Find the displacement of a particle in a+ straight withinsects velocity 4t −the 3 ft/s overpopulation the time interval days,[2, assuming that there are 35 insects at t = 0. 5]. SOLUTION

The increase in the insect population over three days is " 3

1 200 + 10t + t 2 dt = 4 0

200t + 5t 2 +

1 3 3 2589 t = = 647.25. 12 4 0

Accordingly, the population after 3 days is 35 + 647.25 = 682.25 or 682 insects. 2 − t bicycles per week (t in weeks). How many bicycles were 5. A factory produces bicycles at a rate of 95 +is0.1t A survey shows that a mayoral candidate gaining votes at a rate of 2,000t + 1,000 votes per day, where t is produced from day 8 to 21? the number of days since she announced her candidacy. How many supporters will the candidate have after 60 days,

assuming The that rate she had no supporters at t = = 95 0? + 10 t 2 SOLUTION of production is r (t) 1

− t bicycles per week and the period between days 8 and 21 corresponds to the second and third weeks of production. Accordingly, the number of bikes produced between days 8 and 21 is " 3 " 3 1 3 1 2 3 2803 1 2 r (t) dt = ≈ 186.87 95 + t − t dt = 95t + t − t = 10 30 2 15 1 1 1 or 187 bicycles. 7. A cat falls from a tree (with zero initial velocity) at time t = 0. How far does the cat fall between t = 0.5 and Find the displacement over the time interval [1, 6] of a helicopter whose (vertical) velocity at time t is v(t) = t = 1 s? Use Galileo’s formula v(t) = −32t ft/s. 0.02t 2 + t ft/s.

274

CHAPTER 5

THE INTEGRAL

Given v(t) = −32 ft/s, the total distance the cat falls during the interval [ 12 , 1] is

SOLUTION

" 1

|v(t)| dt =

1/2

" 1 1/2

1 32t dt = 16t 2

= 16 − 4 = 12 ft.

1/2

In Exercises 9–12, assume that with a particle in avelocity straight100 linem/s. withUse given total−displacement and A projectile is released initial moves (vertical) thevelocity. formulaFind v(t) the = 100 9.8t for velocity totaltodistance traveled over the time interval, and draw a motion diagram like Figure 3 (with distance and time labels). determine the distance traveled during the first 15 s. 9. 12 − 4t ft/s,

[0, 5]

" 5

Total displacement is given by

SOLUTION

0

" 5 0

|12 − 4t| dt =

" 3 0

5 (12 − 4t) dt = (12t − 2t 2 ) = 10 ft, while total distance is given by 0

(12 − 4t) dt +

" 5 3

3 5 (4t − 12) dt = (12t − 2t 2 ) + (2t 2 − 12t) = 26 ft. 0

3

The displacement diagram is given here. t =5 t =3 t =0 Distance 0

11. t −2 − 1 m/s, [0.5, 2] 32 − 2t 2 ft/s, [0, 6] SOLUTION

10

" 2

Total displacement is given by

.5

18

2 (t −2 − 1) dt = (−t −1 − t) = 0 m, while total distance is given by .5

1 2 " 2 " 2 " 1 −2 −2 −2 −1 −1 (t − 1) dt + (1 − t ) dt = (−t − t) + (t + t ) = 1 m. t − 1 dt = .5

.5

.5

1

1

The displacement diagram is given here. t =2 t =1 t =0 Distance 0

0.5

13. The rate (in liters per minute) at which water drains from a tank is recorded at half-minute intervals. Use the average cos t m/s, [0, 4π ] of the left- and right-endpoint approximations to estimate the total amount of water drained during the first 3 min.

SOLUTION

t (min)

0

0.5

1

1.5

2

2.5

3

l/min

50

48

46

44

42

40

38

Let t = .5. Then R N = .5(48 + 46 + 44 + 42 + 40 + 38) = 129.0 liters L N = .5(50 + 48 + 46 + 44 + 42 + 40) = 135.0 liters

The average of R N and L N is 12 (129 + 135) = 132 liters.

" t2 of a car is recorded at half-second intervals (in feet per second). the average thechange left- and 15. LetThe a(t)velocity be the acceleration of an object in linear motion at time t. Explain why Use a(t) dt is theofnet in right-endpoint approximations to estimate the total distance traveled during the first 4t1s. velocity over [t1 , t2 ]. Find the net change in velocity over [1, 6] if a(t) = 24t − 3t 2 ft/s2 . t 0 of 0.5 1 in 1.5 2.5at time 3 t. 3.5 4 be the velocity of the object. SOLUTION Let a(t) be the acceleration an object linear 2motion Let v(t) We know that v (t) = a(t). By FTC, v(t) 0 12 20 29 38 44 32 35 30 t2 " t2 a(t) dt = (v(t) + C) = v(t2 ) + C − (vt1 + C) = v(t2 ) − v(t1 ), t1

t1

which is the net change in velocity over [t1 , t2 ]. Let a(t) = 24t − 3t 2 . The net change in velocity over [1, 6] is 6 " 6 (24t − 3t 2 ) dt = (12t 2 − t 3 ) = 205 ft/s. 1

1

Show that if acceleration a is constant, then the change in velocity is proportional to the length of the time interval.

Net or Total Change as the Integral of a Rate

S E C T I O N 5.5

275

17. The traffic flow rate past a certain point on a highway is q(t) = 3,000 + 2,000t − 300t 2 , where t is in hours and t = 0 is 8 AM. How many cars pass by during the time interval from 8 to 10 AM? SOLUTION

The number of cars is given by " 2 0

q(t) dt =

" 2 0

2 (3000 + 2000t − 300t 2 ) dt = 3000t + 1000t 2 − 100t 3 0

= 3000(2) + 1000(4) − 100(8) = 9200 cars. 19. Carbon Tax To encourage manufacturers to reduce pollution, a carbon tax on each ton of CO2 released into the 0.6x the + 350 dollars. What is the cost Suppose that the marginal of producing video is 0.001x 2 −study atmosphere has been proposed. Tocost model the effectsx of suchrecorders a tax, policymakers marginal cost of abatement ofdefined producing 300cost unitsofifincreasing the setup cost $20,000 (see 4)? production 300 units,tons—Figure what is the cost fromExample x to x + 1 If tons (in units isofset tenatthousand 4). B(x), as the CO2isreduction " 3 of producing 20 additional units? B(t) dt? Which quantity is represented by 0 y Dollars/ton

100 75 50 25 x 1 2 3 Tons reduced (in ten thousands)

FIGURE 4 Marginal cost of abatement B(x).

" 3 SOLUTION

The quantity 0

sphere by 3 tons.

B(t) dt represents the total cost of reducing the amount of CO2 released into the atmo-

Figure shows the migration rate per M(t) of time. Ireland during theofperiod This×is10the rate at Figure which 21. 9 J/hour. Power is the 6rate of energy consumption unit A megawatt power1988–1998. is 106 W or 3.6 people (in thousands persupplied year) move in California or out of the country. 5 shows the"power by the power grid over a typical 1-day period. Which quantity is represented 1991 by the area under the graph? (a) What does M(t) dt represent? 1988

Migration (in thousands)

(b) Did migration over the 11-year period 1988–1998 result in a net influx or outflow of people from Ireland? Base your answer on a rough estimate of the positive and negative areas involved. (c) During which year could the Irish prime minister announce, “We are still losing population but we’ve hit an inflection point—the trend is now improving.” 30 20 10 0 −10 −20 −30 −40 −50

1994 1988

1990

1992

1996

1998

2000

FIGURE 6 Irish migration rate (in thousands per year). SOLUTION

" 1991

(a) The amount

M(t) dt represents the net migration in thousands of people during the period from 1988–1991. 1988

(b) Via linear interpolation and using the midpoint approximation with n = 10, the migration (in thousands of people) over the period 1988 – 1998 is estimated to be 1 · (−43 − 33.5 − 12 + 0.5 − 2.5 − 6 − 3.5 + 3 + 11.5 + 19) = −66.5 That is, there was a net outflow of 66,500 people from Ireland during this period. (c) “The trend is now improving” implies that the population is decreasing, but that the rate of decrease is approaching zero. The population is decreasing with an improving trend in part of the years 1989, 1990, 1991, 1993, and 1994. “We’ve hit an inflection point” implies that the rate of population has changed from decreasing to increasing. There are two years in which the trend improves after it was getting worse: 1989 and 1993. During only one of these, 1989, was the population declining for the entire previous year. 23. Heat Capacity The heat capacity C(T ) of a substance is the amount of energy (in joules) required to raise the Figure 7 shows the graph of Q(t), the rate of retail truck sales in the United States (in thousands sold per year). temperature of 1 g by 1◦ C at temperature T . (a) What does the area under the graph over the interval [1995, 1997] represent? (b) Express the total number of trucks sold in the period 1994–1997 as an integral (but do not compute it). (c) Use the following data to compute the average of the right- and left-endpoint approximations as an estimate for

276

CHAPTER 5

THE INTEGRAL

(a) Explain why the energy required to raise the temperature from T1 to T2 is the area under the graph of C(T ) over [T1 , T2 ]. √ (b) How much energy is required to raise the temperature from 50 to 100◦ C if C(T ) = 6 + 0.2 T ? SOLUTION

(a) Since C(T ) is the energy required to raise the temperature of one gram of a substance by one degree when its temperature is T , the total energy required to raise the temperature from T1 to T2 is given by the definite integral " T2 C(T ) d T . As C(T ) > 0, the definite integral also represents the area under the graph of C(T ). T1 √ (b) If C(T ) = 6 + .2 T = 6 + 15 T 1/2 , then the energy required to raise the temperature from 50◦ C to 100◦ C is ! 100 50 C(T ) d T or " 100 50

6+

1 1/2 T 5

2 2 3/2 100 3/2 − 6(50) + 2 (50)3/2 = 6(100) + T (100) 15 15 15 50 √ 1300 − 100 2 = ≈ 386.19 Joules 3

dT =

6T +

In Exercises 24 and 25, consider the following. Paleobiologists have studied the extinction of marine animal families during the phanerozoic period, which began 544 million years ago. A recent study suggests that the extinction rate r (t) may be modeled by the function r (t) = 3,130/(t + 262) for 0 ≤ t ≤ 544. Here, t is time elapsed (in millions of years) since the beginning of the phanerozoic period. Thus, t = 544 refers to the present time, t = 540 is 4 million years ago, etc. the total number of extinct families from t = 0 to the present, using M N with N = 544. Use REstimate N or L N with N = 10 (or their average) to estimate the total number of families that became extinct in the periods 100 t≤ 150 and 350 ≤ t ≤ 400. SOLUTION We≤are estimating

25.

" 544 0

using M N with N = 544. If N = 544, t = M N = t

3130 dt (t + 262)

544 − 0 = 1 and {ti∗ }i=1,...N = it − (t/2) = i − 12 . 544

N #

r (ti∗ ) = 1 ·

i=1

544 #

3130 = 3517.3021. 261.5 +i i=1

Thus, we estimate that 3517 families have become extinct over the past 544 million years. output is the rate R of volume of blood pumped by the heart per unit time (in liters per minute). Doctors Further Cardiac Insights and Challenges measure R by injecting A mg of dye into a vein leading into the heart at t = 0 and recording the concentration c(t) 27. A particle located at the origin at t = 0 moves along the x-axis with velocity v(t) = (t + 1)−2 . Show that the of dye (in milligrams per liter) pumped out at short regular time intervals (Figure 8). particle will never pass the point x = 1.

SOLUTION The particle’s velocity is v(t) = s (t) = (t + 1)−2 , an antiderivative for which is F(t) = −(t + 1)−1 . Hence its position at time t is " tout in a small timetinterval [t, t + t] is approximately Rc(t)t. Explain why. (a) The quantity of dye pumped 1 " T s (u) du = F(u) = F(t) − F(0) = 1 − 0, then x a x 1 (b) Show that the areas under the hyperbola over the intervals [1, 2], [2, 4], [4, 8], . . . are all equal.

Chapter Review Exercises y

283

y

1

1

y = 1 − x2

y = cos2 u

x

p 2

1 (A)

u

(B)

FIGURE 4

!1 SOLUTION The area of the region in Figure 4(A) is given by 0 1 − x 2 d x. Let x = sin u. Then d x = cos u du and 1 − x 2 = 1 − sin2 u = cos u. Hence, " 1 " π /2 " π /2 1 − x2 dx = cos u · cos u du = cos2 u du. 0

0

0

This last integral represents the area of the region in Figure 4(B). The two regions in Figure 4 therefore have the same area. ! π /2 Let’s now focus on the definite integral 0 cos2 u du. Using the trigonometric identity cos2 u = 12 (1 + cos 2u), we have π /2 " " π /2 1 π /2 1 1 π 1 π cos2 u du = 1 + cos 2u du = = · −0= . u + sin 2u 2 0 2 2 2 2 4 0 0 83. Area of an Ellipse Prove the formula A = π ab for the area of the ellipse with equation Area of a Circle The number π is defined as one-half the circumference of the unit circle. Prove that the area of a circle of radius r is A = π r 2 . The case r = 1x 2follows y 2 from Exercise 81. Prove it for all r > 0 by showing that + =1 " r a2 b2 " 1 " a r2 − x2 dx = r2 1 − x 2 d x. 2 0 0 use the formula for the area of a circle (Figure 5). 1 − (x/a) d x, change variables, and Hint: Show that A = 2b −a

y b

−a

a

x

−b

FIGURE 5 Graph of 2

x2 y2 + 2 = 1. 2 a b

2

Consider the ellipse with equation x 2 + y2 = 1; here a, b > 0. The area between the part of the ellipse in a b

!a 2 f (x) d x. By symmetry, the part of the elliptical the upper half-plane, y = f (x) = b2 1 − x 2 , and the x-axis is −a SOLUTION

a

region in the lower half-plane has the same area. Accordingly, the area enclosed by the ellipse is " a " a " a x2 2 f (x) d x = 2 b2 1 − 2 d x = 2b 1 − (x/a)2 d x a −a −a −a Now, let u = x/a. Then x = au and a du = d x. Accordingly, " a " 1

x 2

π = π ab 2b 1− d x = 2ab 1 − u 2 du = 2ab a 2 −a −1 !1 Here we recognized that −1 1 − u 2 du represents the area of the upper unit semicircular disk, which by Exercise 81 is 2( π4 ) = π2 .

CHAPTER REVIEW EXERCISES In Exercises 1–4, refer to the function f (x) whose graph is shown in Figure 1.

284

CHAPTER 5

THE INTEGRAL y 3 2 1 x 1

2

3

4

FIGURE 1

1. Estimate L 4 and M4 on [0, 4]. SOLUTION

and

With n = 4 and an interval of [0, 4], x = 4−0 4 = 1. Then, 1 5 23 L 4 = x( f (0) + f (1) + f (2) + f (3)) = 1 +1+ +2 = 4 2 4 1 3 5 7 1 9 9 M4 = x f + f + f + f =1 +2+ + = 7. 2 2 2 2 2 4 4

Estimate R4 , [a, L 4 ,b]and is larger than 3. Find an interval on M which R4 3]. 4 on [1,

" b a

f (x) d x. Do the same for L 4 .

! In general, R N is larger than ab f (x) d x on any interval [a, b] over which f (x) is increasing. Given the ! graph of f (x), we may take [a, b] = [0, 2]. In order for L 4 to be larger than ab f (x) d x, f (x) must be decreasing over the interval [a, b]. We may therefore take [a, b] = [2, 3]. SOLUTION

In Exercises 5–8, let f (x) = x 2 + 4x. Justify " 2 Sketch the graph of f (x) and the corresponding rectangles 5. Calculate R6 , M6 , and L 6 for f (x) on the interval 9 3 [1, 4]. f (x) d x ≤ ≤ for each approximation. 2 4 1

SOLUTION

Let f (x) = x 2 + 4x. A uniform partition of [1, 4] with N = 6 subintervals has x =

4−1 1 = , 6 2

and x ∗j = a +

x j = a + jx = 1 +

j−

j , 2

j 1 3 x = + . 2 4 2

Now, R6 = x 1 = 2

6 #

f (x j ) =

j=1

1 2

3 5 7 f + f (2) + f + f (3) + f + f (4) 2 2 2

33 65 105 463 + 12 + + 21 + + 32 = . 4 4 4 8

The rectangles corresponding to this approximation are shown below. y 35 30 25 20 15 10 5 x 1

2

3

4

Next, M6 = x 1 = 2

6 #

j=1

f (x ∗j ) =

1 2

5 7 9 11 13 15 f + f + f + f + f + f 4 4 4 4 4 4

105 161 225 297 377 465 + + + + + 16 16 16 16 16 16

=

The rectangles corresponding to this approximation are shown below.

1630 815 = . 32 16

Chapter Review Exercises y 35 30 25 20 15 10 5 x 1

2

3

4

Finally, L 6 = x

5 #

=

f (x j ) =

j=0

1 2

f (1) + f

3 5 7 + f (2) + f + f (3) + f 2 2 2

1 33 65 105 5+ + 12 + + 21 + 2 4 4 4

=

355 . 8

The rectangles corresponding to this approximation are shown below. y 35 30 25 20 15 10 5 x 1

2

3

4

" 2 " 4 7. Find a formula for L N for f (x) on [0, 2] and compute f (x) d x by taking the limit. Find a formula for R N for f (x) on [1, 4] and compute f (x) d x by taking the limit. 0 SOLUTION

1 Let f (x) = x 2 + 4x and N be a positive integer. Then

x =

2 2−0 = N N

and x j = a + jx = 0 +

2j 2j = N N

for 0 ≤ j ≤ N . Thus, L N = x

N −1 #

f (x j ) =

j=0

=

−1 N −1 −1 # 4 j2 2 N# 8j 8 N# 2 + 16 + j j = N j=0 N 2 N N 3 j=0 N 2 j=0

32 12 4 4(N − 1)(2N − 1) 8(N − 1) + . = + + N 3 N 3N 2 3N 2

Finally, " 2 0

4 32 12 + + N N →∞ 3 3N 2

f (x) d x = lim

=

32 . 3

9. Calculate R6 , M6 , and L 6 for f (x)" =x (x 2 + 1)−1 on the interval [0, 1]. Use FTC I to evaluate A(x) = f (t) dt. SOLUTION Let f (x) = (x 2 + 1)−1 . A uniform partition of [0, 1] with N = 6 subintervals has −2 x =

1 1−0 = , 6 6

and x ∗j = a +

j−

x j = a + jx =

j , 6

1 2j − 1 x = . 2 12

Now, R6 = x

6 # j=1

f (x j ) =

1 6

1 1 1 2 5 f + f + f + f + f + f (1) 6 3 2 3 6

285

286

CHAPTER 5

THE INTEGRAL

=

1 6

36 9 4 9 36 1 + + + + + 37 10 5 13 61 2

≈ 0.742574.

Next, M6 = x =

1 6

6 #

f (x ∗j ) =

j=1

1 6

1 1 5 7 3 11 f + f + f + f + f + f 12 4 12 12 4 12

144 16 144 144 16 144 + + + + + 145 17 169 193 25 265

≈ 0.785977.

Finally, L 6 = x

5 #

=

f (x j ) =

j=0

1 6

1 1 1 2 5 + f + f + f + f 6 3 2 3 6

f (0) + f

1 36 9 4 9 36 1+ + + + + 6 37 10 5 13 61

≈ 0.825907.

11. Which approximation to the area is represented by the shaded rectangles in Figure 3? Compute R5 and L 5 . Let R N be the N th right-endpoint approximation for f (x) = x 3 on [0, 4] (Figure 2). y 64(N + 1)2 . (a) Prove that R N = 30 N2 (b) Prove that the area of the region below the right-endpoint rectangles and above the graph is equal to 18

64(2N + 1) . N2

6

x 1

2

3

4

5

FIGURE 3 SOLUTION There are five rectangles and the height of each is given by the function value at the right endpoint of the subinterval. Thus, the area represented by the shaded rectangles is R5 . From the figure, we see that x = 1. Then

R5 = 1(30 + 18 + 6 + 6 + 30) = 90

and

L 5 = 1(30 + 30 + 18 + 6 + 6) = 90.

In Exercises 13–34, evaluate the integral. Calculate any two Riemann sums for f (x) = x 2 on the interval [2, 5], but choose partitions with at least five " subintervals of unequal widths and intermediate points that are neither endpoints nor midpoints. 13. (6x 3 − 9x 2 + 4x) d x " 3 (6x 3 − 9x 2 + 4x) d x = x 4 − 3x 3 + 2x 2 + C. SOLUTION 2 " " 13 − 1)2 d x 15. (2x (4x 3 − 2x 5 ) d x " " 0 4 (2x 3 − 1)2 d x = (4x 6 − 4x 3 + 1) d x = x 7 − x 4 + x + C. SOLUTION 7 " 4 x" 4+ 1 17. dx x 2(x 5/2 − 2x −1/2 ) d x 1 " 4 " 1 x +1 SOLUTION d x = (x 2 + x −2 ) d x = x 3 − x −1 + C. 2 3 x " 4 " 42 |x − 9| d x 19. r −2 dr −1 1

SOLUTION

" 4 −1

|x 2 − 9| d x =

" 3 −1

(9 − x 2 ) d x +

= (27 − 9) − −9 + " 21.

" 32 θ d θ csc [t] dt 1

" 4 3

1 3

(x 2 − 9) d x =

+

9x −

4 1 3 3 1 3 x + x − 9x 3 3 −1 3

64 − 36 − (9 − 27) = 30. 3


" csc2 θ d θ = − cot θ + C.

SOLUTION

"

" π2/4 (9t − 4) dt sec sec t tan t dt 0 SOLUTION Let u = 9t − 4. Then du = 9dt and " " 1 1 1 sec2 u du = tan u + C = tan(9t − 4) + C. sec2 (9t − 4) dt = 9 9 9

23.

"

" π−/34)11 dt (9t sin 4θ d θ 0 SOLUTION Let u = 9t − 4. Then du = 9dt and " " 1 1 12 1 u 11 du = (9t − 4)11 dt = u +C = (9t − 4)12 + C. 9 108 108

25.

"

" 22 (3 sin θ ) cos(3θ ) d θ 4y + 1 d y 6 SOLUTION Let u = sin(3θ ). Then du = 3 cos(3θ )d θ and " " 1 1 1 u 2 du = u 3 + C = sin3 (3θ ) + C. sin2 (3θ ) cos(3θ ) d θ = 3 9 9

27.

" 29.

(2x " π3/2+ 3x) d x 2 )5 θ ) sin θ d θ (3x 4 +sec 9x2 (cos 0

SOLUTION

Let u = 3x 4 + 9x 2 . Then du = (12x 3 + 18x) d x = 6(2x 3 + 3x) d x and "

" 31.

1 (2x 3 + 3x) d x = 6 (3x 4 + 9x 2 )5

"

u −5 du = −

1 −4 1 u + C = − (3x 4 + 9x 2 )−4 + C. 24 24

√ " −2 sin θ 412x − cos d xθ dθ −4 (x 2 + 2)3

SOLUTION

" 33.

Let u = 4 − cos θ . Then du = sin θ d θ and " " √ 2 2 sin θ 4 − cos θ d θ = u 1/2 du = u 3/2 + C = (4 − cos θ )3/2 + C. 3 3

/3 + sin y" π2y 3 dθ y 0

SOLUTION

dθ cos2/3 θ Let u = 2y + 3. Then du = 2d y, y = 12 (u − 3) and " " " √ 1 1 (u − 3) u du = (u 3/2 − 3u 1/2 ) du y 2y + 3 d y = 4 4 1 1 2 5/2 1 − 2u 3/2 + C = = u (2y + 3)5/2 − (2y + 3)3/2 + C. 4 5 10 2

35. Combine to write as a single integral " 8 √ " 8 t 2 t + 8 dt 1

0 SOLUTION

f (x) d x +

" 0 −2

f (x) d x +

" 6 8

f (x) d x

First, rewrite " 8 0

f (x) d x =

" 6 0

f (x) d x +

" 8 6

f (x) d x

and observe that " 6 8

f (x) d x = −

" 8 6

f (x) d x.

287

288

CHAPTER 5

THE INTEGRAL

Thus, " 8 0

f (x) d x +

" 6 8

f (x) d x =

" 6 0

f (x) d x.

Finally, " 8 0

f (x) d x +

" 0 −2

f (x) d x +

" 6 8

f (x) d x =

" 6 0

f (x) d x +

" 0 −2

f (x) d x =

" 6 −2

f (x) d x.

" x t dt " x . 37. Find inflection points of A(x) = Let A(x) = f (x) d x, where + is 1 the function shown in Figure 4. Indicate on the graph of f where the 3 ft 2(x) SOLUTION Let local minima,

0

maxima, and points of inflection of A(x) occur and identify the intervals where A(x) is increasing, decreasing, concave up, or concave down. " x t dt . A(x) = 3 t2 + 1

Then x A (x) = 2 x +1 and A (x) =

(x 2 + 1)(1) − x(2x) 1 − x2 = 2 . 2 2 (x + 1) (x + 1)2

Thus, A(x) is concave down for |x| > 1 and concave up for |x| < 1. A(x) therefore has inflection points at x = ±1. 2 (in thousands of gallons per hour), 39. On A a typical city watert at ratemoves of r (t) = velocity 100 + 72t − as 3t shown particleday, startsa at theconsumes origin at time = the 0 and with v(t) in Figure 5. where t is the number of hours past midnight. What is the daily water consumption? How much water is consumed (a) How many times does the particle return to the origin in the first 12 s? between 6 PM and midnight? (b) Where is the particle located at time t = 12? SOLUTION With a consumption rate of r (t) = 100 + 72t − 3t 2 thousand gallons per hour, the daily consumption of (c) At which time t is the particle’s distance to the origin at a maximum? water is " 24 24 (100 + 72t − 3t 2 ) dt = 100t + 36t 2 − t 3 = 100(24) + 36(24)2 − (24)3 = 9312, 0

0

or 9.312 million gallons. From 6 PM to midnight, the water consumption is " 24 18

24

(100 + 72t − 3t 2 ) dt = 100t + 36t 2 − t 3 18

= 100(24) + 36(24)2 − (24)3 − 100(18) + 36(18)2 − (18)3 = 9312 − 7632 = 1680,

or 1.68 million gallons. 41. Cost engineers at NASA have the task of projecting the cost P of major space projects. It has been found that the per bicycle), which The learning curve for producing bicycles certain = 12x, −1/5 cost C of developing a projection increases with Pinata the rate factory dC/d Pis≈L(x) 21P −0.65 where(in C hours is in thousands of dollars means that it takes a bike mechanic L(n) hours to assemble the nth bicycle. If 24 bicycles are produced, and P in millions of dollars. What is the cost of developing a projection for a project whose cost turns out to be how P =long $35 does it take to produce the second batch of 12? million? SOLUTION Assuming it costs nothing to develop a projection for a project with a cost of $0, the cost of developing a projection for a project whose cost turns out to be $35 million is

" 35 0

35 21P −0.65 d P = 60P 0.35 = 60(35)0.35 ≈ 208.245, 0

or $208,245. 43. Let fof(x) a positive increasing continuous function b], where 0 ≤ a 0) 4 # 1 N →∞ N k+1 lim 4k 2 N N →∞ + ) k=1 (3 SOLUTION Observe that N

49.

lim

1 1k + 2k + 3k + · · · + N k = N N k+1

)

k k k * N k 2 3 N 1 # j 1 k + + + ··· . = N N N N N j=1 N

290

CHAPTER 5

THE INTEGRAL

Now, let f (x) = x k and N be a positive integer. A uniform partition of the interval [0, 1] with N subintervals has x =

1 N

xj =

and

j N

for 0 ≤ j ≤ N . Then N k N # 1 # j = x f (x j ) = R N ; N j=1 N j=1

consequently, 1 " 1 N k 1 1 1 # j = xk dx = x k+1 = . N k + 1 k + 1 N →∞ N 0 0 j=1 lim

" 1 " π /4 9 f (x) dxx, dassuming that f (x) is an even continuous function such that 51. Evaluate x , using the properties of odd functions. Evaluate 0 2 cos x −π /4 " 2 " 1 f (x) d x = 5, f (x) d x = 8 −2

1

Using the given information

SOLUTION

" 2 −2

f (x) d x =

" 1 −2

f (x) d x +

" 2 1

f (x) d x = 13.

Because f (x) is an even function, it follows that " 0 −2

f (x) d x =

" 2 0

f (x) d x,

so " 2 0

f (x) d x =

13 . 2

Finally, " 1 0

f (x) d x =

" 2 0

f (x) d x −

" 2 1

f (x) d x =

13 3 −5= . 2 2

53. Show that Plot the graph of f (x) = sin mx sin " nx on [0, π ] for the pairs (m, n) = (2, 4), (3, 5) and in each case guess " π f (x) d x =more x F(x) − G(x) f (x) d x. Experiment xwith a few values (including two cases with m = n) and formulate the value of I = 0

" a conjecture for when I is zero. where F (x) = f (x) and G (x) = F(x). Use this to evaluate x cos x d x. Suppose F (x) = f (x) and G (x) = F(x). Then

SOLUTION

d (x F(x) − G(x)) = x F (x) + F(x) − G (x) = x f (x) + F(x) − F(x) = x f (x). dx Therefore, x F(x) − G(x) is an antiderivative of x f (x) and " x f (x) d x = x F(x) − G(x) + C. To evaluate

55.

!

x cos x d x, note that f (x) = cos x. Thus, we may take F(x) = sin x and G(x) = − cos x. Finally, " x cos x d x = x sin x + cos x + C.

" 2 Prove Plot the graph of f (x) = x −2 sin x and show that 0.2 ≤ f (x) d x ≤ 0.9. 1 " " 2 2 1 1 2x d x ≤ 4 and 3−x d x ≤ . 2≤ ≤ 9 3 1 1

Chapter Review Exercises SOLUTION

291

Let f (x) = x −2 sin x. From the figure below, we see that 0.2 ≤ f (x) ≤ 0.9

for 1 ≤ x ≤ 2. Therefore, 0.2 =

" 1 0

0.2 d x ≤

" 1 0

f (x) d x ≤

" 1 0

0.9 d x = 0.9.

y 1 0.8 0.6

x−2sin x

0.4 0.2 x 0.5

1

1.5

2

In Exercises 57–62, find the derivative. " 1 " xbounds for Find upper and lower f (x) d x, where f (x) has the graph shown in Figure 8. 57. A (x), where A(x) = sin(t 3 ) dt 0 3

SOLUTION

Let A(x) =

" x 3

sin(t 3 ) dt. Then A (x) = sin(x 3 ).

" y d " x 3x d x cos t (π ), d y A−2 where A(x) = dt 2 1+t " y d SOLUTION 3x d x = 3 y . d y −2 " x3 "√ sin x 61. G (2), where G(x) = t + 13 dt t dt G (x), where G(x) = 0 −2 " x3 √ SOLUTION Let G(x) = t + 1 dt. Then 59.

0

G (x) =

x3 + 1

d 3 x = 3x 2 x 3 + 1 dx

√ and G (2) = 3(2)2 8 + 1 = 36. " 9 If f (x) is increasing and concave up on [a, b], then L N is more accurate than R N . Explain with a graph: 63. 1 (1), where H (x) = dt Which isHmore accurate if f (x) is increasing and concave down? 4x 2 t SOLUTION Consider the figure below, which displays a portion of the graph of an increasing, concave up function. y

x

The shaded rectangles represent the differences between the right-endpoint approximation R N and the left-endpoint approximation L N . In particular, the portion of each rectangle that lies below the graph of y = f (x) is the amount by which L N underestimates the area under the graph, whereas the portion of each rectangle that lies above the graph of y = f (x) is the amount by which R N overestimates the area. Because the graph of y = f (x) is increasing and concave up, the lower portion of each shaded rectangle is smaller than the upper portion. Therefore, L N is more accurate (introduces less error) than R N . By similar reasoning, if f (x) is increasing and concave down, then R N is more accurate than L N . Explain with a graph: If f (x) is linear on [a, b], then the

" b a

f (x) d x =

1 (R N + L N ) for all N . 2

APPLICATIONS OF 6 THE INTEGRAL 6.1 Area Between Two Curves Preliminary Questions 1. What is the area interpretation of

" b a

f (x) − g(x) d x if f (x) ≥ g(x)?

!b

SOLUTION Because f (x) ≥ g(x), a ( f (x) − g(x)) d x represents the area of the region bounded between the graphs of y = f (x) and y = g(x), bounded on the left by the vertical line x = a and on the right by the vertical line x = b. " b 2. Is f (x) − g(x) d x still equal to the area between the graphs of f and g if f (x) ≥ 0 but g(x) ≤ 0?

a

SOLUTION

Yes. Since f (x) ≥ 0 and g(x) ≤ 0, it follows that f (x) − g(x) ≥ 0.

3. Suppose that f (x) ≥ g(x) on [0, 3] and g(x) ≥ f (x) on [3, 5]. Express the area between the graphs over [0, 5] as a sum of integrals. SOLUTION Remember that to calculate an area between two curves, one must subtract the equation for the lower curve from the equation for the upper curve. Over the interval [0, 3], y = f (x) is the upper curve. On the other hand, over the interval [3, 5], y = g(x) is the upper curve. The area between the graphs over the interval [0, 5] is therefore given by " 3 " 5 ( f (x) − g(x)) d x + (g(x) − f (x)) d x.

0

3

4. Suppose that the graph of x = f (y) lies to the left of the y-axis. Is

" b a

f (y) d y positive or negative?

If the graph of x = f (y) lies to the left of the y-axis, then for each value of y, the corresponding value of x ! is less than zero. Hence, the value of ab f (y) d y is negative. SOLUTION

Exercises 1. Find the area of the region between y = 3x 2 + 12 and y = 4x + 4 over [−3, 3] (Figure 8). y y = 3x 2 + 12

50

25 y = 4x + 4

−2 −3

−1

x 1

2

3

FIGURE 8

As the graph of y = 3x 2 + 12 lies above the graph of y = 4x + 4 over the interval [−3, 3], the area between the graphs is " 3 " 3

3 (3x 2 + 12) − (4x + 4) d x = (3x 2 − 4x + 8) d x = x 3 − 2x 2 + 8x = 102. SOLUTION

−3

−3

−3

3. Let f (x) = x and g(x) = 2 − x 2 [Figure 9(B)]. Compute the area of the region in Figure 9(A), which lies between y = 2 − x 2 and y = −2 over [−2, 2]. (a) Find the points of intersection of the graphs. (b) Find the area enclosed by the graphs of f and g. SOLUTION

(a) Setting f (x) = g(x) yields 2 − x 2 = x, which simplifies to 0 = x 2 + x − 2 = (x + 2)(x − 1). Thus, the graphs of y = f (x) and y = g(x) intersect at x = −2 and x = 1.

S E C T I O N 6.1

Area Between Two Curves

293

(b) As the graph of y = x lies below the graph of y = 2 − x 2 over the interval [−2, 1], the area between the graphs is 1 " 1

1 1 9 (2 − x 2 ) − x d x = 2x − x 3 − x 2 = . 3 2 2 −2 −2 In Exercises 5–7, find the area between y = sin x and y = cos x over the interval. Sketch the curves if necessary. Let f (x) = 8x − 10 and g(x) = x 2 − 4x + 10. π 5. (a) 0, Find the points of intersection of the graphs. (b) 4Compute the area of the region below the graph of f and above the graph of g. π SOLUTION Over the interval [0, 4 ], the graph of y = sin x lies below that of y = cos x. Hence, the area between the two curves is √ √ " π /4 π /4 √ 2 2 = + − (0 + 1) = 2 − 1. (cos x − sin x) d x = (sin x + cos x) 2 2 0 0 7. [0, π ] π π , π π SOLUTION 4 2Over the interval [0, 4 ], the graph of y = sin x lies below that of y = cos x, while over the interval [ 4 , π ], the orientation of the graphs is reversed. The area between the graphs over [0, π ] is then " π /4 0

(cos x − sin x) d x +

" π π /4

(sin x − cos x) d x

π π /4 + (− cos x − sin x) = (sin x + cos x) 0

√

π /4

√ √ 2 2 2 2 = + − (0 + 1) + (1 − 0) − − − = 2 2. 2 2 2 2 √

√

In Exercises 8–10, let f (x) = 20 + x − x 2 and g(x) = x 2 − 5x. 9. Find the area of the region enclosed by the two graphs. Find the area between the graphs of f and g over [1, 3]. SOLUTION Setting f (x) = g(x) gives 20 + x − x 2 = x 2 − 5x, which simplifies to 0 = 2x 2 − 6x − 20 = 2(x − 5)(x + 2). Thus, the curves intersect at x = −2 and x = 5. With y = 20 + x − x 2 being the upper curve, the area between the two curves is 5 " 5

" 5

2 343 (20 + x − x 2 ) − (x 2 − 5x) d x = 20 + 6x − 2x 2 d x = 20x + 3x 2 − x 3 = . 3 3 −2 −2 −2 In Exercises 11–14, of the between shaded region ingraphs the figure. Compute thefind areathe of area the region the two over [4, 8] as a sum of two integrals. 11.

y

y = x 3 − 2x 2 + 10 −2

x y=

3x 2

2 + 4x − 10

FIGURE 10 SOLUTION

As the graph of y = x 3 − 2x 2 + 10 lies above the graph of y = 3x 2 + 4x − 10, the area of the shaded

region is " 2

" 2

(x 3 − 2x 2 + 10) − (3x 2 + 4x − 10) d x = x 3 − 5x 2 − 4x + 20 d x −2

−2

=

2 1 4 5 3 160 x − x − 2x 2 + 20x = . 4 3 3 −2

294

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L

13.

y

1 y=− x 2

−1

x

1 y = x 1 − x2

FIGURE 12 √ √ 3 3 1 1 SOLUTION Setting 2 x = x 1 − x 2 yields x = 0 or 2 = 1 − x 2 , so that x = ± 2 . Over the interval [− 2 , 0], √ y = 12 x is the upper curve but over the interval [0, 23 ], y = x 1 − x 2 is the upper curve. The area of the shaded region is then " 0 " √3/2 1 2 dx + 2 − 1 x dx 1 − x 1 − x x − x x √ 2 − 3/2 2 0

=

0 √ 5 5 1 1 2 3/2 5 1 2 1 2 3/2 2 3/2 x + (1 − x ) + = . + − (1 − x ) − x = √ 4 3 3 4 48 48 24 − 3/2 0

15. Find the area of the region enclosed by the curves y = x 3 − 6x and y = 8 − 3x 2 . Setting x 3 − 6x = 8 − 3x 2 yields (x + 1)(x + 4)(x − 2) = 0, so the two curves intersect at x = −4, x = −1 and x = 2. Over the interval [−4, −1], y = x 3 − 6x is the upper curve, while y = 8 − 3x 2 is the upper curve over the interval [−1, 2]. The area of the region enclosed by the two curves is then SOLUTION

" 2

" −1

(x 3 − 6x) − (8 − 3x 2 ) d x + (8 − 3x 2 ) − (x 3 − 6x) d x −4

=

−1

−1 2 1 4 81 1 81 81 x − 3x 2 − 8x + x 3 + 8x − x 3 − x 4 + 3x 2 = + = . 4 4 4 4 2 −4 −1

In Exercises 17–18, find the area between the graphs of x = sin y and x = 21 − cos y over the given interval (Figure 14). Find the area of the region enclosed by the semicubical parabola y = x 3 and the line x = 1. y x = sin y x = 1 − cos y 2

x −

2

FIGURE 14

π 17. 0 ≤ y ≤ 2 SOLUTION As shown in the figure, the graph on the right is x = sin y and the graph on the left is x = 1 − cos y. Therefore, the area between the two curves is given by

" π /2 0

π /2 π π = − + 1 − (−1) = 2 − . (sin y − (1 − cos y)) d y = (− cos y − y + sin y) 2 2 0

2 19. Find the π area ofπthe region lying to the right of x = y + 4y − 22 and the left of x = 3y + 8. − ≤y≤ 2 Setting2 y 2 + 4y − 22 = 3y + 8 yields SOLUTION

0 = y 2 + y − 30 = (y + 6)(y − 5), so the two curves intersect at y = −6 and y = 5. The area in question is then given by " 5

−6

(3y + 8) − (y 2 + 4y − 22)

5 " 5

y3 1331 y2 2 −y − y + 30 d y = − dy = − + 30y = . 3 2 6 −6 −6

y-axis and as an integral 21. Calculate the area enclosed by x = 9 − y 2 and x = 5 in 2two ways: as an integral along the Find the area of the region lying to the right of x = y − 5 and the left of x = 3 − y 2 . along the x-axis.



295

Along the y-axis, we have points of intersection at y = ±2. Therefore, the area enclosed by the two curves

is 2 " 2

" 2

32 1 . 9 − y2 − 5 d y = 4 − y 2 d y = 4y − y 3 = 3 3 −2 −2 −2 Along the x-axis, we have integration limits of x = 5 and x = 9. Therefore, the area enclosed by the two curves is 9 " 9 √ 4 32 32 2 9 − x d x = − (9 − x)3/2 = 0 − − = . 3 3 3 5 5 In Exercises 23–24, find the area of the region using the method (integration along either the x- or y-axis) that requires Figure 15 shows the graphs of x = y 3 − 26y + 10 and x = 40 − 6y 2 − y 3 . Match the equations with the curve you to evaluate just one integral. and compute the area of the shaded region. 23. Region between y 2 = x + 5 and y 2 = 3 − x From the figure below, we see that integration along the x-axis would require two integrals, but integration along the y-axis requires only one integral. Setting y 2 − 5 = 3 − y 2 yields points of intersection at y = ±2. Thus, the area is given by SOLUTION

2 " 2

" 2

64 2 . (3 − y 2 ) − (y 2 + 5) d y = 8 − 2y 2 d y = 8y − y 3 = 3 3 −2 −2 −2 y 2

x = 3 − y2

1 −4

x

−2

−1

x = y2 − 5

2

−2

In Exercises 25–41, sketch thex region enclosed by the and compute its area as an integral along the x- or y-axis. Region between y= and x + y = 8 over [2,curves 3] 25. y = 4 − x 2 ,

y = x2 − 4

Setting 4 − x 2 = x 2 − 4 yields 2x 2 = 8 or x 2 = 4. Thus, the curves y = 4 − x 2 and y = x 2 − 4 intersect at x = ±2. From the figure below, we see that y = 4 − x 2 lies above y = x 2 − 4 over the interval [−2, 2]; hence, the area of the region enclosed by the curves is SOLUTION

2 " 2 2 64 (8 − 2x 2 ) d x = 8x − x 3 = . (4 − x 2 ) − (x 2 − 4) d x = 3 3 −2 −2 −2

" 2

y y = 4 − x2

4 2 −2

−1

x −2 −4

2 27. x =y sin = xy,2 −x6,= πy y= 6 − x 3 , SOLUTION

1

2

y = x2 − 4

y-axis

Here, integration along the y-axis will require less work than integration along the x-axis. The curves

π intersect when 2y π = sin y or when y = 0, ± 2 . From the graph below, we see that both curves are symmetric with respect to the origin. It follows that the portion of the region enclosed by the curves in the first quadrant is identical to the region enclosed in the third quadrant. We can therefore determine the total area enclosed by the two curves by doubling the area enclosed in the first quadrant. In the first quadrant, x = sin y lies to the right of x = 2y π , so the total area enclosed by the two curves is

2

π /2 " π /2

1 π π 2 =2 0− − (−1 − 0) = 2 − . sin y − y d y = 2 − cos y − y 2 π π 4 2 0 0

296

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L y x=

2

y

1 x = sin y x

−1

1 −1

−3 , y = 4 − x, y = x 29. y =x 3x +y= 4, x − y = 0, y 3+ 3x = 4

The curves y = 3x −3 and y = 4 − x intersect at x = 1, the curves y = 3x −3 and y = x/3 intersect at x = 3 and the curves y = 4 − x and y = x/3 intersect at x = 3. From the graph below, we see that the top of the −3 region enclosed √ by the three curves is√always bounded by y = 4 − x. The bottom of the region is bounded by y = 3x for 1 ≤ x ≤ 3 and by y = x/3 for 3 ≤ x ≤ 3. The total area of the region is then SOLUTION

√

√ " √3

" 3 2 2 3 1 1 2 3 −2 3 −3 (4 − x) − 3x d x + √ (4 − x) − x d x = 4x − x + x + 4x − x √ = 2. 3 2 2 3 1 3 3 1 y 3 y= 3 x3

2

1 y=

y=4−x

x 3 x 1

0

2

3

√ √ 2, y = −x√ x − 2, x = 4 31. y = x x − √ y = 2 − x, y = x, √ x =0 √ SOLUTION Note that y = x x − 2 and y = −x x − 2 are the upper and lower branches, respectively, of the curve y 2 = x 2 (x − 2). The area enclosed by this curve and the vertical line x = 4 is " 4 " 4

√ √ √ x x − 2 − (−x x − 2) d x = 2x x − 2 d x. 2

2

Substitute u = x − 2. Then du = d x, x = u + 2 and √ " 2

4 5/2 8 3/2 2 128 2 3/2 1/2 u 2u du = 2x x − 2 d x = 2(u + 2) u du = + 4u + u = 15 . 5 3 2 0 0 0

" 4

" 2

√

√

y 4

y 2 = x 2 (x − 2)

2 x −2

1

2

3

4

−4

33. x = |y|, x = 6 − y22 y = |x|, y = x − 6 SOLUTION From the graph below, we see that integration along the y-axis will require less work than integration along the x-axis. Moreover, the region is symmetric with respect to the x-axis, so the total area can be determined by doubling the area of the upper portion of the region. For y > 0, setting y = 6 − y 2 yields 0 = y 2 + y − 6 = (y − 2)(y + 3), so the curves intersect at y = 2. Because x = 6 − y 2 lies to the right of x = |y| = y in the first quadrant, we find the total area of the region is 2

2 " 2 1 1 8 44 6 − y 2 − y d y = 2 6y − y 3 − y 2 = 2 12 − − 2 = . 3 2 3 3 0 0

S E C T I O N 6.1


297

y 2 x = |y| 1 x 1

−1

2

3

4

5

6

x = 6 − y2

−2

35. x = 12 − y, x = y, x = 2y x = |y|, x = 1 − |y| SOLUTION From the graph below, we see that the bottom of the region enclosed by the three curves is always bounded by y = x2 . On the other hand, the top of the region is bounded by y = x for 0 ≤ x ≤ 6 and by y = 12 − x for 6 ≤ x ≤ 8. The area of the region is then 8 " 6

" 8

x x 3 1 6 x− (12 − x) − dx + d x = x 2 + 12x − x 2 = 9 + (96 − 48) − (72 − 27) = 12. 2 2 4 0 4 0 6 6 y 6

y = 12 − x

4

y=x

2

y=

x 2 x

0

2

4

6

8

37. x = 2y, 3x + 1 = (y − 1)2 x = y − 18y, y + 2x = 0 SOLUTION Setting 2y = (y − 1)2 − 1 yields 0 = y 2 − 4y = y(y − 4), so the two curves intersect at y = 0 and at y = 4. From the graph below, we see that x = 2y lies to the right of x + 1 = (y − 1)2 over the interval [0, 4] along the y-axis. Thus, the area of the region enclosed by the two curves is 4 " 4

" 4

1 32 . 2y − ((y − 1)2 − 1) d y = 4y − y 2 d y = 2y 2 − y 3 = 3 3 0 0 0 y 4

x + 1 = ( y − 1) 2

3 x = 2y

2 1

x 2

4

6

8

+ x 2 (in1/2 the region x > 0) 39. y = 6, y = x −2 1/2 x + y = 1, x +y =1 −2 SOLUTION Setting 6 = x + x 2 yields 0 = x 4 − 6x 2 + 1, which is a quadratic equation in the variable x 2 . By the quadratic formula, √ √ 6 ± 36 − 4 = 3 ± 2 2. x2 = 2 Now, √ √ √ √ and 3 − 2 2 = ( 2 − 1)2 , 3 + 2 2 = ( 2 + 1)2 √ √ so the two curves intersect at x = 2 − 1 and x = 2 + 1. Note there are also two points of intersection with x < 0, but as the problem specifies the region is for x > 0, we neglect these other two values. From the graph below, we see that y = 6 lies above y = x −2 + x 2 , so the area of the region enclosed by the two curves for x > 0 is √ " √2+1

1 3 2+1 16 −2 2 −1 6 − (x + x ) d x = 6x + x − x √ = . √ 3 3 2−1 2−1

298

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L y 6

4 y = x −2 + x 2 2

x 0

1

2

3

π 3π 41. y = sin x, y = csc2 x, x = , x = 2π y = cos x, y = cos(2x), 4x = 0, x 4= 3 π 3π SOLUTION Over the interval [ 4 , 4 ], y = csc2 x lies above y = sin x. The area of the region enclosed by the two curves is then √ √ " 3π /4 √ 2 3π /4 2 2 csc x − sin x d x = − cot x + cos x = 1− − −1 + = 2 − 2. 2 2 π /4 π /4 y 2 1.5

y = csc 2 x

1

y = sin x

0.5 x 0

0.5

1

1.5

2

43. Sketch a region whose x area is represented2by Plot y = the same set of axes. Use a computer algebra system to find the points and y = (x − 1) on " √2/2 x2 + 1 2 −curves. of intersection numerically and compute the √ area between 1 − x the |x| d x − 2/2

and evaluate using geometry.

1 − x 2 − |x| with the yTOP − yBOT template for calculating area, we see that the region in question is bounded along the top by the curve y = 1 − x 2 (the upper half of the unit circle) and is bounded along the bottom by the curve y = |x|. Hence, the region is 14 of the unit circle (see the figure below). The area of the region must then be SOLUTION

Matching the integrand

1 π π (1)2 = . 4 4 y

0.8

0.4 x

−0.4

0.4

45. Express the area (not signed) of the shaded region in Figure 16 as a sum of three integrals involving the functions f Two athletes run in the same direction along a straight track with velocities v1 (t) and v2 (t) (in ft/s). Assume and g. that y " 5 " 20 f(x)(v1 (t) − v2 (t)) dt = 5 (v1 (t) − v2 (t)) dt = 2, 0

" 35 30

0

x 3

5

9

(v1 (t) − v2 (t)) dt = −2 g(x)

" t2 16v2 (t) dt. v1 (t) − (a) Give a verbal interpretation of the integral FIGURE t1

(b) Is enough information given to determine the distance between the two runners at time t = 5 s? (c) If the runners begin at the same time and place, how far ahead is runner 1 at time t = 20 s? (d) Suppose that runner 1 is 8 ft ahead at t = 30 s. How far is she ahead or behind at t = 35 s?

S E C T I O N 6.1


299

SOLUTION Because either the curve bounding the top of the region or the curve bounding the bottom of the region or both change at x = 3 and at x = 5, the area is calculated using three integrals. Specifically, the area is " 3 " 5 " 9 ( f (x) − g(x)) d x + ( f (x) − 0) d x + (0 − f (x)) d x

0

=

" 3 0

3

( f (x) − g(x)) d x +

" 5 3

5

f (x) d x −

" 9 5

f (x) d x.

47. Set up (but do not evaluate) an integral that expresses the 2area between the circles x 2 + y 2 = 2 and x 2 + 2 (y − 1)2Find = 1.the area enclosed by the curves y = c − x and y = x − c as a function of c. Find the value of c for which this area is equal to 1. SOLUTION Setting 2 − y 2 = 1 − (y − 1)2 yields y = 1. The two circles therefore intersect at the points (1, 1) and (−1, 1). From the graph below, we see that over the interval [−1, 1], the upper half of the circle x 2 + y 2 = 2 lies above the lower half of the circle x 2 + (y − 1)2 = 1. The area enclosed by the two circles is therefore given by the integral " 1 2 − x 2 − (1 − 1 − x 2 ) d x. −1

y x2 + y2 = 2 1 x 2 + (y − 1)2 = 1 x 1

−1

49. Find a numerical approximation to the area above y = 1 − (x/π ) and below y = sin x (find the points of Set up (but do not evaluate) an integral that expresses the area between the graphs of y = (1 + x 2 )−1 and y = x 2 . intersection numerically). SOLUTION The region in question is shown in the figure below. Using a computer algebra system, we find that y = 1 − x/π and y = sin x intersect on the left at x = 0.8278585215. Analytically, we determine the two curves intersect on the right at x = π . The area above y = 1 − x/π and below y = sin x is then " π

x sin x − 1 − d x = 0.8244398727, π 0.8278585215 where the definite integral was evaluated using a computer algebra system. y 1 y = sin x

y=1−

x x

1

0

2

3

51. Use a computer algebra system to find a numerical approximation to the number c (besides zero) in [0, π2 ], Find a numerical approximation to the area above y = |x| and below y = cos x. where the curves y = sin x and y = tan2 x intersect. Then find the area enclosed by the graphs over [0, c]. SOLUTION The region in question is shown in the figure below. Using a computer algebra system, we find that y = sin x and y = tan2 x intersect at x = 0.6662394325. The area of the region enclosed by the two curves is then " 0.6662394325

sin x − tan2 x d x = 0.09393667698, 0

where the definite integral was evaluated using a computer algebra system. y

0.8 0.6 0.4

y = sin x y = tan 2 x

0.2 x 0

0.2

0.4

0.6

The back of Jon’s guitar (Figure 17) has a length 19 in. He measured the widths at 1-in. intervals, beginning and ending 12 in. from the ends, obtaining the results

300

CHAPTER 6


Further Insights and Challenges 53. Find the line y = mx that divides the area under the curve y = x(1 − x) over [0, 1] into two regions of equal area. SOLUTION

First note that " 1 0

x(1 − x) d x =

" 1

0

1 2 1 3 1 1 x − x2 dx = x − x = . 2 3 6 0

Now, the line y = mx and the curve y = x(1 − x) intersect when mx = x(1 − x), or at x = 0 and at x = 1 − m. The area of the region enclosed by the two curves is then " 1−m 0

1−m " 1−m

x2 1 1 3 2 (1 − m)x − x d x = (1 − m) = (1 − m)3 . − x (x(1 − x) − mx) d x = 2 3 6 0 0

To have 16 (1 − m)3 = 12 · 16 requires m =1−

55.

1/3 1 ≈ 0.206299. 2

calculation) following for any n >by 0: the line y = cx (Figure LetExplain c be thegeometrically number such (without that the area under y why = sinthe x over [0, π ] holds is divided in half 18). Find an equation for c and solve this"equation numerically using a computer algebra system. " 1 1 xn dx + x 1/n d x = 1 0

SOLUTION

0

Let A1 denote the area of region 1 in the figure below. Define A2 and A3 similarly. It is clear from the

figure that A1 + A2 + A3 = 1. Now, note that x n and x 1/n are inverses of each other. Therefore, the graphs of y = x n and y = x 1/n are symmetric about the line y = x, so regions 1 and 3 are also symmetric about y = x. This guarantees that A1 = A3 . Finally, " 1 0

xn dx +

" 1 0

x 1/n d x = A3 + ( A2 + A3 ) = A1 + A2 + A3 = 1. y 1 1

2 3 x 0

1

6.2 Setting Up Integrals: Volume, Density, Average Value Preliminary Questions 1. What is the average value of f (x) on [1, 4] if the area between the graph of f (x) and the x-axis is equal to 9? Assuming that f (x) ≥ 0 over the interval [1, 4], the fact that the area between the graph of f and the x-axis ! is equal to 9 indicates that 14 f (x) d x = 9. The average value of f over the interval [1, 4] is then SOLUTION

!4

1 f (x) d x = 9 = 3.

4−1

3

2. Find the volume of a solid extending from y = 2 to y = 5 if the cross section at y has area A(y) = 5 for all y. SOLUTION Because the cross-sectional area of the solid is constant, the volume is simply the cross-sectional area times the length, or 5 × 3 = 15.

3. Describe the horizontal cross sections of an ice cream cone and the vertical cross sections of a football (when it is held horizontally).

Setting Up Integrals: Volume, Density, Average Value

S E C T I O N 6.2

301

SOLUTION The horizontal cross sections of an ice cream cone, as well as the vertical cross sections of a football (when held horizontally), are circles.

4. What is the formula for the total population within a circle of radius R around a city center if the population has a radial function? SOLUTION

Because the population density is a radial function, the total population within a circle of radius R is 2π

" R 0

r ρ(r ) dr,

where ρ(r ) is the radial population density function. 5. What is the definition of flow rate? SOLUTION

The flow rate of a fluid is the volume of fluid that passes through a cross-sectional area at a given point per

unit time. 6. Which assumption about fluid velocity did we use to compute the flow rate as an integral? SOLUTION To express flow rate as an integral, we assumed that the fluid velocity depended only on the radial distance from the center of the tube.

Exercises 1. Let V be the volume of a pyramid of height 20 whose base is a square of side 8. (a) Use similar triangles as in Example 1 to find the area of the horizontal cross section at a height y. (b) Calculate V by integrating the cross-sectional area. SOLUTION

(a) We can use similar triangles to determine the side length, s, of the square cross section at height y. Using the diagram below, we find 8 s = 20 20 − y

s=

or

2 (20 − y). 5

4 (20 − y)2 . The area of the cross section at height y is then given by 25

20 − y 20

s

8

(b) The volume of the pyramid is 20 " 20 4 1280 4 (20 − y)2 d y = − (20 − y)3 = . 75 3 0 25 0 3. Use the method of Exercise 2 to find the formula for the volume of a right circular cone of height h whose base is a Let V be the volume of a right circular cone of height 10 whose base is a circle of radius 4 (Figure 16). circle of radius r (Figure 16). SOLUTION

(a) From similar triangles (see Figure 16), (a) Use similar triangles to find the area of a horizontal cross section at a height y. r h (b) Calculate V by integrating the cross-sectional area. = , h−y r0 where r0 is the radius of the cone at a height of y. Thus, r0 = r − rhy . (b) The volume of the cone is h " h

r y 3 r y 2 −h π r − h hπ r 3 πr 2h r− π dy = = = . h r 3 r 3 3 0 0

Calculate the volume of the ramp in Figure 17 in three ways by integrating the area of the cross sections: (a) Perpendicular to the x-axis (rectangles) (b) Perpendicular to the y-axis (triangles)

302

CHAPTER 6


5. Find the volume of liquid needed to fill a sphere of radius R to height h (Figure 18). y

R h

FIGURE 18 Sphere filled with liquid to height h. SOLUTION

The radius r at any height y is given by r =

R 2 − (R − y)2 . Thus, the volume of the filled portion of

the sphere is

π

" h 0

r2 dy =

π

" h

0

R 2 − (R − y)2

dy = π

" h 0

(2Ry − y 2 ) d y =

π

h h3 2 . = π Rh − 3 3

y3 Ry 2 −

0

7. Derive a formula for the volume of the wedge in Figure 19(B) in terms of the constants a, b, and c. Find the volume of the wedge in Figure 19(A) by integrating the area of vertical cross sections. The line from c to a is given by the equation (z/c) + (x/a) = 1 and the line from b to a is given by (y/b) + (x/a) = 1. The cross sections perpendicular to the x-axis are right triangles with height c(1 − x/a) and base b(1 − x/a). Thus we have " a

1 1 x 3 a 1 2 bc (1 − x/a) d x = − abc 1 − = 6 abc. 2 6 a 0 0 SOLUTION

In Exercises 9–14, find the volume of the solid with given2base 2and cross sections. Let B be the solid whose base is the unit circle x + y = 1 and whose vertical cross √ sections perpendicular to thatthe thecross vertical crossperpendicular sections haveto area =are3(1 − x 2 ) whose and compute 9. the Thex-axis base isare theequilateral unit circletriangles. x 2 + y 2 Show = 1 and sections theA(x) x-axis triangles height the volume of B. and base are equal. SOLUTION At each location x, the side of the triangular cross section that lies in the base of the solid extends from the 2 2 top half of the unit circle (with y = 1 − x ) to the bottom half (with y = − 1 − x ). The triangle therefore has base 2 and height equal to 2 1 − x 2 and area 2(1 − x ). The volume of the solid is then 1 " 1 1 8 2(1 − x 2 ) d x = 2 x − x 3 = . 3 3 −1 −1

2 where −3 ≤ x ≤ 3. The cross sections perpendicular to the x-axis are 9− 11. TheThe basebase is the semicircle y = is the triangle enclosed byx x, + y = 1, the x-axis, and the y-axis. The cross sections perpendicular to squares. the y-axis are semicircles. SOLUTION For each x, the base of the square cross section extends from the semicircle y = 9 − x 2 to the x-axis.

2 9 − x 2 = 9 − x 2 . The volume of the solid is The square therefore has a base with length 9 − x 2 and an area of then 3 " 3

1 9 − x 2 d x = 9x − x 3 = 36. 3 −3 −3 y =interval 3. The cross to the y-axis are squares. 13. TheThe basebase is the enclosed y =sides x 2 and is aregion square, one of by whose is the [0, ] sections along theperpendicular √x-axis. The cross sections perpendicular 2. SOLUTION At any location y,of theheight distance to = thexparabola from the y-axis is y. Thus the base of the square will have to the x-axis are rectangles f (x) √ length 2 y. Therefore the volume is " 3 " 3 3 √ √ 4y d y = 2y 2 = 18. 2 y 2 y dy = 0

0

0

15. Find the volume of the solid whose base is the region |x| + |y| ≤ 1 and whose vertical cross sections perpendicular The base is the region(with enclosed by y = x 2 and y = 3. The cross sections perpendicular to the y-axis are rectangles to the y-axis are semicircles diameter along the base). of height y 3 . SOLUTION The region R in question is a diamond shape connecting the points (1, 0), (0, −1), (−1, 0), and (0, 1). Thus, in the lower half of the x y-plane, the radius of the circles is y + 1 and in the upper half, the radius is 1 − y. Therefore, the volume is " " π 0 π 1 π 1 1 π + = . (y + 1)2 d y + (1 − y)2 d y = 2 −1 2 0 2 3 3 3


S E C T I O N 6.2

17. Find the volume V of a regular tetrahedron whose face is an equilateral triangle of side s (Figure 20). Show that the volume of a pyramid of height h whose base is an equilateral triangle of side s is equal to

303

√

3 2 hs . 12

s

s

FIGURE 20 Regular tetrahedron. SOLUTION Our first task is to determine the relationship between the height of the tetrahedron, h, and the side length of the equilateral triangles, s. Let B be the orthocenter of the tetrahedron (the point directly below the apex), and let b denote the distance from B to each corner of the base triangle. By the Law of Cosines, we have

s 2 = b2 + b2 − 2b2 cos 120◦ = 3b2 , so b2 = 13 s 2 . Thus 2 h 2 = s 2 − b2 = s 2 3

or

h=s

2 . 3

Therefore, using similar triangles, the side length of the equilateral triangle at height z above the base is h−z z s =s−√ . h 2/3 The volume of the tetrahedron is then given by √ √ √ 2 3 s 2/3 " s √2/3 √ s3 2 z z 3 2 dz = − = s−√ s−√ . 4 12 12 2/3 2/3 0 0

19. A frustum of a pyramid is a pyramid with its top cut off [Figure 22(A)]. Let V be the volume of a frustum of height The area of an ellipse is π ab, where a and b are the lengths of the semimajor and semiminor axes (Figure 21). h whose base is a square of side a and top is a square of side b with a > b ≥ 0. Compute the volume of a cone of height 12 whose base is an ellipse with semimajor ha axis a = 6 and semiminor axis (a) Show [Figure 22(B)]. b = 4.that if the frustum were continued to a full pyramid, it would have height a−b (b) Show that the cross section at height x is a square of side (1/ h)(a(h − x) + bx). (c) Show that V = 13 h(a 2 + ab + b2 ). A papyrus dating to the year 1850 BCE indicates that Egyptian mathematicians had discovered this formula almost 4,000 years ago.

b h

a (A)

(B)

FIGURE 22 SOLUTION

(a) Let H be the height of the full pyramid. Using similar triangles, we have the proportion H H −h = a b which gives H=

ha . a−b

(b) Let w denote the side length of the square cross section at height x. By similar triangles, we have a w = . H H −x Substituting the value for H from part (a) gives w=

a(h − x) + bx . h

304

CHAPTER 6


(c) The volume of the frustrum is " h 1 0

h

(a(h − x) + bx)

2

" h

1 a 2 (h − x)2 + 2ab(h − x)x + b2 x 2 d x dx = 2 h 0 h 1 a2 2 1 h 2 a + ab + b2 . = 2 − (h − x)3 + abhx 2 − abx 3 + b2 x 3 = 3 3 3 3 h 0

21. Figure 24 shows the solid S obtained by intersecting two cylinders of radius r whose axes are perpendicular. A plane inclined at an angle of 45◦ passes through a diameter of the base of a cylinder of radius r . Find the (a) The horizontal cross section of each cylinder at distance y from the central axis is a rectangular strip. Find the strip’s volume of the region within the cylinder and below the plane (Figure 23). width. (b) Find the area of the horizontal cross section of S at distance y. (c) Find the volume of S as a function of r .

S y

FIGURE 24 Intersection of two cylinders intersecting at right angles. SOLUTION

(a) The horizontal cross section at distance y from the central axis (for −r ≤ y ≤ r ) is a square of width w = 2 r 2 − y2. (b) The area of the horizontal cross section of S at distance y from the central axis is w2 = 4(r 2 − y 2 ). (c) The volume of the solid S is then r " r

1 16 3 r . r 2 − y 2 d y = 4 r 2 y − y 3 = 4 3 3 −r −r 23. Calculate the volume of a cylinder inclined at an angle θ = 30◦ whose height is 10 and whose base is a circle of Let S be the solid obtained by intersecting two cylinders of radius r whose axes intersect at an angle θ . Find the radius 4 (Figure 25). volume of S as a function of r and θ . 4

10 30°

FIGURE 25 Cylinder inclined at an angle θ = 30◦ . SOLUTION

The area of each circular cross section is π (4)2 = 16π , hence the volume of the cylinder is " 10 0

10 16π d x = (16π x) = 160π 0

25. Find the total mass of a 2-m rod whose linear density function is ρ(x) = 1 + 0.5 sin(π x) kg/m for 0 ≤ x ≤ 2. Find the total mass of a 1-m rod whose linear density function is ρ(x) = 10(x + 1)−2 kg/m for 0 ≤ x ≤ 1. SOLUTION The total mass of the rod is cos π x 2 ρ(x) d x = (1 + .5 sin π x) d x = x − .5 = 2 kg, π 0 0 0

" 2

" 2

27. Calculate the population within a 10-mile radius of the city center if the radial population density is ρ(r ) = 4(1 + A mineral deposit along a strip of length 6 cm has density s(x) = 0.01x(6 − x) g/cm for 0 ≤ x ≤ 6. Calculate r 2 )1/3 (in thousands per square mile). the total mass of the deposit.



305

The total population is 2π

" 10 0

r · ρ(r ) dr = 2π

" 10 0

10

4r (1 + r 2 )1/3 dr = 3π (1 + r 2 )4/3

0

≈ 4423.59 thousand ≈ 4.4 million. 29. Table 1 lists the population density (in people per squared kilometer) as a function of distance r (in kilometers) from Odzala National Park in the Congo has a high density of gorillas. Suppose that the radial population density is the center of a rural town. Estimate the total population within a 2-km radius of the center by taking the average of the ρ(r ) = 52(1 + r 2 )−2 gorillas per square kilometer, where r is the distance from a large grassy clearing with a source left- and right-endpoint approximations. of food and water. Calculate the number of gorillas within a 5-km radius of the clearing. TABLE 1

SOLUTION

Population Density

r

ρ(r )

r

ρ(r )

0.0

125.0

1.2

37.6

0.2

102.3

1.4

30.8

0.4

83.8

1.6

25.2

0.6

68.6

1.8

20.7

0.8

56.2

2.0

16.9

1.0

46.0

The total population is given by 2π

" 2 0

r · ρ(r ) dr.

With r = 0.2, the left- and right-endpoint approximations to the required definite integral are L 10 = .2(2π )[0(125) + (.2)(102.3) + (.4)(83.8) + (.6)(68.6) + (.8)(56.2) + (1)(46) + (1.2)(37.6) + (1.4)(30.8) + (1.6)(25.2) + (1.8)(20.7)] = 442.24; R10 = .2(2π )[(.2)(102.3) + (.4)(83.8) + (.6)(68.6) + (0.8)(56.2) + (1)(46) + (1.2)(37.6) + (1.4)(30.8) + (1.6)(25.2) + (1.8)(20.7) + (2)(16.9)] = 484.71. This gives an average of 463.475. Thus, there are roughly 463 people within a 2-km radius of the town center. 2 + 2)−2 deer per km2 , where r is the distance (in ρ(rcm ) =whose 150(r mass 31. TheFind density deer in aofforest is the plate radialoffunction the of total mass a circular radius 20 density is the radial function ρ(r ) = 0.03 + kilometers) to a small meadow. Calculate the number of deer in the region 2 ≤ r ≤ 5 km. 0.01 cos(π r 2 ) g/cm2 . SOLUTION The number of deer in the region 2 ≤ r ≤ 5 km is

2π

5

−2 1 = −150π 1 − 1 ≈ 61 deer. r (150) r 2 + 2 dr = −150π 27 6 r 2 + 2 2 2

" 5

33. Find the flow rate through a tube of radius 4 cm, assuming that the velocity of fluid particles at a distance r cm from 4 the center is v(r ) =a 16 − r 2 cm/s. g/cm has finite total mass, even Show that circular plate of radius 2 cm with radial mass density ρ(r ) = r though theThe density SOLUTION flow becomes rate is infinite at the origin. 2π

" R 0

r v(r ) dr = 2π

4 " 4

1 cm3 r 16 − r 2 dr = 2π 8r 2 − r 4 = 128π . 4 s 0 0

35. A solid rod of radius 1 cm is placed in a pipe of radius 3 cm so that their axes are aligned. Water flows through −5 Use function Poiseuille’s ) be the of the blood in rate an arterial capillaryofofthe radius 4 × 10 the pipeLet andv(raround thevelocity rod. Find flow if the velocity waterR is=given by them.radial v(r Law ) = 6 (m-s)−1 to determine the velocity at the center of the capillary and the flow rate (use (Example 6) with k = 10 0.5(r − 1)(3 − r ) cm/s. correct units). SOLUTION The flow rate is 2π

" 3 1

r (.5)(r − 1)(3 − r ) dr = π

" 3

1

3 4 3 8π cm3 1 −r 3 + 4r 2 − 3r dr = π − r 4 + r 3 − r 2 = . 4 3 2 3 s 1

In Exercises 37–44,the calculate over the given interval. Bureau of Fisheries created the depth contour map in To estimate volumethe V average of Lake Nogebow, the Minnesota Figure 26 and determined the area of the cross section of the lake at the depths recorded in the table below. Estimate V by taking the average of the right- and left-endpoint approximations to the integral of cross-sectional area.

306

CHAPTER 6


37. f (x) = x 3 , SOLUTION

[0, 1]

The average is " 1 " 1 1 1 1 1 x3 dx = x 3 d x = x 4 = . 1−0 0 4 0 4 0

39. f (x) = cos x, [0, π2 ] f (x) = x 3 , [−1, 1] SOLUTION The average is " π /2 π /2 2 1 2

sin t 0 = . cos x d x = π /2 − 0 0 π π 41. f (s) = s −2 , 2[2, 5] f (x) = sec x, [0, π4 ] SOLUTION The average is " 5 1 −1 5 1 1 −2 s ds = − s = . 5−2 2 3 10 2 2 , [−1, 3] − 3x 43. f (x) = 2x 3sin( π /x) , [1, 2] f (x) = SOLUTION The average is x2

3 " 3

1 1 4 1 2x 3 − 3x 2 d x = x − x 3 = 3. 3 − (−1) −1 4 2 −1 45. Let M be thenaverage value of f (x) = x 3 on [0, A], where A > 0. Which theorem guarantees that f (c) = M has a f (x) = x , [0, 1] solution c in [0, A]? Find c. SOLUTION

The Mean Value Theorem for Integrals guarantees that f (c) = M has a solution c in [0, A]. With f (x) =

x 3 on [0, A], M=

" A 1 1 4 A A3 1 x3 dx = x = . A−0 0 A 4 0 4

3

Solving f (c) = c3 = A4 for c yields A c= √ . 3 4 47. Which of f (x) = x sin2 x and g(x) = x 2 sin2 x has a larger average value over [0, 1]? Over [1, 2]? Let f (x) = 2 sin x − x. Use a computer algebra system to plot f (x) and estimate: 2 SOLUTION The functions f and f (x).g differ only in the power of x multiplying sin x. It is also important to note that (a) The positive root α of 2 2 x ≥ The 0 foraverage all x. Now, x ∈ on (0,[0, 1),αx].> x so sin (b) valuefor M each of f (x) (c) A value c ∈ [0, α ] such that f (c)f = (x)M. = x sin2 x > x 2 sin2 x = g(x). Thus, over [0, 1], f (x) will have a larger average value than g(x). On the other hand, for each x ∈ (1, 2), x 2 > x, so g(x) = x 2 sin2 x > x sin2 x = f (x). Thus, over [1, 2], g(x) will have the larger average value. 49. Sketch the graph of a function f (x)sin such x that fπ(x) ≥ 0 on [0, 1] and f (x) ≤ 0 on [1, 2], whose average on that the average value of f (x) = over [ 2 , π ] is less than 0.41. Sketch the graph if necessary. [0, 2] isShow negative. x SOLUTION

Many solutions will exist. One could be y 1 x 1 −1 −2

2

S E C T I O N 6.2


307

π 51. TheFind temperature T (t)ofatf time (in + hours) in the an art museum varies T (t)M=are 70arbitrary + 5 cos constants. t . Find the the average (x) =t ax b over interval [−M, M], according where a, b,toand 12 average over the time periods [0, 24] and [2, 6]. SOLUTION

• The average temperature over the 24-hour period is

" 24

π

π 24 1 1 60 70 + 5 cos sin t dt = 70t + t = 70◦ F. 24 − 0 0 12 24 π 12 0 • The average temperature over the 4-hour period is

" 6

π

π 6 1 1 60 70 + 5 cos sin t dt = 70t + t = 72.4◦ F. 6−2 2 12 4 π 12 2 53. What is the average area of the circles whose radii vary from 0 to 1? A ball is thrown in the air vertically from ground level with initial velocity 64 ft/s. Find the average height of the SOLUTION Thetime average area extending is ball over the interval from the time of the ball’s release to its return to ground level. Recall that the height at time t is h(t) = 64t − 16t 2 . " 1 π 1 π 1 π r 2 dr = r 3 = . 1−0 0 3 0 3 55. The acceleration of a particle is a(t) = t − t 3 m/s2 for 0 ≤ t ≤ 1. Compute2the average acceleration and average . Find its average velocity during the object withinterval zero initial velocity accelerates a constant ratevelocity of 10 m/s velocityAn over the time [0, 1], assuming that theatparticle’s initial is zero. first 15 s. SOLUTION The average acceleration is " 1

1 1 2 1 4 1 1 t − t = m/s2 . t − t 3 dt = 1−0 0 2 4 4 0 An acceleration a(t) = t − t 3 with zero initial velocity gives v(t) =

1 2 1 4 t − t . 2 4

Thus the average velocity is given by 1 " 1 1 7 1 2 1 4 1 3 1 t − t t − t 5 = m/s. dt = 1−0 0 2 4 6 20 60 0 √ 57. Let f (x) = x. Find a value of c in [4, 9] such that f (c) is equal to the average of f on [4, 9]. Let M be the average value of f (x) = x 4 on [0, 3]. Find a value of c in [0, 3] such that f (c) = M. SOLUTION The average value is " 9 " √ 1 9√ 2 3/2 9 38 1 x dx = x dx = x = . 9−4 4 5 4 15 15 4 Then f (c) =

√

c = 38 15 implies c=

38 2 1444 = ≈ 6.417778. 15 225

Give an example of a function (necessarily discontinuous) that does not satisfy the conclusion of the MVT for Further Insights and Challenges Integrals.

59. An object is tossed in the air vertically from ground level with initial velocity v0 ft/s at time t = 0. Find the average speed of the object over the time interval [0, T ], where T is the time the object returns to earth. The height is given by h(t) = v0 t − 16t 2 . The ball is at ground level at time t = 0 and T = v0 /16. The velocity is given by v(t) = v0 − 32t and thus the speed is given by s(t) = |v0 − 32t|. The average speed is SOLUTION

" v0 /16 " " 1 16 v0 /32 16 v0 /16 |v0 − 32t| dt = (v0 − 32t) dt + (32t − v0 ) dt v0 /16 − 0 0 v0 0 v0 v0 /32 v0 /32 16

v0 /16 16

v0 t − 16t 2 16t 2 − v0 t + = v0 /2. = 0 v0 /32 v0 v0 Review the MVT stated in Section 4.3 (Theorem 1, p. 192) and show how it can be used, together with the Fundamental Theorem of Calculus, to prove the MVT for integrals.

308

CHAPTER 6


6.3 Volumes of Revolution Preliminary Questions 1. Which of the following is a solid of revolution? (a) Sphere (b) Pyramid

(c) Cylinder

(d) Cube

SOLUTION The sphere and the cylinder have circular cross sections; hence, these are solids of revolution. The pyramid and cube do not have circular cross sections, so these are not solids of revolution.

2. True or false? When a solid is formed by rotating the region under a graph about the x-axis, the cross sections perpendicular to the x-axis are circular disks. SOLUTION

True. The cross sections will be disks with radius equal to the value of the function.

3. True or false? When a solid is formed by rotating the region between two graphs about the x-axis, the cross sections perpendicular to the x-axis are circular disks. SOLUTION

False. The cross sections may be washers.

4. Which of the following integrals expresses the volume of the solid obtained by rotating the area between y = f (x) and y = g(x) over [a, b] around the x-axis [assume f (x) ≥ g(x) ≥ 0]? " b 2 (a) π f (x) − g(x) d x a " b (b) π f (x)2 − g(x)2 d x a

SOLUTION The correct answer is (b). Cross sections of the solid will be washers with outer radius f (x) and inner radius g(x). The area of the washer is then π f (x)2 − π g(x)2 = π ( f (x)2 − g(x)2 ).

Exercises In Exercises 1–4, (a) sketch the solid obtained by revolving the region under the graph of f (x) about the x-axis over the given interval, (b) describe the cross section perpendicular to the x-axis located at x, and (c) calculate the volume of the solid. 1. f (x) = x + 1,

[0, 3]

SOLUTION

(a) A sketch of the solid of revolution is shown below: y 2 x 1

2

3

−2

(b) Each cross section is a disk with radius x + 1. (c) The volume of the solid of revolution is

π

" 3 0

(x + 1)2 d x = π

" 3 0

(x 2 + 2x + 1) d x = π

3 1 3 x + x 2 + x = 21π . 3 0

√ 3. f (x) = x + 1, [1, 4] f (x) = x 2 , [1, 3] SOLUTION

(a) A sketch of the solid of revolution is shown below: y 2 1 x −1 −2

(b) Each cross section is a disk with radius

√

x + 1.

1

2

3

4

S E C T I O N 6.3

Volumes of Revolution

309

(c) The volume of the solid of revolution is 4 " 4 " 4 √ 21π 1 2 ( x + 1)2 d x = π (x + 1) d x = π x + x = . 2 2 1 1 1

π

In Exercises 5–12,−1find the volume of the solid obtained by rotating the region under the graph of the function about the f (x) = x , [1, 2] x-axis over the given interval. 5. f (x) = x 2 − 3x, SOLUTION

[0, 3]

The volume of the solid of revolution is

π

" 3 0

(x 2 − 3x)2 d x =

π

" 3 0

(x 4 − 6x 3 + 9x 2 ) d x =

π

3 81π 1 5 3 4 3 x − x + 3x = . 5 2 10 0

7. f (x) = x 5/31, [1, 8] f (x) = 2 , [1, 4] SOLUTION The x volume of the solid of revolution is " 8 " 8 3π 13/3 8 3π 13 24573π π (x 5/3 )2 d x = π x 10/3 d x = x = 13 (2 − 1) = 13 . 13 1 1 1 2 9. f (x)f (x) = = 4 −, x 2[1, , 3] [0, 2] x +1 SOLUTION

The volume of the solid of revolution is

π

" 3 1

3 2 " 3 2 d x = 4π (x + 1)−2 d x = −4π (x + 1)−1 = π . x +1 1 1

√ x + 1, [0, π ] 11. f (x) = cos f (x) = x 4 + 1, [1, 3] SOLUTION The volume of the solid of revolution is

π

π " π " π √ ( cos x + 1)2 d x = π (cos x + 1) d x = π (sin x + x) = π 2 . 0

0

0

√ (a) sketch the region enclosed by the curves, (b) describe the cross section perpendicular to the In Exercises 13–18, f (x) = cos x sin x, [0, π /2] x-axis located at x, and (c) find the volume of the solid obtained by rotating the region about the x-axis. 13. y = x 2 + 2,

y = 10 − x 2

SOLUTION

(a) Setting x 2 + 2 = 10 − x 2 yields 2x 2 = 8, or x 2 = 4. The two curves therefore intersect at x = ±2. The region enclosed by the two curves is shown in the figure below. y y = 10 − x 2 8 4 y = x2 + 2 −2

x

−1

1

2

(b) When the region is rotated about the x-axis, each cross section is a washer with outer radius R = 10 − x 2 and inner radius r = x 2 + 2. (c) The volume of the solid of revolution is " 2

" 2

2 (10 − x 2 )2 − (x 2 + 2)2 d x = π π (96 − 24x 2 ) d x = π 96x − 8x 3 = 256π . −2

15. y = 16 − x, y = 3x + 12, y = x 2 , y = 2x + 3

−2

−2

x = −1

SOLUTION

(a) Setting 16 − x = 3x + 12, we find that the two lines intersect at x = 1. The region enclosed by the two curves is shown in the figure below.

310

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L y

y = 3x + 12

10

−1

y = 16 − x

x

−0.5

0.5

1

(b) When the region is rotated about the x-axis, each cross section is a washer with outer radius R = 16 − x and inner radius r = 3x + 12. (c) The volume of the solid of revolution is

π

1 " 1

" 1 8 656π (16 − x)2 − (3x + 12)2 d x = π (112 − 104x − 8x 2 ) d x = π 112x − 52x 2 − x 3 = . 3 3 −1 −1 −1

π 17. y = sec x, 1 y = 0,5 x = − 4 , y = , y = −x x 2

x=

π 4

SOLUTION

(a) The region in question is shown in the figure below. y

y = sec x

1.2 0.8 0.4 x

−0.4

0.4

(b) When the region is rotated about the x-axis, each cross section is a circular disk with radius R = sec x. (c) The volume of the solid of revolution is π /4 " π /4 2 π (sec x) d x = π (tan x) = 2π . −π /4

−π /4

In Exercises 19–22, find the volume of the solid obtained by rotating the region enclosed by the graphs about the y-axis π = secinterval. x, y = csc x, y = 0, x = 0, and x = . over theygiven 2 √ 19. x = y, x = 0; 1 ≤ y ≤ 4 SOLUTION When the region in question (shown in the figure below) is rotated about the y-axis, each cross section is a √ disk with radius y. The volume of the solid of revolution is

" 4 √ 2 15π π y 2 4 π y dy = = 2 . 2 1 1 y 4 3 x=

y

2 1 x 0

0.5

1

1.5

2

√ 21. x = y 2 , √ x = y; 0 ≤ y ≤ 1 x = sin y, x = 0; 0 ≤ y ≤ π SOLUTION When the region in question (shown in the figure below) is rotated about the y-axis, each cross section is a √ washer with outer radius R = y and inner radius r = y 2 . The volume of the solid of revolution is 1 " 1

y2 3π y 5 √ 2 2 2 ( y) − (y ) d y = π π − . = 2 5 10 0 0

S E C T I O N 6.3


311

y 1 x = y2 x=

y

x 0

1

In Exercises 23–28, find the volume of the solid obtained by rotating region A in Figure 10 about the given axis. x = 4 − y, x = 16 − y 2 ; −3 ≤ y ≤ 4 y

y = x2 + 2

6 A 2

B x 1

2

FIGURE 10

23. x-axis Rotating region A about the x-axis produces a solid whose cross sections are washers with outer radius R = 6 and inner radius r = x 2 + 2. The volume of the solid of revolution is 2 " 2

" 2 1 704π 4 (6)2 − (x 2 + 2)2 d x = π π (32 − 4x 2 − x 4 ) d x = π 32x − x 3 − x 5 = . 3 5 15 0 0 0

SOLUTION

25. y = 2 y = −2 SOLUTION Rotating the region A about y = 2 produces a solid whose cross sections are washers with outer radius R = 6 − 2 = 4 and inner radius r = x 2 + 2 − 2 = x 2 . The volume of the solid of revolution is 2 " 2

128π 1 π . 42 − (x 2 )2 d x = π 16x − x 5 = 5 5 0 0 27. x = −3 y-axis

Rotating region A about x = −3 produces a solid whose cross sections are washers with outer radius √ y − 2 − (−3) = y − 2 + 3 and inner radius r = 0 − (−3) = 3. The volume of the solid of revolution is

SOLUTION

R=

√

π

" 6

2

(3 +

6 " 6 1 y − 2)2 − (3)2 d y = π (6 y − 2 + y − 2) d y = π 4(y − 2)3/2 + y 2 − 2y = 40π . 2 2 2

In Exercises x = 229–34, find the volume of the solid obtained by rotating region B in Figure 10 about the given axis. 29. x-axis Rotating region B about the x-axis produces a solid whose cross sections are disks with radius R = x 2 + 2. The volume of the solid of revolution is 2 " 2 " 2 376π 1 5 4 3 π (x 2 + 2)2 d x = π (x 4 + 4x 2 + 4) d x = π x + x + 4x = . 5 3 15 0 0 0 SOLUTION

31. y = 6 y = −2 SOLUTION Rotating region B about y = 6 produces a solid whose cross sections are washers with outer radius R = 6 − 0 = 6 and inner radius r = 6 − (x 2 + 2) = 4 − x 2 . The volume of the solid of revolution is " 2

" 2

824π 8 3 1 5 2 2 2 2 2 4 6 − (4 − x ) d y = π 20 + 8x − x d y = π 20x + x − x = π . 3 5 15 0 0 0 33. x = 2 y-axis Hint for Exercise 32: Express the volume as a sum of two integrals along the y-axis, or use Exercise 26.

312

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L SOLUTION Rotating region B about x = 2 produces a solid with two different cross sections. For each y ∈ [0, 2], the √ cross section is a disk with radius R = 2; for each y ∈ [2, 6], the cross section is a disk with radius R = 2 − y − 2. The volume of the solid of revolution is " 2 " 6 " 2 " 6 π (2)2 d y + π (2 − y − 2)2 d y = π 4 dy + π (2 + y − 4 y − 2) d y

0

2

0

2

2 6 1 2 8 3/2 = 32π . = π (4y) + π 2y + y − (y − 2) 2 3 3 0 2

In Exercises 35–46, find the volume of the solid obtained by rotating the region enclosed by the graphs about the given x = −3 axis. 35. y = x 2 ,

y = 12 − x,

x = 0,

about y = −2

Rotating the region enclosed by y = x 2 , y = 12 − x and the y-axis (shown in the figure below) about y = −2 produces a solid whose cross sections are washers with outer radius R = 12 − x − (−2) = 14 − x and inner radius r = x 2 − (−2) = x 2 + 2. The volume of the solid of revolution is " 3

" 3 (14 − x)2 − (x 2 + 2)2 d x = π π (192 − 28x − 3x 2 − x 4 ) d x SOLUTION

0

0

3 1 1872π = π 192x − 14x 2 − x 3 − x 5 = . 5 5 0 y y = 12 − x

12

8

4

y = x2 x

0

1

2

3

37. y = 16 − x, y = 3x + 12, x = 0, about y-axis y = x 2 , y = 12 − x, x = 0, about y = 15 SOLUTION Rotating the region enclosed by y = 16 − x, y = 3x + 12 and the y-axis (shown in the figure below) about the y-axis produces a solid with two different cross sections. For each y ∈ [12, 15], the cross section is a disk with radius R = 13 (y − 12); for each y ∈ [15, 16], the cross section is a disk with radius R = 16 − y. The volume of the solid of revolution is 2 " 15 " 16 1 π (16 − y)2 d y (y − 12) d y + π 3 12 15 " 15 " 16 1 2 (y 2 − 32y + 256) d y (y − 24y + 144) d y + π =π 12 9 15 15 16 4 π 1 3 1 3 2 2 = y − 12y + 144y + π y − 16y + 256y = π . 9 3 3 3 12 15 y 16 14 12 10 8 6 4 2

y = 16 − x y = 3x + 12

x 0

9 y= , +about 39. y =y =2 ,16 − x, 10y−=x 23x 12, x-axis x = 0, x

0.2 0.4 0.6 0.8

1

about x = 2

SOLUTION The region enclosed by the two curves is shown in the figure below. Note that the region consists of two pieces that are symmetric with respect to the y-axis. We may therefore compute the volume of the solid of revolution by

S E C T I O N 6.3


313

considering one of the pieces and doubling the result. Rotating the portion of the region in the first quadrant about the x-axis produces a solid whose cross sections are washers with outer radius R = 10 − x 2 and inner radius r = 9x −2 . The volume of the solid of revolution is " 3

" 3

(10 − x 2 )2 − (9x −2 )2 d x = 2π 100 − 20x 2 + x 4 − 81x −4 d x 2π 1

1

= 2π

3 20 3 1 5 −3 = 1472π . x + x + 27x 100x − 3 5 15 1

y

y = 10 − x 2

8 6 4 y = 92 x

2 −3 −2 −1

1

x 2

3

1 5 41. y = , 9y = − x, about y-axis y x= 2 , y2 = 10 − x 2 , about y = 12 x 5 SOLUTION Rotating the region enclosed by y = x −1 and y = 2 − x (shown in the figure below) about the y-axis produces a solid whose cross sections are washers with outer radius R = 52 − y and inner radius r = y −1 . The volume of the solid of revolution is 2 " 2 " 2 25 5 −1 2 dy = π π − y − (y ) − 5y + y 2 − y −2 d y 2 4 1/2 1/2 = π

2 5 1 9π 25 y − y 2 + y 3 + y −1 . = 4 2 3 8 1/2

y 2 1.5

y = 2.5 − x

1 0.5

y=

1 x x

0

0.5

1

1.5

2

43. y = x 3 , y = x 1/3 , about y-axis x = 2, x = 3, y = 16 − x 4 , y = 0, about y-axis SOLUTION Rotating the region enclosed by y = x 3 and y = x 1/3 (shown in the figure below) about the y-axis produces a solid whose cross sections are washers with outer radius R = y 1/3 and inner radius r = y 3 . The volume of the solid of revolution is " 1

" 1 32π 3 5/3 1 7 1 (y 1/3 )2 − (y 3 )2 d y = π π (y 2/3 − y 6 ) d y = π − y = y . 5 7 35 −1 −1 −1 y 1

y = x 1/3 y = x3 x

−1

1

−1

about x-axis 45. y 2 = 4x, 3 y = x, 1/3 y = x , y = x , about x = −2 SOLUTION Rotating the region enclosed by y 2 = 4x and y = x (shown in the figure below) about the x-axis produces √ a solid whose cross sections are washers with outer radius R = 2 x and inner radius r = x. The volume of the solid of

314

CHAPTER 6


revolution is " 4

√ 2 1 3 4 32π 2 2 (2 x) − x d x = π 2x − x = π . 3 3 0 0 y 4 3

y 2 = 4x

2

y=x

1 x 0

1

2

3

4

47. Sketch the hypocycloid x 2/3 + y 2/3 = 1 and find the volume of the solid obtained by revolving it about the 2= y 4x, y = x, about y = 8 x-axis. SOLUTION

A sketch of the hypocycloid is shown below. y 1

x

−1

1

−1

3/2 For the hypocycloid, y = ± 1 − x 2/3 . Rotating this region about the x-axis will produce a solid whose cross 3/2

sections are disks with radius R = 1 − x 2/3 . Thus the volume of the solid of revolution will be 1 " 1

2 −x 3 32π 9 7/3 9 5/3 2/3 3/2 (1 − x ) π dx = π − x +x = + x . 3 7 5 105 −1 −1

49. A “bead” is formed by removing a cylinder of radius r from the center of a sphere of radius R (Figure 12). Find the 2 2 solid generated rotating volume The of the bead with r =by 1 and R =the 2. region between the branches of the hyperbola y − x = 1 about the x-axis is called a hyperboloid (Figure 11). Find the volume of the hyperboloid for −a ≤ x ≤ a. y

y

h r

R

x

x

FIGURE 12 A bead is a sphere with a cylinder removed.

The equation of the outer circle is x 2 + y 2 = 22 , and the inner cylinder intersects the sphere when √ y = ± 3. Each cross section of the bead is a washer with outer radius 4 − y 2 and inner radius 1, so the volume is given by 2 " √3 " √3

√ 2 2 π √ 4− y −1 d y = π √ 3 − y 2 d y = 4π 3. SOLUTION

− 3


− 3

x 2 y 2 + = 1 πaround the x-axis is 51. The solid generated by rotating the region inside the ellipse with equation 3 Find the volume V of the bead (Figure 12) in terms of r and R. Thena show that b V = h , where h is the 6 4 2 is theVvolume the ellipseinisterms rotated called an ellipsoid. ShowThis thatformula the ellipsoid volumeconsequence: height of the bead. has a has surprising can beifexpressed of around h alone,the it 3 π ab . WhatSince y-axis? follows that two beads of height 2 in., one formed from a sphere the size of an orange and the other the size of the earth, would have the same volume! Can you explain intuitively how this is possible?

S E C T I O N 6.3


315

SOLUTION

• Rotating the ellipse about the x-axis produces an ellipsoid whose cross sections are disks with radius R = b 1 − (x/a)2 . The volume of the ellipsoid is then

π

2 a " a " a 4 1 1 d x = b2 π b 1 − (x/a)2 1 − 2 x 2 d x = b2 π x − 2 x 3 = π ab2 . 3 a 3a −a −a −a

• Rotating the ellipse about the y-axis produces an ellipsoid whose cross sections are disks with radius R = a 1 − (y/b)2 . The volume of the ellipsoid is then

2 b " b " b 4 1 1 d y = a2 π a 1 − (y/b)2 1 − 2 y 2 d y = a 2 π y − 2 y 3 = π a 2 b. 3 b 3b −b −b −b 53. The curve y = f (x) in Figure 14, called a tractrix, has the following property: the tangent line at each point (x, y) A doughnut-shaped solid is called a torus (Figure 13). Use the washer method to calculate the volume of the on the curve has slope torus obtained by rotating the region inside the circle with equation (x − a)2 + y 2 = b2 around the y-axis (assume that a > b). Hint: Evaluate the integral by interpreting it as −ythe area of a circle. dy = . dx 1 − y2 y 1

c

R

y = f (x) x a

2

FIGURE 14 The tractrix.

Let R be the shaded region under the graph of 0 ≤ x ≤ a in Figure 14. Compute the volume V of the solid obtained by revolving R around the x-axis in terms of the constant c = f (a). Hint: Use the disk method and the substitution u = f (x) to show that V =π

" 1 u 1 − u 2 du c

SOLUTION Let y = f (x) be the tractrix depicted in Figure 14. Rotating the region R about the x-axis produces a solid whose cross sections are disks with radius f (x). The volume of the resulting solid is then " a V =π [ f (x)]2 d x.

0

Now, let u = f (x). Then − f (x) −u dx = d x; du = f (x) d x = 2 1 − [ f (x)] 1 − u2 hence, 1 − u2 du, dx = − u and V =π

" c

u2

1

1 − u2 − du u

=π

" 1 u 1 − u 2 du. c

Carrying out the integration, we find 1 π π V = − (1 − u 2 )3/2 = (1 − c2 )3/2 . 3 3 c Verify the formula " x2 x1

1 (x − x 1 )(x − x 2 ) d x = (x 1 − x2 )3 6

316

CHAPTER 6


55. Let R be the region in the unit circle lying above the cut with the line y = mx + b (Figure 16). Assume the points where the line intersects the circle lie above the x-axis. Use the method of Exercise 54 to show that the solid obtained by π rotating R about the x-axis has volume V = hd 2 , with h and d as in the figure. 6 y = mx + b

y

R

d h

x2 + y2 = 1 x

FIGURE 16

Let x 1 and x 2 denote the x-coordinates of the points of intersection between the circle x 2 + y 2 = 1 and the line y = mx + b with x 1 < x 2 . Rotating the regionenclosed by the two curves about the x-axis produces a solid whose cross sections are washers with outer radius R = 1 − x 2 and inner radius r = mx + b. The volume of the resulting solid is then " x2

V =π (1 − x 2 ) − (mx + b)2 d x SOLUTION

x1

Because x1 and x 2 are roots of the equation (1 − x 2 ) − (mx + b)2 = 0 and (1 − x 2 ) − (mx + b)2 is a quadratic polynomial in x with leading coefficient −(1 + m 2 ), it follows that (1 − x 2 ) − (mx + b)2 = −(1 + m 2 )(x − x 1 )(x − x 2 ). Therefore, " x2 π (x − x 1 )(x − x 2 ) d x = (1 + m 2 )(x 2 − x 1 )3 . V = −π (1 + m 2 ) 6 x1 From the diagram, we see that h = x 2 − x 1 . Moreover, by the Pythagorean theorem, d 2 = h 2 + (mh)2 = (1 + m 2 )h 2 . Thus, π π π V = (1 + m 2 )h 3 = h (1 + m 2 )h 2 = hd 2 . 6 6 6

6.4 The Method of Cylindrical Shells Preliminary Questions 1. Consider the region R under the graph of the constant function f (x) = h over the interval [0, r ]. What are the height and radius of the cylinder generated when R is rotated about: (a) the x-axis (b) the y-axis SOLUTION

(a) When the region is rotated about the x-axis, each shell will have radius h and height r. (b) When the region is rotated about the y-axis, each shell will have radius r and height h. 2. Let V be the volume of a solid of revolution about the y-axis. (a) Does the Shell Method for computing V lead to an integral with respect to x or y? (b) Does the Disk or Washer Method for computing V lead to an integral with respect to x or y? SOLUTION

(a) The Shell method requires slicing the solid parallel to the axis of rotation. In this case, that will mean slicing the solid in the vertical direction, so integration will be with respect to x. (b) The Disk or Washer method requires slicing the solid perpendicular to the axis of rotation. In this case, that means slicing the solid in the horizontal direction, so integration will be with respect to y.

Exercises In Exercises 1–10, sketch the solid obtained by rotating the region underneath the graph of the function over the given interval about the y-axis and find its volume. 1. f (x) = x 3 ,

[0, 1]

The Method of Cylindrical Shells


317

A sketch of the solid is shown below. Each shell has radius x and height x 3 , so the volume of the solid is 2π

" 1 0

x · x 3 d x = 2π

" 1 0

x 4 d x = 2π

2 1 5 1 x = π. 5 5 0

y 1

x

−1

1

3. f (x) = 3x + √ 2, [2, 4] f (x) = x, [0, 4] SOLUTION A sketch of the solid is shown below. Each shell has radius x and height 3x + 2, so the volume of the solid is " 4 " 4

4 2π x(3x + 2) d x = 2π (3x 2 + 2x) d x = 2π x 3 + x 2 = 136π . 2

2

2

y 14 12 10 8 6 4 2 x

−4

4

5. f (x) = 4 − x 2 , 2[0, 2] f (x) = 1 + x , [1, 3] SOLUTION A sketch of the solid is shown below. Each shell has radius x and height 4 − x 2 , so the volume of the solid is 2 " 2 " 2 1 2π x(4 − x 2 ) d x = 2π (4x − x 3 ) d x = 2π 2x 2 − x 4 = 8π . 4 0 0 0 y

2

x

−2

2

√ 2 ), [0, π ] 7. f (x) = sin(x f (x) = x 2 + 9, [0, 3] SOLUTION A sketch of the solid is shown below. Each shell has radius x and height sin x 2 , so the volume of the solid is " √π 2π x sin(x 2 ) d x. 0

Let u = x 2 . Then du = 2x d x and π " √π " π 2π x sin(x 2 ) d x = π sin u du = −π (cos u) = 2π . 0

0

0

y 1 0.5

−2

−1

x 1

2

318

CHAPTER 6


9. f (x) = x + 1−1− 2x 2 , [0, 1] f (x) = x , [1, 3] SOLUTION A sketch of the solid is shown below. Each shell has radius x and height x + 1 − 2x 2 , so the volume of the solid is " 1 " 1 1 3 1 2 1 4 1 2 2π x(x + 1 − 2x 2 ) d x = 2π (x 2 + x − 2x 3 ) d x = 2π x + x − x = π. 3 2 2 3 0 0 0 y 1 0.5 x

−1

1

In Exercises 11–14, xuse the Shell Method to compute the volume of the solids obtained by rotating the region enclosed f (x) =of , [1, 4] the y-axis. by the graphs the functions about 1 + x3 11. y = x 2 ,

y = 8 − x 2,

x =0

The region enclosed by y = x 2 , y = 8 − x 2 and the y-axis is shown below. When rotating this region about the y-axis, each shell has radius x and height 8 − x 2 − x 2 = 8 − 2x 2 . The volume of the resulting solid is SOLUTION

2π

" 2 0

x(8 − 2x 2 ) d x = 2π

2 1 (8x − 2x 3 ) d x = 2π 4x 2 − x 4 = 16π . 2 0 0

" 2

y 8 y = 8 − x2 6 4 2

y = x2 x

0

0.5

1

1.5

2

√ x2 13. y = x, y = y = 8 − x 3 , y = 8 − 4x √ SOLUTION The region enclosed by y = x and y = x 2 is shown below. When rotating this region about the y-axis, √ each shell has radius x and height x − x 2 . The volume of the resulting solid is 2π

" 1 0

√ x( x − x 2 ) d x = 2π

" 1 0

(x 3/2 − x 3 ) d x = 2π

3 2 5/2 1 4 1 − x = π. x 5 4 10 0

y 1 y=

x y = x2

x 0

1

In Exercises y= 1 − |x −15–16, 1|, yuse = 0the Shell Method to compute the volume of rotation of the region enclosed by the curves about the y-axis. Use a computer algebra system or graphing utility to find the points of intersection numerically. 15. y = 12 x 2 ,

y = sin(x 2 )

The region enclosed by y = 12 x 2 and y = sin x 2 is shown below. When rotating this region about the y-axis, each shell has radius x and height sin x 2 − 12 x 2 . Using a computer algebra system, we find that the x-coordinate of the point of intersection on the right is x = 1.376769504. Thus, the volume of the resulting solid of revolution is " 1.376769504 1 x sin x 2 − x 2 d x = 1.321975576. 2π 2 0 SOLUTION

S E C T I O N 6.4


319

y 1 y = sin x 2 y=

x2 2 x

0

1

In Exercises 17–22, sketch the solid obtained by rotating the region underneath the graph of the function over the interval y = 1 − x 4 , y = x, x = 0 about the given axis and calculate its volume using the Shell Method. 17. f (x) = x 3 , [0, 1], SOLUTION

x =2

A sketch of the solid is shown below. Each shell has radius 2 − x and height x 3 , so the volume of the solid

is 2π

" 1 0

(2 − x) x 3

d x = 2π

" 1 0

(2x 3 − x 4 ) d x = 2π

x4 x5 − 2 5

1 3π . = 5 0

y 1

x 0

4

19. f (x) = x −4 ,3 [−3, −1], x = 4 f (x) = x , [0, 1], x = −2 SOLUTION A sketch of the solid is shown below. Each shell has radius 4 − x and height x −4 , so the volume of the solid is " −1 " −1

280π 1 −2 4 −3 −1 −4 −4 −3 d x = 2π 2π (4x − x ) d x = 2π x − x (4 − x) x = 81 . 2 3 −3 −3 −3 y

0.8

0.4 x

−2

10

21. f (x) = a − bx, [0, a/b], x = −1, a, b > 0 1 2], isxshown = 0 below. Each shell has radius x − (−1) = x + 1 and height a − bx, so the f (x) = SOLUTION A sketch of,the[0, solid 2 x +1 volume of the solid is " a/b " a/b

a + (a − b)x − bx 2 d x (x + 1) (a − bx) d x = 2π 2π 0

0

a − b 2 b 3 a/b x − x 2 3 0 2 2 3 a a 2 (a + 3b) a a (a − b) = = 2π − π. + b 2b2 3b2 3b2 = 2π ax +

y a

−2 − a /b

x −2 −1

a/b

320

CHAPTER 6


In Exercises 23–28, use the Shell Method to calculate the volume of rotation about the x-axis for the region underneath f (x) = 1 − x 2 , [−1, 1], x = c (with c > 1) the graph. 23. y = x,

0≤x ≤1

SOLUTION When the region shown below is rotated about the x-axis, each shell has radius y and height 1 − y. The volume of the resulting solid is

2π

" 1 0

y(1 − y) d y = 2π

" 1 0

(y − y 2 ) d y = 2π

π 1 2 1 3 1 y − y = . 2 3 3 0

y 1 0.8 0.6

y=x

0.4 0.2 x 0

0.2 0.4 0.6 0.8

1

25. y = x 1/3 − 2,2 8 ≤ x ≤ 27 y =4−x , 0≤ x ≤2 SOLUTION When the region shown below is rotated about the x-axis, each shell has radius y and height 27 − (y + 2)3 . The volume of the resulting solid is 2π

" 1 0

" 1

19y − 12y 2 − 6y 3 − y 4 d y y · 27 − (y + 2)3 d y = 2π 0

= 2π

1 3 1 38π 19 2 y − 4y 3 − y 4 − y 5 = . 2 2 5 5 0

y 1 0.8 0.6

y=

3

5

10 15 20 25 30

x−2

0.4 0.2 x 0

27. y = x −2 ,−12 ≤ x ≤ 4 y = x , 1 ≤ x ≤ 4. Sketch the region and express the volume as a sum of two integrals. SOLUTION When the region shown below is rotated about the x-axis, two different shells are generated. For each 1 ], the shell has radius y and height 4 − 2 = 2; for each y ∈ [ 1 , 1 ], the shell has radius y and height √1 − 2. y ∈ [0, 16 16 4 y The volume of the resulting solid is 2π

" 1/16 0

2y d y + 2π

" 1/4 1/16

y(y −1/2 − 2) d y = 2π

" 1/16 0

2y d y + 2π

" 1/4 1/16

(y 1/2 − 2y) d y

1/4

1/16 2 3/2 = 2π y 2 + 2π − y 2 y 0 3 1/16 =

11π 7π π + = . 128 384 192

y 0.25 0.2 y = 12 x

0.15 0.1 0.05

x 0

y=

√

x, 1 ≤ x ≤ 4

1

2

3

4


S E C T I O N 6.4

321

29. Use both the Shell and Disk Methods to calculate the volume of the solid obtained by rotating the region under the graph of f (x) = 8 − x 3 for 0 ≤ x ≤ 2 about: (a) the x-axis (b) the y-axis SOLUTION

(a) x-axis: Using the disk method, the cross sections are disks with radius R = 8 − x 3 ; hence the volume of the solid is

π

" 2 0

(8 − x 3 )2 d x = π

64x − 4x 4 +

1 7 2 576π x = . 7 7 0

With the shell method, each shell has radius y and height (8 − y)1/3 . The volume of the solid is 2π

" 8 0

y (8 − y)1/3 d y

Let u = 8 − y. Then d y = −du, y = 8 − u and 2π

" 8 0

y (8 − y)1/3 d y = 2π

" 8 0

= 2π

(8 − u) · u 1/3 du = 2π

" 8 0

(8u 1/3 − u 4/3 ) du

8 3 576π . 6u 4/3 − u 7/3 = 7 7 0

(b) y-axis: With the shell method, each shell has radius x and height 8 − x 3 . The volume of the solid is 2π

" 2 0

2 1 96π . 4x 2 − x 5 = 5 5 0

x(8 − x 3 ) d x = 2π

Using the disk method, the cross sections are disks with radius R = (8 − y)1/3 . The volume is then given by

π

8 3π 96π 2/3 5/3 (8 − y) d y = − (8 − y) = . 5 5 0 0

" 8

31. Assume that the graph in Figure 9(A) can be described by both y = f (x) and x = h(y). Let V be the volume of the Sketch the solid of rotation about the y-axis for the region under the graph of the constant function f (x) = c solid obtained by rotating the region under the curve about the y-axis. (where c > 0) for 0 ≤ x ≤ r . (a) Describe the figures generated by rotating segments AB and C B about the y-axis. (a) Find the volume without using integration. (b) Set up integrals that compute V by the Shell and Disk Methods. (b) Use the Shell Method to compute the volume. y

y y = f(x) x = h(y)

1.3

y = g(x)

B

A

A'

B'

x C

x

2

C'

(A)

2 (B)

FIGURE 9 SOLUTION

(a) When rotated about the y-axis, the segment AB generates a disk with radius R = h(y) and the segment C B generates a shell with radius x and height f (x). (b) Based on Figure 9(A) and the information from part (a), when using the Shell Method, V = 2π

" 2 0

x f (x) d x;

when using the Disk Method, V =π

" 1.3 0

(h(y))2 d y.

In Exercises 33–38, use the Shell Method to find the volume of the solid obtained by rotating region A in Figure 10 about Let W be the volume of the solid obtained by rotating the region under the curve in Figure 9(B) about the the given axis. y-axis. (a) Describe the figures generated by rotating segments A B and A C about the y-axis. (b) Set up an integral that computes W by the Shell Method. (c) Explain the difficulty in computing W by the Washer Method.

322

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L y

y = x2 + 2

6 A 2

B x 1

2

FIGURE 10

33. y-axis When rotating region A about the y-axis, each shell has radius x and height 6 − (x 2 + 2) = 4 − x 2 . The volume of the resulting solid is SOLUTION

2π

" 2 0

x(4 − x 2 ) d x = 2π

2 1 (4x − x 3 ) d x = 2π 2x 2 − x 4 = 8π . 4 0 0

" 2

35. x = 2 x = −3 SOLUTION When rotating region A about x = 2, each shell has radius 2 − x and height 6 − (x 2 + 2) = 4 − x 2 . The volume of the resulting solid is " 2

1 4 2 40π 2 3 2 2 3 2 . 8 − 2x − 4x + x d x = 2π 8x − x − 2x + x = 2π (2 − x) 4 − x d x = 2π 3 4 3 0 0 0 " 2

37. y = −2 x-axis

When rotating region A about y = −2, each shell has radius y − (−2) = y + 2 and height volume of the resulting solid is SOLUTION

2π

" 6 2

√

y − 2. The

(y + 2) y − 2 d y

Let u = y − 2. Then du = d y, y + 2 = u + 4 and 2π

" 4 √ 1024π 2 5/2 8 3/2 4 (y + 2) y − 2 d y = 2π (u + 4) u du = 2π + u = u . 5 3 15 2 0 0

" 6

In Exercises y = 639–44, use the Shell Method to find the volumes of the solids obtained by rotating region B in Figure 10 about the given axis. 39. y-axis When rotating region B about the y-axis, each shell has radius x and height x 2 + 2. The volume of the resulting solid is SOLUTION

2π

" 2 0

x(x 2 + 2) d x = 2π

" 2 0

(x 3 + 2x) d x = 2π

2 1 4 x + x 2 = 16π . 4 0

41. x = 2 x = −3 SOLUTION When rotating region B about x = 2, each shell has radius 2 − x and height x 2 + 2. The volume of the resulting solid is 2π

" 2 0

(2 − x) x 2 + 2

2 " 2

32π 2 3 1 4 2 3 2 d x = 2π x − x + 4x − x = . 2x − x + 4 − 2x d x = 2π 3 4 3 0 0

43. y = −2 x-axis

SOLUTION When rotating region B about y = −2, two different shells are generated. For each y ∈ [0, 2], the resulting shell has radius √ y − (−2) = y + 2 and height 2; for each y ∈ [2, 6], the resulting shell has radius y − (−2) = y + 2 and height 2 − y − 2. The volume of the solid is then

2π

" 2 0

2(y + 2) d y + 2π

" 6 2

(y + 2)(2 −

y − 2) d y = 2π

" 6 0

2(y + 2) d y − 2π

= 120π − 2π

" 6 2

" 6 2

(y + 2) y − 2 d y

(y + 2) y − 2 d y.

S E C T I O N 6.4


323

In the remaining integral, let u = y − 2, so du = d y and y + 2 = u + 4. Then 2π

" 4 √ 2 5/2 8 3/2 4 1024π (y + 2) y − 2 d y = 2π (u + 4) u du = 2π + u u = 15 . 5 3 2 0 0

" 6

Finally, the volume of the solid is 1024π 776π = . 15 15

120π −

45. Use the Shell Method to compute the volume of a sphere of radius r . y=8 SOLUTION A sphere of radius r can be generated by rotating the region under the semicircle y = r 2 − x 2 around the x-axis. Each shell has radius y and height

2 2 2 2 r − y − − r − y = 2 r 2 − y2. Thus, the volume of the sphere is 2π

" r 0

2y r 2 − y 2 d y.

Let u = r 2 − y 2 . Then du = −2y d y and 2π

2 " r2

√ 2 3/2 r 4 3 u 2y r 2 − y 2 d y = 2π u du = 2π = 3 πr . 3 0 0 0

" r

2+ 47. UseUse the the Shell Method to compute the volume of the by rotating the interior of the circle (x −ra)from Shell Method to calculate the volume V torus of theobtained “bead” formed by removing a cylinder of radius y 2 =the r 2center about of thea y-axis, where a > r . Hint: Evaluate the integral by interpreting part of it as the area of a circle. sphere of radius R (compare with Exercise 49 in Section 6.3). SOLUTION When rotating the region enclosed by the circle (x − a)2 + y 2 = r 2 about the y-axis each shell has radius x and height

r 2 − (x − a)2 − − r 2 − (x − a)2 = 2 r 2 − (x − a)2 .

The volume of the resulting torus is then 2π

" a+r a−r

2x r 2 − (x − a)2 d x.

Let u = x − a. Then du = d x, x = u + a and " a+r " r 2 2 2π 2x r − (x − a) d x = 2π 2(u + a) r 2 − u 2 du −r

a−r

= 4π

" r

−r

" r u r 2 − u 2 du + 4a π r 2 − u 2 du. −r

Now, " r −r

u r 2 − u 2 du = 0

because the integrand is an odd function and the integration interval is symmetric with respect to zero. Moreover, the other integral is one-half the area of a circle of radius r ; thus, " r 1 r 2 − u 2 du = π r 2 . 2 −r Finally, the volume of the torus is 4π (0) + 4a π

1 2 πr 2

= 2π 2 ar 2 .

49. Use the most convenient method to compute the volume of the solid obtained by rotating the region in Figure 12 Use the Shell or Disk Method (whichever is easier) to compute the volume of the solid obtained by rotating the about the axis: region in Figure 11 about: (a) x = 4 (b) y = −2 (a) the x-axis (b) the y-axis

324

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L y y = x3 + 2 y = 4 − x2 x 1

2

FIGURE 12

Examine Figure 12. If the indicated region is sliced vertically, then the top of the slice lies along the curve y = x 3 + 2 and the bottom lies along the curve y = 4 − x 2 . On the other hand, the left end of a horizontal slice switches from y = 4 − x 2 to y = x 3 + 2 at y = 3. Here, vertical slices will be more convenient. (a) Now, suppose the region in Figure 12 is rotated about x = 4. Because a vertical slice is parallel to x = 4, we will calculate the volume of the resulting solid using the shell method. Each shell has radius 4 − x and height x 3 + 2 − (4 − x 2 ) = x 3 + x 2 − 2, so the volume is SOLUTION

2π

" 2 1

(4 − x)(x 3 + x 2 − 2) d x = 2π

2 3 4 563π 1 . − x 5 + x 4 + x 3 + x 2 − 8x = 5 4 3 30 1

(b) Now suppose the region is rotated about y = −2. Because a vertical slice is perpendicular to y = −2, we will calculate the volume of the resulting solid using the disk method. Each cross section is a washer with outer radius R = x 3 + 2 − (−2) = x 3 + 4 and inner radius r = 4 − x 2 − (−2) = 6 − x 2 , so the volume is

π

2 " 2

1748π 1 7 1 5 (x 3 + 4)2 − (6 − x 2 )2 d x = π x − x + 2x 4 + 4x 3 − 20x = . 7 5 35 1 1


x 2 y 2 = 1 to (Figure Show that obtained rotating 51. Let R The be the region bounded by theofellipse surface area of a sphere radius r ais 4π+r 2 . bUse this derive13). the formula forthe thesolid volume V of abysphere of 4 2 radius R in a new way. R about the y-axis (called an ellipsoid) has volume 3 π a b. (a) Show that the volume of a thin spherical shell of inner radius r and thickness x is approximately 4π r 2 x. y (b) Approximate V by decomposing the sphere of radius R into N thin spherical shells of thickness x = R/N . b (c) Show that the approximation is a Riemann sum which converges to an integral. Evaluate the integral. R a

FIGURE 13 The ellipse

x 2 a

+

x

y 2 b

= 1.

SOLUTION Let’s slice the portion of the ellipse in the first and fourth quadrants horizontally and rotate the slices about the y-axis. The resulting ellipsoid has cross sections that are disks with radius a2 y2 R = a2 − 2 . b

Thus, the volume of the ellipsoid is b " b a2 y2 a 2 y 3 2 2 a − 2 dy = π a y − π b 3b2 −b

−b

) =π

a2 b a2b − 3

−

a2 b −a 2 b + 3

* =

4 2 π a b. 3

The bell-shaped curve in Figure 14 is the graph of a certain function y = f (x) with the following property: The tangent line at a point (x, y) on the graph has slope d y/d x = −x y. Let R be the shaded region under the graph for 6.50 Work Energy ≤ x ≤ and a in Figure 14. Use the Shell Method and the substitution u = f (x) to show that the solid obtained by revolving R around the y-axis has volume V = 2π (1 − c), where c = f (a). Observe that as c → 0, the region R becomes infinite but the volume V approaches 2π . Preliminary Questions 1. Why is integration needed to compute the work performed in stretching a spring? SOLUTION Recall that the force needed to extend or compress a spring depends on the amount by which the spring has already been extended or compressed from its equilibrium position. In other words, the force needed to move a spring is variable. Whenever the force is variable, work needs to be computed with an integral.

2. Why is integration needed to compute the work performed in pumping water out of a tank but not to compute the work performed in lifting up the tank?

S E C T I O N 6.5

Work and Energy

325

SOLUTION To lift a tank through a vertical distance d, the force needed to move the tank remains constant; hence, no integral is needed to calculate the work done in lifting the tank. On the other hand, pumping water from a tank requires that different layers of the water be moved through different distances, and, depending on the shape of the tank, may require different forces. Thus, pumping water from a tank requires that an integral be evaluated.

3. Which of the following represents the work required to stretch a spring (with spring constant k) a distance x beyond its equilibrium position: kx, −kx, 12 mk 2 , 12 kx 2 , or 12 mx 2 ? SOLUTION

The work required to stretch a spring with spring constant k a distance x beyond its equilibrium position is x " x 1 1 ky d y = ky 2 = kx 2 . 2 2 0 0

Exercises 1. How much work is done raising a 4-kg mass to a height of 16 m above ground? SOLUTION

The force needed to lift a 4-kg object is a constant (4 kg)(9.8 m/s2 ) = 39.2 N.

The work done in lifting the object to a height of 16 m is then (39.2 N)(16 m) = 627.2 J. In Exercises work (in joules) required to stretch or ftcompress a spring as indicated, assuming that the How 3–6, muchcompute work is the done raising a 4-lb mass to a height of 16 above ground? spring constant is k = 150 kg/s2 . 3. Stretching from equilibrium to 12 cm past equilibrium SOLUTION

The work required to stretch the spring 12 cm past equilibrium is " .12

.12 150x d x = 75x 2 = 1.08 J. 0

0

5. Stretching from 5 to 15 cm past equilibrium Compressing from equilibrium to 4 cm past equilibrium SOLUTION The work required to stretch the spring from 5 cm to 15 cm past equilibrium is " .15 .05

.15 150x d x = 75x 2 = 1.5 J. .05

7. If 5 J of work are needed to stretch a spring 10 cm beyond equilibrium, how much work is required to stretch it Compressing the spring 4 more cm when it is already compressed 5 cm 15 cm beyond equilibrium? SOLUTION

First, we determine the value of the spring constant as follows: " 0.1 0

kx d x =

1 2 0.1 kx = 0.005k = 5 J. 2 0

Thus, k = 1000 kg/s2 . Next, we calculate the work required to stretch the spring 15 cm beyond equilibrium: " 0.15 0

0.15 1000x d x = 500x 2 = 11.25 J. 0

9. If 10 ft-lb of work are needed to stretch a spring 1 ft beyond equilibrium, how far will the spring stretch if a 10-lb If 5 J of work are needed to stretch a spring 10 cm beyond equilibrium, how much work is required to compress weight is attached to its end? it 5 cm beyond equilibrium? SOLUTION First, we determine the value of the spring constant as follows: 1 2 1 1 kx d x = kx = k = 10 ft-lb. 2 2 0 0

" 1

Thus k = 20 lb/ft. Balancing the forces acting on the weight, we have 10 lb = kd = 20d, which implies d = 0.5 ft. A 10-lb weight will therefore stretch the spring 6 inches. In Exercises 11–14, calculate the work against gravity required to build the structure out of brick using the method of Show that the work required to stretch a spring from position a to position b is 12 k(b2 − a 2 ), where k is the Examples 2 and 3. Assume that brick has density 80 lb/ft3 . spring constant. How do you interpret the negative work obtained when |b| < |a|?

326

CHAPTER 6


11. A tower of height 20 ft and square base of side 10 ft SOLUTION The volume of one layer is 100y ft3 and so the weight of one layer is 8000y lb. Thus, the work done against gravity to build the tower is

W =

" 20 0

20 8000y d y = 4000y 2 = 1.6 × 106 ft-lb. 0

13. A 20-ft-high tower in the shape of a right circular cone with base of radius 4 ft A cylindrical tower of height 20 ft and radius 10 ft y 2 2 SOLUTION From similar triangles, the area of one layer is π 4 − 5 ft , so the volume of each small layer is y 2 y 2 3 π 4 − 5 y ft . The weight of one layer is then 80π 4 − 5 y lb. Finally, the total work done against gravity to build the tower is " 20

y 2 128,000π ft-lb. 80π 4 − y dy = 5 3 0 15. Built around 2600 BCE, the Great Pyramid of Giza in Egypt is 485 ft high (due to erosion, its current height is slightly A structure in the shape of a hemisphere of radius 4 ft less) and has a square base of side 755.5 ft (Figure 6). Find the work needed to build the pyramid if the density of the stone is estimated at 125 lb/ft3 .

FIGURE 6 The Great Pyramid in Giza, Egypt. SOLUTION

From similar triangles, the area of one layer is 755.5 −

755.5 2 2 ft , y 485

so the volume of each small layer is 755.5 2 755.5 − y y ft3 . 485 The weight of one layer is then 755.5 2 125 755.5 − y y lb. 485 Finally, the total work needed to build the pyramid was " 485 0

755.5 2 125 755.5 − y y d y = 1.399 × 1012 ft-lb. 485

In Exercises 16–20, calculate the work (in joules) required to pump all of the water out of the tank. Assume that the tank is full, distances are measured in meters, and the density of water is 1,000 kg/m3 . 17. The hemisphere in Figure 8; water exits from the spout as shown. The box in Figure 7; water exits from a small hole at the top. 10

2

FIGURE 8

S E C T I O N 6.5

Work and Energy

327

SOLUTION Place the origin at the center of the hemisphere, and let the positive y-axis point downward. The radius of a layer of water at depth y is 100 − y 2 m, so the volume of the layer is π (100 − y 2 )y m3 , and the force needed to lift the layer is 9800π (100 − y 2 )y N. The layer must be lifted y + 2 meters, so the total work needed to empty the tank is

" 10 0

9800π (100 − y 2 )(y + 2) d y =

112700000π J ≈ 1.18 × 108 J. 3

19. The horizontal cylinder in Figure 10; water exits from a small hole at the top. Hint: Evaluate the integral by interThe conical tank in Figure 9; water exits through the spout as shown. preting part of it as the area of a circle. Water exits here

r

FIGURE 10

Place the origin along the axis of the cylinder. At location y, the layer of water is a rectangular slab of length , width 2 r 2 −y 2 and thickness y. Thus, the volume of the layer is 2 r 2 − y 2 y, and the force needed to lift the layer is 19600 r 2 − y 2 y. The layer must be lifted a distance r − y, so the total work needed to empty the tank is given by " r " r " r

19600 r 2 − y 2 (r − y) d y = 19600r r 2 − y 2 d y − 19600 y r 2 − y 2 d y. SOLUTION

−r

−r

−r

Now, " r −r

y r 2 − y 2 du = 0

because the integrand is an odd function and the integration interval is symmetric with respect to zero. Moreover, the other integral is one-half the area of a circle of radius r ; thus, " r 1 r 2 − y2 d y = πr 2. 2 −r Finally, the total work needed to empty the tank is 1 2 π r − 19600(0) = 9800π r 3 J. 19600r 2 21. Find the work W required to empty the tank in Figure 7 if it is half full of water. The trough in Figure 11; water exits by pouring over the sides. SOLUTION Place the origin on the top of the box, and let the positive y-axis point downward. Note that with this coordinate system, the bottom half of the box corresponds to y values from 2.5 to 5. The volume of one layer of water is 32y m3 , so the force needed to lift each layer is (9.8)(1000)32y = 313600y N. Each layer must be lifted y meters, so the total work needed to empty the tank is " 5 2.5

5 313600y d y = 156800y 2 = 2.94 × 106 J. 2.5

23. Find the work required to empty the tank in Figure 9 if it is half full of water. Assume the tank in Figure 7 is full of water and let W be the work required to pump out half of the water. Do SOLUTION origin the vertex of theininverted and let the y-axis point you expectPlace W tothe equal theatwork computed Exercisecone, 21? Explain andpositive then compute W . upward. Consider a layer of water at a height of y meters. From similar triangles, the area of the layer is

π

y 2 2

m2 ,

so the volume is

π

y 2 2

y m3 .

CHAPTER 6


Thus the weight of one layer is 9800π

y 2 2

y N.

The layer must be lifted 12 − y meters, so the total work needed to empty the half-full tank is " 5 0

9800π

y 2 2

(12 − y) d y =

1684375π J ≈ 2.65 × 106 J. 2

25. Assume that the tank in Figure 9 is full. Assume the tank in Figure 9 is full of water and find the work required to pump out half of the water. (a) Calculate the work F(y) required to pump out water until the water level has reached level y. Plot F(y). (b) (c)

What is the significance of F (y) as a rate of change?

(d)

If your goal is to pump out all of the water, at which water level y0 will half of the work be done?

SOLUTION

(a) Place the origin at the vertex of the inverted cone, and let the positive y-axis point upward. Consider a layer of water at a height of y meters. From similar triangles, the area of the layer is

π

y 2 2

m2 ,

so the volume is

π

y 2 2

y m3 .

Thus the weight of one layer is 9800π

y 2 2

y N.

The layer must be lifted 12 − y meters, so the total work needed to pump out water until the water level has reached level y is " 10 y

9800π

y 2 2

(12 − y) d y = 3675000π − 9800π y 3 +

1225π 4 y J. 2

(b) A plot of F(y) is shown below. y 1 × 10 −7 Work, J

328

8 × 10 −6 6 × 10 −6 4 × 10 −6 2 × 10 −6 x 0

2

4 6 8 Depth, m

10

(c) First, note that F (y) < 0; as y increases, less water is being pumped from the tank, so F(y) decreases. Therefore, when the water level in the tank has reached level y, we can interpret −F (y) as the amount of work per meter needed to remove the next layer of water from the tank. In other words, −F (y) is a “marginal work” function. (d) The amount of work needed to empty the tank is 3675000π J. Half of this work will be done when the water level reaches height y0 satisfying 3675000π − 9800π y03 +

1225π 4 y0 = 1837500π . 2

Using a computer algebra system, we find y0 = 6.91 m. 27. How much work is done lifting a 3-m chain over the side of a building if the chain has mass density 4 kg/m? How much work is done lifting a 25-ft chain over the side of a building (Figure 12)? Assume that the chain has SOLUTION segment of the of length ysegments, located a estimate distance the y j meters from the top the building. a density Consider of 4 lb/ft.a Hint: Break up chain the chain into N work performed on of a segment, and The compute work needed to liftasthis segment chain to the top of the building is approximately the limit N→ ∞ as of anthe integral. W j ≈ (4y)(9.8)y j J.

S E C T I O N 6.5

Work and Energy

329

Summing over all segments of the chain and passing to the limit as y → 0, it follows that the total work is " 3 0

3 4 · 9.8y d y = 19.6y 2 = 176.4 J. 0

29. A 20-foot chain with mass density 3 lb/ft is initially coiled on the ground. How much work is performed in lifting An 8-ft chain weighs 16 lb. Find the work required to lift the chain over the side of a building. the chain so that it is fully extended (and one end touches the ground)? SOLUTION Consider a segment of the chain of length y that must be lifted y j feet off the ground. The work needed to lift this segment of the chain is approximately

W j ≈ (3y)y j ft-lb. Summing over all segments of the chain and passing to the limit as y → 0, it follows that the total work is " 20 0

3y d y =

3 2 20 y = 600 ft-lb. 2 0

31. A 1,000-lb wrecking ball hangs from a 30-ft cable of density 10 lb/ft attached to a crane. Calculate the work done if How much work is done lifting a 20-ft chain with mass density 3 lb/ft (initially coiled on the ground) so that its the crane lifts the ball from ground level to 30 ft in the air by drawing in the cable. top end is 30 ft above the ground? SOLUTION We will treat the cable and the wrecking ball separately. Consider a segment of the cable of length y that must be lifted y j feet. The work needed to lift the cable segment is approximately W j ≈ (10y)y j ft-lb. Summing over all of the segments of the cable and passing to the limit as y → 0, it follows that lifting the cable requires " 30 0

30 10y d y = 5y 2 = 4500 ft-lb. 0

Lifting the 1000 lb wrecking ball 30 feet requires an additional 30,000 ft-lb. Thus, the total work is 34,500 ft-lb. In Exercises 32–34, use Newton’s Universal Law of Gravity, according to which the gravitational force between two objects of mass m and M separated by a distance r has magnitude G Mm/r 2 , where G = 6.67 × 10−11 m3 kg−1 s−1 . Although the Universal Law refers to point masses, Newton proved that it also holds for uniform spherical objects, where r is the distance between their centers. 33. Use the result of Exercise 32 to calculate the work required to place a 2,000-kg satellite in an orbit 1,200 km above Two spheres of mass M and m are separated by a distance r1 . Show that 24 the work required to increase6 the the surface of the earth. Assume that the earth is a sphere−1 of mass −1 Me = 5.98 × 10 kg and radius re = 6.37 × 10 m. to aasdistance r2 is equal to W = G Mm(r1 − r2 ). Treatseparation the satellite a point mass. The satellite will move from a distance r1 = re to a distance r2 = re + 1200000. Thus, from Exercise 32, 1 1 ≈ 1.99 × 1010 J. − W = (6.67 × 10−11 )(5.98 × 1024 )(2000) 6.37 × 106 6.37 × 106 + 1200000

SOLUTION

35. Assume that the pressure P and volume V of the gas in a 30-in. cylinder of radius 3 in. with a movable piston are Use the result of Exercise 32 to compute the work required to move a 1,500-kg satellite from an orbit 1,000 to related by P V 1.4 = k, where k is a constant (Figure 13). When the cylinder is full, the gas pressure is 200 lb/in.2 . 1,500 km above the surface of the earth. (a) Calculate k. (b) Calculate the force on the piston as a function of the length x of the column of gas (the force is P A, where A is the piston’s area). (c) Calculate the work required to compress the gas column from 30 to 20 in.

3 x

FIGURE 13 Gas in a cylinder with a piston. SOLUTION

(a) We have P = 200 and V = 270π . Thus k = 200(270π )1.4 = 2.517382 × 106 lb-in2.2 .

330

CHAPTER 6


(b) The area of the piston is A = 9π and the volume of the cylinder as a function of x is V = 9π x, which gives P = k/V 1.4 = k/(9π x)1.4 . Thus F = PA =

k 9π = k(9π )−0.4 x −1.4 . (9π x)1.4

(c) Since the force is pushing against the piston, in order to calculate work, we must calculate the integral of the opposite force, i.e., we have " 20 1 −0.4 20 x −1.4 d x = −k(9π )−0.4 x W = −k(9π )−0.4 = 74677.8 in-lb. −0.4 30 30

Further Insights and Challenges 37. Work-Kinetic Energy Theorem The kinetic energy of an object of mass m moving with velocity v is KE = 1 mv 2 . A 20-foot chain with linear mass density 2 (x) = 0.02x(20 − x) lb/ft the time interval [t1 , t2 ] due to a net force F(x) acting along the (a) Suppose that the object moves from x1 to x 2ρduring interval [x 1 , x2 ]. Let x(t) be the position of the object at time t. Use the Change of Variables formula to show that the on the ground. worklies performed is equal to (a) How much work is done lifting the chain extended (and one end touches the ground)? " x2 so that it is "fully t2 (b) How much work is done liftingWthe so that has a height d xits=top end F(x(t))v(t) dt of 30 ft? = chain F(x) x1

t1

(b) By Newton’s Second Law, F(x(t)) = ma(t), where a(t) is the acceleration at time t. Show that d 1 mv(t)2 = F(x(t))v(t) dt 2 (c) Use the FTC to show that the change in kinetic energy during the time interval [t1 , t2 ] is equal to " t2 F(x(t))v(t) dt t1

(d) Prove the Work-Kinetic Energy Theorem: The change in KE is equal to the work W performed. SOLUTION

(a) Let x 1 = x(t1 ) and x2 = x(t2 ), then x = x(t) gives d x = v(t) dt. By substitution we have " x2 " t2 W = F(x) d x = F(x(t))v(t) dt. x1

t1

(b) Knowing F(x(t)) = m · a(t), we have d 1 2 m · v(t) = m · v(t) v (t) dt 2

(Chain Rule)

= m · v(t) a(t) = v(t) · F(x(t))

(Newton’s 2nd law)

(c) From the FTC, 1 m · v(t)2 = 2

" F(x(t)) v(t) dt.

Since K E = 12 m v 2 , K E = K E(t2 ) − K E(t1 ) =

(d)

W =

" x2 x1

F(x) d x =

" t2 1 1 F(x(t)) v(t) dt. m v(t2 )2 − m v(t1 )2 = 2 2 t1 " t2 t1

F(x(t)) v(t) dt

(Part (a))

= K E(t2 ) − K E(t1 )

(Part (c))

= K E

(as required)

39. With what initial velocity v0 must we fire a rocket so it attains a maximum height r above the earth? Hint: Use the A model train of mass 0.5 kg is placed at one end of a straight 3-m electric track. Assume that a force F(x) = results of Exercises 32 and 37. As the rocket reaches its maximum height, its KE decreases from 12 mv02 to zero. (Exercise 37) to 3x − x 2 N acts on the train at distance x along the track. Use the Work-Kinetic Energy Theorem determine the velocity of the train when it reaches the end of the track.


331

The work required to move the rocket a distance r from the surface of the earth is 1 1 W (r ) = G Me m − . re r + re

As the rocket climbs to a height r , its kinetic energy is reduced by the amount W (r ). The rocket reaches its maximum height when its kinetic energy is reduced to zero, that is, when 1 2 1 1 − mv = G Me m . 2 0 re r + re Therefore, its initial velocity must be v0 =

2G Me

1 1 − . re r + re

41. Calculate escape velocity, the minimum initial velocity of an object to ensure that it will continue traveling into With what initial velocity must we fire a rocket so it attains a maximum height of r = 20 km above the surface space and never fall back to earth (assuming that no force is applied after takeoff). Hint: Take the limit as r → ∞ in of the earth? Exercise 39. SOLUTION The result of the previous exercise leads to an interesting conclusion. The initial velocity v0 required to reach a height r does not increase beyond all bounds as r tends to infinity; rather, it approaches a finite limit, called the escape velocity: 1 2G Me 1 − vesc = lim 2G Me = r →∞ re r + re re

In other words, vesc is large enough to insure that the rocket reaches a height r for every value of r ! Therefore, a rocket fired with initial velocity vesc never returns to earth. It continues traveling indefinitely into outer space. Now, let’s see how large escape velocity actually is: vesc =

2 · 6.67 × 10−11 · 5.989 × 1024 6.37 × 106

1/2 ≈ 11,190 m/sec.

Since one meter per second is equal to 2.236 miles per hour, escape velocity is approximately 11,190(2.236) = 25,020 miles per hour.

CHAPTER REVIEW EXERCISES In Exercises 1–4, find the area of the region bounded by the graphs of the functions. 1. y = sin x,

y = cos x,

0≤x ≤

5π 4

The region bounded by the graphs of y = sin x and y = cos x over the interval [0, 54π ] is shown below. For π x ∈ [0, 4 ], the graph of y = cos x lies above the graph of y = sin x, whereas, for x ∈ [ π4 , 54π ], the graph of y = sin x SOLUTION

lies above the graph of y = cos x. The area of the region is therefore given by " 5π /4 " π /4 (cos x − sin x) d x + (sin x − cos x) d x π /4

0

5π /4 π /4 + (− cos x − sin x) = (sin x + cos x) π /4

0

√ √ √ √ √ 2 2 2 2 2 2 = + − (0 + 1) + + − − − = 3 2 − 1. 2 2 2 2 2 2 √

√

y 1 0.5

y = sin x x 1

−0.5 −1

2

y = cos x

3

4

332

CHAPTER 6


= x 2 − 1, h(x) = x 2 + x − 2 3. f (x) = x 2 +32x, g(x) f (x) = x − 2x 2 + x, g(x) = x 2 − x SOLUTION The region bounded by the graphs of y = x 2 + 2x, y = x 2 − 1 and y = x 2 + x − 2 is shown below. For each x ∈ [−2, − 12 ], the graph of y = x 2 + 2x lies above the graph of y = x 2 + x − 2, whereas, for each x ∈ [− 12 , 1], the graph of y = x 2 − 1 lies above the graph of y = x 2 + x − 2. The area of the region is therefore given by " −1/2

" 1

(x 2 + 2x) − (x 2 + x − 2) d x + (x 2 − 1) − (x 2 + x − 2) d x −2

−1/2

−1/2 1 1 2 1 = + − x 2 + x x + 2x 2 2 −2 −1/2 1 1 1 1 9 = − 1 − (2 − 4) + − + 1 − − − = . 8 2 8 2 4 y = x 2 + 2x −2

y

y = x2 − 1

−1

1

x

−2 y = x2 + x − 2

In Exercises 5–8, sketch the region bounded π by the graphs of the functions and find its area. f (x) = sin x, g(x) = sin 2x, ≤ x ≤ π 3 5. f (x) = x 3 − x 2 − x + 1, g(x) = 1 − x 2 , 0 ≤ x ≤ 1 Hint: Use geometry to evaluate the integral. SOLUTION The region bounded by the graphs of y = x 3 − x 2 − x + 1 and y = 1 − x 2 is shown below. As the 3 2 2 graph of y = 1 − x lies above the graph of y = x − x − x + 1, the area of the region is given by " 1 1 − x 2 − (x 3 − x 2 − x + 1) d x. 0

Now, the region below the graph of y = circle; thus,

1 − x 2 but above the x-axis over the interval [0, 1] is one-quarter of the unit " 1 0

1 − x2 dx =

1 π. 4

Moreover, " 1 0

(x 3 − x 2 − x + 1) d x =

1 5 1 4 1 3 1 2 1 1 1 x − x − x + x = − − + 1 = . 4 3 2 4 3 2 12 0

Finally, the area of the region shown below is 1 5 π− . 4 12 y 1 0.8

y=

1 − x2

0.6 0.4 0.2

y = x3 − x2 − x + 1 x

0

0.2 0.4 0.6 0.8

1

7. y = 4 −1x 2 , y = 3x, y=4 x = 1 − y 2 , by0the ≤ ygraphs ≤ 1 of y = 4 − x 2 , y = 3x and y = 4 is shown below. For x ∈ [0, 1], the y, x = y SOLUTION 2The region bounded graph of y = 4 lies above the graph of y = 4 − x 2 , whereas, for x ∈ [1, 43 ], the graph of y = 4 lies above the graph of y = 3x. The area of the region is therefore given by 4/3 " 1 " 4/3 1 1 3 1 16 8 3 1 (4 − (4 − x 2 )) d x + (4 − 3x) d x = x 3 + 4x − x 2 = + − − 4− = . 3 0 2 3 3 3 2 2 0 1 1


333

y y=4

4

y = 3x

y = 4 − x2

3 2 1

x 0

0.2 0.4 0.6 0.8 1 1.2

Use graphing utility to locate the points of intersection of y = e−x and y = 1 − x 2 and find the area 9. 3 −a 2y 2 = ytwo y, x = y 2 − y betweenx the curves+(approximately). The region bounded by the graphs of y = e−x and y = 1 − x 2 is shown below. One point of intersection clearly occurs at x = 0. Using a computer algebra system, we find that the other point of intersection occurs at x = 0.7145563847. As the graph of y = 1 − x 2 lies above the graph of y = e−x , the area of the region is given by " 0.7145563847

1 − x 2 − e−x d x = 0.08235024596 SOLUTION

0

y 1 y = 1 − x2

0.8 y = e −x

0.6 0.4 0.2

x 0

0.2

0.4

0.6

0.8

11. Find the total weight of a 3-ft metal rod of linear density Figure 1 shows a solid whose horizontal cross section at height y is a circle of radius (1 + y)−2 for 0 ≤ y ≤ H . 2 Find the volume of the solid. ρ(x) = 1 + 2x + x 3 lb/ft. 9 SOLUTION

The total weight of the rod is " 3 0

ρ(x) d x =

x + x2 +

1 4 3 9 33 x =3+9+ = lb. 18 2 2 0

In Exercises the(in average valueunits) of the through functionaover Find 13–16, the flowfind rate the correct pipethe ofinterval. diameter 6 cm if the velocity of fluid particles at a distance r from the center of the pipe is v(r ) = (3 − r ) cm/s. 3 13. f (x) = x − 2x + 2, [−1, 2] SOLUTION

The average value is

2 " 2

1 1 4 1 1 9 1 x 3 − 2x + 2 d x = x − x 2 + 2x = (4 − 4 + 4) − −1−2 = . 2 − (−1) −1 3 4 3 4 4 −1 15. f (x) = |x|, [−4, 4] f (x) = 9 − x 2 , [0, 3] Hint: Use geometry to evaluate the integral. SOLUTION The average value is " " 4 " 4 0 1 1 1 1 2 4 1 1 2 0 |x| d x = (−x) d x + x dx = − x + x = [(0 + 8) + (8 − 0)] = 2. 4 − (−4) −4 8 8 2 2 0 8 −4 0 −4 (x) = x[x], [0,g(t) 3] on [2, 5] is 9. Find 17. Thefaverage value of SOLUTION

" 5 g(t) dt. 2

The average value of the function g(t) on [2, 5] is given by " 5 " 1 5 1 g(t) dt = g(t) dt. 5−2 2 3 2

Therefore, " 5 2

g(t) dt = 3(average value) = 3(9) = 27.

334

CHAPTER 6


19. UseFor theall Shell to find the volume of the obtained the region between y = x 2 and y = mx x ≥Method 0, the average value of R(x) oversolid [0, x] is equalby torevolving x. Find R(x). about the x-axis (Figure 2). y y = x2 y = mx

x

FIGURE 2 SOLUTION Setting x 2 = mx yields x(x − m) = 0, so the two curves intersect at (0, 0) and (m, m 2 ). To use the shell method, we must slice the solid parallel to the axis of rotation; as we are revolving about the x-axis, this implies a √ horizontal slice and integration in y. For each y ∈ [0, m 2 ], the shell has radius y and height y − my . The volume of the solid is therefore given by

2π

" m2 y

√

0

y y− d y = 2π m

2 5/2 y3 − y 5 3m

m 2 2m 5 m5 2π 5 − = m . = 2π 5 3 15 0

21. Let R be the intersection of the circles of radius 1 centered at (1, 0) and (0, 1). Express as an integral (but do not Use(a)the to find the volume of the solid byx-axis. revolving the region between y = x 2 and evaluate): thewasher area ofmethod R and (b) the volume of revolution of Robtained about the y = mx about the y-axis (Figure 2). SOLUTION The region R is shown below. y 1

(x − 1)2 + y2 = 1

0.8 0.6 0.4 0.2

x2 + (y − 1)2 = 1 x

0

0.2 0.4 0.6 0.8

1

(a) A vertical slice of R has its top along the upper left arc of the circle (x − 1)2 + y 2 = 1 and its bottom along the lower right arc of the circle x 2 + (y − 1)2 = 1. The area of R is therefore given by " 1 1 − (x − 1)2 − (1 − 1 − x 2 ) d x. 0

(b) and use the washer method, each cross section is a washer with outer radius If we revolve R about the x-axis 1 − (x − 1)2 and inner radius 1 − 1 − x 2 . The volume of the solid is therefore given by

π

" 1 0

(1 − (x − 1)2 ) − (1 −

1 − x 2 )2 d x.

In Exercises 23–31, the volume of the solid obtained theexpressing region enclosed by theofcurves about the given Use the Shellfind Method to set up an integral (but doby notrotating evaluate) the volume the solid obtained by axis.rotating the region under y = cos x over [0, π /2] about the line x = π . 23. y = 2x,

y = 0,

x = 8;

x-axis

SOLUTION The region bounded by the graphs of y = 2x, y = 0 and x = 8 is shown below. Let’s choose to slice the region vertically. Because a vertical slice is perpendicular to the axis of rotation, we will use the washer method to calculate the volume of the solid of revolution. For each x ∈ [0, 8], the cross section is a circular disk with radius R = 2x. The volume of the solid is therefore given by

π

" 8 0

(2x)2 d x =

4π 3 8 2048π x = . 3 3 0


335

y 16 12 y = 2x

8 4

x 0

2

4

6

8

1, 8;axis x= −2−3 25. y =y x=2 − 2x,1, yy==0,2x − x= axis x= SOLUTION The region bounded by the graphs of y = x 2 − 1 and y = 2x − 1 is shown below. Let’s choose to slice the region vertically. Because a vertical slice is parallel to the axis of rotation, we will use the shell method to calculate the volume of the solid of revolution. For each x ∈ [0, 2], the shell has radius x − (−2) = x + 2 and height (2x − 1) − (x 2 − 1) = 2x − x 2 . The volume of the solid is therefore given by 2π

2 1 (x + 2)(2x − x 2 ) d x = 2π 2x 2 − x 4 = 2π (8 − 4) = 8π . 4 0 0

" 2

y 3 2 y = 2x − 1

1

x 1 −1

2

y = x2 − 1

27. y 2 = x 3 , 2 y = x, x = 8; axis x = −1 y = x − 1, y = 2x − 1, axis y = 4 SOLUTION The region bounded by the graphs of y 2 = x 3 , y = x and x = 8 is composed of two components, shown below. Let’s choose to slice the region vertically. Because a vertical slice is parallel to the axis of rotation, we will use the shell method to calculate the volume of the solid of revolution. For each x ∈ [0, 1], the shell as radius x − (−1) = x + 1 and height x − x 3/2 ; for each x ∈ [1, 8], the shell also has radius x + 1, but the height is x 3/2 − x. The volume of the solid is therefore given by 2π

" 1 0

(x + 1)(x − x 3/2 ) d x + 2π

" 8 1

(x + 1)(x 3/2 − x) d x =

y

√ 2π (12032 2 − 7083). 35

y

1

20

0.8 15 0.6 0.4

y=x

y2 = x3

10

y2 = x3

5

0.2

y=x x

x 0

0.2 0.4 0.6 0.8

1

0

2

4

6

8

29. y = 2−x 2 +−14x − 3, y = 0; axis y = −1 y = x , x = 1, x = 3; axis y = −3 SOLUTION The region bounded by the graph of y = −x 2 + 4x − 3 and the x-axis is shown below. Let’s choose to slice the region vertically. Because a vertical slice is perpendicular to the axis of rotation, we will use the washer method to calculate the volume of the solid of revolution. For each x ∈ [1, 3], the cross section is a washer with outer radius R = −x 2 + 4x − 3 − (−1) = −x 2 + 4x − 2 and inner radius r = 0 − (−1) = 1. The volume of the solid is therefore given by

π

3 " 3

20 3 1 5 (−x 2 + 4x − 2)2 − 1 d x = π x − 2x 4 + x − 8x 2 + 3x 5 3 1 1 1 20 56π 243 =π − 162 + 180 − 72 + 9 − −2+ −8+3 = . 5 5 3 15

336

CHAPTER 6

A P P L I C AT I O N S O F T H E I N T E G R A L y y = −x 2 + 4x − 3

1 0.8 0.6 0.4 0.2

x 0

0.5 1 1.5 2 2.5 3

31. y 2 = x −1 , x 3= 1, x = 3; axis x = −3 x = 4y − y , y = 0, y = 2; y-axis SOLUTION The region bounded by the graphs of y 2 = x −1 , x = 1 and x = 3 is shown below. Let’s choose to slice the region vertically. Because a vertical slice is parallel to the axis of rotation, we will use the shell method to calculate the volume of the solid of revolution. For each x ∈ [1, 3], the shell has radius x − (−3) = x + 3 and height 1 1 2 = √ . √ − −√ x x x The volume of the solid is therefore given by 2π

" 3

2 (x + 3) √ d x = 2π x 1

3 √ 40 4 3/2 + 12x 1/2 = 2π 16 3 − x . 3 3 1

y 1

x 1

2

3

1 x

y2 =

−1

In Exercises 32–34, the regions refer to the graph of the hyperbola y 2 − x 2 = 1 in Figure 3. Calculate the volume of revolution about both the x- and y-axes. y 3 y=x

2 1 −c

−1 −2

x

c

y2 − x 2 = 1

−3

FIGURE 3

33. The region between the upper branch of the hyperbola and the line y = x for 0 ≤ x ≤ c. The shaded region between the upper branch of the hyperbola and the x-axis for −c ≤ x ≤ c. SOLUTION

• x-axis: Let’s choose to slice the region vertically. Because a vertical slice is perpendicular to the axis of rotation,

we will use the washer method to calculate the volume of the solid of revolution. For each x ∈ [0, c], cross sections are washers with outer radius R = 1 + x 2 and inner radius r = x. The volume of the solid is therefore given by c " c

2 2 (1 + x ) − x d x = π x = cπ . π 0

0

• y-axis: Let’s choose to slice the region vertically. Because a vertical slice is parallel to the axis of rotation, we will

use the shell method to calculate the volume of the solid of revolution. For each x ∈ [0, c], the shell has radius x and height 1 + x 2 − x. The volume of the solid is therefore given by " c c 2π

2π

(1 + x 2 )3/2 − x 3 = (1 + c2 )3/2 − c3 − 1 . 2π x 1 + x2 − x dx = 3 3 0 0 2 35. LetThe a >region 0. Show that when the branch region between y = a and x −y ax between the upper of the hyperbola = 2.and the x-axis is rotated about the x-axis, the resulting volume is independent of the constant a.


337

Setting a x − ax 2 = 0 yields x = 0 and x = 1/a. Using the washer method, cross sections are circular disks with radius R = a x − ax 2 . The volume of the solid is therefore given by SOLUTION

π

" 1/a 0

a 2 (x − ax 2 ) d x =

π

π 1 2 2 1 3 3 1/a 1 1 =π a x − a x − = , 2 3 2 3 6 0

which is independent of the constant a. In Exercises 37–38, water is pumpedlength into aisspherical tank aofforce radius a source locatedto2 20 ft below a hole the A spring whose equilibrium 15 cm exerts of 550ftNfrom when it is stretched cm. Find theat work 3. bottom (Figure 4). The density of water is 64.2 lb/ft required to stretch the spring from 22 to 24 cm.

5

2 Water source

FIGURE 4

37. Calculate the work required to fill the tank. SOLUTION Place the origin at the base of the sphere with the positive y-axis pointing upward. The equation for the great is then x 2 + (y − 5)2 = 25. At location y, the horizontal cross section is a circle of radius circle of the sphere 25 − (y − 5)2 = 10y − y 2 ; the volume of the layer is then π (10y − y 2 )y ft3 , and the force needed to lift the layer is 64.2π (10y − y 2 )y lb. The layer of water must be lifted y + 2 feet, so the work required to fill the tank is given by " 10 " 10 (y + 2)(10y − y 2 ) d y = 64.2π (8y 2 + 20y − y 3 ) d y 64.2π

0

0

= 64.2π

10 1 8 3 y + 10y 2 − y 4 3 4 0

= 74900π ≈ 235,305 ft-lb. 3 of water. The container is raised vertically at a constant speed of 39. A container lb required is filled with Calculateweighing the work 50 F(h) to fill20 thefttank to height h ft from the bottom of the sphere. 2 ft/s for 1 min, during which time it leaks water at a rate of 13 ft3 /s. Calculate the total work performed in raising the container. The density of water is 64.2 lb/ft3 . SOLUTION Let t denote the elapsed time of the ascent of the container, and let y denote the height of the container. Given that the speed of ascent is 2 ft/s, y = 2t; moreover, the volume of water in the container is

1 1 20 − t = 20 − y ft3 . 3 6 The force needed to lift the container and its contents is then 1 50 + 64.2 20 − y = 1334 − 10.7y lb, 6 and the work required to lift the container and its contents is " 120 0

120 (1334 − 10.7y) d y = (1334y − 5.35y 2 ) = 83040 ft-lb. 0

Let W be the work (against the sun’s gravitational force) required to transport an 80-kg person from Earth to Mars when the two planets are aligned with the sun at their minimal distance of 55.7 × 106 km. Use Newton’s Universal Law of Gravity (see Exercises 32–34 in Section 6.5) to express W as an integral and evaluate it. The sun has mass Ms = 1.99 × 1030 kg, and the distance from the sun to the earth is 149.6 × 106 km.

THE EXPONENTIAL 7 FUNCTION 7.1 Derivative of bx and the Number e Preliminary Questions 1. Which of the following equations is incorrect? (a) 32 · 35 = 37

√ (b) ( 5)4/3 = 52/3

(c) 32 · 23 = 1

(d) (2−2 )−2 = 16

SOLUTION

(a) (b) (c) (d)

This equation is correct: 32 · 35 = 32+5 = 37 . √ This equation is correct: ( 5)4/3 = (51/2 )4/3 = 5(1/2)·(4/3) = 52/3 . This equation is incorrect: 32 · 23 = 9 · 8 = 72 = 1. this equation is correct: (2−2 )−2 = 2(−2)·(−2) = 24 = 16.

2. Which of the following functions can be differentiated using the Power Rule? (b) 2e (c) x e (a) x 2

(d) e x

SOLUTION The Power Rule applies when the function has a variable base and a constant exponent. Therefore, the Power Rule applies to (a) x 2 and (c) x e .

3. For which values of b does b x have a negative derivative? SOLUTION

The function b x has a negative derivative when 0 0 except b = 1.

5. Which point lies on the graph of y = b x for all b? SOLUTION

The point (0, 1) lies on the graph of y = b x for all b.

6. Which of the following statements is not true? (a) (e x ) = e x eh − 1 (b) lim =1 h h→0 (c) The tangent line to y = e x at x = 0 has slope e. (d) The tangent line to y = e x at x = 0 has slope 1. SOLUTION

(a) This statement is true: (e x ) = e x . (b) This statement is true: d x eh − 1 = e = e0 = 1. lim h d x x=0 h→0 (c) This statement is false: the tangent line to y = e x at x = 0 has slope e0 = 1. (d) This statement is true: the tangent line to y = e x at x = 0 has slope e0 = 1.

Exercises 1. Rewrite as a whole number (without using a calculator): (a) 70 (c)

(43 )5

(45 )3 (e) 8−1/3 · 85/3

SOLUTION

(a) 70 = 1.

(b) 102 (2−2 + 5−2 ) (d) 274/3 (f) 3 · 41/4 − 12 · 2−3/2

S E C T I O N 7.1

(b) (c) (d) (e) (f)

Derivative of bx and the Number e

339

102 (2−2 + 5−2 ) = 100(1/4 + 1/25) = 25 + 4 = 29. (43 )5 /(45 )3 = 415 /415 = 1. (27)4/3 = (271/3 )4 = 34 = 81. 8−1/3 · 85/3 = (81/3 )5 /81/3 = 25 /2 = 24 = 16. 3 · 41/4 − 12 · 2−3/2 = 3 · 21/2 − 3 · 22 · 2−3/2 = 0.

In Exercises 2–10, solve for the unknown variable. 3. e2x = e x+1 92x = 98 SOLUTION If e2x = e x+1 then 2x = x + 1, and x = 1. x+1 5. 3x =t 2 13 4t−3 e =e 1 SOLUTION Rewrite ( 3 ) x+1 as (3−1 ) x+1 = 3−x−1 . Then 3 x = 3−x−1 , which requires x = −x − 1. Thus, x = −1/2. 7. 4−x √ = 2x+1 ( 5)x = 125 SOLUTION Rewrite 4−x as (22 )−x = 2−2x . Then 2−2x = 2 x+1 , which requires −2x = x + 1. Solving for x gives x = −1/3. 9. k 3/2 4= 27 12 b = 10 SOLUTION Raise both sides of the equation to the two-thirds power. This gives k = (27)2/3 = (271/3 )2 = 32 = 9. In Exercises 11–14, determine the limit. (b2 )x+1 = b−6 11. lim 4x x→∞

SOLUTION

lim 4x = ∞.

x→∞

1 −x 13. limlim 4−x x→∞ 4 x→∞ SOLUTION

lim

x→∞

1 −x 4

= lim 4x = ∞. x→∞

In Exercises 15–18, 2 find the equation of the tangent line at the point indicated. lim e x−x x→∞ 15. y = 4e x , x 0 = 0 Let f (x) = 4e x . Then f (x) = 4e x and f (0) = 4. At x0 = 0, f (0) = 4, so the equation of the tangent line is y = 4(x − 0) + 4 = 4x + 4. SOLUTION

17. y = e x+24x , x 0 = −1 y = e , x0 = 0 SOLUTION Let f (x) = e x+2 . Then f (x) = e x+2 and f (−1) = e1 . At x 0 = −1, f (−1) = e, so the equation of the tangent line is y = e(x + 1) + e = ex + 2e. In Exercises 19–41, find the derivative. 2 y = e x , x0 = 1 19. f (x) = 7e2x + 3e4x SOLUTION

d (7e2x + 3e4x ) = 14e2x + 12e4x . dx

21. f (x) = eπ x f (x) = e−5x d πx SOLUTION e = π eπ x . dx 23. f (x) = 4e−x−4x+9 + 7e−2x f (x) = e d SOLUTION (4e−x + 7e−2x ) = −4e−x − 14e−2x . dx 25. f (x) = x 2 e2x 2 ex f (x) =d 2 2x 2x 2 2x SOLUTION x e = 2xe + 2x e . dx −2x )4 27. f (x) = (2e3x + 2e f (x) = (1 + e x )4 d SOLUTION (2e3x + 2e−2x )4 = 4(2e3x + 2e−2x )3 (6e3x − 4e−2x ). dx

29. f (x) = e1/x 2 f (x) = e x +2x−3

340

CHAPTER 7

THE EXPONENTIAL FUNCTION

SOLUTION

−e1/x d 1/x = . e dx x2

31. f (x) = esin x3 f (x) = e d sin x SOLUTION = cos xesin x . e dx 33. f (x) = sin(e x )2 2 f (x) =de(x +2x+3) x SOLUTION sin(e ) = e x cos(e x ). dx 1√ 35. f (t)f (t) = = e −3t t 1−e d 1 d SOLUTION = (1 − e−3t )−1 = −(1 − e−3t )−2 (3e−3t ). dt 1 − e−3t dt −2t ) 37. f (t) = cos(te f (t) = et+1/2t−1 d SOLUTION cos(te−2t ) = − sin(te−2t )(t (−2e−2t ) + e−2t ) = (2te−2t − e−2t ) sin(te−2t ). dt

) 39. f (x) = tan(e5−6x ex f (x) =d 3xtan(e + 15−6x ) = −6e5−6x sec2 (e5−6x ). SOLUTION dx x

41. f (x) = ee x+1 e +x f (x) =d exx x 2e − SOLUTION e = 1e x ee . dx In Exercises 42–47, find the critical points and determine whether they are local minima, maxima, or neither. 43. f (x) = x + ex −x f (x) = e − x SOLUTION Setting f (x) = 1 − e−x equal to zero and solving for x gives e−x = 1 which is true if and only if x = 0. f (x) = e−x , so f (0) = e0 = 1 > 0. Therefore, x = 0 corresponds to a local minimum. 45. f (x) = x 2 e x x e for x > 0 f (x) = SOLUTION Setting f (x) = (x 2 + 2x)e x equal to zero and solving for x gives (x 2 + 2x)e x = 0 which is true if and x only if x = 0 or x = −2. Now, f (x) = (x 2 + 4x + 2)e x . Because f (0) = 2 > 0, x = 0 corresponds to a local minimum. On the other hand, f (−2) = (4 − 8 + 2)e−2 = −2/e2 < 0, so x = −2 corresponds to a local maximum. t 47. g(t) = (t 3 − 2t)e et g(t) = 2 SOLUTION t +1

g (t) = (t 3 − 2t)(et ) + et (3t 2 − 2) = et (t 3 + 3t 2 − 2t − 2). √ The critical points occur when t 3 + 3t 2 − 2t − 2 = 0. This occurs when t = 1, −2 ± 2. g (t) = et (3t 2 + 6t − 2) + et (t 3 + 3t 2 − 2t − 2) = et (t 3 + 6t 2 + 4t − 4). Thus, • g (1) = 7e > 0, so t = 1 corresponds to a local minimum; √ √ • g (−2 + 2) ≈ −2.497 < 0, so t = −2 + 2 corresponds to a local maximum; and √ √ • g (−2 − 2) ≈ .4108 > 0, so t = −2 − 2 corresponds to a local minimum.

In Exercises 48–53, find the critical points and points of inflection. Then sketch the graph. 49. y = e−x +−x ex y = xe SOLUTION Let f (x) = e−x + e x . Then f (x) = −e−x + e x =

e2x − 1 . ex

Thus, x = 0 is a critical point, and f (x) is increasing for x > 0 and decreasing for x < 0. Observe that f (x) = e−x + e x > 0 for all x, so f (x) is concave up for all x and there are no points of inflection. A graph of y = f (x) is shown below.

S E C T I O N 7.1


341

y

6 4

−2

x

−1

1

2

−x 51. y =y e= e−x cos x on [−π /2, π /2] 2

Let f (x) = e−x . Then f (x) = −2xe−x and x = 0 is the only critical point. Further, f (x) is increasing for x < 0 and decreasing for x > 0. Now, 2

SOLUTION

2

f (x) = 4x 2 e−x − 2e−x = 2e−x (2x 2 − 1), 2

2

√

2

√

√

so f (x) is concave up for |x| > 22 , is concave down for |x| < 22 and has points of inflection at x = ± 22 . A graph of y = f (x) is shown below. y 1

0.4 0.2 −2

x

−1

1

2

on [0, 10] 53. y = x 2 e−x y = ex − x SOLUTION Let f (x) = x 2 e−x . Then f (x) = −x 2 e−x + 2xe−x = x(2 − x)e−x , so x = 0 and x = 2 are critical points. Also, f (x) is increasing for 0 < x < 2 and decreasing for 2 < x < 10. Since f (x) = −(2x − x 2 )e−x + (2 − 2x)e−x = (x 2 − 4x + 2)e−x , √ √ √ √ we find f (x) is concave up for 0 < x√< 2 − 2 and for 2 + 2 < x < 10, is concave down for 2 − 2 < x < 2 + 2, and has inflection points at x = 2 ± 2. A graph of y = f (x) is shown below. y 0.6 0.4 0.2 x 0

2

4

6

8

10

In Exercises 55–68, evaluate integral. Use Newton’s Methodthe to indefinite find the two solutions of e x = 5x to three decimal places (Figure 7). " 55. (e x + 2) d x " SOLUTION

" 57.

" 2 dy ye y 4x e dx

SOLUTION

" 59.

(e x + 2) d x = e x + 2x + C.

Use the substitution u = y 2 , du = 2y d y. Then " " 2 1 1 1 2 eu du = eu + C = e y + C. ye y d y = 2 2 2

" dt e−9t 4x (e + 1) d x

342

CHAPTER 7

THE EXPONENTIAL FUNCTION SOLUTION

" 61.

" dt et ext + 1−x (e + e ) d x

SOLUTION

" 63.

Use the substitution u = −9t, du = −9 dt. Then " " 1 1 1 eu du = − eu + C = − e−9t + C. e−9t dt = − 9 9 9

Use the substitution u = et + 1, du = et dt. Then " " √ 2 2 et et + 1 dt = u du = u 3/2 + C = (et + 1)3/2 + C. 3 3

" 10x ) d x (7 − e−x (e − 4x) d x

SOLUTION

First, observe that " " " " (7 − e10x ) d x = 7 d x − e10x d x = 7x − e10x d x.

In the remaining integral, use the substitution u = 10x, du = 10 d x. Then " " 1 1 u 1 10x eu du = e10x d x = + C. e +C = e 10 10 10 Finally, " (7 − e10x ) d x = 7x − " 65.

2 " −4x d xe4x xe e2x −

SOLUTION

" 67.

1 10x + C. e 10

dx ex Use the substitution u = −4x 2 , du = −8x d x. Then " " 2 2 1 1 1 eu du = − eu + C = − e−4x + C. xe−4x d x = − 8 8 8

" ex d xx √ x eex +cos(e 1 ) dx

SOLUTION

Use the substitution u = e x + 1, du = e x d x. Then " √

ex ex + 1

" dx =

√ √ du √ = 2 u + C = 2 e x + 1 + C. u

69. Find" an approximation to m 4 using the limit definition and estimate the slope of the tangent line to y = 4x at x = 0 and x = 2.e x (e2x + 1)3 d x SOLUTION

Recall

4h − 1 . h h→0

m 4 = lim Using a table of values, we find h

4h − 1 h

.01 .001 .0001 .00001

1.39595 1.38726 1.38639 1.38630

Thus m 4 ≈ 1.386. Knowing that y (x) = m 4 · 4x , it follows that y (0) ≈ 1.386 and y (2) ≈ 1.386 · 16 = 22.176. 71. Find the area between y = e x and y = e−x over [0, 2]. Find the area between y = e x and y = e2x over [0, 1].



343

Over [0, 2], the graph of y = e x lies above the graph of y = e−x . Hence, the area between the graphs is 2 1 1 (e x − e−x ) d x = (e x + e−x ) = e2 + 2 − (1 + 1) = e2 − 2 + 2 . e e 0 0

" 2

73. Find the volume obtained by revolving y = e x about the x-axis for 0 ≤ x ≤ 1. Find the area bounded by y = e2 , y = e x , and x = 0. SOLUTION Each cross section of the solid is a disk with radius R = e x . The volume is then

π

" 1 0

e2x d x =

π 2x 1 π e = (e2 − 1). 2 2 0

x f (x). Show that if g(x) is another function satisfying g (x) = g(x), then g(x) = Ce forx some constant C. Hint: Compute the derivative of g(x)e−x . 75. Prove that f (x) = e is not a polynomial function. Hint: Differentiation lowers the degree of a polynomial by 1.

TheInsights function fand (x) =Challenges e satisfies f (x) = Further x

Assume f (x) = e x is a polynomial function of degree n. Then f (n+1) (x) = 0. But we know that any derivative of e x is e x and e x = 0. Hence, e x cannot be a polynomial function. SOLUTION

77. Generalize Exercise 76; that is, use induction (if you are familiar with this method of proof) to prove that for all Recall the following property of integrals: If f (t) ≥ g(t) for all t ≥ 0, then for all x ≥ 0, n ≥ 0, " x " x 1 2 f (t)1 dt3 ≥ 1 x e ≥ 1 + x + x + x + · · · +g(t) xdtn (x ≥ 0) 0 20 6 n! t

x ≥ 1 + x for x ≥ 0. Then prove, by e76. > Assume 1. Use (5) prove thatis etrue thetostatement for n = k. We need to prove the successive the1.following inequalities (for x ≥ 0): statement is trueintegration, for n = k + By the Induction Hypothesis, x ≥ 1for The inequality because SOLUTION For n e=≥1,1eholds + tx ≥by0 Exercise

x 2 k /k!. 1 1 2 1 3 x ≥ · · ·1+ x + x +xx + e x ≥ 1e +≥x 1++ xx 2+, x /2e+ 2 2 6 Integrating both sides of this inequality yields " x et dt = e x − 1 ≥ x + x 2 /2 + · · · + x k+1 /(k + 1)! 0

or e x ≥ 1 + x + x 2 /2 + · · · + x k+1 /(k + 1)! as required. 79. Calculate the first three derivatives xof f (x) = xe x . Then guess the formula for f (n) (x) (use induction to prove it if x ex e you are Use familiar with this of proof). ≥ = ∞. Then use Exercise 77 to prove more and conclude that lim Exercise 76 tomethod show that x→∞ x 2 6 x2 x x x x x x x x SOLUTION f (x) = e e + xe , f (x) = e + e + xe = 2e + xe , f (x) = 2e x + e x + xe x = 3e x + xe x . So all n. generally that lim(n) n = ∞ for x→∞ one would guess that f x(x) = ne x + xe x . Assuming this is true for f (n) (x), we verify that f (n+1) (x) = ( f (n) (x)) = x x x x x ne + e + xe = (n + 1)e + xe . 81. Prove in two ways that the numbers m a satisfy Consider the equation e x = λ x, where λ is a constant. m a + mdraw intuition, a graph of y = e x and the line y = λ x. (a) For which λ does it have a unique solution?mFor ab = b (b) For which λ does it have at least one solution? (a) First method: Use the limit definition of m b and ah − 1 (ab)h − 1 bh − 1 h =b + h h h (b) Second method: Apply the Product Rule to a x b x = (ab)x . SOLUTION

(ab)h − 1 bh (a h − 1) bh − 1 ah − 1 bh − 1 = lim + = lim bh lim + lim h h h h h h→0 h→0 h→0 h→0 h→0

(a) m ab = lim

ah − 1 bh − 1 + lim = ma + mb. h h h→0 h→0

= lim

So, m ab = m a + m b . (b) m ab (ab)x = ((ab)x ) = (a x b x ) = (a x ) b x + (b x ) a x = m a a x b x + m b a x b x = a x b x (m a + m b ) = (ab)x (m a + m b ). Therefore, we have m ab (ab)x = (ab)x (m a + m b ). Dividing both sides by (ab)x , we see that m ab = m a + m b .

344

CHAPTER 7


7.2 Inverse Functions Preliminary Questions 1. (a) (c) (e)

Which of the following satisfy f −1 (x) = f (x)? f (x) = x f (x) = 1 f (x) = |x|

SOLUTION

(b) f (x) = 1 − x √ (d) f (x) = x (f) f (x) = x −1

The functions (a) f (x) = x, (b) f (x) = 1 − x and (f) f (x) = x −1 satisfy f −1 (x) = f (x).

2. The graph of a function looks like the track of a roller coaster. Is the function one-to-one? Because the graph looks like the track of a roller coaster, there will be several locations at which the graph has the same height. The graph will therefore fail the horizontal line test, meaning that the function is not one-to-one. SOLUTION

3. Consider the function f that maps teenagers in the United States to their last names. Explain why the inverse of f does not exist. SOLUTION Many different teenagers will have the same last name, so this function will not be one-to-one. Consequently, the function does not have an inverse.

4. View the following fragment of a train schedule for the New Jersey Transit System as defining a function f from towns to times. Is f one-to-one? What is f −1 (6:27)?

SOLUTION

Trenton

6:21

Hamilton Township

6:27

Princeton Junction

6:34

New Brunswick

6:38

This function is one-to-one, and f −1 (6:27) = Hamilton Township.

5. A homework problem asks for a sketch of the graph of the inverse of f (x) = x + cos x. Frank, after trying but failing to find a formula for f −1 (x), says it’s impossible to graph the inverse. Bianca hands in an accurate sketch without solving for f −1 . How did Bianca complete the problem? SOLUTION

The graph of the inverse function is the reflection of the graph of y = f (x) through the line y = x.

Exercises 1. Show that f (x) = 7x − 4 is invertible by finding its inverse. y+4 x +4 SOLUTION Solving y = 7x − 4 for x yields x = . Thus, f −1 (x) = . 7 7 3. What is the largest interval containing zero on which f (x) = sin x is one-to-one? Is f (x) = x 2 + 2 one-to-one? If not, describe a domain on which it is one-to-one. SOLUTION Looking at the graph of sin x, the function is one-to-one on the interval [−π /2, π /2]. 5. Verify that f (x) = x 3 x+− 3 2and g(x) = (x − 3)1/3 are inverses by showing that f (g(x)) = x and g( f (x)) = x. Show that f (x) = is invertible by finding its inverse. SOLUTION x +3

(a) What is the domain of3 f (x)? The range of f −1 (x)? • f (g(x)) = (x − 3)1/3 +−13 = x − 3 + 3 = x. (b) What is the

domain of f (x)?

The range of f (x)? • g( f (x)) =

x3 + 3 − 3

1/3

= x3

1/3

= x.

t + 1R is v(R) = t + 1 M and radius 7. TheRepeat escapeExercise velocity 5from and g(t) = . for af planet (t) = of mass t − 1 constant. Find the inverse of v(R) as a function of R. t − 1 SOLUTION

To find the inverse, we solve y=

2G M R

R=

2G M . y2

for R. This yields

Therefore, v −1 (R) =

2G M . R2

2G M , where G is the universal gravitational R

S E C T I O N 7.2

Inverse Functions

345

In Exercises find avertically domain in onthe which f islands one-to-one and alater. formula forofthe of f functions restrictedare to one-to-one? this domain. A ball9–16, is tossed air and T seconds Which theinverse following Sketch the graphs of f and f −1 . (a) The ball’s height s(t) for 0 ≤ t ≤ T ball’s 9. (b) f (x)The = 3x − 2velocity v(t) for 0 ≤ t ≤ T (c) The ball’s maximum height as a function of initial velocity SOLUTION The linear function f (x) = 3x − 2 is one-to-one for all real numbers. Solving y = 3x − 2 for x gives x = (y + 2)/3. Thus,

f −1 (x) =

x +2 . 3 y

f −1(x) =

x+2 3

−2

1 x

−1

1 −1 −2

f(x) = 3x − 2

1 11. f (x)f (x) = =4−x x +1 The graph of f (x) = 1/(x + 1) given below shows that f passes the horizontal line test, and is therefore 1 1 1 one-to-one, on its entire domain {x : x = −1}. Solving y = for x gives x = − 1. Thus, f −1 (x) = − 1. x +1 y x SOLUTION

y

y

4

4 y = f −1(x) 2

y = f(x) x

−4

2

−4

4

x

−2

−4

1 13. f (s) = 2 f (x)s=

1 7x − 3 SOLUTION To make f (s) = s −2 one-to-one, we must restrict the domain to either {s : s > 0} or {s : s < 0}. If we 1 1 1 choose the domain {s : s > 0}, then solving y = 2 for s yields s = √ . Hence, f −1 (s) = √ . Had we chosen the y s s 1 −1 domain {s : s < 0}, the inverse would have been f (s) = − √ . s y

y 8 6

4

4 y = f(s)

2 −2

−1

y = f −1(s)

2 s

1

2

s 1

2

3

4

15. f (z) = z 3 1 f (x) = 3 3 SOLUTION The x function 2 + 1 f (z) = z is one-to-one over its entire domain (see the graph below). Solving y = z for z 1/3 −1 1/3 yields y = z. Thus, f (z) = z .

346

CHAPTER 7

THE EXPONENTIAL FUNCTION y y = f(z) 3 2 y = f −1(z)

1

z

−3 −2 −1 −1

1

2

3

−2 −3

17. For each function shown in Figure 15, sketch the graph of the inverse (restrict the function’s domain if necessary). f (x) = x 3 + 9 y

y

y

x

x

x (A)

(B)

(C)

y

y

y

x x

x (E)

(D)

(F)

FIGURE 15

Here, we apply the rule that the graph of f −1 is obtained by reflecting the graph of f across the line y = x. For (C) and (D), we must restrict the domain of f to make f one-to-one. y (b) y (c) y (a) SOLUTION

x

x (A)

(d)

x

(B)

(e)

y

(C)

(f)

y

y

x x x (D)

(E)

(F)

19. Let n be a non-zero integer. Find a domain on which f (x) = (1 − x n )1/n coincides with its inverse. Hint: The Which ofon thewhether graphs ninisFigure 16odd. is the graph of a function satisfying f −1 = f ? answer depends even or SOLUTION

First note

1/n n 1/n = 1 − (1 − x n ) = (x n )1/n = x, f ( f (x)) = 1 − (1 − x n )1/n

so f (x) coincides with its inverse. For the domain and range of f , let’s first consider the case when n > 0. If n is even, then f (x) is defined only when 1 − x n ≥ 0. Hence, the domain is −1 ≤ x ≤ 1. The range is 0 ≤ y ≤ 1. If n is odd, then f (x) is defined for all real numbers, and the range is also all real numbers. Now, suppose n < 0. Then −n > 0, and 1/−n 1 −1/−n x −n f (x) = 1 − −n = . x x −n − 1 If n is even, then f (x) is defined only when x −n − 1 > 0. Hence, the domain is |x| > 1. The range is y > 1. If n is odd, then f (x) is defined for all real numbers except x = 1. The range is all real numbers except y = 1. 21. Show that the inverse of f (x) = e−x exists (without finding it explicitly). What is the domain of f −1 ? Let f (x) = x 7 + x + 1. SOLUTION (a) Show that f −1 exists (but do not attempt to find it). Hint: Show that f is strictly increasing. (b) What is the domain of f −1 ? (c) Find f −1 (3).

S E C T I O N 7.2

Inverse Functions

347

y

6 4 2

−2

y = e−x x

−1

1

2

Notice that the graph of f (x) = e−x (shown above) passes the horizontal line test. That is, f (x) = e−x is one-to-one. By Theorem 3, f −1 (x) exists and the domain of f −1 (x) is the range of f (x), namely (0, ∞) 23. Let f (x) = x 2 − 2x. Determine a domain on which f −1 exists and find a formula for f −1 on this domain. Show that f (x) = (x 2 + 1)−1 is one-to-one on (−∞, 0] and find a formula for f −1 for this domain. SOLUTION From the graph of y = x 2 − 2x shown below, we see that if the domain of f is restricted to either x ≤ 1 or x ≥ 1, then f is one-to-one and f −1 exists. To find a formula for f −1 , we solve y = x 2 − 2x for x as follows: y + 1 = x 2 − 2x + 1 = (x − 1)2 x −1=± y+1 x =1± y+1 −1 If the √ domain of f is restricted to x ≤ 1, then we choose the negative sign in front of the radical and f −1(x) = 1 − √ x + 1. If the domain of f is restricted to x ≥ 1, we choose the positive sign in front of the radical and f (x) = 1 + x + 1. y y = x 2 − 2x 6 4 2 x

−1

2

3

two ways: using Theorem 2 and 25. Find the inverse g(x) of f (x) −1 = x 2 + 9 with domain x ≥ 0 and calculate g (x) in −1 Show that f (x) = x + x is one-to-one on [1, ∞) and find a formula for f on this domain. What is the by direct calculation. −1 domain of f ? SOLUTION To find a formula for g(x) = f −1 (x), solve y = x 2 + 9 for x. This yields x = ± y 2 − 9. Because the domain of f was restricted to x ≥ 0, we must choose the positive sign in front of the radical. Thus g(x) = f −1 (x) = x 2 − 9. Because x 2 + 9 ≥ 9 for all x, it follows that f (x) ≥ 3 for all x. Thus, the domain of g(x) = f −1 (x) is x ≥ 3. The range of g is the restricted domain of f : y ≥ 0. By Theorem 2, g (x) =

1 . f (g(x))

With f (x) = it follows that

x x2 + 9

x2 − 9 = f (g(x)) = ( x 2 − 9)2 + 9

,

x2 − 9 = √ x2

since the domain of g is x ≥ 3. Thus, g (x) =

1 f (g(x))

=

x x2 − 9

.

This agrees with the answer we obtain by differentiating directly: x 2x = . g (x) = 2 2 2 x −9 x −9

x2 − 9 x

348

CHAPTER 7


In Exercises 27–32, use Theorem 2 to calculate3 g (x), where g(x) is the inverse of f (x). Let g(x) be the inverse of f (x) = x + 1. Find a formula for g(x) and calculate g (x) in two ways: using Theorem 27. f (x) = 7x2 and + 6 then by direct calculation. Let f (x) = 7x + 6 then f (x) = 7. Solving y = 7x + 6 for x and switching variables, we obtain the inverse g(x) = (x − 6)/7. Thus, SOLUTION

1 1 = . f (g(x)) 7

g (x) =

29. f (x) = x −5√ f (x) = 3 − x SOLUTION Let f (x) = x −5 , then f (x) = −5x −6 . Solving y = x −5 for x and switching variables, we obtain the inverse g(x) = x −1/5 . Thus, g (x) =

1 1 = − x −6/5 . 5 −5(x −1/5 )−6

x 31. f (x) = f (x) x=+4x13 − 1 SOLUTION

x , then Let f (x) = x+1

(x + 1) − x 1 = . (x + 1)2 (x + 1)2

f (x) =

x for x and switching variables, we obtain the inverse g(x) = x . Thus Solving y = x+1 1−x

g (x) = 1

1 1 = . (x/(1 − x) + 1)2 (1 − x)2

33. Let g(x) be the inverse of f (x) = x 3 + 2x + 4. Calculate g(7) [without finding a formula for g(x)] and then −1 f (x) = 2 + x calculate g (7). SOLUTION

Let g(x) be the inverse of f (x) = x 3 + 2x + 4. Because f (1) = 13 + 2(1) + 4 = 7,

it follows that g(7) = 1. Moreover, f (x) = 3x 2 + 2, and g (7) =

1 1 1 = = . f (g(7)) f (1) 5

In Exercises 35–40, calculate g(b) and g (b), where g is the inverse of f (in the given domain, if indicated). x3 Find g (− 12 ), where g(x) is the inverse of f (x) = 2 . x +1 35. f (x) = x + cos x, b = 1 f (0) = 1, so g(1) = 0. f (x) = 1 − sin x so f (g(1)) = f (0) = 1 − sin 0 = 1. Thus, g (1) = 1/1 = 1. 37. f (x) = x 2 +3 6x for x ≥ 0, b = 4 f (x) = 4x − 2x, b = −2 SOLUTION To determine g(4), we solve f (x) = x 2 + 6x = 4 for x. This yields: SOLUTION

x 2 + 6x = 16 x 2 + 6x − 16 = 0 (x + 8)(x − 2) = 0 or x = −8, 2. Because the domain of f has been restricted to x ≥ 0, we have g(4) = 2. With f (x) =

x +3 x 2 + 6x

,

it then follows that g (4) = 1 1 = for x ≤ −6, 39. f (x)f (x) = = x,2 +b 6x x +1 4

b=4

1

1

4

= = . f (g(4)) f (2) 5


f (3) = 1/4, so g(1/4) = 3. f (x) =

g (1/4) = −16.

Logarithms and Their Derivatives

349

−1 so f (g(1/4)) = f (3) = −1 2 = −1/16. Thus, (x+1)2 (3+1)

41. Let f (x) = xxn and g(x) = x 1/n . Compute g (x) using Theorem 2 and check your answer using the Power Rule. f (x) = e , b = e SOLUTION Note that g(x) = f −1 (x) (see problem 55 from 6.1). Therefore, g (x) =

1 f (g(x))

=

x 1/n−1 1 1 1 1 = = = = (x 1/n−1 ) n−1 1/n n−1 1−1/n n n n(g(x)) n(x ) n(x )

which agrees with the Power Rule. graphical reasoning to determine if the following statements are true or false. If false, modify the statement Further Use Insights and Challenges

to make it correct. 43. Show that if f (x) is odd and f −1 (x) exists, then f −1 (x) is odd. Show, on the other hand, that an even function does (a) If (x) is strictly increasing, then f −1 (x) is strictly decreasing. not have anfinverse. (b) If f (x) is strictly decreasing, then−1f −1 (x) is strictly decreasing. SOLUTION Suppose f (x) is odd and f (x) exists. Because f (x) is odd, f (−x) = − f (x). Let y = f −1 (x), then (c) If f (x) is concave up, then f −1 (x) is concave up. f (y) = x. Since f (x) is odd, f (−y) = − f (y) = −x. Thus f −1 (−x) = −y = − f −1 (x). Hence, f −1 is odd. is concave (d)the If other f (x) hand, is concave down, thenthen f −1 (x) On if f (x) is even, f (−x) = f (x).up. Hence, f is not one-to-one and f −1 does not exist. (e) Linear functions f (x) = ax + b (a = 0) are always one-to-one. (x). thatone-to-one. g (x) x −1 . We will this[a, in the next 45. Let gQuadratic bethe theIntermediate inverse of a function f satisfying f (x) + show bx + c (a= are always (f) Use polynomials f (x) = ax 2xto Value Theorem that if = ff0) (x) isShow continuous and= one-to-one on anapply interval b], then section to show that the inverse of f (x) = e (the natural logarithm) is an antiderivative of x −1 . f(g) (x)sin is xanisincreasing or a decreasing function. not one-to-one. SOLUTION

g (x) =

1 1 1 1 = −1 = = . f (g(x)) x f ( f (x)) f ( f −1 (x))

7.3 Logarithms and Their Derivatives Preliminary Questions 1. Compute logb2 (b4 ). SOLUTION

Because b4 = (b2 )2 , logb2 (b4 ) = 2.

2. When is ln x negative? SOLUTION

ln x is negative for 0 < x < 1.

3. What is ln(−3)? Explain. SOLUTION

ln(−3) is not defined.

4. Explain the phrase “the logarithm converts multiplication into addition.” SOLUTION

This phrase is a verbal description of the general property of logarithms that states log(ab) = log a + log b.

5. What are the domain and range of ln x? SOLUTION

The domain of ln x is x > 0 and the range is all real numbers.

6. Does x −1 have an antiderivative for x < 0? If so, describe one. SOLUTION

Yes, ln(−x) is an antiderivative of x −1 for x < 0.

Exercises In Exercises 1–12, calculate directly (without using a calculator). 1. log3 27 SOLUTION

log3 27 = log3 33 = 3 log3 3 = 3.

3. log2 (25/31) log5 25 SOLUTION

log2 25/3 =

5. log64 4 log2 (85/3 )

5 5 log2 2 = . 3 3

350

CHAPTER 7


SOLUTION

log64 4 = log64 641/3 =

1 1 log64 64 = . 3 3

7. log8 2 + log4 2 log7 (492 ) 1/3 + log 41/2 = 1 + 1 = 5 . SOLUTION log8 2 + log4 2 = log8 8 4 3 2 6 9. log4 48 − log4 12 log25 30 + log25 65 48 SOLUTION log4 48 − log4 12 = log4 = log4 4 = 1. 12 11. ln(e3 )√ + ln(e4 ) ln( e · e7/5 ) SOLUTION ln(e3 ) + ln(e4 ) = 3 + 4 = 7. 13. Write as the natural log of a single expression: 4 (a) 2 ln 5log +23 3ln+ 4 log2 24

(b) 5 ln(x 1/2 ) + ln(9x)

SOLUTION

(a) 2 ln 5 + 3 ln 4 = ln 52 + ln 43 = ln 25 + ln 64 = ln(25 · 64) = ln 1600. (b) 5 ln x 1/2 + ln 9x = ln x 5/2 + ln 9x = ln(x 5/2 · 9x) = ln(9x 7/2 ). In Exercises 15–20, solve 2for the unknown. Solve for x: ln(x + 1) − 3 ln x = ln(2). 15. 7e5t = 100 SOLUTION

Divide the equation by 7 and then take the natural logarithm of both sides. This gives 1 100 100 or t = ln . 5t = ln 7 5 7

2 17. 2x −2x =8 6e−4t = 2 SOLUTION Since 8 = 23 , we have x 2 − 2x − 3 = 0 or (x − 3)(x + 1) = 0. Thus, x = −1 or x = 3.

19. ln(x 42t+1 ) − ln(x 2 1−t )=2 e = 9e SOLUTION

ln(x 4 ) − ln(x 2 ) = ln

x4 x2

= ln(x 2 ) = 2 ln x. Thus, 2 ln x = 2 or ln x = 1. Hence, x = e.

21. The population of a city millions) at time t (years) is P(t) = 2.4e0.06t , where t = 0 is the year 2000. When 2 ) =(in log y + 3 log (y 14 3 3 will the population double from its size at t = 0? SOLUTION

Population doubles when 4.8 = 2.4e.06t . Thus, 0.06t = ln 2 or t =

ln 2 ≈ 11.55 years. 0.06

In Exercises 23–40, find theGutenberg–Richter derivative. According to the law, the number N of earthquakes worldwide of Richter magnitude M approximately satisfies a relation ln N = 16.17 − bM for some constant b. Find b, assuming that there are 800 23. y = x ln x earthquakes of magnitude M = 5 per year. How many earthquakes of magnitude M = 7 occur per year? d x SOLUTION x ln x = ln x + = ln x + 1. dx x 2 25. y =y (ln = tx)ln t − t 1 2 d SOLUTION (ln x)2 = (2 ln x) = ln x. dx x x

27. y = ln(x 3 +23x + 1) y = ln(x ) d 3x 2 + 3 . SOLUTION ln(x 3 + 3x + 1) = 3 dx x + 3x + 1 29. y = ln(sin t s+ 1) y = ln(2 ) cos t d ln(sin t + 1) = . SOLUTION dt sin t + 1 ln x 31. y =y = x 2 ln x x SOLUTION

1 (x) − ln x 1 − ln x d ln x = . = x dx x x2 x2

33. y = ln(ln x) 2 y = e(ln x)

S E C T I O N 7.3

SOLUTION


351

d 1 ln(ln x) = . dx x ln x

35. y = ln((ln x)3 ) 3 y = (ln(ln x)) 3(ln x)2 3 d ln((ln x)3 ) = . SOLUTION = dx x ln x x(ln x)3 Alternately, because ln((ln x)3 ) = 3 ln(ln x), d 1 d ln((ln x)3 ) = 3 ln(ln x) = 3 · . dx dx x ln x 37. y = ln(tan x) y = ln((x + 1)(2x + 9)) d 1 1 SOLUTION ln(tan x) = · sec2 x = . dx tan x sin x cos x 39. y = 5x x +1 y = lnd x3 + ln 1 5 · 5x . SOLUTION 5x = dx In Exercises 41–44, compute the derivative using Eq. (1). 2 y = 5x −x 41. f (x), f (x) = log2 x SOLUTION

f (x) = log2 x =

ln x 1 1 . Thus, f (x) = · . ln 2 x ln 2

43. f (3), f (x) = log5 x d 2 log10 (x 3 + ln xx ) 1 1 d SOLUTIONx f (x) = , so f (x) = . Thus, f (3) = . ln 5 x ln 5 3 ln 5 In Exercises d 45–56, find an equation of the tangent line at the point indicated. log (sin t) dt= 4x3, x = 3 45. f (x) SOLUTION Let f (x) = 4 x . Then f (3) = 43 = 64. f (x) = ln 4 · 4 x , so f (3) = ln 4 · 64. Therefore, the equation of the tangent line is y = 64 ln 4(x − 3) + 64.

√ 47. s(t) = 37t , √t =x2 f (x) = ( 2) , x = 2 SOLUTION Let s(t) = 37t . Then s(2) = 314 . s (t) = 7 ln 3 · 37t , so s (2) = 7 ln 3 · 314 . Therefore the equation of the tangent line is y = 7 ln 3 · 314 (t − 2) + 314 . 49. f (x) = 5x −2x+9 , x =1 f (x) = π 3x+9 , x = 1 2 2 SOLUTION Let f (x) = 5 x −2x+9 . Then f (1) = 58 . f (x) = ln 5 · 5 x −2x+9 (2x − 2), so f (1) = ln 5(0) = 0. Therefore, the equation of the tangent line is y = 58 . 2

51. s(t) = ln(8 − 4t), t = 1 s(t) = ln t, t = 5 −4 SOLUTION Let s(t) = ln(8 − 4t). Then s(1) = ln(8 − 4) = ln 4. s (t) = 8−4t , so s (1) = −4/4 = −1. Therefore the equation of the tangent line is y = −1(t − 1) + ln 4. 53. f (x) = 4 ln(9x + 2), x = 2 f (x) = ln(x 2 ), x = 4 36 SOLUTION Let f (x) = 4 ln(9x + 2). Then f (2) = 4 ln 20. f (x) = 9x+2 , so f (2) = 9/5. Therefore the equation of the tangent line is y = (9/5)(x − 2) + 4 ln 20. 55. f (x) = log5 x, x = 2 π f (x) = ln(sin x), x = 2 1 1 4 SOLUTION Let f (x) = log5 x. Then f (2) = log5 2 = ln ln 5 . f (x) = x ln 5 , so f (2) = 2 ln 5 . Therefore the equation of the tangent line is y=

1 ln 2 . (x − 2) + 2 ln 5 ln 5

In Exercises 57–60, find the−1 local extreme values in the domain {x : x > 0} and use the Second Derivative Test to f (x) = log2 (x + x ), x = 1 determine whether these values are local minima or maxima. 57. g(x) =

ln x x

352

CHAPTER 7


Let g(x) = lnxx . Then g (x) =

x(1/x) − ln x 1 − ln x = . 2 x x2

We know g (x) = 0 when 1 − ln x = 0, or when x = e. g (x) = so g (e) =

x 2 (−1/x) − (1 − ln x)(2x) −3 + 2 ln x = x4 x3

−1 < 0. Thus, g(e) is a local maximum. e3

59. g(x) = x − ln x ln x g(x) Let = g(x) 2 SOLUTION x 2 = x − ln x. Then g (x) = 1 − 1/x, and g (x) = 0 when x = 1. g (x) = 1/x , so g (1) = 1 > 0. Thus g(1) is a local minimum. In Exercises g(x)61–66, = x lnfind x the derivative using the methods of Example 9. 61. f (x) = x 2x SOLUTION

Method 1: x 2x = e2x ln x , so d 2x x = e2x ln x (2 + 2 ln x) = x 2x (2 + 2 ln x). dx

Method 2: Let y = x 2x . Then, ln y = 2x ln x. By logarithmic differentiation 1 y = 2x · + 2 ln x, y x so y = y(2 + 2 ln x) = x 2x (2 + 2 ln x) . x

e 63. f (x)f (x) = x= x cos x x x SOLUTION Method 1: x e = ee ln x , so x d ex x = ee ln x dx

x x x e e + e x ln x = x e + e x ln x . x x

x Method 2: Let y = x e . Then ln y = e x ln x. By logarithmic differentiation

y 1 = e x · + e x ln x, y x so y = y

x x x e e + e x ln x = x e + e x ln x . x x

x

65. f (x) = x 2 x 2 f (x) = x x x SOLUTION Method 1: x 2 = e2 ln x , so x x x ln x x 2 2 d 2x 2 x 2 x x =e + (ln x)(ln 2)2 = x + (ln x)(ln 2)2 . dx x x x Method 2: Let y = x 2 . Then ln y = 2x ln x. By logarithmic differentiation

1 y = 2x + (ln x)(ln 2)2x , y x so x y = x 2

x 2 x + (ln x)(ln 2)2 . x

In Exercises 67–74,x xevaluate the derivative using logarithmic differentiation as in Example 8. f (x) = e


S E C T I O N 7.3

353

67. y = (x + 2)(x + 4) SOLUTION Let y = (x + 2)(x + 4). Then ln y = ln((x + 2)(x + 4)) = ln(x + 2) + ln(x + 4). By logarithmic differentiation

y 1 1 = + y x +2 x +4 or y = (x + 2)(x + 4)

1 1 + x +2 x +4

= (x + 4) + (x + 2) = 2x + 6.

x(x + 1)3 + 2)(x + 4) 69. y =y = (x + 1)(x (3x − 1)2 SOLUTION

3 Let y = x(x+1)2 . Then ln y = ln x + 3 ln(x + 1) − 2 ln(3x − 1). By logarithmic differentiation

(3x−1)

y 1 3 6 = + − , y x x + 1 3x − 1 so y =

(x + 1)3 3x(x + 1)2 6x(x + 1)3 + − . (3x − 1)2 (3x − 1)2 (3x − 1)3

√ 2) x − 9 71. y = (2x + 1)(4x 2 x(x + 1) y = Let √ y = (2x + 1)(4x 2 )√x − 9. Then SOLUTION x +1 ln y = ln(2x + 1) + ln 4x 2 + ln(x − 9)1/2 = ln(2x + 1) + ln 4 + 2 ln x +

1 ln(x − 9). 2

By logarithmic differentiation 2 2 1 y = + + , y 2x + 1 x 2(x − 9) so

√ y = (2x + 1)(4x 2 ) x − 9

2 2 1 + + . 2x + 1 x 2(x − 9)

73. y = (x 2 + 1)(x 2 + 2)(x 2 + 3)2 x(x + 2) y = Let y = (x 2 + 1)(x 2 + 2)(x 2 + 3)2 . Then ln y = ln(x 2 + 1) + ln(x 2 + 2) + 2 ln(x 2 + 3). By logarithmic SOLUTION (2x + 1)(2x + 2) differentiation y 2x 2x 4x = 2 + 2 + 2 , y x +1 x +2 x +3 so y = (x 2 + 1)(x 2 + 2)(x 2 + 3)2

2x 2x 4x + + . x2 + 1 x2 + 2 x2 + 3

In Exercises 75–78, x cosfind x the local extrema and points of inflection, and sketch the graph of y = f (x) over the given interval. y = (x + 1) sin x 75. y = x 2 − ln x, SOLUTION

[ 12 , 2]

Let y = x 2 − ln x. Then y = 2x −

1 2x 2 − 1 = x x

√

√

√

and y = 2 + 12 . Thus, the function is decreasing on ( 12 , 22 ), is increasing on ( 22 , 2), has a local minimum at x = 22 , x and is concave up over the entire interval ( 12 , 2). A graph of y = x 2 − ln x is shown below.

354

CHAPTER 7

THE EXPONENTIAL FUNCTION y 3 2 1 x 0

0.5

1

1.5

2

2 + 4), [0, ∞) 77. y = ln(xln t y = 3 , [1, ∞)2 SOLUTION Let t y = ln(x + 4). Then

2x y = 2 x +4

y =

and

(x 2 + 4)(2) − 2x(2x) 8 − 2x 2 = 2 . 2 2 (x + 4) (x + 4)2

Thus, the function is increasing on (0, ∞), is concave up on (0, 2), is concave down on (2, ∞) and has a point of inflection at x = 2. A graph of y = ln(x 2 + 4) is shown below. y 4 3 2 1 x 0

2

4

6

8

In Exercises 79–96, evaluate the indefinite integral, using substitution if necessary. y = 2−1/t , (0, 4) " dx 79. 2x + 4 SOLUTION

81.

Let u = 2x + 4. Then du = 2 d x, and " " 1 1 1 dx = du = ln |2x + 4| + C. 2x + 4 2 u 2

" "2 x dtxdt x3 + t 2 2+ 4

SOLUTION

Let u = x 3 + 2. Then du = 3x 2 d x, and "

x2 dx 1 = 3 3 x +2

"

du 1 = ln |x 3 + 2| + C. u 3

"

" tan(4x (3x+−1)1)d dx x 9 − 2x + 3x 2 ! ! sin(4x+1) SOLUTION First we rewrite tan(4x + 1) d x as cos(4x+1) d x. Let u = cos(4x + 1). Then du = −4 sin(4x + 1) d x, and " " sin(4x + 1) 1 du 1 dx = − = − ln | cos(4x + 1)| + C. cos(4x + 1) 4 u 4 83.

" 85.

" cos x dx 2 sin cotx x+d3x

SOLUTION

Let u = 2 sin x + 3. Then du = 2 cos x d x, and " " cos x 1 du 1 dx = = ln(2 sin x + 3) + C, 2 sin x + 3 2 u 2

where we have used the fact that 2 sin x + 3 ≥ 1 to drop the absolute value. " 4" ln x + 5 87. ln x d x x dx x



355

Let u = 4 ln x + 5. Then du = (4/x)d x, and " " 1 1 1 4 ln x + 5 u du = u 2 + C = (4 ln x + 5)2 + C. dx = x 4 8 8

"

"d x 2 x ln(ln x x) d x x SOLUTION Let u = ln x. Then du = (1/x)d x, and " " dx 1 = du = ln |u| + C = ln | ln x| + C. x ln x u 89.

"

ln(ln x) " dx dx x ln x (4x − 1) ln(8x − 2) 1 1 · d x and SOLUTION Let u = ln(ln x). Then du = ln x x " " ln(ln x) u2 (ln(ln x))2 d x = u du = +C = + C. x ln x 2 2 91.

"

" 3x d x cot x ln(sin x) d x " 3x 3x d x = SOLUTION + C. ln 3 " " x dx 95. cos x 3xsin 2 x3 d x 93.

SOLUTION

Let u = sin x. Then du = cos x d x, and "

" cos x 3sin x d x =

3u du =

3u 3sin x +C = + C. ln 3 ln 3

In Exercises evaluate the definite integral. " 97–102, 1 3x+2 dx " 2 1 2 97. dx 1 x 2 " 2 1 SOLUTION d x = ln |x| = ln 2 − ln 1 = ln 2. x 1 1 " e 1 " 12 99. d x1 1 x dx e 4 "x e 1 SOLUTION d x = ln |x| = ln e − ln 1 = 1. x 1

" −e " 41 dtdt 101. −e2 t 4 2 3t "+ −e SOLUTION

1

−e e 1 = ln | − e| − ln | − e2 | = ln 2 = ln(1/e) = −1. dt = ln |t| 2 e −e2 t −e

103. " e2 Find a good numerical approximation to the coordinates of the point on the graph of y = ln x − x closest 1 10). to the origin (Figure dt e t ln t y 0.5 x 1

2

3

4

5

FIGURE 10 Graph of y = ln x − x.

356

CHAPTER 7


The distance from the origin to the point (x, ln x − x) on the graph of y = ln x − x is d = As usual, we will minimize d 2 . Let d 2 = f (x) = x 2 + (ln x − x)2 . Then 1 f (x) = 2x + 2(ln x − x) −1 . x SOLUTION

x 2 + (ln x − x)2 .

To determine x, we need to solve 4x +

2 ln x − 2 ln x − 2 = 0. x

This yields x ≈ .632784. Thus, the point on the graph of y = ln x − x that is closest to the origin is approximately (0.632784, −1.090410). x 105. the formula (ln of f (x)) forf x(x) > to 0. show that ln x and ln(2x) have the same derivative. Is there a Find Use the minimum value f (x)== fx (x)/ simpler explanation of this result? SOLUTION

Observe (ln x) =

1 x

and (ln 2x) =

2 1 = . 2x x

As an alternative explanation, note that ln(2x) = ln 2 + ln x. Hence, ln x and ln(2x) differ by a constant, which implies the two functions have the same derivative. 107. What is b if the slope of the tangent line to the curve y = b x at x = 0 is 4? Why is the area under the hyperbola y = 1/x between 1 and 3 equal to the area under the same hyperbola = ln b · b x so y (0) = ln b. Setting this equal to 4 and solving for b yields b = e4 . SOLUTION between y5 and 15? For which value of b is it also equal to the area between 10 and b? 109. The energy E (in joules) radiated as seismic waves from an earthquake of Richter magnitude M is given by the Let f E (x)==4.8 ln + x. 1.5M. Use the Linear Approximation to estimate f = ln(e2 + 0.1) − ln(e2 ). formula log 10 (a) Express E as a function of M. (b) Show that when M increases by 1, the energy increases by a factor of approximately 31. (c) Calculate d E/d M. SOLUTION

(a) Solving log10 E = 4.8 + 1.5M for E yields E = 104.8+1.5M . (b) Using the formula from part (a), we find 106.3+1.5M 104.8+1.5(M+1) E(M + 1) = = 101.5 ≈ 31.6228. = E(M) 104.8+1.5M 104.8+1.5M (c) Again using the formula from part (a), dE = 1.5(ln 10)104.8+1.5M . dM

Further Insights and Challenges 111. Show that loga b logb a = 1. (a) Show that if f and g are differentiable, then ln b ln a ln b ln a and logb a = . Thus loga b · logb a = · = 1. SOLUTION loga b = (x)b f (x)ln a g ln ln a ln bd ln( f (x)g(x)) = + f (x) g(x) 113. Use Exercise 112 to verify the formula d x loga x Verify the formula logb x = for a, b > 0. d 1 ( f (x)g(x)) log b logb x =that the left-hand side of Eq. (5) is equal to . (b) Give a new proof of the Producta Rule by observing dx (ln b)x f (x)g(x) SOLUTION

d ln x 1 d logb x = = . dx d x ln b (ln b)x

115. Show that if f (x) is strictly increasing and satisfies f (x y) = f (x) + f (y), then its inverse g(x) satisfies g(x + Defining ln x as an Integral Define a function ϕ (x) in the domain x > 0: y) = g(x)g(y). " x SOLUTION Let x = f (w) and y = f (z). Then 1 ϕ (x) = dt t g(x + y) = g( f (w) + f (z)) = g( f 1(wz)) = wz = g(x) · g(y). This exercise proceeds as if we didn’t know that ϕ (x) = ln x and shows directly that ϕ (x) has all the basic properties Exercises provide a the mathematically elegant approach to the exponential and logarithm functions, which avoids of the114–116 logarithm. Prove following statements: x for irrational " bofisdefining " eab This a continuation of the previous exercises. g(x) be the inverse of ϕ (x). Show that the problem x and two of proving thatLet e x is differentiable. 1 1 (a) g(x)g(y) dt = dt for all a, b > 0. Hint: Use the substitution u = t/a. = g(x + y). t 1 t a r number. (b) g(r ) ==e ϕfor (ab) (a)any + ϕrational (b). Hint: Break up the integral from 1 to ab into two integrals and use (a). (b) ϕ

S E C T I O N 7.4

Exponential Growth and Decay

357

7.4 Exponential Growth and Decay Preliminary Questions 1. Two quantities increase exponentially with growth constants k = 1.2 and k = 3.4, respectively. Which quantity doubles more rapidly? SOLUTION Doubling time is inversely proportional to the growth constant. Consequently, the quantity with k = 3.4 doubles more rapidly.

2. If you are given both the doubling time and the growth constant of a quantity that increases exponentially, can you determine the initial amount? SOLUTION

No. To determine the initial amount, we need to know the amount at one instant in time.

3. A cell population grows exponentially beginning with one cell. Does it take less time for the population to increase from one to two cells than from 10 million to 20 million cells? SOLUTION Because growth from one cell to two cells and growth from 10 million to 20 million cells both involve a doubling of the population, both increases take exactly the same amount of time.

4. Referring to his popular book A Brief History of Time, the renowned physicist Stephen Hawking said, “Someone told me that each equation I included in the book would halve its sales.” If this is so, write a differential equation satisfied by the sales function S(n), where n is the number of equations in the book. SOLUTION

Let S(0) denote the sales with no equations in the book. Translating Hawking’s observation into an equation

yields S(n) =

S(0) . 2n

Differentiating with respect to n then yields dS d = S(0) 2−n = − ln 2S(0)2−n = − ln 2S(n). dn dn 5. Carbon dating is based on the assumption that the ratio R of C14 to C12 in the atmosphere has been constant over the past 50,000 years. If R were actually smaller in the past than it is today, would the age estimates produced by carbon dating be too ancient or too recent? SOLUTION If R were actually smaller in the past than it is today, then we would be overestimating the amount of decay and therefore overestimating the age. Our estimates would be too ancient.

Exercises 1. A certain bacteria population P obeys the exponential growth law P(t) = 2,000e1.3t (t in hours). (a) How many bacteria are present initially? (b) At what time will there be 10,000 bacteria? SOLUTION

(a) P(0) = 2000e0 = 2000 bacteria initially. (b) We solve 2000e1.3t = 10, 000 for t. Thus, e1.3t = 5 or t=

1 ln 5 ≈ 1.24 hours. 1.3

3. A certain RNA molecule has a doubling time of 3 minutes. Find the growth constant k and the differential equation A quantity P of obeys the exponential P(t) = e5tStarting (t in years). for the number N (t) molecules present atgrowth time t law (in minutes). with one molecule, how many will be present At what time t is P = 10? after(a) 10 min? ln 2 (b) At what time t is P = 20? ln 2 SOLUTION The doubling time is so k = . Thus, the differential equation is N (t) = k N (t) = doubling time (c) What is the doubling time forkP? ln 2 N (t). With one molecule initially, 3 N (t) = e(ln 2/3)t = 2t/3 . Thus, after ten minutes, there are N (10) = 210/3 ≈ 10.079, or 10 molecules present. A quantity P obeys the exponential growth law P(t) = Cekt (t in years). Find the formula for P(t), assuming that the doubling time is 7 years and P(0) = 100.

358

CHAPTER 7


5. The decay constant of Cobalt-60 is 0.13 years−1 . What is its half-life? SOLUTION

Half-life =

ln 2 ≈ 5.33 years. 0.13

7. Find all solutions to the differential equation y = −5y. Which solution satisfies the initial condition y(0) = 3.4? Find the decay constant of Radium-226, given that its half-life is 1,622 years. SOLUTION y = −5y, so y(t) = Ce−5t for some constant C. The initial condition y(0) = 3.4 determines C = 3.4. Therefore, y(t) = 3.4e−5t . y(2) = 4. 9. Find the solution to y = 3y satisfying √ Find the solution to y = 2y satisfying y(0) = 20. 4 SOLUTION y = 3y, so y(t) = Ce3t for some constant C. The initial condition y(2) = 4 determines C = . e6 4 3t Therefore, y(t) = 6 e = 4e3(t−2) . e 11. (a) (b) (c)

The population of a city is P(t) = 2 · e0.06t (in millions), where t is measured in years. Find the function y = f (t) that satisfies the differential equation y = −0.7y and initial condition y(0) = 10. Calculate the doubling time of the population. How long does it take for the population to triple in size? How long does it take for the population to quadruple in size?

SOLUTION

(a) Since k = 0.06, the doubling time is ln 2 ≈ 11.55 years. k (b) The tripling time is calculated in the same way as the doubling time. Solve for in the equation P(t + ) = 3P(t) 2 · e0.06(t+) = 3(2e0.06t ) 2 · e0.06t e0.06 = 3(2e0.06t ) e0.06 = 3 0.06 = ln 3, or = ln 3/0.06 ≈ 18.31 years. (c) Since the population doubles every 11.55 years, it quadruples after 2 × 11.55 = 23.10 years. 13. Assuming that population growth is approximately exponential, which of the two sets of data is most likely to The population of Washington state increased from 4.86 million in 1990 to 5.89 million in 2000. Assuming represent the population (in millions) of a city over a 5-year period? exponential growth, (a) What will the population be in 2010? Year 2000 2001 2002 2003 2004 (b) What is the doubling time? Data I 3.14 3.36 3.60 3.85 4.11 Data II 3.14 3.24 3.54 4.04 4.74 SOLUTION If the population growth is approximately exponential, then the ratio between successive years’ data needs to be approximately the same.

Year

2000

2001

2002

2003

2004

Data I Ratios

3.14 3.36 3.60 3.85 4.11 1.07006 1.07143 1.06944 1.06753

Data II Ratios

3.14 3.24 3.54 4.04 4.74 1.03185 1.09259 1.14124 1.17327

As you can see, the ratio of successive years in the data from “Data I” is very close to 1.07. Therefore, we would expect exponential growth of about P(t) ≈ (3.14)(1.07t ). 15. The Beer–Lambert Law is used in spectroscopy to determine the molar absorptivity α or the concentration c of Lightdissolved intensityin The intensity of light passing through an 10). absorbing medium exponentially withasthe a compound a solution at low concentrations (Figure The law states decreases that the intensity I of light it distance traveled. Suppose the decay constant for a certain plastic block is k = 2 when the distance is measured in passes through the solution satisfies ln(I /I0 ) = α cx, where I0 is the initial intensity and x is the distance traveled by the How must the block be toequation reduce the byI afor factor one-third? light.feet. Show thatthick I satisfies a differential d I intensity /d x = −k someofconstant k.

S E C T I O N 7.4


359

Intensity I

Distance

x

Solution

I0 0

x

FIGURE 10 Light of intensity passing through a solution.

SOLUTION

ln

I I0

= α cx so

I = eα cx or I = I0 eα cx . Therefore, I0 dI = I0 eα cx (α c) = I (α c) = −k I, dx

where k = −α c is a constant. 17. A 10-kg quantity of a radioactive isotope decays to 3 kg after 17 years. Find the decay constant of the isotope. An insect population triples in size after 5 months. Assuming exponential growth, when will it quadruple in size? ln(3/10) SOLUTION P(t) = 10e−kt . Thus P(17) = 3 = 10e−17k , so k = ≈ 0.071 years−1 . −17 19. Chauvet Caves In 1994, rock climbers in southern France stumbled on a cave containing prehistoric cave paint14 to C12 ratio showedout that samplearcheologist of sheepskinHelene parchment discovered bythat archaeologists had byaFrench Valladas showed the paintings area C between 29,700 ings. A Measurements C14 -analysis carried equal to 40% of that found in the atmosphere. Approximately how old is the parchment? and 32,400 years old, much older than any previously known human art. Given that the C14 to C12 ratio of the atmosphere is R = 10−12 , what range of C14 to C12 ratios did Valladas find in the charcoal specimens? SOLUTION

The C14 -C12 ratio found in the specimens ranged from 10−12 e−0.000121(32400) ≈ 1.98 × 10−14

to 10−12 e−0.000121(29700) ≈ 2.75 × 10−14 . 21. Atmospheric Pressure The atmospheric pressure P(h) (in pounds per square inch) at a height h (in miles) above A paleontologist has discovered the remains of animals that appear to have died at the onset of the Holocene ice sea level on earth satisfies a differential equation P = −k P for some positive constant k. age. She applies carbon dating to test her theory that the Holocene age started between 10,000 and 12,000 years ago. (a) Measurements with a barometer show that P(0) =to14.7 P(10) = 2.13. What is the decay constant k? 14 to C12 ratio would she expect find and in the animal remains? What range of C (b) Determine the atmospheric pressure 15 miles above sea level. SOLUTION

(a) Because P = −k P for some positive constant k, P(h) = Ce−kh where C = P(0) = 14.7. Therefore, P(h) = 14.7e−kh . We know that P(10) = 14.7e−10k = 2.13. Solving for k yields 2.13 1 k = − ln ≈ 0.193 miles−1 . 10 14.7 (b) P(15) = 14.7e−0.193(15) ≈ 0.813 pounds per square inch. 23. A quantity P increases exponentially with doubling time 6 hours. After how many hours has P increased by 50%? Inversion of Sugar When cane sugar is dissolved in water, it converts to invert sugar over a period of several ln 2 −1 . P willexponentially. −0.2 f. hours. TheThe percentage (t) ofisunconverted at time t decreases Suppose thatwhen f =1.5P SOLUTION doublingf time have increased by 50% = 6 socane k ≈sugar 0.1155 hours 0 = k What percentage of cane sugar remains after 5 hours? After 10 hours? ln 1.5 P0 e0.1155t , or when t = ≈ 3.5 hours. 0.1155 25. Moore’s Law In 1965, Gordon Moore predicted that the number N of transistors on a microchip would increase Two bacteria colonies are cultivated in a laboratory. The first colony has a doubling time of 2 hours and the second exponentially. a doubling time of 3 hours. Initially, the first colony contains 1,000 bacteria and the second colony 3,000 bacteria. (a) Does thetime tablet of data below confirm Moore’s prediction for the period from 1971 to 2000? If so, estimate the growth At what will sizes of the colonies be equal? constant k. Plot the data in the table. (b) (c) Let N (t) be the number of transistors t years after 1971. Find an approximate formula N (t) ≈ Cekt , where t is the number of years after 1971. (d) Estimate the doubling time in Moore’s Law for the period from 1971 to 2000. (e) If Moore’s Law continues to hold until the end of the decade, how many transistors will a chip contain in 2010?

360

CHAPTER 7


(f) Can Moore have expected his prediction to hold indefinitely? Transistors

Year

No. Transistors

4004 8008 8080 8086 286 386 processor 486 DX processor Pentium processor Pentium II processor Pentium III processor Pentium 4 processor

1971 1972 1974 1978 1982 1985 1989 1993 1997 1999 2000

2,250 2,500 5,000 29,000 120,000 275,000 1,180,000 3,100,000 7,500,000 24,000,000 42,000,000

SOLUTION

(a) Yes, the graph looks like an exponential graph especially towards the latter years. We estimate the growth constant by setting 1971 as our starting point, so P0 = 2250. Therefore, P(t) = 2250ekt . In 2000, t = 29. Therefore, P(29) = 2250e29k = 42000000, so k = ln 18666.67 ≈ 0.339. Note: A better estimate can be found by calculating k for each time 29 period and then averaging the k values. (b) y 4×10 7 3×10 7 2×10 7 1×10 7 x 1980 1985 1990 1995 2000

(c) (d) (e) (f)

N (t) = 2250e0.339t The doubling time is ln 2/0.339 ≈ 2.04 years. In 2010, t = 39 years. Therefore, N (39) = 2250e0.339(39) ≈ 1,241,623,327. No, you can’t make a microchip smaller than an atom.

In Exercises 27–28, the Gompertz differential equation: Assume that we in aconsider certain country, the rate at which jobs are created is proportional to the number of people who million jobs 3 months later, how many jobs will

y15.1 already have jobs. If there are 15 million jobs daty t = 0 and = ky ln there be after two years? dt M (where M and k are constants), introduced in 1825 by the English mathematician Benjamin Gompertz and still used today to model aging and mortality. kt

27. Show that y = Meae is a solution for any constant a. SOLUTION

kt

Let y = Meae . Then kt dy = M(kaekt )eae dt

and, since ln(y/M) = aekt , we have kt

ky ln(y/M) = Mkaekt eae =

dy . dt

A certain quantity quadratically: P(t) = P0 ta2scientist . 29. To model mortality in aincreases population of 200 laboratory rats, assumes that the number P(t) of rats alive (Figure at time t at (intime months) the Gompertz M in = size? 204 and k= 0.15 months 1, how long will it takeequation for P to with double How long will it take−1 starting at 11). t0 =Find 2 or P(t) 3? (a) Starting t0 = satisfies [note that P(0) = 200] and determine the population after 20 months. (b) In general, starting at time t0 , how long will it take for P to double in size? SOLUTION

S E C T I O N 7.4


(a) Starting from t0 = 1, P doubles when P(t) = 2P(1) = 2P0 . Thus, P0 t 2 = 2P0 and t = P doubles when

√

361

2. Starting from t0 = 2,

P(t) = P0 t 2 = 2P(2) = 8P0 . √ Thus, t = 2 2. Finally, starting from t0 = 3, P doubles when √ Thus, t = 3 2. (b) Starting from t = t0 , P doubles when

P(t) = P0 t 2 = 2P(3) = 18P0 .

P(t) = P0 t 2 = 2P(t0 ) = 2P0 t02 .

√ Thus, t = t0 2.

Verify that the half-life of a quantity that decays exponentially with decay constant k is equal to ln 2/k. Further Insights and Challenges 31. Isotopes for Dating Which of the following isotopes would be most suitable for dating extremely old rocks: Carbon-14 (half-life 5,570 years), Lead-210 (half-life 22.26 years), and Potassium-49 (half-life 1.3 billion years)? Explain why. SOLUTION For extremely old rocks, you need to have an isotope that decays very slowly. In other words, you want a very large half-life such as Potassium-49; otherwise, the amount of undecayed isotope in the rock sample would be too small to accurately measure.

33. of Decay Physicists use the radioactive decay R =constant R0 e−ktk.toShow compute average Let PAverage = P(t) Time be a quantity that obeys an exponential growth law withlaw growth that the P increases −kt m-fold after interval of (ln m)/kLet years. or mean time M an until an atom decays. F(t) = R/R0 = e be the fraction of atoms that have survived to time t without decaying. (a) Find the inverse function t (F). (b) The error in the following approximation tends to zero as N → ∞: N 1 # j M = mean time to decay ≈ t N j=1 N

Argue that M =

" 1

t (F) d F. " (c) Verify the formula ln x d x = x ln x − x by differentiation and use it to show that for c > 0, 0

" 1 c

t (F) d F =

1 1 + (c ln c − c) k k

(d) Verify numerically that lim (c − c ln c) = 0 (we will prove this in Section 7.7). c→0

(e) The integral defining M is “improper” because t (0) is infinite. Show that M = 1/k by computing the limit " 1 M = lim t (F) d F c→0 c

(f) What is the mean time to decay for Radon (with a half-life of 3.825 days)? SOLUTION

(a) F = e−kt so ln F = −kt and t (F) =

ln F −k

1 $N j=1 t ( j/N ). For the interval [0, 1], from the approximation given, the subinterval length is 1/N and thus N the right-hand endpoints have x-coordinate ( j/N ). Thus we have a Riemann sum and by definition, (b) M ≈

" 1 N 1 # t ( j/N ) = t (F)d F. N →∞ N 0 j=1 lim

(c)

d (x ln x − x) = x dx

1 + ln x − 1 = ln x. Thus x 1 1 " 1 1 1 t (F) d F = − (F ln F − F) = (F − F ln F) k k c c c =

1 1 1 (1 − 1 ln 1 − (c − c ln c)) = + (c ln c − c). k k k

362

CHAPTER 7


(d) Let g(c) = c ln c − c. Then, c

0.01

0.001

0.0001

0.00001

g(c)

−0.056052

−0.007908

−0.001021

−0.000125

Thus, as c → 0+, it appears that g(c) → 0. " 1 1 1 1 (e) M = lim t (F)d F = lim + (c ln c − c) = . k k c→0 c c→0 k ln 2 1 (f) Since the half-life is 3.825 days, k = and = 5.52. Thus, M = 5.52 days. 3.825 k

7.5 Compound Interest and Present Value Preliminary Questions 1. Which is preferable: an interest rate of 12% compounded quarterly, or an interest rate of 11% compounded continuously? SOLUTION To answer this question, we need to determine the yearly multiplier associated with each interest rate. The multiplier associated with an interest rate of 12% compounded quarterly is

1+

0.12 4 ≈ 1.1255, 4

while the multiplier associated with an interest rate of 11% compounded continuously is e0.11 ≈ 1.11627. Thus, the compounded quarterly rate is preferable. 2. Find the yearly multiplier if r = 9% and interest is compounded (a) continuously and (b) quarterly. SOLUTION

With r = 9%, the yearly multiplier for continuously compounded interest is e0.09 ≈ 1.09417,

and the yearly multiplier for compounded quarterly interest is

0.09 4 ≈ 1.09308. 1+ 4

3. The PV of N dollars received at time T is (choose the correct answer): (a) The value at time T of N dollars invested today (b) The amount you would have to invest today in order to receive N dollars at time T SOLUTION The correct response is (b): the PV of N dollars received at time T is the amount you would have to invest today in order to receive N dollars at time T .

4. A year from now, $1 will be received. Will its PV increase or decrease if the interest rate goes up? SOLUTION

If the interest rate goes up, the present value of $1 a year from now will decrease.

5. Xavier expects to receive a check for $1,000 1 year from today. Explain, using the concept of PV, whether he will be happy or sad to learn that the interest rate has just increased from 6% to 7%. SOLUTION If the interest rate goes up, the present value of $1,000 one year from today decreases. Therefore, Xavier will be sad is the interest rate has just increased from 6 to 7%.

Exercises 1. Compute the balance after 10 years if $2,000 is deposited in an account paying 9% interest and interest is compounded (a) quarterly, (b) monthly, and (c) continuously. SOLUTION

(a) P(10) = 2000(1 + .09/4)4(10) = $4870.38 (b) P(10) = 2000(1 + .09/12)12(10) = $4902.71 (c) P(10) = 2000e.09(10) = $4919.21 Suppose $500 is deposited into an account paying interest at a rate of 7%, continuously compounded. Find a formula for the value of the account at time t. What is the value of the account after 3 years?

S E C T I O N 7.5

Compound Interest and Present Value

363

3. A bank pays interest at a rate of 5%. What is the yearly multiplier if interest is compounded (a) yearly? (b) three times a year? (c) continuously? SOLUTION

(a) P(t) = P0 (1 + 0.05)t , so the yearly multiplier is 1.05. 0.05 3 0.05 3t (b) P(t) = P0 1 + , so the yearly multiplier is 1 + ≈ 1.0508. 3 3 (c) P(t) = P0 e0.05t , so the yearly multiplier is e0.05 ≈ 1.0513. 5. Show that if interest is compounded continuously at a rate r , then an account doubles after (ln 2)/r years. How long will it take for $4,000 to double in value if it is deposited in an account bearing 7% interest, continuln 2 ously compounded? SOLUTION The account doubles when P(t) = 2P0 = P0 er t , so 2 = er t and t = . r 7. An investment increases in value at a continuously compounded rate of 9%. How large must the initial investment Howtomuch invested today over in order to receive period? $20,000 after 5 years if interest is compounded continuously be in order buildmust up a be value of $50,000 a seven-year at the rate r = 9%? SOLUTION Solving 50,000 = P0 e0.09(7) for P0 yields P0 =

50000 ≈ $26,629.59. e0.63

9. Is it better to receive $1,000 today or $1,300 in 4 years? Consider r = 0.08 and r = 0.03. Compute the PV of $5,000 received in 3 years if the interest rate is (a) 6% and (b) 11%. What is the PV in these SOLUTION compounding, if r = 0.08, then the present value of $1300 four years from now is two casesAssuming if the sum continuous is instead received in 5 years? 1300e−0.08(4) = $943.99. It is better to get $1,000 now. On the other hand, if r = 0.03, the present value of $1300 four years from now is 1300e−0.03(4) = $1153.00, so it is better to get the $1,300 in four years. 11. If a company invests $2 million to upgrade its factory, it will earn additional profits of $500,000/year for 5 years. Is Find theworthwhile, interest rateassuming r if the PV $8,000rate to be in 1 year is $7,300. the investment anofinterest of received 6% (assume that the savings are received as a lump sum at the end of each year)? SOLUTION

The present value of the stream of additional profits is 500,000(e−0.06 + e−0.12 + e−0.18 + e−0.24 + e−0.3 ) = $2,095,700.63.

This is more than the $2 million cost of the upgrade, so the upgrade should be made. 13. After winning $25 million in the state lottery, Jessica learns that she will receive five yearly payments of $5 million A new computer system costing $25,000 will reduce labor costs by $7,000/year for 5 years. beginning immediately. (a) Is it a good investment r =if8%? (a) What is the PV of Jessica’s if prize r = 6%? (b) How much money will the company save?amount were paid today? (b) How much more would the prize be worthactually if the entire SOLUTION

(a) The present value of the prize is 5,000,000(e−0.24 + e−0.18 + e−0.12 + e−0.06 + e−.06(0) ) = $22,252,915.21. (b) If the entire amount were paid today, the present value would be $25 million, or $2,747,084.79 more than the stream of payments made over five years. 15. Use Eq. (2) to compute the PV of an income stream paying out R(t) = $5,000/year continuously for 10 years and An investment group purchased an office building in 1998 for $17 million and sold it 5 years later for $26 million. r = 0.05. 10 on this investment. Calculate the annual " 10(continuously compounded) rate of return −0.05t −0.05t = $39,346.93. SOLUTION P V = 5,000e dt = −100,000e 0

0

17. Find the PV of an investment that produces income continuously at a rate of $800/year for 5 years, assuming an the0.08. PV of an income stream if income is paid out continuously at a rate R(t) = $5,000e0.1t /year for 5 interest Compute rate of r = 5 " 5 years and r = 0.05. SOLUTION P V = 800e−0.08t dt = −10,000e−0.08t = $3296.80. 0

0

19. Show that the PV of an investment that pays out R dollars/year continuously for T years is R(1 − e−r T )/r , where The rate of yearly income generated by a commercial property is $50,000/year at t = 0 and increases at a r is the interest rate. continuously compounded rate of 5%. Find the PV of the income generated in the first four years if r = 8%. SOLUTION The present value of an investment that pays out R dollars/year continuously for T years is PV =

" T

Re−r t dt.

0

Let u = −r t, du = −r dt. Then " R −r T R R 1 −r T Reu du = − eu = − (e−r T − 1) = (1 − e−r T ). PV = − r 0 r r r 0

364

CHAPTER 7


21. Suppose that r = 0.06. Use the result of Exercise 20 to estimate the payout rate R needed to produce an income Explain statement: If T that is very large, then the PV the income described in Exercise 19 is stream whose PV isthis $20,000, assuming the stream continues for of a large number stream of years. approximately R/r . R R SOLUTION From Exercise 20, P V = so 20000 = or R = $1200. r .06 23. Use Eq. (5) to compute the PV of an investment that pays out income continuously at a rate R(t) = (5,000 + Verify by differentiation 1,000t)e0.02t dollars/year for 10 years and r = 0.08. " e−r t (1 + r t) SOLUTION te−r t dt = − +C " 10 " 10r 2 " 10 0.02t )e−0.08t dt = −0.06t dt + −0.06t dt P V = (5000 + 1000t)(e 5000e 1000te Use Eq. (5) to compute the PV of an investment that pays out income continuously at a rate R(t) = (5,000 + 1,000t) 0 0 0 dollars/year for 5 years and r = 0.05. 1 e−0.06(10) (1 + 0.06(10)) 5000 −0.06(10) + 1000 − 1) − 1000 (e = −0.06 (0.06)2 (0.06)2 = 37599.03 − 243916.28 + 277777.78 ≈ $71,460.53. 70 Bankers have a rule of thumb that if you receive Further Banker’s InsightsRule andofChallenges

R percent interest, continuously compounded, then your money doubles after approximately 70/R years. For example, at R = 5%, your money 25. The text after proves thator e= lim (1Use + n1the )n . concept Use a change of variables to show any x, Rule. (Note: Sometimes, doubles 70/5 14n→∞ years. of doubling time to justifythat thefor Banker’s the approximation 72/R is used. It is less accurate but easier toapply because 72 is divisible by more numbers than

x n 1 nx 70.) = lim 1 + lim 1 + n→∞ n→∞ n n Use this to conclude that e x = lim (1 + nx )n . n→∞

SOLUTION

Let t = x/n. Then

x n 1 tx 1 nx lim 1 + = lim 1 + = lim 1 + . n→∞ n→∞ t→∞ n t n

1 n Since e = limn→∞ 1 + , n e x = lim

n→∞

1+

x n 1 nx = lim 1 + . n→∞ n n

27. A bank pays interest at the rate r , compounded M times yearly. The effective interest rate re is the rate at which Use Eq. (3) to prove that for n > 0, interest, if compounded annually, would have to be paid to produce the same yearly return. (a) Find re if r = 9% compounded monthly. 1 n+1 1 n M r ≤ e ≤ 1 + 1 + (b) Show that re = (1 + r/M) − 1 and that re = ne − 1 if interest isncompounded continuously. (c) Find re if r = 11% compounded continuously. (d) Find the rate r , compounded weekly, that would yield an effective rate of 20%. SOLUTION

(a) Compounded monthly, P(t) = P0 (1 + r/12)12t . By the definition of re , P0 (1 + 0.09/12)12t = P0 (1 + re )t so (1 + 0.09/12)12t = (1 + re )t

or re = (1 + 0.09/12)12 − 1 = 0.0938,

or 9.38% (b) In general, P0 (1 + r/M) Mt = P0 (1 + re )t , so (1 + r/M) Mt = (1 + re )t or re = (1 + r/M) M − 1. If interest is compounded continuously, then P0 er t = P0 (1 + re )t so er t = (1 + re )t or re = er − 1. (c) Using part (b), re = e0.11 − 1 ≈ 0.1163 or 11.63%. (d) Solving

r 52 −1 0.20 = 1 + 52 for r yields r = 52(1.21/52 − 1) = 0.1826 or 18.26%.

S E C T I O N 7.6

Models Involving y = k ( y − b)

365

7.6 Models Involving y = k ( y − b) Preliminary Questions 1. What is the general solution to y = −k(y − b)? SOLUTION

The general solution is y(t) = b + Ce−kt for any constant C.

2. Write down a solution to y = 4(y − 5) that tends to −∞ as t → ∞. The general solution is y(t) = 5 + Ce4t for any constant C; thus the solution tends to −∞ as t → ∞ whenever C < 0. One specific example is y(t) = 5 − e4t . SOLUTION

3. Does there exist a solution of y = −4(y − 5) that tends to ∞ as t → ∞? The general solution is y(t) = 5 + Ce−4t for any constant C. As t → ∞, y(t) → 5. Thus, there is no solution of y = −4(y − 5) that tends to ∞ as t → ∞. SOLUTION

4. True or false? If k > 0, then all solutions of y = −k(y − b) approach the same limit as t → ∞. True. The general solution of y = −k(y − b) is y(t) = b + Ce−kt for any constant C. If k > 0, then y(t) → b as t → ∞.

SOLUTION

5. Suppose that material A cools more rapidly than material B. Which material has a larger cooling constant k in Newton’s Law of Cooling, y = −k(y − T0 )? SOLUTION

Because material A cools more rapidly, material A has the larger cooling constant.

Exercises 1. Find the general solution of y = 2(y − 10). Then find the two solutions satisfying y(0) = 25 and y(0) = 5, and sketch their graphs. The general solution of y = 2(y − 10) is y(t) = 10 + Ce2t for any constant C. If y(0) = 25, then 10 + C = 25, or C = 15; therefore, y(t) = 10 + 15e2t . On the other hand, if y(0) = 5, then 10 + C = 5, or C = −5; therefore, y(t) = 10 − 5e2t . Graphs of these two functions are given below. SOLUTION

y

y 0.5

800

1

1.5

x

−50

600

−100

y(0) = 25

400

−150

y(0) = 5

−200

200 x 0.5

1

−250

1.5

3. Verify directly that y = b + Cekt satisfies y = k(y − b) for any constant C. Find the general solution of y = −3(y − 12). Then find the two solutions satisfying y(0) = 20 and y(0) = 0, SOLUTION y(t) = b + Cekt for any constant C. Then y = kCekt and and sketchLet their graphs. k(y − b) = k(b + Cekt − b) = kCekt = y . 5. A hot metal bar is submerged in a large reservoir of water whose temperature is 60◦ F. The temperature of ◦ the bar Letsubmersion F(t) be theistemperature hot object submergedhas in cooled a large to pool 20 s after 100◦ F. Afterof1amin, the temperature 80◦of F. water whose temperature is 70 F. (a) What isthe thecooling differential equation (a) Determine constant k. satisfied by F(t) if the cooling constant is k = 1.5? ◦ F. (b) Find a formula for F(t) if the object’sby initial temperatureF(t) is 250 (b) What is the differential equation satisfied the temperature of the bar? (c) What is the formula for F(t)? (d) Determine the temperature of the bar at the moment it is submerged. SOLUTION

With T0 = 60◦ F, the temperature of the bar is given by F(t) = 60 + Ce−kt for some constants C and k.

(a) Given F(20) = 100◦ F and F(60) = 80◦ F, it follows that Ce−20k = 40 and Ce−60k = 20. Dividing the first equation by the second leaves e40k = 2, so k=

1 ln 2 ≈ 0.017 seconds−1 . 40

(b) F (t) = −0.017(F(t) − 60). (c) The general solution of the equation in part (b) is F(t) = 60 + Ce−0.017t for some constant C. With F(20) = 100 = 60 + Ce−0.017(20) we have 40 = Ce−0.017(20) so C ≈ 56.2. Thus, F(t) = 60 + 56.2e−0.017t . (d) Using the formula found in part (c), F(0) = 60 + 56.2 = 116.2◦ F. A hot metal rod is placed in a water bath whose temperature is 40◦ F. The rod cools from 300 to 200◦ F in 1 min. How long will it take for the rod to cool to 150◦ F?

366

CHAPTER 7


7. When a hot object is placed in a water bath whose temperature is 25◦ C, it cools from 100 to 50◦ C in 150 s. In another bath, the same cooling occurs in 120 s. Find the temperature of the second bath. With T0 = 25◦ C, the temperature of the object is given by F(t) = 25 + Ce−kt for some constants C and k. From the initial condition, F(0) = 25 + C = 100, so C = 75. After 150 seconds, F(150) = 25 + 75e−150k = 50, so 25 1 ln ≈ 0.0073 seconds−1 . k=− 150 75 SOLUTION

If we place the same object with a temperature of 100◦ C into a second bath whose temperature is T0 , then the temperature of the object is given by F(t) = T0 + (100 − T0 )e−0.0073t . To cool from 100◦ C to 50◦ C in 120 seconds, T0 must satisfy T0 + (100 − T0 )e−0.0073(120) = 50. Thus, T0 = 14.32◦ C. min−1 . 9. A cup of coffee, cooling off◦in a room at temperature 20◦ C, has cooling constant k = 0.09 A cold metal bar at −30 C is submerged in a pool maintained at a temperature of 40◦ C. Half a minute later, the (a) How fast is the coffee degrees when is T = 80◦ C? temperature of the bar cooling is 20◦ C.(in How long per willminute) it take for the its bartemperature to attain a temperature of 30◦ C? (b) Use the Linear Approximation to estimate the change in temperature over the next 6 s when T = 80◦ C. (c) The coffee is served at a temperature of 90◦ C. How long should you wait before drinking it if the optimal temperature is 65◦ C? SOLUTION

(a) According to Newton’s Law of Cooling, the coffee will cool at the rate k(T − T0 ), where k is the cooling constant of the coffee, T is the current temperature of the coffee and T0 is the temperature of the surroundings. With k = 0.09 min−1 , T = 80◦ C and T0 = 20◦ C, the coffee is cooling at the rate 0.09(80 − 20) = 5.4◦ C/min. (b) Using the result from part (a) and the Linear Approximation, we estimate that the coffee will cool (5.4◦ C/min)(0.1 min) = 0.54◦ C over the next 6 seconds. (c) With T0 = 20◦ C and an initial temperature of 90◦ C, the temperature of the coffee at any time t is T (t) = 20 + 70e−0.09t . Solving 20 + 70e−0.09t = 65 for t yields 1 45 t =− ln ≈ 4.91 minutes. 0.09 70 In Exercises 11–14, use the model for free-fall with air resistance discussed in this section. If the weight w of an object Two identical objects are heated to different temperatures T1 and T2 . Both are submerged in a cold bath of (in pounds) is given rather◦ than its mass (as in Example 2), then the differential equation for free fall with air resistance temperature T = 40 C at t = 0. Measurements show that T2 = 400◦ C and the cooling rate of object 1 is twice as is v = −(kg/w)(v0 + w/k), where g = 32 ft/s2 . large as the cooling rate of object 2 (at each time t). Find T1 . 11. A 60-kg skydiver jumps out of an airplane. What is her terminal velocity in miles per hour, assuming that k = 10 kg/s for free-fall (no parachute)? SOLUTION

The free-fall terminal velocity is −9.8(60) −gm = = −58.8 m/s ≈ −192.9 ft/s ≈ −131.5 mph. k 10

13. A 175-lb skydiver jumps out of an airplane (with zero initial velocity). Assume that k = 0.7 lb-s/ft with a closed Find the terminal velocity of a 192-lb skydiver if k = 1.2 lb-s/ft. How long does it take him to reach half of his parachute and k = 5 lb-s/ft with an open parachute. What is the skydiver’s velocity at t = 25 s if the parachute opens terminal velocity if his initial velocity is zero? after 20 seconds of free fall? SOLUTION

The skydiver’s velocity v(t) satisfies the differential equation v = −(kg/w)(v + w/k), which has general

solution v(t) = −

w + Ce−(kg/w)t . k

While the parachute is closed, k = 0.7. Since w = 175 lbs, we have 175 w = = 250 ft/sec, k 0.7

kg (0.7)(32) = = 0.128 sec−1 w 175

S E C T I O N 7.6

Models Involving y = k ( y − b)

367

and thus v(t) = −250 + Ce−0.128t . From the initial condition v(0) = −250 + C = 0, we find that C = 250 and v(t) = −250 + 250e−0.128t = −250(1 − e−0.128t ). At t = 20 sec, the skydiver’s velocity is then v(20) = −250(1 − e−0.128(20) ) ≈ 231 ft/sec. After the parachute opens, we have k = 5, so 175 w = = 35 ft/sec, k 5

kg (5)(32) = ≈ 0.91 sec−1 w 175

and v(t) = −35 + Ce−0.91t , where t = 0 now corresponds to the time when the parachute opens, Because the skydiver’s velocity is 231 ft/sec when the parachute opens, we have v(0) = −35 + C = 231

so that C = 266 ft/sec.

Therefore v(t) = −35 + 266e−0.91t . Finally, five seconds after the parachute opens, the skydiver’s velocity is v(t) = −35 + 266e−0.91(5) = −35 + 266e−4.55 ≈ −32 ft/sec. 15. A continuous annuity with withdrawal rate N = $1,000/year and interest rate r = 5% is funded by an initial deposit Does a heavier or lighter skydiver reach terminal velocity faster? P0 . (a) When will the annuity run out of funds if P0 = $15,000? (b) Which initial deposit P0 yields a constant balance? SOLUTION

(a) Let P(t) denote the balance of the annuity at time t measured in years. Then P(t) =

N 1000 + Cer t = + Ce0.05t = 20000 + Ce0.05t r 0.05

for some constant C. If P0 = P(0) = 15000, then 15000 = 20000 + C and C = −5000. To determine when the annuity runs out, we set P(t) = 0 and solve for t. This yields t=

1 ln 4 ≈ 28 years. 0.05

(b) From part (a), we know that P(t) = 20000 + Ce0.05t . The balance of the annuity will remain constant provided C = 0. Then P(t) = 20000 for all t. Hence, P0 = $20000 leads to a constant balance. 17. Find the minimum initial deposit that will allow an annuity to pay out $500/year indefinitely if it earns interest at a Show that a continuous annuity with withdrawal rate N = $5,000/year and interest rate r = 8%, funded by an rate of 5%. initial deposit of P0 = $75,000, never runs out of money. SOLUTION Let P(t) denote the balance of the annuity at time t measured in years. Then P(t) =

500 N + Cer t = + Ce0.05t = 10000 + Ce0.05t r 0.05

for some constant C. To fund the annuity indefinitely, we must have C ≥ 0. If the initial deposit is P0 , then P0 = 10000 + C and C = P0 − 10000. Thus, to fund the annuity indefinitely, we must have P0 ≥ $10000. 19. An initial deposit of $5,000 is placed in a bank account. What is the minimum interest rate the bank must pay to What is the minimum initial deposit necessary to fund an annuity for 30 years if the annuity earns interest at the allow continuous withdrawals at a rate of $500/year to continue indefinitely? rate r = 0.07 and withdrawals are made at a rate of $2,000/year? SOLUTION Let P(t) denote the balance of the annuity at time t measured in years. Then P(t) =

500 N + Cer t = + Cer t r r

for some constant C. To fund the annuity indefinitely, we need C ≥ 0. If the initial deposit is $5000, then 5000 = 500 r +C 500 500 and C = 5000 − r . Thus, to fund the annuity indefinitely, we need 5000 − r ≥ 0, or r ≥ 0.1. The bank must pay at least 10%.

368

CHAPTER 7


$10,000 from never a bankruns at anout interest rate of pays back the loan continuously at atorate 21. ShowJulie that borrows a continuous annuity of money if 9% the and initial balance is greater than or equal N /rof, N dollars/year. P(t) denoterate the and amount owedrate. at time t. where N is Let the withdrawal r thestill interest (a) Explain why y = P(t) satisfies the differential equation y = 0.09y − N (b) How long will it take Julie to pay back the loan if N = $1,200? (c) Will she ever be able to pay back the loan if N = $800? SOLUTION

(a) Rate of Change of Loan = (Amount still owed)(Interest rate) − (Payback rate) N = P(t) · r − N = r P − . r Therefore, if y = P(t), y = r

y−

N r

= ry − N

(b) From the differential equation derived in part (a), we know that P(t) = Nr + Cer t = 13333.33 + Ce0.09t . Since $10000 was initially borrowed, P(0) = 13333.33 + C = 10000, and C = −3333.33. The loan is paid off when P(t) = 13333.33 − 3333.33e0.09t = 0. This yields 1 13333.33 t= ln ≈ 15.4 years. 0.09 3333.33 (c) If the annual rate of payment is $800, then P(t) = 800/0.09 + Ce0.09t = 8888.89 + Ce.09t . With P(0) = 8888.89 + C = 10000, it follows that C = 1111.11. Since C > 0 and e0.09t → ∞ as t → ∞, P(t) → ∞, and the loan will never be paid back. 23. Current in a Circuit The electric current flowing in the circuit in Figure 6 (consisting of a battery of V volts, a Let N (t) be the fraction of the population who have heard a given piece of news t hours after its initial release. resistor of R ohms, and an inductor) satisfies According to one model, the rate N (t) at which the news spreads is equal to k times the fraction of the population d I k. that has not yet heard the news, for some constant = −k(I − b) (a) Determine the differential equation satisfieddtby N (t). (b) Find the solution thiskdifferential equation with the Iinitial condition aNmaximum (0) = 0 inlevel termsVof for some constants k and bofwith > 0. Initially, I (0) = 0 and (t) approaches /Rk.as t → ∞. (c) Suppose that half (a) What is the value of b?of the population is aware of an earthquake 8 hours after it occurs. Use the model to calculate k andaestimate that (b) Find formula the for percentage I (t) in terms ofwill k, Vknow , and about R. the earthquake 12 hours after it occurs. (c) Show that I (t) reaches approximately 63% of its maximum value at time t = 1/k.

Battery

V

R (resistor)

FIGURE 6 Current flow approaches the level Imax = V /R. SOLUTION

(a) Since the general solution has the form I (t) = b + Ce−kt , we see that b is the final current level as t → ∞. Therefore b = V /R. V V V (b) I (t) = − e−kt = (1 − e−kt ). R R R (c) As 1 − e−kt is an increasing function of t, it follows that I (t) achieves its maximum value as t → ∞: maximum value = lim

V

t→∞ R

(1 − e−kt ) =

At time t = 1/k, I or roughly 63% of its maximum value.

1 V V = (1 − e−1 ) ≈ 0.6321 , k R R

V . R

S E C T I O N 7.7

ˆ L’Hopital’s Rule

369

Further Insights and Challenges 25. Show that by Newton’s Law of Cooling, the time required to cool an object from temperature A to temperature B is Show that the cooling constant of an object can be determined from two temperature readings y(t1 ) and y(t2 ) at times t1 = t2 by the formula 1 A − T0 t = ln k1 B−y(t T0 ) − T 2 0 ln k= t1 − t 2 y(t1 ) − T0 where T0 is the ambient temperature. SOLUTION At any time t, the temperature of the object is y(t) = T0 + Ce−kt for some constant C. Suppose the object is initially at temperature A and reaches temperature B at time t. Then A = T0 + C, so C = A − T0 . Moreover,

B = T0 + Ce−kt = T0 + ( A − T0 )e−kt . Solving this last equation for t yields t=

1 ln k

A − T0 B − T0

.

Air Resistance A projectile of mass m = 1 travels straight up from ground level with initial velocity v0 . Suppose that the velocity v satisfies v = −g − kv. 7.7 (a)L’Hˆ opital’s Rule Find a formula for v(t). (b) Show that the projectile’s height h(t) is given by


g 1. Which of the following two limits can be evaluated opital’s ) − Rule? t h(t) = using C(1 −L’Hˆ e−kt k 3x − 12 12x − 3 , lim lim 2 where C = k −2 (g + kv0 ). x→4 x − 16 x→4 x − 4 (c) Show that the projectile reaches its maximum height at time tmax = k −1 ln(1 + kv0 /g). SOLUTION x → 4,of air resistance, the maximum height is reached at time t = v /g. In view of this, explain why (d) In theAs absence 0 we should expect that 3x − 12 x 2 −+16kv0 ) ln(1 v g = 0 lim 0 g is of the form , so L’Hôpital’s Rule can be used tok→0 evaluate k 0

kv0 1/k 3x − 12 (e) Verify Eq. (7). Hint: Use Theorem 1 in Section = ev0 /g or use L’Hôpital’s lim7.5 2to show. that lim 1 + g k→0 x→4 x − 16 Rule. As x → 4, 12x − 3 x −4 is not of the form 00 or ∞ ∞ , so L’Hôpital’s Rule cannot be used to evaluate lim

12x − 3

x→4 x − 4

2. What is wrong with evaluating x 2 − 2x x→0 3x − 2 lim

using L’Hôpital’s Rule? SOLUTION

As x → 0, x 2 − 2x 3x − 2

is not of the form 00 or ∞ ∞ , so L’Hôpital’s Rule cannot be used.

.

370

CHAPTER 7


Exercises In Exercises 1–10, show that L’Hôpital’s Rule is applicable and use it to evaluate the limit. 2x 2 + x − 3 x −1 x→1

1. lim

SOLUTION

The functions 2x 2 + x − 3 and x − 1 are differentiable, but the quotient is indeterminate at x = 1, 2+1−3 0 2x 2 + x − 3 = = , x −1 1 − 1 0 x=1

so L’Hôpital’s Rule applies. We find 2x 2 + x − 3 4x + 1 = lim = 4 + 1 = 5. x −1 1 x→1 x→1 lim

6x 3 +2 13x 2 + 9x + 2 lim 2x + x − 10 lim 6x 3 − x 2 − 5x + 2 x→−1 x→2 2x 5 − 40x + 16 SOLUTION The functions 6x 3 + 13x 2 + 9x + 2 and 6x 3 − x 2 − 5x + 2 are differentiable, but the quotient is indeterminate at x = −1, 6x 3 + 13x 2 + 9x + 2 −6 + 13 − 9 + 2 0 = = , −6 − 1 + 5 + 2 0 6x 3 − x 2 − 5x + 2 x=−1 3.

so L’Hôpital’s Rule applies. We find 6x 3 + 13x 2 + 9x + 2 18x 2 + 26x + 9 18 − 26 + 9 1 = lim = = . 3 2 2 18 + 2 − 5 15 x→−1 6x − x − 5x + 2 x→−1 18x − 2x − 5 lim

√ x −3 sin 4x 2 lim x→9 2x − 17x − 9 x→0 x 2 + 3x √ SOLUTION The functions x − 3 and 2x 2 − 17x − 9 are differentiable, but the quotient is indeterminate at x = 9, √ 3−3 0 x −3 = = , 162 − 153 − 9 0 2x 2 − 17x − 9 x=9 5. lim

so L’Hôpital’s Rule applies. We find √

√ x −3 1/(2 x) 1/6 1 lim = = . = lim 36 − 17 114 x→9 2x 2 − 17x − 9 x→9 4x − 17 x2 3 7. lim x 1 − cos x x→0lim x→0 sin x − x SOLUTION The functions x 2 and 1 − cos x are differentiable, but the quotient is indeterminate at x = 0, x2 0 0 = = , 1 − cos x x=0 1−1 0 so L’Hôpital’s Rule applies. Here, we use L’Hôpital’s Rule twice to find x2 2x 2 = lim = lim = 2. x→0 1 − cos x x→0 sin x x→0 cos x lim

cos x − cos2 x 9. limlim cos 2x − 1 sin x x→0 x→0 sin 5x SOLUTION The functions cos x − cos2 x and sin x are differentiable, but the quotient is indeterminate at x = 0, 1−1 0 cos x − cos2 x = = , sin x 0 0 x=0 so L’Hôpital’s Rule applies. We find cos x − cos2 x − sin x + 2 cos x sin x 0 = lim = = 0. sin x cos x 1 x→0 x→0 lim

S E C T I O N 7.7


371

√ In Exercises 11–16, x + 1show − 2 that L’Hôpital’s Rule is applicable to the limit as x → ±∞ and evaluate. lim 3 x→33xx −−1 7x − 6 11. lim x→∞ 7 − 12x 3x − 1 ∞ SOLUTION As x → ∞, the quotient is of the form , so L’Hôpital’s Rule applies. We find 7 − 12x ∞ lim

3x − 1

x→∞ 7 − 12x

3

= lim

x→∞ −12

1 =− . 4

x 13. lim x ln x x→∞ e lim √ x→∞ x x ∞ SOLUTION As x → ∞, the quotient is of the form , so L’Hôpital’s Rule applies. We find x e ∞ x

1

= lim x = 0. lim x→∞ e x x→∞ e x2 x→∞lim ex

ln(x 4 + 1) x→−∞ x

15. lim

SOLUTION

twice to find

x2 ∞ As x → ∞, the quotient x is of the form , so L’Hôpital’s Rule applies. Here, we use L’Hôpital’s Rule e ∞ x2 2x 2 = lim x = lim x = 0. x→∞ e x x→∞ e x→∞ e lim

1 In Exercises 17–48, apply L’Hôpital’s Rule to evaluate the limit. In some cases, it may be necessary to apply it more than lim x sin once. x→−∞ x 17. lim

sin 4x

x→0 sin 3x

sin 4x 4 cos 4x 4 = lim = . 3 x→0 sin 3x x→0 3 cos 3x tan x 19. lim tan 3x x x→0 lim x→π /2 tan 5x tan x sec2 x SOLUTION lim = lim = 1. 1 x→0 x x→0 1 √ 21. lim cot x8 − + xx − 3x 1/4 x→0lim 2 x→1 x + 3x − 4 lim

SOLUTION

SOLUTION

lim

x→0

cot x −

1 x

x cos x − sin x −x sin x + cos x − cos x −x sin x = lim = lim x sin x x cos x + sin x x→0 x→0 x→0 x cos x + sin x

= lim

0 −x cos x − x = = 0. 2 x→0 −x sin x + cos x + cos x

= lim

23.

3x − 2

lim x 2 − e4 x→−∞ 1e − 5x lim x→2 x − 2 3x − 2 lim

SOLUTION

x→−∞ 1 − 5x

=

lim

3

x→−∞ −5

3 =− . 5

7x 22 + 4x lim x + 4x x→−∞ lim 9 − 3x 2 x→∞ 9x 3 + 4 7 7x 2 + 4x 14x + 4 14 SOLUTION lim = lim = lim =− . 2 x→−∞ 9 − 3x x→−∞ −6x x→−∞ −6 3

25.

3x 3 + 4x 2 27. limlim tan3 π x ln x x→∞ 4x − 27 x→1 18 3 3x 3 + 4x 2 9x 2 + 8x 18x + 8 = lim SOLUTION lim = lim = = . 3 x→∞ 4x − 7 x→∞ 12x 2 x→∞ 24x 24 4

4 1 lim √ − x −4 x −2 x→4

372

CHAPTER 7


29. lim

x→1

x(ln x − 1) + 1 (x − 1) ln x

SOLUTION

1 x( 1x ) + (ln x − 1) 1 x(ln x − 1) + 1 ln x 1 x = = lim = . = lim = lim 1+1 2 x→1 (x − 1) ln x x→1 (x − 1)( 1 ) + ln x x→1 1 − 1 + ln x x→1 12 + 1 x x x

lim

x

cos(x + π2 )π 31. lim lim x − sin x 2 tan x x→0 x→π /2 SOLUTION

cos(x + π2 ) − sin(x + π2 ) = lim = −1. sin x cos x x→0 x→0 lim

sin xx− x cos x 33. lim e −1 x→0lim x − sin x x→0 sin x SOLUTION

sin x − x cos x x sin x sin x + x cos x cos x + cos x − x sin x = lim = lim = lim = 2. x − sin x sin x cos x x→0 x→0 1 − cos x x→0 x→0 lim

e x −e 1 2x − csc x→1limln x x→∞ x 2 e ex − e ex SOLUTION lim = lim −1 = = e. 1 x→1 ln x x→1 x cos x 37. lim x2 x→π /2 sin(2x) lim x→0 1 − cos x cos x 1 − sin x SOLUTION lim = lim = . 2 x→π /2 sin(2x) x→π /2 2 cos(2x) 35. lim

39. lim e−x2x (x 3 − x 2 + 9) x→∞ e − 1 lim x x→0 SOLUTION lim e−x (x 3 − x 2 + 9) = lim

x→∞

x→∞

x3 − x2 + 9 3x 2 − 2x 6x − 2 6 = lim = lim = lim x = 0. x x x x→∞ x→∞ x→∞ e e e e

x 1/3 x −1 41. lim 2 a − 1 lim x→1 x + 3x − 4 (a > 0) x x→0

2

SOLUTION

1 1 −3 x 1/3 − 1 3x 3 = 1 . = lim = 5 15 x→1 x 2 + 3x − 4 x→1 2x + 3

lim

x 1/x(sec x − tan x) 43. lim lim x→∞ x→π /2

SOLUTION

1 1 ln x ln x = lim = lim x = 0. Hence, x→∞ x x→∞ x x→∞ 1

lim ln x 1/x = lim

x→∞

lim x 1/x = lim eln(x

x→∞

x→∞

(1/x) )

= e0 = 1.

45. lim (sin t)(ln t) t→0+ lim (1 + ln x)1/(x−1) x→1

SOLUTION

1 ln t − sin2 t −2 sin t cos t t = lim = lim = lim = 0. t→0+ csc t t→0+ − csc t cot t t→0+ t cos t t→0+ cos t − t sin t

lim (sin t)(ln t) = lim

t→0+

47. lim x sin xsin x x→0lim x x→0

S E C T I O N 7.7


373

SOLUTION

lim ln(x sin x ) = lim sin x(ln x) = lim

x→0

x→0

x→0

x→0

Hence, lim x sin x = lim e x→0

x→0

1 sin x

1 x

= lim

x→0 − cos x(sin x)−2

sin2 x 2 sin x cos x = lim − = 0. x cos x x→0 −x sin x + cos x

= lim − ln(x sin x )

ln x

= e0 = 1.

cos mx 49. Evaluate limx x , where m and n are nonzero whole numbers. cos nx x→ π /2 lim x→∞ x + 1 SOLUTION Suppose m and n are even. Then there exist integers k and l such that m = 2k and n = 2l and cos mx

lim

x→π /2 cos nx

=

cos k π = (−1)k−l . cos l π

Now, suppose m is even and n is odd. Then cos mx x→π /2 cos nx lim

does not exist (from one side the limit tends toward −∞, while from the other side the limit tends toward +∞). Third, suppose m is odd and n is even. Then lim

cos mx

x→π /2 cos nx

= 0.

Finally, suppose m and n are odd. This is the only case when the limit is indeterminate. Then there exist integers k and l such that m = 2k + 1, n = 2l + 1 and, by L’Hôpital’s Rule, lim

cos mx

x→π /2 cos nx

−m sin mx

= lim

x→π /2 −n sin nx

= (−1)k−l

m . n

To summarize, ⎧ (−1)(m−n)/2 , ⎪ ⎪ ⎪ ⎨ cos mx does not exist, lim = ⎪0 x→π /2 cos nx ⎪ ⎪ ⎩ (−1)(m−n)/2 m n,

m, n even m even, n odd m odd, n even m, n odd

Can L’Hôpital’s Rule be applied to lim x sin(1/x) ? Does a graphical or numerical investigation suggest that xn − 1 lim m for any numbersx→0+ n, m = 0. the limitEvaluate exists? x→1 x −1

51.

SOLUTION Since sin(1/x) oscillates as x → 0+, L’Hˆ opital’s Rule cannot be applied. Both numerical and graphical investigations suggest that the limit does not exist due to the oscillation.

x

1

0.1

0.01

0.001

0.0001

0.00001

x sin(1/x)

1

3.4996

10.2975

0.003316

16.6900

0.6626

y 20 15 10 5 x 0

0.1

0.2

0.3

0.4

53. Let f (x) = x 1/x in the domainln{xx : x > 0}. Sketch the by determining the increasing/decreasing behavior, concavity, and asymptotic f (x)of andy = lim f (x). (a) Calculate lim graph x→∞x x→0+ behavior as in Section 4.5. (b) Find the maximum value of f (x) and determine the intervals on which f (x) is increasing or decreasing.

374

CHAPTER 7


(c) Use (a) and (b) to prove that x 1/x = c has a unique solution if 0 < c ≤ 1 or c = e1/e , two solutions if 1 < c < e1/e , and no solutions if c > e1/e . Plot the graph of f (x). Explain how the graph confirms the conclusions in (c). (d) SOLUTION

(a) Let f (x) = x 1/x . Note that limx→0+ x 1/x is not indeterminate. As x → 0+, the base of the function tends toward 0 and the exponent tends toward +∞. Both of these factors force x 1/x toward 0. Thus, limx→0+ f (x) = 0. On the other hand, limx→∞ f (x) is indeterminate. We calculate this limit as follows: lim ln f (x) = lim

x→∞

ln x

x→∞ x

1

= lim

x→∞ x

= 0,

so limx→∞ f (x) = e0 = 1. (b) Again, let f (x) = x 1/x , so that ln f (x) = 1x ln x. To find the derivative f , we apply the derivative to both sides: d 1 d ln f (x) = ln x dx dx x 1 ln x 1 f (x) = − 2 + 2 f (x) x x ln x 1 x 1/x f (x) = f (x) − 2 + 2 = 2 (1 − ln x) x x x Thus, f is increasing for 0 < x < e, is decreasing for x > e and has a maximum at x = e. The maximum value is f (e) = e1/e ≈ 1.444668. (c) Because (e, e1/e ) is the only maximum, no solution exists for c > e1/e and only one solution exists for c = e1/e . Moreover, because f (x) increases from 0 to e1/e as x goes from 0 to e and then decreases from e1/e to 1 as x goes from e to +∞, it follows that there are two solutions for 1 < c < e1/e , but only one solution for 0 < c ≤ 1. (d) Observe that if we sketch the horizontal line y = c, this line will intersect the graph of y = f (x) only once for 0 < c ≤ 1 and c = e1/e and will intersect the graph of y = f (x) twice for 1 < c < e1/e . There are no points of intersection for c > e1/e . y

1 0.5 x

0 5

10

15

20

√ 2 4 x 1/10 . (ln fx)(or 55. Show that (ln x) Determine if f gxorand g neither) for the functions f (x) = log10 x and g(x) = ln x. SOLUTION

•

√

x (ln x)2 : √

1 1 √ √ √ √ x x x 2 x 2 x lim = lim 2 = lim 4 = lim = ∞. = lim x→∞ (ln x)2 x→∞ ln x x→∞ 4 ln x x→∞ x→∞ 8 x x • x 1/10 (ln x)4 : 1

1

x 1/10 x 1/10 x 1/10 10x 9/10 10x 9/10 = lim = lim = lim = lim x→∞ (ln x)4 x→∞ 4 (ln x)3 x→∞ 40(ln x)3 x→∞ 120 (ln x)2 x→∞ 1200(ln x)2 x x lim

1

1

x 1/10 x 1/10 9/10 10x 9/10 = lim 2400 = lim = lim 10x = ∞. = lim 24000 x→∞ x→∞ 24000 ln x x→∞ x→∞ 240000 x (ln x) x 57. Show x) N x afunctions for all N are anddistinguished all a > 0. Justthat as (ln exponential by their rapid rate of increase, the logarithm functions grow a for all a > 0. particularly slowly. Show that ln x x SOLUTION xa ax a−1 ax a = lim = lim = ··· x→∞ (ln x) N x→∞ N (ln x) N −1 x→∞ N (ln x) N −1 x lim

S E C T I O N 7.7


375

If we continue in this manner, L’Hôpital’s Rule will give a factor of x a in the numerator, but the power on ln x in the denominator will eventually be zero. Thus, xa = ∞, x→∞ (ln x) N lim

so x a (ln x) N for all N and for all a > 0. all whole√numbers√n > 0. 59. Show that lim x n e−x√= 0 for √ x→∞ Determine whether x e ln x or e ln x x. Hint: Do not use L’Hôpital’s Rule. Instead, make a substituSOLUTION tion u = ln x. xn nx n−1 = lim x→∞ e x x→∞ e x

lim x n e−x = lim

x→∞

n(n − 1)x n−2 x→∞ ex

= lim .. . = lim

n!

x→∞ e x

= 0.

61. Use Eq. (2) of Section 7.5 to show that the PV of an investment which pays out income continuously at a constant −r Tx) and g(x) = x 2 + 1. Assumptions Matter Let f (x) = x(2 1 −+esin rate of R dollars/year for T years is PV = R , where r is the interest rate. Use L’Hôpital’s Rule to prove that (a) Show directly that lim f (x)/g(x) = 0. r x→∞ the PV approaches RT as r → 0. (b) Show that lim f (x) = lim g(x) = ∞, but lim f (x)/g (x) does not exist. x→∞ x→∞ x→∞ SOLUTION By Eq. (2) of Section 7.5, Do (a) and (b) contradict L’Hôpital’s Rule? Explain. " T R −r t T R −r t Re dt = PV = e = (1 − e−r T ). −r r 0 0 Using L’Hôpital’s Rule, R(1 − e−r T ) RT e−r T = lim = RT. r 1 r →0 r →0 lim

63. Show that bylim t k e−t = 0 for all k. Hint: Compare with lim t k e−t = 0. Verify differentiation: t→∞ t→∞ 2

T as t → +∞, Because we are interested in the"limit will er T (rwe T− 1)restrict + 1 attention to t > 1. Then, for all k, ter t dt = 2 r 0 2 0 ≤ t k e−t ≤ t k e−t . Then use L’Hôpital’s Rule to show that the limit of the right-hand side as r → 0 is equal to the value of the integral k e−t for r t= 0. = 0, it follows from the Squeeze Theorem that As lim SOLUTION

t→∞

2 lim t k e−t = 0.

t→∞

In Exercises 64–66, let 2 e−1/x f (x) = 0

for x = 0 for x = 0

These exercises show that f (x) has an unusual property: All of its higher derivatives at x = 0 exist and are equal to zero. 65. Show that f (0) exists and is equal to zero. Also, verify that f (0) exists and is equal to zero. f (x) Show that lim from = 0 for all k. Hint: Let t = x −1 and apply the result of Exercise 63. SOLUTION Working x→0 x k the definition, f (x) − f (0) f (x) = lim =0 x −0 x→0 x→0 x

f (0) = lim

by the previous exercise. Thus, f (0) exists and is equal to 0. Moreover,

2 2 for x = 0 e−1/x x3 f (x) = 0 for x = 0

376

CHAPTER 7


Now, 2 f (x) − f (0) f (0) = lim = lim e−1/x x −0 x→0 x→0

2 x4

= 2 lim

f (x)

x→0 x 4

=0

by the previous exercise. Thus, f (0) exists and is equal to 0. Show that for k ≥ 1 and x = 0, Further Insights and Challenges

2 x + cos x P(x)e , but that −1/x it is of no help. Then evaluate the limit directly. x→∞ x f−(k) (x)x = cos r

67. Show that L’Hôpital’s Rule applies to lim

x x + cos x As x → ∞, both x + cos x and x − cos x tend toward infinity, so L’Hôpital’s Rule applies to(k)lim ; for some polynomial P(x) and some exponent r ≥ 1. Use the result of Exercise 64 to show that f x→∞ (0) exists andx x − cos 1 − sin x is equal zero forlimit, all k ≥lim 1. , does not exist due to the oscillation of sin x. However, however, theto resulting x→∞ 1 + sin x SOLUTION

1 + cosx x x + cos x = 1, = lim x→∞ x − cos x x→∞ 1 − cos x x lim

cos x = 0. x 69. Resonance A spring oscillates with a natural frequency λ /2π . If we drive the spring with a sinusoidal force L’Hôpital’s evaluate following limit, assuming that f is differentiable and f is continuous: C sin(ωUse t), where ω = λRule , thentothe springthe oscillates according to since lim

x→∞

y(t) =

C lim f (x) − f (a) λ sin( xω −t)a − ω sin(λ t)

x→a λ 2 − ω2

(a) Use L’Hôpital’s Rule to determine y(t) in the limit as ω → λ . (b) Define y0 (t) = lim y(t). Show that y0 (t) ceases to be periodic and that its amplitude |y0 (t)| tends to infinity as ω →λ

t → ∞ (the system is said to be in resonance; eventually, the spring is stretched beyond its limits). (c) Plot y(t) for λ = 1 and ω = 0.5, 0.8, 0.9, 0.99, and 0.999. How do the graphs change? Do the graphs confirm your conclusion in (b)? SOLUTION

(a) d

lim y(t) = lim C

ω →λ

ω →λ

= C lim

ω →λ

(λ sin(ω t) − ω sin(λ t)) λ sin(ω t) − ω sin(λ t) = C lim d ω d 2 2 2 2 λ −ω ω →λ d ω (λ − ω )

λ t cos(ω t) − sin(λ t) λ t cos(λ t) − sin(λ t) =C −2ω −2λ

(b) From part (a) y0 (t) = lim y(t) = C ω →λ

λ t cos(λ t) − sin(λ t) . −2λ

This may be rewritten as λ 2t 2 + 1 cos(λ t + φ ), y0 (t) = C −2λ 1 λt where cos φ = and sin φ = . Since the amplitude varies with t, y0 (t) is not periodic. Also note 2 2 2 λ t +1 λ t2 + 1 that λ 2t 2 + 1 → ∞ as t → ∞. C −2λ (c) The graphs below were produced with C = 1 and λ = 1. Moving from left to right and from top to bottom, ω = 0.5, 0.8, 0.9, 0.99, 0.999, 1. y 1 100 −1

y

y 4 2 x −2 −4

5 100 x

100 x −5

60

Inverse Trigonometric Functions

S E C T I O N 7.8 y

y

40

y

40

20

40

20

100 x

377

20

100 x

−20

−20

−20

−40

−40

−40

100 x

71. Suppose that f and g are polynomials such that f (a) = g(a) = 0. In this case, it is a fact from algebra that sin x f (x) = (xWe − expended a) f 1 (x) and g(x) = (x to − evaluate a)g1 (x) for and gthat to have verifyevaluated L’Hôpital’s Rule a lot of effort limsome polynomials in Chapter 2.f 1Show we this could it easily 1 . Use x x→0 directly for f and g. using L’Hôpital’s Rule. Then explain why this method would involve circular reasoning. SOLUTION As in the problem statement, let f (x) and g(x) be two polynomials such that f (a) = g(a) = 0, and let f 1 (x) and g1 (x) be the polynomials such that f (x) = (x − a) f 1 (x) and g(x) = (x − a)g1 (x). By the product rule, we have the following facts, f (x) = (x − a) f 1 (x) + f 1 (x) g (x) = (x − a)g1 (x) + g1 (x) so lim f (x) = f 1 (a)

and

x→a

lim g (x) = g1 (a).

x→a

L’Hôpital’s Rule stated for f and g is: if limx→a g (x) = 0, so that g1 (a) = 0, lim

f (x)

x→a g(x)

= lim

f (x)

x→a g (x)

f (a) = 1 . g1 (a)

Suppose g1 (a) = 0. Then, by direct computation, lim

f (x)

x→a g(x)

(x − a) f 1 (x) f (x) f (a) = lim 1 = 1 , x→a (x − a)g1 (x) x→a g1 (x) g1 (a)

= lim

exactly as predicted by L’Hôpital’s Rule. x n+1 − 1 " The integralnon the left in Exercise 72 is equal to f n (x) = . Investigate the limit graphically by x n+1 n + 1 Rule to show that the exceptional case The formula x d x = + C is valid for n = −1. Use L’Hôpital’s n + 1 plotting f n (x) for n = 0, −0.3, −0.6, and −0.9 together with ln x on a single plot. n = −1 is a limit of the general case in the following sense: For fixed x > 0, SOLUTION " x " x t n dt = t −1 dt lim

73.

yn→−1 1

2

Note that the integral on the left is equal to 1

1

n=0

x n+1 − 1 . n+1

n = −0.3 n = −0.6 n = −0.9 y = ln x x

1

2

3

4

5

−1

Patience Required: Use L’Hôpital’s Rule to evaluate and check your answers numerically: 2 1 sin x 1/x 1 7.8 (a)Inverse Trigonometric Functions (b) lim lim − 2 x x→0+ x→0 sin2 x x

Preliminary Questions 1. Which of the following quantities is undefined? (b) cos−1 (2) (a) sin−1 − 12 (d) csc−1 (2) (c) csc−1 12 1 π π SOLUTION (b) and (c) are undefined. sin−1 − 2 = − 6 and csc−1 (2) = 6 . 2. Give an example of an angle θ such that cos−1 (cos θ ) = θ . Does this contradict the definition of inverse function? SOLUTION

Any angle θ < 0 or θ > π will work. No, this does not contradict the definition of inverse function.

3. What is the geometric interpretation of the identity sin−1 x + cos−1 x = π /2? SOLUTION

Angles whose sine and cosine are x are complementary.

378

CHAPTER 7


Exercises In Exercises 1–6, evaluate without using a calculator. 1. cos−1 1 SOLUTION

cos−1 1 = 0.

3. cot−1 −1 1 sin 21 π SOLUTION cot−1 1 = 4 . √ 5. tan−1 3 2 sec−1 √ √ √3/2 π SOLUTION tan3−1 3 = tan−1 1/2 = 3 . In Exercises 7–16, compute without using a calculator. sin−1 (−1)

π 7. sin−1 sin 3 sin−1 (sin π3 ) = π3 . 3π 9. cos−1 −1cos 4π sin sin2 3 3π π 3π 3π SOLUTION cos−1 (cos 2 ) = cos−1 (0) = 2 . The answer is not 2 because 2 is not in the range of the inverse cosine function. 3π 11. tan−1 −1tan 5π 4 − sin sin 6 3π π 3π 3π SOLUTION tan−1 (tan 4 ) = tan−1 (−1) = − 4 . The answer is not 4 because 4 is not in the range of the inverse tangent function. SOLUTION

13. sec−1 (sec 3π ) tan−1 (tan π ) SOLUTION sec−1 (sec 3π ) = sec−1 (−1) = π . The answer is not 3π because 3π is not in the range of the inverse secant function. π) 15. csc−1 csc(− π 3 sec−1 sec 1 1 2 since csc(−π ) = sin(− SOLUTION No inverse π ) = 0 −→ ∞.

simplify In Exercises 17–20, π by referring to the appropriate triangle or trigonometric identity. cot−1 cot − 4 17. tan(cos−1 x) SOLUTION

Let θ = cos−1 x. Then cos θ = x and we generate the triangle shown below. From the triangle, 1 − x2 −1 tan(cos x) = tan θ = . x

19. cot(sec−1 x) cos(tan−1 x) SOLUTION Let θ = sec−1 x. Then sec θ = x and we generate the triangle shown below. From the triangle, cot(sec−1 x) = cot θ =

cot(sin−1 x)

1 x2 − 1

.

S E C T I O N 7.8


379

In Exercises 21–28, refer to the appropriate triangle or trigonometric identity to compute the given value. 21. cos sin−1 32 SOLUTION

Let θ = sin−1 32 . Then sin θ = 23 and we generate the triangle shown below. From the triangle, √ 2 5 −1 cos sin = cos θ = . 3 3

23. tan sin−1 0.8 tan cos−1 32 4 SOLUTION Let θ = sin−1 0.8. Then sin θ = 0.8 = 5 and we generate the triangle shown below. From the triangle, tan(sin−1 0.8) = tan θ =

4 . 3

25. cot csc−1 2−1 cos cot 1 √ π π SOLUTION csc−1 2 = 6 . Hence, cot(csc−1 2) = cot 6 = 3. 27. cot tan−1 20 tan sec−1 (−2) 1 1 SOLUTION Let θ = tan−1 20. Then tan θ = 20, so cot(tan−1 20) = cot θ = tan θ = 20 . 29–32, compute In Exercises the derivative at the point indicated without using a calculator. sin csc−1 20 29. y = sin−1 x, SOLUTION

x = 35

Let y = sin−1 x. Then y = √ 1

1−x 2

y 31. y = sec−1 −1 x, x = 4 y = tan x, x = 12 SOLUTION Let y = sec−1 x. Then y =

and

1 5 3 1 = √ = . = 5 4/5 4 1 − 9/25

√1

|x|

x 2 −1

and

1 y (4) = √ . 4 15 In Exercises 33–48, find the derivative. y = arccos(4x), x = 15 33. y = sin−1 (7x) SOLUTION

1 d d 7 · . sin−1 (7x) = 7x = 2 dx d x 1 − (7x) 1 − (7x)2

35. y = cos−1 (x 2 ) x y = arctan −1 d 2 −2x d 3 SOLUTION · . cos−1 (x 2 ) = x = dx 1 − x4 dx 1 − x4 37. y = x tan−1−1x y = sec (t + 1)

380

CHAPTER 7


SOLUTION

d x tan−1 x = x dx

1 1 + x2

+ tan−1 x.

39. y = arcsin(e x ) cos−1 x 1 d x ex y = d−1 −1 x SOLUTION sin sin (e ) = · . e = x dx 1 − e2x d x 1 − e2x −1 −1 z−1 41. y =y tan = csc (x 2) 1−z (1 − z 2 ) − z(−2z) 1 + z2 d z 1 −1 · = SOLUTION . tan =

2 dx 1 − z2 (1 − z 2 )2 z 2 + (1 − z 2 )2 z +1 2 1−z

t 43. y = cos−1 t −1− sec−1 1+t −1 y = tand − tt −1 − sec−1 t)= SOLUTION (cos1−1

−1

1 − (1/t)2

dx

−1 t2

−

1 |t| t 2 − 1

1 1 1 1 = − = 0. = − t4 − t2 |t| t 2 − 1 |t| t 2 − 1 |t| t 2 − 1 Alternately, let t = sec θ . Then t −1 = cos θ and cos−1 t −1 − sec−1 t = θ − θ = 0. Consequently, d (cos−1 t −1 − sec−1 t) = 0. dx 45. y = cos−1 (x−1+ sin−1 x) y = ecos x 1 −1 d −1 −1

1+ . SOLUTION cos (x + sin x) = dx 1 − x2 1 − (x + sin−1 x)2 47. y = 1 − t 2 + sin−1 t y = ln(sin −1 t) 1 1 −t 1 1−t d SOLUTION 1 − t 2 + sin−1 t = (1 − t 2 )−1/2 (−2t) + = + = . 2 2 2 dx 2 1−t 1−t 1−t 1 − t2 1 −1 −1 49. UseyFigure = x sin9 tox prove that (cos x) = − 1 − x 2 . 1

1 − x2

x

FIGURE 9 Right triangle with θ = cos−1 x. SOLUTION

Let θ = cos−1 x. Then cos θ = x and

1 dθ dθ 1 =− = 1 or =− . dx dx sin θ sin(cos−1 x) From Figure 9, we see that sin(cos−1 x) = sin θ = 1 − x 2 ; hence, − sin θ

1 1 d = − . cos−1 x = dx − sin(cos−1 x) 1 − x2 51. Let θ = sec−1 x.−1 Show that tan θ = x 2 − 1 if x ≥ 1 and tan θ = − x 2 − 1 if x ≤ −1. Hint: tan θ ≥ 0 on Show that (tan x) = cos2 (tan−1 x) and then use Figure 10 to prove that (tan−1 x) = (x 2 + 1)−1 . [0, π /2) and tan θ ≤ 0 on (π /2, π ]. SOLUTION In general, 1 + tan2 θ = sec2 θ , so tan θ = ± sec2 θ − 1. With θ = sec−1 x, it follows that sec θ = x, so tan θ = ± x 2 − 1. Finally, if x ≥ 1 then θ = sec−1 x ∈ [0, π /2) so tan θ is positive; on the other hand, if x ≤ 1 then θ = sec−1 x ∈ (−π /2, 0] so tan θ is negative. 53. Let f (x) = tan−1 x. Compute f (x) and f (x), and determine 1 the increasing/decreasing and concavity behavior of . f (x). Use Exercise 51 to verify the formula (sec−1 x) = |x| x 2 − 1 d 1 > 0 for all x, so tan−1 x is always increasing. Moreover, d 2 tan−1 x = −2x , so SOLUTION d x tan−1 x = 2 2 2 2 1+x

tan−1 x is concave down for x > 0 and is concave up for x < 0.

dx

(1+x )

Sketch the graph of g(x) = x − tan−1 x by determining the increasing/decreasing behavior and concavity as in Section 4.5.

S E C T I O N 7.8


381

55. Find the minimum value of y = tan−1 (x 2 − x). SOLUTION

Let f (x) = tan−1 (x 2 − x). Then f (x) =

2x − 1 . (x 2 − x)2 + 1

Hence, f (x) is decreasing for all x < 12 , is increasing for all x > 12 and has an absolute minimum at x = 12 . The minimum value is 1 1 f = tan−1 − ≈ −0.244979. 2 4 57. Use the Linear Approximation to estimate Approximate the critical points of g(x) = x arccos x and estimate the maximum value of g(x). tan−1 (1.05) − tan−1 1 d 1 −1 −1 1 ≈ SOLUTION tan (1.05) − tan (0.05) ≈ (0.05) = 0.025. tan−1 x dx 2 x=1 Find the linearization of y = sin−1 x at x = 12 . In Exercises 59–62, evaluate the limit using L’Hôpital’s Rule if necessary. sin−1 x x x→0

59. lim

SOLUTION

sin−1 x lim = lim x x→0 x→0

√1

1−x 2

1

= 1.

tan−1 x−1− π4 tan x tan π4 x−1− 1 x→1lim x→0 sin x 1 1 tan−1 x − π4 1+x 2 2 = 1. SOLUTION lim = lim π = π π x→1 tan(π x/4) − 1 x→1 4 sec2 (π x/4) 2

61. lim

In Exercises 63–66, −1 calculate the definite integral. lim ln x tan x x→0 " 1/2 1 63. dx 0 1 − x2 1/2 " 1/2 1 1 π SOLUTION d x = sin−1 x = sin−1 − sin−1 0 = . 2 6 0 1 − x2 0 " √3/2 " 2 1 d x dx 65. 2 1/2 1 − x 2 x + 4 0 √ √3/2 " √3/2 1 3 1 π π π − sin−1 = − = . SOLUTION d x = sin−1 x = sin−1 2 2 2 3 6 6 1/2 1−x 1/2 67. Use"the √ substitution u = x/3 to prove 2 dx √ 2/ 3 x x 2 − 1 SOLUTION

"

dx 1 x = tan−1 + C 3 3 9 + x2

Let u = x/3. Then, x = 3u, d x = 3 du, 9 + x 2 = 9(1 + u 2 ), and " " " 1 1 1 x du dx 3 du = = = tan−1 u + C = tan−1 + C. 3 3 3 3 9 + x2 9(1 + u 2 ) 1 + u2

" In Exercises 69–80, calculate the indefinite integral. dx . " Use the substitution u = 2x to evaluate 2 4x + 1 dt 69. 16 − t 2 Let t = 4u. Then dt = 4 du, and " " " " t 4 du 4 du du dt −1 −1 = = = = sin u + C = sin + C. 2 2 2 2 4 16 − (4u) 16 − t 4 1−u 1−u

SOLUTION

"

dt 2

382

CHAPTER 7


" 71.

dt 25 − 4t 2

SOLUTION

Let t = (5/2)u. Then dt = (5/2) du, and " " " " dt (5/2)du 5/2 du

= = du = 2 2 5 2 25 − 4t 25 − 25u 2 1 − u2 25 − 4( 2 u) 1 1 2t = sin−1 u + C = sin−1 + C. 2 2 5

"

" dx dx 1 −4x 2 2 x 1 − 4x SOLUTION Let u = 2x. Then du = 2 d x, and " " 1 1 dx du = = sin−1 u + C = sin−1 (2x) + C. 2 2 2 2 1 − 4x 2 1−u

73.

"

(x " + 1)d x dx 1 − x 22 4+x SOLUTION Observe that

75.

"

(x + 1) d x = 1 − x2

"

x dx + 1 − x2

"

dx . 1 − x2

In the first integral on the right, we let u = 1 − x 2 , du = −2x d x. Thus " " " 1 1 dx (x + 1) d x du =− + = − 1 − x 2 + sin−1 x + C. 1/2 2 2 2 u 1−x 1−x "

"e x d x dx 2x 1 + e x x4 − 1 SOLUTION Let u = e x . Then du = e x d x, and

77.

"

ex = 1 + e2x

"

du = tan−1 u + C = tan−1 e x + C. 1 + u2

"

tan " −1 x d x −1 ln(cos x) d x 1 + x2 (cos−1 x) 1 − x 2 SOLUTION Let u = tan−1 x. Then du =

79.

dx , and 1 + x2 " " 1 2 (tan−1 x)2 tan−1 x d x = u du = + C = u + C. 2 2 1 + x2

81. Use"Figure 11 to prove the formula dx " x 1 1 (tan−1 x)(1 + x 2 ) 1 − t 2 dt = x 1 − x 2 + sin−1 x 2 2 0 Hint: The area represented by the integral is the sum of a triangle and a sector. y

x

FIGURE 11

1

x

S E C T I O N 7.8

SOLUTION

The definite integral


383

" x 1 − t 2 dt represents the area of the region under the upper half of the unit circle 0

from 0 to x. The region consists of a sector of the circle and a right triangle. The sector has a central angle of π2 − θ , where cos θ = x. Hence, the sector has an area of 1 1 2 π (1) − cos−1 x = sin−1 x. 2 2 2 The right triangle has a base of length x, a height of 1 − x 2 , and hence an area of 12 x 1 − x 2 . Thus, " x 1 1 1 − t 2 dt = x 1 − x 2 + sin−1 x. 2 2 0

83. A painting of length b is located at a height h above eye level (Figure 12). Find the distance x at which the viewing − t 2 + t with sin−1Exercise t is an antiderivative of 4.6; sin−1solve t. it this time using inverse trigonometric that G(t) (this = 1coincides is maximized 47 in Section angle θShow functions).

C b B q

P

Q

h

x

FIGURE 12 SOLUTION

From the figure, we see that

θ (x) = tan−1

b+h h − tan−1 . x x

Thus, h b+h θ (x) = 2 − 2 . 2 x +h x + (b + h)2 Setting θ (x) = 0 yields b+h h = 2 , x 2 + h2 x + (b + h)2 which simplifies to x =

bh + h 2 .

UseInsights the result and of Exercise 83 to show that θ Further Challenges

is maximized at the value of x for which the angles Q P B and

QC P are equal. 85. A cylindrical tank of radius R and length L lying horizontally as in Figure 13 is filled with oil to height h. (a) Show that the volume V (h) of oil in the tank as a function of height h is h − (R − h) 2h R − h 2 V (h) = L R 2 cos−1 1 − R dV = 2L h(2R − h). dh (c) Suppose that R = 4 ft and L = 30 ft, and that the tank is filled at a constant rate of 10 ft3 /min. How fast is the height h increasing when h = 5?

(b) Show that

R L h

FIGURE 13 Oil in the tank has level h.

384

CHAPTER 7


(a) From Figure 13, we see that the volume of oil in the tank, V (h), is equal to L times A(h), the area of that portion of the circular cross section occupied by the oil. Now, A(h) = area of sector − area of triangle =

R2 θ R 2 sin θ − , 2 2

where θ is the central angle of the sector. Referring to the diagram below, θ θ 2h R − h 2 R−h cos = and sin = . 2 R 2 R /2

R

R−h

2hR − h2

Thus,

θ

= 2 cos−1

sin θ = 2 sin

and

h 1− R

,

θ θ (R − h) 2h R − h 2 , cos = 2 2 2 R2

h V (h) = L R 2 cos−1 1 − − (R − h) 2h R − h 2 . R

(b) Recalling that ddx cos−1 u = − √ 1

du , 1−x 2 d x

d

h R 2 cos−1 1 − − (R − h) 2h R − h 2 R dh −1 (R − h)2 L −R + 2h R − h 2 − 1 − (1 − (h/R))2 2h R − h 2 R 2 − 2Rh + h 2 R2 2 + 2h R − h − L 2h R − h 2 2h R − h 2 R 2 + (2h R − h 2 ) − (R 2 − 2Rh + h 2 ) L 2h R − h 2 2(2h R − h 2 ) 4h R − 2h 2 =L = 2L 2h R − h 2 . L 2 2 2h R − h 2h R − h

dV =L dh = = = =

(c)

d dh

dV d V dh dh 1 dV = , so = . From part (b) with R = 4, L = 30 and h = 5, dt dh dt dt d V /dh dt

√ dV = 2(30) 2(5)(4) − 52 = 60 15 ft2 . dh

Thus, √ 1 15 dh = √ (10) = ≈ 0.043 ft/min. dt 90 60 15 87. Tom drives with his friend Ali along a highway " represented by the graph of a differentiable function y = f (x) dx as in Figure 14.substitution During the utrip, Alixviews a billboard represented by Show the segment BC along the y-axis. Let Q be the . Hint: that Use the = tan to evaluate 2x 1 + y-intercept of the tangent line to y = f (x). Show that θ issin maximized at the value of x for which the angles Q P B and QC P are equal. This is a generalization of Exercise 84 [which corresponds to the case f (x) = 0]. Hints: dx du = (a) Compute d θ /d x and check that it equals 2 1 + 2u 2 1 + sin x (b − c) ·

(x 2 + (x f (x))2 ) − (b − ( f (x) − x f (x)))(c − ( f (x) − x f (x))) (x 2 + (b − f (x))2 )(x 2 + (c − f (x))2 )

S E C T I O N 7.8


385

(b) Show that the y-coordinate of Q is f (x) − x f (x). dθ = 0 is equivalent to (c) Show that the condition dx P Q2 = B Q · C Q (d) Use (c) to conclude that triangles QPB and QCP are similar. y C = (0, c)

billboard

B = (0, b)

y = f (x) P = (x, f(x))

x x highway

FIGURE 14 SOLUTION

(a) From the figure, we see that

θ (x) = tan−1

c − f (x) b − f (x) − tan−1 . x x

Then

θ (x) =

b − ( f (x) − x f (x)) c − ( f (x) − x f (x)) − x 2 + (b − f (x))2 x 2 + (c − f (x))2

= (b − c)

x 2 − bc + (b + c)( f (x) − x f (x)) − ( f (x))2 + 2x f (x) f (x) (x 2 + (b − f (x))2 )(x 2 + (c − f (x))2 )

= (b − c)

(x 2 + (x f (x))2 − (bc − (b + c)( f (x) − x f (x)) + ( f (x) − x f (x))2 ) (x 2 + (b − f (x))2 )(x 2 + (c − f (x))2 )

= (b − c)

(x 2 + (x f (x))2 − (b − ( f (x) − x f (x)))(c − ( f (x) − x f (x))) . (x 2 + (b − f (x))2 )(x 2 + (c − f (x))2 )

(b) The point Q is the y-intercept of the line tangent to the graph of f (x) at point P. The equation of this tangent line is Y − f (x) = f (x)(X − x). The y-coordinate of Q is then f (x) − x f (x). (c) From the figure, we see that B Q = b − ( f (x) − x f (x)), C Q = c − ( f (x) − x f (x)) and PQ =

x 2 + ( f (x) − ( f (x) − x f (x)))2 =

Comparing these expressions with the numerator of d θ /d x, it follows that

x 2 + (x f (x))2 .

dθ = 0 is equivalent to dx

P Q 2 = B Q · C Q. (d) The equation P Q 2 = B Q · C Q is equivalent to PQ CQ = . BQ PQ In other words, the sides C Q and P Q from the triangle QC P are proportional in length to the sides P Q and B Q from the triangle Q P B. As P Q B = C Q P, it follows that triangles QC P and Q P B are similar.

386

CHAPTER 7


7.9 Hyperbolic Functions Preliminary Questions 1. Which hyperbolic functions take on only positive values? SOLUTION

cosh x and sech x take on only positive values.

2. Which hyperbolic functions are increasing on their domains? SOLUTION

sinh x and tanh x are increasing on their domains.

3. Describe three properties of hyperbolic functions that have trigonometric analogs. SOLUTION

Hyperbolic functions have the following analogs with trigonometric functions: parity, identities and deriva-

tive formulas. 4. Which hyperbolic derivative formulas differ from their trigonometric counterparts by a minus sign? SOLUTION

The derivatives of cosh x and sech x differ from their trigonometric counterparts by a minus sign.

Exercises 1. Use a calculator to compute sinh x and cosh x for x = −3, 0, 5. SOLUTION

x

−3

0

5

sinh x =

e x − e−x 2

e−3 − e3 = −10.0179 2

e0 − e0 =0 2

e5 − e−5 = 74.203 2

cosh x =

e x + e−x 2

e−3 + e3 = 10.0677 2

e0 + e0 =1 2

e5 + e−5 = 74.210 2

3. For which values of x are y = sinh x and y = cosh x increasing and decreasing? Compute sinh(ln 5) and tanh(3 ln 5) without using a calculator. d SOLUTION sinh x = cosh x. Since e x and e−x are positive for all x, it follows that cosh x = 12 (e x + e−x ) > 0 dx d cosh x = sinh x. Since e x > e−x for x > 0 while for all x. Thus, sinh x is increasing for all x. On the other hand, dx e x < e−x for x < 0, it follows that sinh x > 0 for x > 0 but sinh x < 0 for x < 0. Thus, cosh x is decreasing for x < 0 and is increasing for x > 0. In Exercises derivative. Show5–30, that ycalculate = tanh xthe is an odd function. 5. y = sinh(3x) SOLUTION

d sinh(3x) = 3 cosh(3x). dx

7. y = cosh(1 − 4t) y = sinh(x 2 ) d SOLUTION cosh(1 − 4t) = −4 sinh(1 − 4t). dt √ 1 9. y = cosh x + y = tanh(t 2 + 1) √ 1 d SOLUTION cosh x + 1 = (cosh x + 1)−1/2 sinh x. dx 2 sinh t 11. y =y = sinh x tanh x 1 + cosh t 1 + cosh t 1 d sinh t cosh t (1 + cosh t) − sinh t (sinh t) SOLUTION = = = . 2 2 dt 1 + cosh t 1 + cosh t (1 + cosh t) (1 + cosh t) 13. y = tanh(2x) y = ln(cosh x) d SOLUTION tanh(2x) = 2 sech2 (2x). dx tanh x 15. y =y e= sinh(ln x) d tanh x SOLUTION = sech2 x · etanh x . e dx

17. y = sinh(xe x )x y = tanh(e )

S E C T I O N 7.9

Hyperbolic Functions

d sinh(xe x ) = (e x + xe x ) cosh(xe x ). dx √ 19. y = sech( x) y = sinh(cosh x) √ √ √ 1 d SOLUTION sech( x) = − x −1/2 sech x tanh x. dx 2 21. y = ln(coth x) y = coth(x 2 ) −1 d − csch2 x −1 = SOLUTION ln(coth x) = = . dx coth x sinh x cosh x sinh2 x( cosh x ) SOLUTION

sinh x

23. y = tanh(3x 2 − 9)

y = sech x coth x d SOLUTION tanh(3x 2 − 9) = 6x sech2 (3x 2 − 9). dx

25. y = tanh−1 3x y = x sinh x 3 d SOLUTION . tanh−1 (3x) = dx 1 − 9x 2 3x)4 27. y = (csch−1−1 y = sinh (x 2 ) −1 −4(csch−1 3x)3 d −1 4 −1 3 (3) = SOLUTION . (csch 3x) = 4(csch 3x) dx |3x| 1 + 9x 2 |x| 1 + 9x 2 29. y = sinh−1 ( −1xx2 + 1) cosh y=e 1 d 1 x −1 2 SOLUTION sinh ( x + 1) = (2x) = . 2 2 2 dx x +1+1 2 x +1 x + 2 · x2 + 1 In Exercises 31–42,−1 calculate the integral. y = ln(tanh x) " 31. cosh(3x) d x " 1 cosh(3x) d x = sinh 3x + C. SOLUTION 3 " " 33. x sinh(x 2 + 1) d x sinh(x + 1) d x " 1 x sinh(x 2 + 1) d x = cosh(x 2 + 1) + C. SOLUTION 2 " " 35. sech2 (12− 2x) d x sinh x cosh x d x " 1 SOLUTION sech2 (1 − 2x) d x = − tanh(1 − 2x) + C. 2 " " 37. tanh x sech2 x d x tanh(3x) sech(3x) d x SOLUTION

Let u = tanh x. Then du = sech2 x d x nd "

" tanh x sech2 x d x =

u du =

1 2 tanh2 x u +C = + C. 2 2

"

" tanh x cosh dx x dx 3 sinh " x +4 tanh x d x = ln cosh x + C. SOLUTION

39.

"

" sinhxx d x e−xcosh dx sinh x e x −e−x we can combine the two functions to get SOLUTION Since sinh x = 2 " " " 1 1 1

1 e−x (e x − e−x ) d x = 1 − e−2x d x = x + e−2x + C. e−x sinh x d x = 2 2 2 4

41.

d " (coth x) = − csch2 x. 43. Verify the coshformula x dx d x sinh2 x

387

388

CHAPTER 7


SOLUTION

−1 d cosh x sinh2 x − cosh2 x d = = − csch2 x. coth x = = 2 dx d x sinh x sinh x sinh2 x

Refercosh(ln to the x) graphs to explain 45. Simplify + sinh(ln x). why the equation sinh x = t has a unique solution for every t and cosh x = t has two solutions for every t > 1. SOLUTION

From its graph we see that sinh x is a one-to-one function with

lim sinh x = −∞ and lim sinh x =

x→−∞

x→∞

∞. Thus, for every real number t, the equation sinh x = t has a unique solution. On the other hand, from its graph, we see that cosh x is not one-to-one. Rather, it is an even function with a minimum value of cosh 0 = 1. Thus, for every t > 1, the equation cosh x = t has two solutions: one positive, the other negative. 47. Prove the addition formula for cosh x. Compute cosh x and tanh x, assuming that sinh x = 0.8. SOLUTION

cosh(x + y) =

e x+y + e−(x+y) 2e x+y + 2e−(x+y) = 2 4

e x+y + e−x+y + e x−y + e−(x+y) e x+y − e−x+y − e x−y + e−(x+y) + 4 4 x y x y −x −y −x −y e +e e +e e −e e −e = + 2 2 2 2 =

= cosh x cosh y + sinh x sinh y. In Exercises 49–52, proveformulas the formula. Use the addition to prove 49. cosh(sinh−1 t) = t 2 + 1 SOLUTION

sinh(2x) = 2 cosh x sinh x

Note

cosh(2x) = cosh2 x + sinh2 x d 1 t d = = 1 + t 2, cosh(sinh−1 t) = sinh(sinh−1 t) dt dt 1 + t2 1 + t2 so the functions cosh(sinh−1 t) and 1 + t 2 differ by a constant; substituting t = 0 we find that the constant is 0. Therefore, cosh(sinh−1 t) = t 2 + 1. 51.

d 1 sinh−1 t = −1 2 dt sinh(cosh t)t 2=+ 1t − 1 for t ≥ 1

SOLUTION

Let x = sinh−1 t. Then t = sinh x and 1 = cosh x

dx dt

or

dx 1 = . dt cosh x

Thus, 1 1 d = sinh−1 t = −1 2 dt cosh(sinh t) t +1 by Exercise 49. In Exercises calculate 1 the integral in terms of inverse hyperbolic functions. d 53–60, −1 for t > 1 " 4 dt cosh t = 2 t −1 dx 53. 2 x2 − 1 4 " 4 dx SOLUTION = cosh−1 x = cosh−1 4 − cosh−1 2. 2 x2 − 1 2 " " dx 55. dx 9 + x22 x" − 4 " x dt dx SOLUTION = = sinh−1 + C. 2 2 3 3 1 + (x/3) 9+x " 1/2 " dx dx 57. − x 2 1/3 1 1 + 9x 2

S E C T I O N 7.9


389

1/2 dx 1 1 −1 x = tanh = tanh−1 − tanh−1 . 2 2 3 1/3 1 − x 1/3

" 1/2 SOLUTION

59.

" 10 " 1 dx dx 2−1 2 4x 0 "1 + x 2

10 dx 1 −1 (2x) = 1 (coth−1 4 − coth−1 20). coth = − 2 2 2 2 4x − 1 2 " −1 61. Prove that sinh−1 t = ln(t + t 2 + 1). Hint: Let t = sinh x. Prove that cosh x = t 2 + 1 and use the relation dx sinh x + cosh x = e x . −3 x x 2 + 16 10

SOLUTION

SOLUTION

Let t = sinh x. Then cosh x =

1 + sinh2 x = 1 + t 2 .

Moreover, because sinh x + cosh x =

e x − e−x e x + e−x + = ex , 2 2

it follows that sinh−1 t = x = ln(sinh x + cosh x) = ln(t +

t 2 + 1).

1 1 + t −1 t−1 = 63. Prove that tanh ln 2 for |t| < 1. Prove that cosh t 2= ln(t1+ − t t − 1) for t > 1. SOLUTION

Let A = tanh−1 t. Then t = tanh A =

sinh A e A − e−A . = A cosh A e + e−A

Solving for A yields A=

1 t +1 ln ; 2 1−t

hence, tanh−1 t =

1 t +1 ln . 2 1−t

65. An (imaginary) train moves along a track at velocity u, and a woman walks down the aisle of the train with velocity Use the substitution u = sinh x to prove v in the direction of the train’s motion. Compute the velocity w of the woman relative to the ground using the laws of " both Galileo and Einstein in the following cases. sech x d x = tan−1 (sinh x) + C (a) u = 1,000 mph and v = 50 mph. Is your calculator accurate enough to detect the difference between the two laws? (b) u = 100,000 miles/sec and v = 50 miles/sec. SOLUTION

(a) By Galileo’s law, 1,000 + 50 = 1,050 mph. The speed of light is c = 186,000 miles/sec = 669,600,000 mph. Using Einstein’s law and a calculator, tanh−1

w 1,000 50 = tanh−1 + tanh−1 = 1.568100358 × 10−6 ; c 669,600,000 669,600,000

so w = 1,050 mph. No, the calculator was not accurate enough to detect the difference between the two laws. (b) By Galileo’s law, 100,000 + 50 = 100,050 miles/sec. By Einstein’s law, tanh−1 so w ≈ 100,035.5 miles/sec.

w 100,000 50 = tanh−1 + tanh−1 = 0.601091074, c 186,000 186,000

390

CHAPTER 7


Further Insights and Challenges 67. (a) Use the addition formulas for sinh x and cosh x to prove Show that the linearization of the function y = tanh−1 x at x = 0 is tanh−1 x ≈ x. Use this to explain the tanh[Eq. u + (2)] tanhreduces v following assertion: Einstein’s Law oftanh(u Velocity Addition to Galileo’s Law if the velocities are + v) = 1 + tanh u tanh v small relative to the speed of light. (b) Use (a) to show that Einstein’s Law of Velocity Addition [Eq. (2)] is equivalent to w=

u+v uv 1+ 2 c

SOLUTION

(a) tanh(u + v) = =

sinh(u + v) sinh u cosh v + cosh u sinh v = cosh(u + v) cosh u cosh v + sinh u sinh v tanh u + tanh v sinh u cosh v + cosh u sinh v 1/(cosh u cosh v) · = cosh u cosh v + sinh u sinh v 1/(cosh u cosh v) 1 + tanh u tanh v

(b) Einstein’s law states: tanh−1 (w/c) = tanh−1 (u/c) + tanh−1 (v/c). Thus

w tanh(tanh−1 (v/c)) + tanh(tanh−1 (u/c)) = tanh tanh−1 (u/c) + tanh−1 (v/c) = c 1 + tanh(tanh−1 (v/c)) tanh(tanh−1 (u/c)) v + u (1/c)(u + v) = c v cu = . 1+ c c 1 + uv2 c

Hence, w=

u+v . 1 + uv2 c

dy " = 1 − y 2 with initial condition y(0) = 0. 69. (a) Show that y a= tanh t satisfies the differential equation dt Prove that cosh x sinh x d x = 0 for all a. −a (b) Show that for arbitrary constants A, B, the function y = A tanh(Bt) satisfies B dy = AB − y 2 , dt A

y(0) = 0

(c) Let v(t) be the velocity of a falling object of mass m. For large velocities, air resistance is proportional to the square of velocity v(t)2 . If we choose coordinates so that v(t) > 0 for a falling object, then by Newton’s Law of Motion, there is a constant k > 0 such that k dv = g − v2 dt m √ √ Solve for v(t) by applying the result of (b) with A = gm/k and B = gk/m. (d) Calculate the terminal velocity lim v(t). t→∞

(e) Find k if m = 150 lb and the terminal velocity is 100 mph. SOLUTION

(a) First, note that if we divide the identity cosh2 t − sinh2 t = 1 by cosh2 t, we obtain the identity 1 − tanh2 t = sech2 t. Now, let y = tanh t. Then dy = sech2 t = 1 − tanh2 t = 1 − y 2 . dt Furthermore, y(0) = tanh 0 = 0. (b) Let y = A tanh(Bt). Then

By 2 y2 dy 2 2 = AB sech (Bt) = AB(1 − tanh (Bt)) = AB 1 − 2 = AB − . dt A A

Furthermore, y(0) = A tanh(0) = 0.


S E C T I O N 7.9

391

(c) Matching the differential equation dv k = g − v2 dt m with the template dv B = AB − v 2 dt A from part (b) yields AB = g

and

B k = . A m

and

B=

Solving for A and B gives A=

mg k

Thus v(t) = A tanh(Bt) = (d) lim v(t) = t→∞

mg lim tanh k t→∞

kg mt

=

mg tanh k

kg . m

kg t . m

mg k

(e) Substitute m = 150 lb and g = 32 ft/sec2 = 78545.5 miles/hr2 into the equation for the terminal velocity obtained in part (d) and then solve for k. This gives k=

150(78545.5) = 1178.18 lb/mile. 1002

In Exercises 70–72, a flexible chain of length L is suspended between two poles of equal height separated by a distance x + C. 2M (Figure 9). By Newton’s laws, the chain describes a curve (called a catenary) with equation y = a cosh a M The constant C is arbitrary and a is the number such that L = 2a sinh . The sag s is the vertical distance from the a highest to the lowest point on the chain. y = a cosh(x/a)

2M

FIGURE 9 Chain hanging between two poles describes the curve y = a cosh(x/a).

M − a. to find an approximate value of a Let MSuppose be a fixed constant. Show that M the=sag50. is Experiment given by s =with a cosh that L = 120 ft and your calculator a M (for greater accuracy, use Newton’s method or a computer algebra system). satisfying ds L = 2a sinh . Calculate a da da M Calculate by implicit differentiation using the relation L = 2a sinh . dL a Use (a) and (b) and the Chain Rule to show that

71. (a) (b) (c)

ds da cosh(M/a) − (M/a) sinh(M/a) − 1 ds = = dL da d L 2 sinh(M/a) − (2M/a) cosh(M/a) SOLUTION

The sag in the curve is s = y(M) − y(0) = a cosh

(a)

6

ds = cosh da

M a

−

M sinh a

M a

−1

M a

+ C − (a cosh 0 + C) = a cosh

M a

− a.

392

CHAPTER 7


(b) If we differentiate the relation L = 2a sinh 0=2

M a

da sinh dL

with respect to a, we find

M a

−

2M da cosh a dL

M a

.

Solving for da/d L yields −1 M M da 2M = 2 sinh cosh . − dL a a a (c) By the Chain Rule, ds da ds = · . dL da d L The formula for ds/d L follows upon substituting the results from parts (a) and (b). 73. Prove that every function f (x) is the sum of an even function f + (x) and an odd function f − (x). [Hint: f ± (x) = that M = 50 and L = 160. In this case, a CAS can be used to show that a ≈ 28.46. 1 ( f (x)Assume ± f (−x)).] Express f (x) = 5e x + 8e−x in terms of cosh x and sinh x. 2 (a) Use Eq. (6) and the Linear Approximation to estimate the increase in sag if L is increased from L = 160 to (x)+ f (−x) SOLUTION Let from f + (x)L== f160 and f − (x) = f (x)−2f (−x) . Then f + + f − = 2 f2(x) = f (x). Moreover, L = 161 and to 2 L = 165. (b) If you have a CAS, compute s(161) − s(160) and s(165) − s(160) directly and compare with your estimates f (−x) + f (−(−x)) f (−x) + f (x) in (a). f + (−x) = = = f + (x), 2 2 so f + (x) is an even function, while f − (−x) =

f (−x) − f (−(−x)) f (−x) − f (x) ( f (x) − f (−x)) = =− = − f − (x), 2 2 2

so f − (x) is an odd function. For f (x) = 5e x + 8e−x , we have f + (x) =

5e x + 8e−x + 5e−x + 8e x = 8 cosh x + 5 cosh x = 13 cosh x 2

f − (x) =

5e x + 8e−x − 5e−x − 8e x = 5 sinh x − 8 sinh x = −3 sinh x. 2

and

Therefore, f (x) = f + (x) + f − (x) = 13 cosh x − 3 sinh x. 75. In the Excursion, we discussed the relations Use the method of the previous problem to express cosh(it) = cos t and −3xsinh(it)3x= i sin t f (x) = 7e + 4e Use these relations to show that the identity cos2 t + sin2 t = 1 results from the identity cosh2 x − sinh2 x = 1 by setting in terms of sinh(3x) and cosh(3x). x = it. SOLUTION

Substituting x = it into cosh2 x − sinh2 x = 1 yields cosh2 (it) − sinh2 (it) = 1. Since cosh2 (it) = cos2 t

and sinh2 (it) = (i sin t)2 = − sin2 t, it follows that cos2 t + sin2 t = 1.

CHAPTER REVIEW EXERCISES 1. Match each quantity (a)–(d) with (i), (ii), or (iii) if possible, or state that no match exists. 2a (b) b (a) 2a 3b 3 (c) (2a )b (d) 2a−b 3b−a (i) 2ab

(ii) 6a+b

(iii)

2 a−b 3

SOLUTION

(a) No match. (b) No match. (c) (i): (2a )b = 2ab . (d) (iii): 2a−b 3b−a = 2a−b

a−b a−b 2 1 = . 3 3

Match each quantity (a)–(d) with (i), (ii), or (iii) if possible, or state that no match exists.

a ln a


3. Which of the following is equal to

d x 2 ? dx

(a) 2x

(b) (ln 2)2x 1 x (d) 2 ln 2

(c) x2x−1 SOLUTION

393

The derivative of f (x) = 2x is d x 2 = 2x ln 2. dx

Hence, the correct answer is (b). x −2 5. Find the the inverse of fof(x)f (x) = = x 3and determine its domain and range. Find inverse x − 1 − 8 and determine its domain and range. SOLUTION

x−2 To find the inverse of f (x) = x−2 x−1 , we solve y = x−1 for x as follows:

x − 2 = y(x − 1) = yx − y x − yx = 2 − y x=

2− y . 1− y

Therefore, f −1 (x) =

2−x x −2 = . 1−x x −1

The domain of f −1 is the range of f , namely {x : x = 1}; the range of f −1 is the domain of f , namely {y : y = 1}. x is equal to its inverse on the domain {x : x = −1}. 7. Show that g(x) = 2 Find a domain on which x− 1 h(t) = (t − 3) is one-to-one and determine the inverse on this domain. SOLUTION

x is equal to its inverse, we need to show that for x = 1, To show that g(x) = x−1

g (g(x)) = x. First, we notice that for x = 1, g(x) = 1. Therefore, x x x x g (g(x)) = g = xx−1 = = = x. x −1 − 1 x − (x − 1) 1 x−1 9. Suppose that g(x) is the inverse of f (x). Match the functions (a)–(d) with their inverses (i)–(iv). Describe the graphical interpretation of the relation g (x) = 1/ f (g(x)), where f (x) and g(x) are inverses of (a) f (x) + 1 each other. (b) f (x + 1) (c) 4 f (x) (d) f (4x) (i) (ii) (iii) (iv)

g(x)/4 g(x/4) g(x − 1) g(x) − 1

SOLUTION

(a) (iii): f (x) + 1 and g(x − 1) are inverse functions: f (g(x − 1)) + 1 = (x − 1) + 1 = x; g( f (x) + 1 − 1) = g( f (x)) = x. (b) (iv): f (x + 1) and g(x) − 1 are inverse functions: f (g(x) − 1 + 1) = f (g(x)) = x; g( f (x + 1)) − 1 = (x + 1) − 1 = x. (c) (ii): 4 f (x) and g(x/4) are inverse functions: 4 f (g(x/4)) = 4(x/4) = x; g(4 f (x)/4) = g( f (x)) = x.

394

CHAPTER 7


(d) (i): f (4x) and g(x)/4 are inverse functions: f (4 · g(x)/4) = f (g(x)) = x; 1 1 g( f (4x)) = (4x) = x. 4 4 In Exercises find theg(x) derivative. is the inverse of a differentiable function f (x) such that f (−1) = 8 and f (−1) = 12. Find11–32, g (8) where 11. f (x) = 9e−4x d −4x SOLUTION = −36e−4x . 9e dx

e−x 13. f (x) = = ln(4x 2 + 1) f (x) x e−x (x + 1) d e−x −xe−x − e−x SOLUTION = − . = dx x x2 x2 15. G(s) = (ln(s))2 f (x) = ln(x + e x ) 2 ln s d SOLUTION (ln s)2 = . ds s 2

17. g(t) = e4t−t 2 G(s) = ln(s ) 2 d 4t−t 2 SOLUTION = (4 − 2t)e4t−t . e dt 19. f (θ ) = ln(sin θ ) g (t) =d t 2 e1/t cos θ SOLUTION ln(sin θ ) = = cot θ . dθ sin θ x+ln x 21. f (x)f (= θ )e= sin(ln θ ) SOLUTION

1 x+ln x d x+ln x = 1+ . e e dx x

23. h(y) = 21−y 2 f (x) = esin x d 1−y SOLUTION = −21−y ln 2. 2 dy 25. G(s) = cos−1 (s −1 ) 1 + ey h (y) = −1 d 1 −−1 1 1 y SOLUTION . cos e (s −1 ) = − =

2 2 4 ds s s − s2 1 − 1s x) 27. f (x) = ln(csc−1 √ G(s) = tan−1 ( s) 1 d SOLUTION ln(csc−1 x) = − . 2 dx |x| x − 1 csc−1 x 29. g(t) = sinh(t 2 )−1 f (x) = esec x d SOLUTION sinh(t 2 ) = 2t cosh(t 2 ). dt −1 (e x ) 31. g(x) = tanh h(y) = y tanh(4y) 1 ex d SOLUTION ex = . tanh−1 (e x ) = x 2 dx 1 − (e ) 1 − e2x

f (g(x)) = e x 2 , where g(1) = 2 and g (1) = 4. Find f (2). 33. Suppose that g(t) = t 2 − 1 sinh−1 t 2 SOLUTION We differentiate both sides of the equation f (g(x)) = e x to obtain, f (g(x)) g (x) = 2xe x . 2

Setting x = 1 yields f (g(1)) g (1) = 2e. Since g(1) = 2 and g (1) = 4, we find f (2) · 4 = 2e,


395

or f (2) =

e . 2

35. Find the points of inflection of f (x) =2xln(x 2 + 1) and determine whether the concavity changes from up to down e +1 or vice versa. Find the local extrema of f (x) = x+1 . e SOLUTION With f (x) = ln(x 2 + 1), we find 2x ; and f (x) = 2 x +1

2 x 2 + 1 − 2x · 2x 2(1 − x 2 ) = f (x) = 2 2 (x 2 + 1) (x 2 + 1) Thus, f (x) > 0 for −1 < x < 1, whereas f (x) < 0 for x < −1 and for x > 1. It follows that there are points of inflection at x = ±1, and that the concavity of f changes from down to up at x = −1 and from up to down at x = 1. In Exercises 36–38, let f (x) = xe−x . 37. Show that f (x) has an inverse on [1, ∞). Let g(x) be this inverse. Find the domain and range of g(x) and compute g (2e−2 ). Plot f (x) and use the zoom feature to find two solutions of f (x) = 0.3. Let f (x) = xe−x . Then f (x) = e−x (1 − x). On [1, ∞), f (x) < 0, so f (x) is decreasing and therefore one-to-one. It follows that f (x) has an inverse on [1, ∞). Let g(x) denote this inverse. Because f (1) = e−1 and f (x) → 0 as x → ∞, the domain of g(x) is (0, e−1 ], and the range is [1, ∞). To determine g (2e−2 ), we use the formula g (x) = 1/ f (g(x)). Because f (2) = 2e−2 , it follows that g(2e−2 ) = 2. Then, SOLUTION

g (2e−2 ) =

1 1 1 = = −e2 . = f (2) f (g(2e−2 )) −e−2

In Exercises 39–42, find the local extrema and points of inflection, and sketch the graph over the interval specified. Use Show that f (x) = c has two solutions if 0 < c < e−1 . L’Hôpital’s Rule to determine the limits as x → 0+ or x → ±∞ if necessary. 39. y = x ln x, SOLUTION

x >0

Let y = x ln x. Then y = ln x + x

1 = 1 + ln x, x

and y = 1x . Solving y = 0 yields the critical point x = e−1 . Since y (e−1 ) = e > 0, the function has a local minimum at x = e−1 . y is positive for x > 0, hence the function is concave up for x > 0 and there are no points of inflection. As x → 0+ and as x → ∞, we find ln x x −1 = lim = lim (−x) = 0; x→0+ x −1 x→0+ −x −2 x→0+

lim x ln x = lim

x→0+

lim x ln x = ∞.

x→∞

The graph is shown below:

6 4 2

1

41. y = x(log x)22, x > 0 y = xe−x /2

2

3

4

396

CHAPTER 7


Let y = x(log x)2 . Then y = (log x)2 + x ·

and 2 log x + y = x ln 10

2 log x 2 = (log x) + log x , x ln 10 ln 10

2 1 2 1 · = log x + . ln 10 x ln 10 x ln 10 ln 10

Solving y = 0 yields the critical points x = 1 and x = e−2 . Because y (1) =

2 > 0 and (ln 10)2

y (e−2 ) = −

2(log e)2 < 0, e−2

we conclude that the function has a local minimum at x = 1 and a local maximum at x = e−2 . We see that y > 0 for x > e−1 and y < 0 for 0 < x < e−1 . Therefore, there is a point of inflection at x = e−1 . As x → 0+ and as x → ∞, we find 2 log x ln110 · 1x (log x)2 = lim x→0+ 1/x x→0+ −1/x 2

lim x(log x)2 = lim

x→0+

1

=− =

1

· 2 log x 2 lim =− lim ln 10 2x ln 10 x→0+ 1/x ln 10 x→0+ −1/x

2 lim x = 0; and (ln 10)2 x→0+

lim x(log x)2 = ∞.

x→∞

The graph is shown below:

1 0.8 0.6 0.4 0.2

0.5

1

1.5

2

2.5

3

In Exercises 43–48, use2logarithmic differentiation to find the derivative. x y = tan−13 4 (x + 1) 43. y = (4x − 2)2 SOLUTION

(x + 1)3 . Then (4x − 2)2 (x + 1)3 = ln (x + 1)3 − ln (4x − 2)2 = 3 ln(x + 1) − 2 ln(4x − 2). ln y = ln (4x − 2)2

Let y =

By logarithmic differentiation, 3 2 3 4 y = − ·4= − , y x + 1 4x − 2 x + 1 2x − 1 so y = 45. y = e(x−1) e(x−3) (x + 1)(x + 2)2 y= (x + 3)(x + 4) 2

2

(x + 1)3 (4x − 2)2

4 3 − . x + 1 2x − 1


2 2 Let y = e(x−1) e(x−3) . Then

2 2 2 2 ln y = ln e(x−1) e(x−3) = ln e(x−1) +(x−3) = (x − 1)2 + (x − 3)2 . By logarithmic differentiation, y = 2(x − 1) + 2(x − 3) = 4x − 8, y so 2 2 y = 4e(x−1) e(x−3) (x − 2).

2 e3x (xx − 2) 47. y = x e sin−1 2 y =(x + 1) ln x e3x (x − 2)2 SOLUTION Let y = . Then (x + 1)2 e3x (x − 2)2 ln y = ln = ln e3x + ln (x − 2)2 − ln (x + 1)2 (x + 1)2

= 3x + 2 ln(x − 2) − 2 ln(x + 1). By logarithmic differentiation, 2 2 y =3+ − , y x −2 x +1 so y=

e3x (x − 2)2 (x + 1)2

3+

2 2 − . x −2 x +1

√ In Exercises 49–54, use the given substitution to evaluate the integral. y = x x (x ln x ) " (ln x)2 d x , u = ln x 49. x SOLUTION

Let u = ln x. Then du = dxx , and " " u3 (ln x)3 (ln x)2 d x = u 2 du = +C = + C. x 3 3

"

" dx 2x d x , u = e−x e2x 2− 1 , u = 3 4x + 9 SOLUTION We first rewrite the integrand in terms of e−x . That is, 51.

"

1 dx = e2x − 1

"

1

dx = 2x e 1 − e−2x

"

1 dx = e x 1 − e−2x

"

e−x d x 1 − e−2x

Now, let u = e−x . Then du = −e−x d x, and " " 1 du = −sin−1 u + C = −sin−1 (e−x ) + C. dx = − e2x − 1 1 − u2 "

dt " −1 t dt , u = ln t 2 t (1 cos + (ln t) ) , u = cos−1 t 1 − t2 1 SOLUTION Let u = ln t. Then, du = t dt and " " dt du = tan−1 u + C = tan−1 (ln t) + C. = 2 1 + u2 t (1 + (ln t) ) 53.

" 55–74, calculate the integral. In Exercises dt , u = tanh t 2 cosh t + sinh2 t

397

398

CHAPTER 7


" e9−2x d x

55.

SOLUTION

" 57.

Let u = 9 − 2x. Then du = −2 d x, and " " 1 1 1 eu du = − eu + C = − e9−2x + C. e9−2x d x = − 2 2 2

" −2x ) d x e−2x2sin(e 3 x ex d x

SOLUTION

Let u = e−2x . Then du = −2e−2x d x, and " "

cos u 1 1 sin u du = e−2x sin e−2x d x = − + C = cos e−2x + C. 2 2 2

" e "ln x d x cos(ln x) d x x 1 x dx SOLUTION Let u = ln x. Then du = x and the new limits of integration are u = ln 1 = 0 and u = ln e = 1. Thus, " 1 " e 1 1 1 ln x d x u du = u 2 = . = x 2 0 2 1 0 59.

" 2/3 " ln 3 d x x−e x 1/3 1e− x 2 d x 0 2/3 " 2/3 dx 2 1 −1 SOLUTION = sin x = sin−1 − sin−1 . 2 3 3 1/3 1−x 1/3 " 1 " 12 cosh(2t)ddtx 63. 0 4 x x2 − 1 u du SOLUTION Let u = 2t. Then t = 2 and dt = 2 . The new limits of integration are u = 0 and u = 2. Thus, 61.

" 1 0

cosh(2t) dt =

2 " 1 1 2 1 1 cosh u du = sinh u = (sinh 2 − sinh 0) = sinh 2. 2 0 2 2 2 0

" 3 " x2 d x dt 0 x2 + 9 4t + 12 0 SOLUTION Let u = x 2 + 9. Then du = 2xd x, and the new limits of integration are u = 9 and u = 18. Thus, 65.

18 " " 3 1 x dx 1 1 18 1 1 18 du = ln u = (ln 18 − ln 9) = ln = ln 2. = 2 2 u 2 2 2 9 2 x + 9 0 9 9 "

" x3 d x dx 1 − x4 0 x2 + 9 SOLUTION Let u = x 2 . Then du = 2xd x, and 1 − x 4 = 1 − u 2 . Thus, " " 1 1 1 du x dx = = sin−1 u + C = sin−1 (x 2 ) + C. 4 2 2 2 2 1−x 1−u 67.

69.

" " −1 sin x d x e x 10x d x 1 − x2

SOLUTION

Let u = sin−1 x. Then du = √ 1

1−x 2

"

" 71.

" sinh3 x cosh x d x tanh 5x d x

sin−1 x d x = 1 − x2

d x and

" u du =

1 2 1 2 u + C = (sin−1 x) + C. 2 2


399

Let u = sinh x. Then du = cosh x d x and "

" sinh3 x cosh x d x =

u 3 du =

u4 sinh4 x +C = + C. 4 4

" 4 " 1dx dx 0 2x 2 + 1 2 0 25 − x √ √ √ SOLUTION Let u = 2x. Then du = 2 d x, and the new limits of integration are u = 0 and u = 4 2. Thus, 73.

" 4√2 dx = 0 2x 2 + 1 0

" 4

" 4√2 √1 du du 1 2 = √ u2 + 1 u2 + 1 2 0 4√2 √ √ 1

1 1 = √ tan−1 (4 2) − tan−1 0 = √ tan−1 (4 2). = √ tan−1 u 2 2 2 0

75. The"isotope Thorium-234 has a half-life of 24.5 days. 6 dx (a) Find the differential equation satisfied by the amount y(t) of Thorium-234 in a sample at time t. x 2 + 12 (b) At t =2 0,xa sample contains 2 kg of Thorium-234. How much remains after 1 year? SOLUTION

(a) By the equation for half-life, 24.5 =

ln 2 , k

so k =

ln 2 ≈ 0.028 days−1 . 24.5

Therefore, the differential equation for y(t) is y = −0.028y. (b) If there are 2 kg of Thorium-234 at t = 0, then y(t) = 2e−0.028t . After one year (365 days), the amount of Thorium234 is y(365) = 2e−0.028(365) = 7.29 × 10−5 kg = 0.0729 grams. 77. The C 14 to C 12 ratio of a sample is proportional to the disintegration rate (number of beta particles emitted per The Oldest Snack Food In Bat Cave, New Mexico, archaeologists found ancient human remains, including minute) that is measured directly with a 14 Geiger 12 counter. The disintegration rate of carbon in a living organism is 15.3 ratio equal to 9.5 around of that found in living matter. Estimate the cobs of popping C oftoa C beta particles/min per corn, gram.that Findhad theaage sample that emits beta 48% particles/min per gram. age of the corn cobs. SOLUTION Let t be the age of the sample in years. Because the disintegration rate for the sample has dropped from 15.3 beta particles/min per gram to 9.5 beta particles/min per gram and the C 14 to C 12 ratio is proportional to the disintegration rate, it follows that e−0.000121t =

9.5 , 15.3

so t =−

9.5 1 ln ≈ 3938.5. 0.000121 15.3

We conclude that the sample is approximately 3938.5 years old. 79. In a first-order chemical reaction, the quantity y(t) of reactant at time t satisfies y = −ky, where k > 0. The An of investment pays outT$5,000 at the isend of the yearArrhenius for 3 years. Compute theAePV, an interest rate −Eassuming a /(RT ) , where dependence k on temperature (in kelvins) given by the equation k= E a is the of 8%. −1 −1 −1 12 −1 activation energy (J-mol ), R = 8.314 J-mol -K , and A is a constant. Assume that A = 72 × 10 hour and dk for T = 500 and use the Linear Approximation to estimate the change in k if T is raised E a = 1.1 × 105 . Calculate dT from 500 to 510 K. SOLUTION

Let k = Ae−E a /(RT ) .

Then dk AE a −E a /(RT ) e . = dT RT 2

400

CHAPTER 7


For A = 72 × 1012 , R = 8.314 and E a = 1.1 × 105 we have 5

4

1.1×10 1.32×10 dk 9.53 × 1017 e− T 72 × 1012 · 1.1 × 105 e− 8.314T = . = dT 8.314 T2 T2

The derivative for T = 500 is thus 1.32×104 9.53 × 1017 e− 500 dk = ≈ 12.27 hours−1 K−1 . d T T =500 5002

Using the linear approximation we find dk k ≈ · (510 − 500) = 12.27 · 10 = 122.7 hours−1 . d T T =500 81. Find the solutions to y = −2y + 8 satisfying y(0) = 3 and y(0) = 4, and sketch their graphs. Find the solutions to y = 4(y − 12) satisfying y(0) = 20 and y(0) = 0, and sketch their graphs. SOLUTION First, rewrite the differential equation as y = −2(y − 4); from here we see that the general solution is y(t) = 4 + Ce−2t , for some constant C. If y(0) = 3, then 3 = 4 + Ce0

and C = −1.

Thus, y(t) = 4 − e−2t . If y(0) = 4, then 4 = 4 + Ce0

and C = 0;

hence, y(t) = 4. The graphs of the two solutions are shown below.

y=4 4

y = 4 – e–2t

2

-0.5

0.5

1

1.5

-2

In Exercises 83–86, let P(t) denote the balance at time t (years) of an annuity that earns 5% interest continuously Show that y = sin−1 x satisfies the differential equation y = sec y with initial condition y(0) = 0. compounded and pays out $2000/year continuously. 83. Find the differential equation satisfied by P(t). Since money is withdrawn continuously at a rate of $2000 a year and the growth due to interest is 0.05P, the rate of change of the balance is SOLUTION

P (t) = 0.05P − 2000. Thus, the differential equation satisfied by P(t) is P (t) = 0.05(P − 40, 000). 85. When does the annuity run out of money if P(0) = $2,000? Determine P(2) if P(0) = $5,000. SOLUTION In the previous exercise, we found that P(t) = 40,000 + Ce0.05t . If P(0) = 2000, then 2000 = 40,000 + Ce0.05·0 = 40,000 + C or C = −38,000.


401

Thus, P(t) = 40,000 − 38,000e0.05t . The annuity runs out of money when P(t) = 0; that is, when 40,000 − 38,000e0.05t = 0. Solving for t yields t=

40,000 1 ln ≈ 1.03. 0.05 38,000

The money runs out after roughly 1.03 years. In Exercises 87–98, that L’Hˆ opital’s Rule applies and evaluate theto limit. What is theverify minimum initial balance that will allow the annuity make payments indefinitely? 4x − 12

87. lim

x→3 x 2 − 5x + 6

SOLUTION

The given expression is an indeterminate form of type 00 , therefore L’Hôpital’s Rule applies. We find 4 4x − 12 4 = lim lim 2 = = 4. 2·3−5 x→3 x − 5x + 6 x→3 2x − 5

lim x 1/2 ln x x 3 + 2x 2 − x − 2 lim 4 x . The limit is now an indeterminate form of type ∞ , hence we may apply x→−2 x + 2x 3 −the 4xfunction −8 SOLUTION First rewrite as ln ∞ x −1/2 L’Hôpital’s Rule. We find 89.

x→0+

lim x 1/2 ln x = lim

ln x

x→0+ x −1/2

x→0+

= lim

x −1

x→0+ − 1 x −3/2 2

= lim −2x 1/2 = 0. x→0+

2 sin θ − sin 2θ ln(et + 1) sin θ − θ cos θ θ →0lim t→∞ t 0 SOLUTION The given expression is an indeterminate form of type 0 ; hence, we may apply L’Hˆ opital’s Rule. We find 91. lim

lim

2 sin θ − sin 2θ

θ →0 sin θ − θ cos θ

2 cos θ − 2 cos 2θ

= lim

θ →0 cos θ − (cos θ − θ sin θ )

= lim

θ →0

= lim

θ →0

2 cos θ − 2 cos 2θ θ sin θ

−2 + 8 −2 sin θ + 4 sin 2θ −2 cos θ + 8 cos 2θ = 3. = lim = sin θ + θ cos θ 1+1−0 θ →0 cos θ + cos θ − θ sin θ

ln(t√+ 2) √ 93. lim 4+x −281+x t→∞ lim log2 t x→0 x2 ∞ SOLUTION The limit is an indeterminate form of type ∞ ; hence, we may apply L’Hˆ opital’s Rule. We find 1

ln(t + 2) t ln 2 ln 2 = lim = lim t+2 = lim = ln 2. t→∞ log2 t t→∞ 1 t→∞ t + 2 t→∞ 1 lim

t ln 2

y− sin−1 y 1 ex − y→0lim y 3x x x→0 e − 1 0 SOLUTION The limit is an indeterminate form of type 0 ; hence, we may apply L’Hˆ opital’s Rule. We find 95. lim

sin−1 y − y = lim lim y→0 y→0 y3

√1

1−y 2 3y 2

−1

y(1 − y 2 )−3/2 (1 − y 2 )−3/2 1 = lim = . 6y 6 6 y→0 y→0

= lim

2) sinh(x 97. lim 1 − x2 cosh x − 1 x→0lim x→1 cos−1 x 0 SOLUTION The limit is an indeterminate form of type 0 ; hence, we may apply L’Hˆ opital’s Rule. We find sinh(x 2 ) 2x cosh(x 2 ) 2 cosh(x 2 ) + 4x 2 sinh(x 2 ) 2+0 = lim = lim = = 2. sinh x cosh x 1 x→0 cosh x − 1 x→0 x→0 lim

402

CHAPTER 7


2x − sin x . Evaluate the limit by another Explain tanh x −why sinhL’Hˆ x opital’s Rule gives no information about lim x→∞ 3x + cos 2x lim method. x→0 sin x − x 99.

As x → ∞, both 2x − sin x and 3x + cos 2x tend toward infinity, so L’Hôpital’s Rule applies to 2x − sin x 2 − cos x lim ; however, the resulting limit, lim , does not exist due to the oscillation of sin x and x→∞ 3x + cos 2x x→∞ 3 − 2 sin 2x cos x. To evaluate the limit, we note SOLUTION

2 − sinx x 2 2x − sin x = . = lim x→∞ 3x + cos 2x x→∞ 3 + cos 2x 3 x lim

101. In this exercise, we prove that for all x > 0, Let f (x) be a differentiable function with inverse g(x) such that f (0) = 0 and f (0) = 0. Prove that x2 f (x)+ x) ≤ ≤ ln(1 x x− = f (0)2 2lim g(x) x→0 " x dt (a) Show that ln(1 + x) = for x > 0. 0 1+t 1 ≤ 1 for all t > 0. (b) Verify that 1 − t ≤ 1+t (c) Use (b) to prove Eq. (1). (d) Verify Eq. (1) for x = 0.5, 0.1, and 0.01.

1

SOLUTION

(a) Let x > 0. Then

x dt = ln(1 + t) = ln(1 + x) − ln 1 = ln(1 + x). 1 + t 0 0

" x

1 < 1. Moreover, (1 − t)(1 + t) = 1 − t 2 < 1. Because 1 + t > 0, it follows that (b) For t > 0, 1 + t > 1, so 1+t 1 . Hence, 1 − t < 1+t

1−t ≤

1 ≤ 1. 1+t

(c) Integrating each expression in the result from part (b) from t = 0 to t = x yields x−

x2 ≤ ln(1 + x) ≤ x. 2

(d) For x = 0.5, x = 0.1 and x = 0.01, we obtain the string of inequalities 0.375 ≤ 0.405465 ≤ 0.5 0.095 ≤ 0.095310 ≤ 0.1 0.00995 ≤0.00995033≤ 0.01, respectively. −1 In Exercises Let 103–106, let gd(y) = tan (sinh y) be the so-called gudermannian, which arises in cartography. In a map of the earth constructed by Mercator projection, points " x units from the equator correspond to points on located y radial the globe of latitude gd(y). t 2 − 1 dt F(x) = x x 2 − 1 − 2 1

d 103. Prove that gd(y) = sech y. Prove that dF(x) y and cosh−1 x differ by a constant by showing that they have the same derivative. Then prove they are equal by evaluating both at x = 1. SOLUTION Let gd(y) = tan−1 (sinh y). Then 1 d 1 gd(y) = = sech y, cosh y = dy cosh y 1 + sinh2 y where we have used the identity 1 + sinh2 y = cosh2 y. −1 (tan y) is the inverse of gd(y) for 0 ≤ y < π /2. 105. Show that t (y) = sinh Let f (y) = 2 tan−1 (e y ) − π /2. Prove that gd(y) = f (y). Hint: Show that gd (y) = f (y) and f (0) = gd(0). SOLUTION Let x = gd(y) = tan−1 (sinh y). Solving for y yields y = sinh−1 (tan x). Therefore,

gd −1 (y) = sinh−1 (tan y).


403

107. Let Verify that t (y) in Exercise 105 satisfies " xt (y) = sec y and find a value of a such that dt x F(x) = and" y dt G(x) = ln x 2 ln t t (y) = a cos t F(x) and evaluate L. Verify that L’Hôpital’s Rule may be applied to the limit L = lim x→∞ G(x) SOLUTION

Because t > ln t for t > 2, F(x) =

" x " x dt dt > > ln x. 2 ln t 2 t

Thus, F(x) → ∞ as x → ∞. Moreover, lim G(x) = lim

x→∞

Thus, lim

F(x)

x→∞ G(x)

1

x→∞ 1/x

= lim x = ∞. x→∞

is of the form ∞/∞, and L’Hôpital’s Rule applies. Finally, 1

F(x) ln x x = lim = lim lnlnx−1 = 1. x→∞ G(x) x→∞ x→∞ ln x − 1 2

L = lim

(ln x)

2 Let f (x) = e−Ax /2 , where A > 0. Given any n numbers a1 , a2 , . . . , an , set

(x) = f (x − a1 ) f (x − a2 ) · · · f (x − an ) (a) Assume n = 2 and prove that (x) attains its maximum value at the average x = 12 (a1 + a2 ). Hint: Show that d ln( f (x)) = −Ax and calculate (x) using logarithmic differentiation. dx (b) Show that for any n, (x) attains its maximum value at x = n1 (a1 + a2 + · · · + an ). This fact is related to the role of f (x) (whose graph is a bell-shaped curve) in statistics.

TECHNIQUES OF 8 INTEGRATION 8.1 Numerical Integration Preliminary Questions 1. What are T1 and T2 for a function on [0, 2] such that f (0) = 3, f (1) = 4, and f (2) = 3? SOLUTION

Using the given function values T1 =

1 (2)(3 + 3) = 6 and 2

T2 =

1 (1)(3 + 8 + 3) = 7. 2

2. For which graph in Figure 16 will TN overestimate the integral? What about M N ? y

y

y = f(x)

y = g(x)

x

x

FIGURE 16 SOLUTION TN overestimates the value of the integral when the integrand is concave up; thus, TN will overestimate the integral of y = g(x). On the other hand, M N overestimates the value of the integral when the integrand is concave down; thus, M N will overestimate the integral of y = f (x).

3. How large is the error when the Trapezoidal Rule is applied to a linear function? Explain graphically. The Trapezoidal Rule integrates linear functions exactly, so the error will be zero. " 3 f (x) d x, where | f (x)| ≤ 2 for all x. What is the maximum possible error? 4. Suppose T4 is used to approximate

SOLUTION

0

SOLUTION

The maximum possible error in T4 is max | f (x)|

(b − a)3 2(3 − 0)3 9 = = . 2 2 32 12n 12(4)

5. What are the two graphical interpretations of the Midpoint Rule? SOLUTION The two graphical interpretations of the Midpoint Rule are the sum of the areas of the midpoint rectangles and the sum of the areas of the tangential trapezoids.

Exercises In Exercises 1–12, calculate TN and M N for the value of N indicated. " 2 1.

x 2 d x,

0 SOLUTION

N =4

Let f (x) = x 2 . We divide [0, 2] into 4 subintervals of width x =

2−0 1 = 4 2

with endpoints 0, 0.5, 1, 1.5, 2, and midpoints 0.25, 0.75, 1.25, 1.75. With this data, we get 1 1 2 0 + 2(0.5)2 + 2(1)2 + 2(1.5)2 + 22 = 2.75; and · 2 2 1

0.252 + 0.752 + 1.252 + 1.752 = 2.625. M4 = 2 T4 =

3.

" 4 " 34 x d√x, N = 6 x d x, N = 4 1 0


Numerical Integration

Let f (x) = x 3 . We divide [1, 4] into 6 subintervals of width x =

4−1 1 = 6 2

with endpoints 1, 1.5, 2, 2.5, 3, 3.5, 4, and midpoints 1.25, 1.75, 2.25, 2.75, 3.25, 3.75. With this data, we get 1 1 3 T6 = 1 + 2(1.5)3 + 2(2)3 + 2(2.5)3 + 2(3)3 + 2(3.5)3 + 43 = 64.6875; and 2 2 1

M6 = 1.253 + 1.753 + 2.253 + 2.753 + 3.253 + 3.753 = 63.28125. 2 5.

" 4 "d πx/2 √ , N =6 sin x d x, 1 x 0

SOLUTION

N =8

Let f (x) = 1/x. We divide [1, 4] into 6 subintervals of width x =

4−1 1 = 6 2

with endpoints 1, 1.5, 2, 2.5, 3, 3.5, 4, and midpoints 1.25, 1.75, 2.25, 2.75, 3.25, 3.75. With this data, we get 1 1 1 2 2 2 2 2 1 T6 = + + + + + + ≈ 1.40536; and 2 2 1 1.5 2 2.5 3 3.5 4 1 1 1 1 1 1 1 M6 = + + + + + ≈ 1.37693. 2 1.25 1.75 2.25 2.75 3.25 3.75 " π /4 " −1 secdxxd x, N = 6 , N =5 0 −2 x SOLUTION Let f (x) = sec x. We divide [0, π /4] into 6 subintervals of width 7.

π −0 π x = 4 = 6 24

with endpoints 0,

6π π 2π π , ,... , = , 24 24 24 4

and midpoints

π 3π 11π , ,... , . 48 48 48 With this data, we get 1 π sec(0) + 2 sec(π /24) + 2 sec(2π /24) + · · · + sec(6π /24) ≈ 0.883387; and 2 24 π M6 = sec(π /48) + sec(3π /48) + sec(5π /48) + · · · + sec(11π /48) ≈ 0.880369. 24 T6 =

9.

" 3 " d2x , N =5 2 ln xln x, N = 5 1

SOLUTION

Let f (x) = 1/ ln x. We divide [2, 3] into 5 subintervals of width x =

1 3−2 = = 0.2 5 5

with endpoints 2, 2.2, 2.4, 2.6, 2.8, 3, and midpoints 2.1, 2.3, 2.5, 2.7, 2.9. With this data, we get 1 1 1 2 2 2 2 1 T5 = + + + + + ≈ 1.12096; and 2 5 ln 2 ln 2.2 ln 2.4 ln 2.6 ln 2.8 ln 3 1 1 1 1 1 1 M5 = + + + + ≈ 1.11716. 5 ln 2.1 ln 2.3 ln 2.5 ln 2.7 ln 2.9

11.

" 2 x " 1e −xd2 x, N = 8 0 x +e 1 d x, N = 6 0

405

406

CHAPTER 8

T E C H N I Q U E S O F I N T E G R AT I O N

Let f (x) = e x /(x + 1). We divide [0, 2] into 8 subintervals of width

SOLUTION

2−0 1 = = 0.25 8 4

x =

with endpoints 0, 0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2, and midpoints 0.125, 0.375, 0.625, 0.875, 1.125, 1.375, 1.625, 1.875. With this data, we get 0 e 1 1 2e0.25 2e0.5 2e1.75 e2 T8 = + + + ··· + + ≈ 2.96581; and 2 4 0 + 1 0.25 + 1 0.5 + 1 1.75 + 1 2 + 1 e0.125 1 e0.375 e0.625 e1.875 M8 = + + + ··· + ≈ 2.95302. 4 0.125 + 1 0.375 + 1 0.625 + 1 1.875 + 1 " 1 13–22, calculate S N given by Simpson’s Rule for the value of N indicated. In Exercises 2 e x d x, N = 6 " 4 √ −2 x d x, N = 4 13. 0

Let f (x) =

SOLUTION

√

x. We divide [0, 4] into 4 subintervals of width 4−0 =1 4

x = with endpoints 0, 1, 2, 3, 4. With this data, we get S4 =

15.

√ √ √ √ 1 √ (1) 0 + 4 1 + 2 2 + 4 3 + 4 ≈ 5.25221. 3

" 2 " 5d x , N =4 +− 1 x 2 ) d x, N = 4 0 x 4 (9 3

Let f (x) = 1/(x 4 + 1). We divide [0, 2] into 4 subintervals of length

SOLUTION

x =

2−0 1 = = 0.5 4 2

with endpoints 0, 0.5, 1, 1.5, 2. With this data, we get 4 2 4 1 1 1 1 + + + + S4 = ≈ 1.08055. 3 2 04 + 1 0.54 + 1 14 + 1 1.54 + 1 24 + 1 17.

" 1 2 " −x e 1 d x,2 N = 6 cos(x ) d x, N = 6 0 0

Let f (x) = e−x . We divide [0, 1] into 6 subintervals of length 2

SOLUTION

x =

1−0 1 = 6 6

with endpoints 0, 16 , 26 , . . . , 66 = 1. With this data, we get S6 =

19.

1 3

2 2 2 2 2 2 2 1 e−0 + 4e−(1/6) + 2e−(2/6) + 4e−(3/6) + 2e−(4/6) + 4e−(5/6) + e−(1) ≈ 0.746830. 6

" 4 " 2 ln x d−x x, N = 8 e d x, N = 6 1 1

SOLUTION

Let f (x) = ln x. We divide [1, 4] into 8 subintervals of length x =

4−1 3 = = 0.375 8 8

with endpoints 1, 1.375, 1.75, 2.125, 2.5, 2.875, 3.25, 3.625, 4. With this data, we get 1 3 ln 1 + 4 ln (1.375) + 2 ln (1.75) + · · · + 4 ln (3.625) + ln 4 ≈ 2.54499. S8 = 3 8 " 4 2

x 4 + 1 d x,

N =8

S E C T I O N 8.1

" π /4 21.

407

N = 10

sec x d x, 0

SOLUTION


Let f (x) = sec x. We divide [0, π4 ] into 10 subintervals of width π −0 π = x = 4 10 40

π , 2π , 3π , . . . , 10π = π . With this data, we get with endpoints 0, 40 40 40 40 4

2π 1 π 9π 10π π + 2 sec sec (0) + 4 sec + · · · + 4 sec + sec ≈ 0.881377. S10 = 3 40 40 40 40 40

" 4 23–26, calculate the approximation to the volume of the solid obtained by rotating the graph around the In Exercises given axis. x 4 + 1 d x, N = 10 2 23. y = cos x; 0, π2 ; x-axis; M8 SOLUTION

Using the disk method, the volume is given by V =

" π /2 0

πr 2 d x = π

" π /2 0

(cos x)2 d x

which can be estimated as

π

" π /2 0

(cos x)2 d x ≈ π [M8 ].

Let f (x) = cos2 x. We divide [0, π /2] into 8 subintervals of length π −0 π x = 2 = 8 16

with midpoints

π 3π 5π 15π , , , ... , . 32 32 32 32 With this data, we get

π π2 3π 15π + cos2 V ≈ π [M8 ] = π x(y1 + y2 + · · · + y8 ) = cos2 + · · · + cos2 ≈ 2.46740. 16 32 32 32 −x 2 ; [0, 1]; 25. y =y e= T8 S ; y-axis; cos x; 0, π2 x-axis; 8 SOLUTION Using the disk method, the volume is given by

V =

" 1 0

" 1 " 1 −x 2 2 2 dx = π e−2x d x. e

πr 2 d x = π

0

0

We can use the approximation V =π where f (x) = e

−2x 2

" 1 0

e−2x d x ≈ π [T8 ], 2

. Divide [0, 1] into 8 subintervals of length x =

1 1−0 = , 8 8

with endpoints 0,

1 2 , , . . . , 1. 8 8

With this data, we get V ≈ π [T8 ] = π

y = e−x ; [0, 1]; 2

2 2 2 1 1 −2(02 ) e ≈ 1.87691. + 2e−2(1/8) + · · · + 2e−2(7/8) + e−2(1) · 2 8

y-axis;

S8

408

CHAPTER 8


27. Use S8 to estimate SOLUTION

" π /2 sin x d x, taking the value of (sin x)/x at x = 0 to be 1. x 0

Divide [0, π /2] into 8 subintervals of length π −0 π x = 2 = 8 16

with endpoints 0,

π 2π 8π , ,... , . 16 16 16

Taking the value of (sin x)/x at x = 0 to be 1, we get sin(π /16) sin(2π /16) sin(π /2) 1 π 1+4 +2 + ··· + S8 = ≈ 1.37076. 3 16 π /16 2π /16 π /2 " π /2 " 2 cos x d x. 29. Calculate M6 for the integral I = x 3 d x. Calculate T6 for the integral I =0 0 (a) Is M6 too large or too small? Explain graphically. (a) Is T6 too large or too small? Explain graphically. (b) Show that K 2 = 1 may be used in the Error Bound and find a bound for the error. f (2)| beerror usedisinless the than Errorthe Bound and find a bound (b) ShowI that 2 = |that (c) Evaluate and K check the may actual bound computed in (b).for the error. (c) Evaluate I and check that the actual error is less than the bound computed in (b). SOLUTION Let f (x) = cos x. Divide [0, π /2] into 6 subintervals of length π −0 π = x = 2 6 12

with midpoints 11π π 3π , , ... , . 24 24 24 With this data, we get M6 =

3π π π 11π + cos cos + · · · + cos ≈ 1.0028615. 12 24 24 24

(a) Since f (x) = cos x is concave down on [0, π /2], M6 is too large. (b) We have f (x) = − sin x and f (x) = − cos x. Since | f (x)| = | − cos x| ≤ 1 on [0, π /2], we may take K 2 = 1. Then Error(M6 ) ≤

π3 K 2 (b − a)3 (1)(π /2 − 0)3 = = ≈ 0.00448586. 2 2 6912 24N 24(6)

(c) The exact value is " π /2 0

π /2

π cos x d x = sin x = sin − sin 0 = 1 − 0 = 1. 2 0

We can use this to compute the actual error: Error(M6 ) = |M6 − 1| ≈ |1.0028615 − 1| ≈ 0.0028615. Since 0.0028615 < 0.00448586, the actual error is indeed less than the maximum possible error. In Exercises 30–33, state whether TN or M N underestimates or overestimates the integral and find a bound for the error (but do not calculate TN or M N ). " 2 " −x/4 31. e 4 1 d x, T20 d x, T10 0 1 x SOLUTION Let f (x) = e−x/4 . Then f (x) = −(1/4)e−x/4 and f (x) =

1 −x/4 >0 e 16

on [0, 2], so f (x) is concave up, and T20 overestimates the integral. Since | f (x)| = |(1/16)e−x/4 | has its maximum value on [0, 2] at x = 0, we can take K 2 = |(1/16)e0 | = 1/16, and Error(T20 ) ≤

1 3 K 2 (2 − 0)3 16 (2) = = 1.04167 × 10−4 . 12N 2 12(20)2

S E C T I O N 8.1

33.


409

" π /4 " 4 cos x, M20 ln x d x, M10 0 1

Let f (x) = cos x. Then f (x) = − sin x and f (x) = − cos x < 0 on [0, π /4], so f (x) is concave down, and M20 overestimates the integral. Since | f (x)| = | − cos x| has its maximum value on [0, π /4] at x = 0, we can take K 2 = | − cos(0)| = 1, and SOLUTION

Error(M20 ) ≤

K 2 (π /4 − 0)3 (1)(π /4)3 = = 5.04659 × 10−5 . 24N 2 24(20)2

In Exercises 34–37, use the Error Bound to find a value of N for which Error(TN ) ≤ 10−6 . If you have a computer algebra system, calculate the corresponding approximation and confirm that the error satisfies the required bound. 35.

" π /6 " 1 cos4 x d x x dx 0 0

Let f (x) = cos x. Then f (x) = − sin x and | f (x)| = | − cos x|, which has its maximum value on [0, π /6] at x = 0, so we can take K 2 = | − cos(0)| = 1. Then we have SOLUTION

π3 K 2 (π /6 − 0)3 = . 2 12N 12 · 63 N 2

Error(TN ) ≤

To ensure that the error is at most 10−6 , we must choose N such that

π3 1 ≤ 6. 12 · 63 N 2 10 This gives us N2 ≥

π 3 106 ⇒N≥ 12 · 63

π 3 106 ≈ 109.37. 12 · 63

Thus let N = 110. The exact value of the integral is " π /6 0

π /6 cos x d x = sin x = 0.5. 0

Using a CAS, we find that T110 ≈ 0.4999990559. The error is approximately |0.5 − 0.4999990559| ≈ 9.441 × 10−7 and is indeed less than 10−6 . " 3 " −x e 5 1d x 37. dx 0 2 x SOLUTION Let f (x) = e−x . Then f (x) = −e−x and | f (x)| = |e−x | = e−x , which has its maximum value on [0, 3] at x = 0, so we can take K 2 = e0 = 1. Then we have Error(TN ) ≤

K 2 (3 − 0)3 (1)33 9 = = . 2 12N 12N 2 4N 2

To ensure that the error is at most 10−6 , we must choose N such that 1 9 ≤ 6. 2 4N 10 This gives us 9 · 106 N2 ≥ ⇒N≥ 4

9 · 106 = 1500. 4

Thus let N = 1500. The exact value of the integral is " 3 e−x d x = −e−3 − −e−0 = 1 − e−3 ≈ 0.9502129316. 0

Using a CAS, we find that T1500 ≈ 0.9502132468.

410

CHAPTER 8


The error is approximately |0.9502129316 − 0.9502132468| ≈ 3.152 × 10−7 and is indeed less than 10−6 . " 1 " 5 e−2x d x. 39. (a) Compute S6 for the integral I = (x 3 + 1)−1 d x (use Figure 17 to determine Find a bound for the error in the approximations T10 and M10 to 0 (b) Show that K 4 = 16 may be used in the Error Bound and find a bound 0for the error. a value of K 2 ). Then find a value of N such that the error in M N is at most 10−6 . (c) Evaluate I and check that the actual error is less than the bound for the error computed in (b). SOLUTION

(a) Let f (x) = e−2x . We divide [0, 1] into six subintervals of length x = (1 − 0)/6 = 1/6, with endpoints 0, 1/6, . . . , 5/6, 1. With this data, we get S6 =

1 1 −2(0) + 4e−2(1/6) + 2e−2(2/6) + · · · + e−2(1) ≈ 0.432361. · e 3 6

(b) Taking derivatives, we get f (x) = −2e−2x , f (x) = 4e−2x , f (3) (x) = −8e−2x , f (4) (x) = 16e−2x . Since | f (4) (x)| = |16e−2x | assumes its maximum value on [0, 1] at x = 0, we can set K 4 = |16e0 | = 16. Then we have Error(S6 ) ≤

K 4 (1 − 0)5 16 = ≈ 6.86 × 10−5 . 4 180N 180 · 64

(c) The exact value of the integral is " 1 0

e−2x d x =

e−2x 1 1 − e−2 = ≈ 0.432332. −2 0 2

The actual error is Error(S6 ) ≈ |0.432361 − 0.432332| ≈ 2.9 × 10−5 . The error is indeed less than the maximum possible error. " 5 " 1 ln x d x and find a bound for the error. Then find a value of N such that S N has an error of at 41. Calculate S8 for Calculate S8 for x 4 d x. Use the Error Bound to find a bound for the error and verify that the 1 the integral mostactual 10−6 .error satisfies this bound. 0 SOLUTION Let f (x) = ln x. We divide [1, 5] into eight subintervals of length x = (5 − 1)/8 = 0.5, with endpoints 1, 1.5, 2, . . . , 5. With this data, we get

S8 =

1 1 · ln 1 + 4 ln 1.5 + 2 ln 2 + · · · + 4 ln 4.5 + ln 5 ≈ 4.046655. 3 2

To find the maximum possible error, we first take derivatives: f (x) =

1 1 2 6 , f (x) = − 2 , f (3) (x) = 3 , f (4) (x) = − 4 . x x x x

Since | f (4) (x)| = | − 6x −4 | = 6x −4 , assumes its maximum value on [1, 5] at x = 1, we can set K 4 = 6(1)−4 = 6. Then we have Error(S8 ) ≤

K 4 (5 − 1)5 6 · 45 = ≈ 0.0083333. 4 180N 180 · 84

To ensure that S N has error at most 10−6 , we must find N such that 6 · 45 1 ≤ 6. 4 180N 10 This gives us 6 · 45 · 106 N4 ≥ ⇒N≥ 180

6 · 45 · 106 180

1/4 ≈ 76.435.

Thus let N = 78 (remember that N must be even when using Simpson’s Rule). " 3 Find a bound for the error in the approximation S10 to

0

e−x d x (use Figure 18 to determine a value of K 4 ). 2


S E C T I O N 8.1

Use a computer algebra system to compute and graph f (4) (x) for f (x) = " 5 f (x) d x. the error in the approximation S40 to 43.

SOLUTION

411

1 + x 4 and find a bound for

0 (4) From the graph of f (x) shown below, we see that | f (4) (x)| ≤ 15 on [0, 5]. Therefore we set K 4 = 15.

Now we have Error(S40 ) ≤

15(5 − 0)5 5 = ≈ 1.017 × 10−4 . 49152 180(40)4 y 15 10 5 x 1

−5 −10 −15

2

3

4

5

−9 In Exercises 45–48, use thealgebra Error Bound a valueand of graph N for which ≤ 10 N) = Use a computer systemtotofind compute f (4) (x)Error(S for f (x) tan x .− sec x and find a bound for " " 7 π /4 f (x) d x. x 3/2 dinxthe approximation S40 to 45. the error 0

1 SOLUTION

Let f (x) = x 3/2 . To find K 4 , we first take derivatives: f (x) =

3 1/2 3 3 9 −5/2 x , f (x) = x −1/2 , f (3) (x) = − x −3/2 , f (4) (x) = x . 2 4 8 16

Since f (5) (x) = −(45/32)x −7/2 < 0 on [1, 7], f (4) is decreasing on [1, 7] and therefore assumes its maximum value at x = 1. Thus we can set 9 9 . K 4 = | f (4) (1)| = (1)−5/2 = 16 16 Then we have Error(S N ) ≤

K 4 (7 − 1)5 9 65 = . · 4 16 180N 4 180N

To ensure that S N has error at most 10−9 , we must find N such that 9 · 65 1 ≤ 9. 16 · 180N 4 10 This gives us 9 · 65 · 109 N4 ≥ ⇒N≥ 16 · 180

9 · 65 · 109 16 · 180

1/4 ≈ 394.82.

Thus let N = 396 (remember that N must be even when using Simpson’s Rule). " 1 " x42 e d xx 47. xe d x 0 0

SOLUTION

2

Let f (x) = e x . To find K 4 , we first take derivatives: f (x) = 2xe x

2

f (x) = 4x 2 e x + 2e x 2

2

f (3) (x) = 8x 3 e x + 12xe x 2

2

f (4) (x) = 16x 4 e x + 48x 2 e x + 12e x . 2

2

2

On the interval [0, 1], | f (4) (x)| assumes its maximum value at x = 1. Therefore we set K 4 = | f (4) (1)| = 16e + 48e + 12e = 76e.

412

CHAPTER 8


Now we have Error(S N ) ≤

K 4 (1 − 0)5 76e = . 180N 4 180N 4

To ensure that S N has error at most 10−9 , we must find N such that 1 76e ≤ 9. 180N 4 10 This gives us 76e · 109 N4 ≥ ⇒N≥

180

76e · 109 180

1/4 ≈ 184.06.

Thus we let N = 186 (remember that N must be even when using Simpson’s Rule). " 1 dx π " 4 Show that = [use Eq. (5) in Section 7.8]. 49. 4 sin(ln x) d x 0 1 + x 2 1 (a) Use a computer algebra system to graph f (4) (x) for f (x) = (1 + x 2 )−1 and find its maximum on [0, 1]. (b) Find a value of N such that S N approximates the integral with an error of at most 10−6 . Calculate the corresponding approximation and confirm that you have computed π4 to at least four places. SOLUTION

Recall from Section 7.8 that d 1 tan−1 (x) = . dx 1 + x2

So then 1 dx −1 x = tan−1 (1) − tan−1 (0) = π . = tan 2 4 0 1+x 0

" 1

(a) From the graph of f (4) (x) shown below, we can see that the maximum value of | f (4) (x)| on the interval [0, 1] is 24. y 30 20 10 x −10

0.2

0.4

0.6

0.8

1

(b) From part (a), we set K 4 = 24. Then we have Error(S N ) ≤

24(1 − 0)5 2 = . 180N 4 15N 4

To ensure that S N has error at most 10−6 , we must find N such that 2 1 ≤ 6. 15N 4 10 This gives us 2 · 106 ⇒N≥ N4 ≥ 15

2 · 106 15

1/4 ≈ 19.1.

Thus let N = 20. To compute S20 , let x = (1 − 0)/20 = 0.05. The endpoints of [0, 1] are 0, 0.05, . . . , 1. With this data, we get 1 1 4 2 1 1 S20 = + + + · · · + ≈ 0.785398163242. 3 20 1 + 02 1 + (0.05)2 1 + (0.1)2 1 + 12 The actual error is |0.785398163242 − π /4| = |0.785398163242 − 0.785398163397| = 1.55 × 10−10 .

Let f (x) = sin(x 2 ) and I =

" 1

f (x) d x.

0 (a) Check that f (x) = 2 cos(x 2 ) − 4x 2 sin(x 2 ). Then show that | f (x)| ≤ 6 for x ∈ [0, 1]. Hint: Note that

S E C T I O N 8.1


413

The Error Bound for M N is proportional to 1/N 2 , so the Error Bound decreases by 14 if N is " π sin x d x for N = 4, 8, 16, 32, and 64. Does the actual error increased to 2N . Compute the actual error in M N for 51.

0

seem to decrease by 14 as N is doubled? SOLUTION

The exact value of the integral is " π 0

π sin x d x = − cos x = −(−1) − (1) = 2. 0

To compute M4 , we have x = (π − 0)/4 = π /4, and midpoints π /8, 3π /8, 5π /8, 7π /8. With this data, we get 3π π π 5π 7π + sin M4 = sin + sin + sin ≈ 2.052344. 4 8 8 8 8 The values for M8 , M16 , M32 , and M64 are computed similarly: 3π 15π π π + sin M8 = sin + · · · + sin ≈ 2.012909; 8 16 16 16 π π 3π 31π M16 = sin + sin + · · · + sin ≈ 2.0032164; 16 32 32 32 3π π π 63π + sin M32 = sin + · · · + sin ≈ 2.00080342; 32 64 64 64

3π 127π π π + sin M64 = sin + · · · + sin ≈ 2.00020081. 64 128 128 128 Now we can compute the actual errors for each N : Error(M4 ) = |2 − 2.052344| = 0.052344 Error(M8 ) = |2 − 2.012909| = 0.012909 Error(M16 ) = |2 − 2.0032164| = 0.0032164 Error(M32 ) = |2 − 2.00080342| = 0.00080342 Error(M64 ) = |2 − 2.00020081| = 0.00020081 The actual error does in fact decrease by about 1/4 each time N is doubled. 1 why the ErrorforBound for S Nhas decreases to large 2N . Compute theBound actual 12 in thebydenominator) is twice as as the Error Observe Explain that the Error Bound TN (which 16 if N is increased " π " π 1 as N is doubled? d x in forthe N denominator). = 4, 8, 16, 32,Compute and 64. Does the actual errorfor in SMNN for (whichsin hasx 24 the actual errorerror in Tseem sin x d xbyfor16 N = 4, 8, 16, 32, N for to decrease

53.

0

0

and 64 and compare with the calculations of Exercise 51. Does the actual error in TN seem to be roughly twice as If we plug in 2N for N in the formula for the error bound for S N , we get large as the error in M N in this case? K 4 (b − a)5 1 K 4 (b − a)5 K 4 (b − a)5 . = = 16 180(2N )4 180 · 24 · N 4 180N 4

SOLUTION

Thus we see that, since N is raised to the fourth power in the denominator, the Error Bound for S N decreases by 1/16 if N is increased to 2N . The exact value of the integral is π " π sin x d x = − cos x = −(−1) − (1) = 2. 0

0

To compute S4 , we have x = (π − 0)/4 = π /4, and endpoints 0, π /4, 2π /4, 3π /4, π . With this data, we get

π 2π 1 π 3π + 2 sin S4 = · sin(0) + 4 sin + 4 sin + sin(π ) ≈ 2.004560. 3 4 4 4 4 The values for S8 , S16 , S32 , and S64 are computed similarly:

π 1 π 2π 7π S8 = · sin(0) + 4 sin + 2 sin + · · · + 4 sin + sin(π ) ≈ 2.0002692; 3 8 8 8 8

2π 15π π 1 π + 2 sin S16 = · sin(0) + 4 sin + · · · + 4 sin + sin(π ) ≈ 2.00001659; 3 16 16 16 16

2π 1 π 31π π + 2 sin S32 = · sin(0) + 4 sin + · · · + 4 sin + sin(π ) ≈ 2.000001033; 3 32 32 32 32

414

CHAPTER 8


S64 =

π 2π 1 π 63π + 2 sin · sin(0) + 4 sin + · · · + 4 sin + sin(π ) ≈ 2.00000006453. 3 64 64 64 64

Now we can compute the actual errors for each N : Error(S4 ) = |2 − 2.004560| = 0.004560 Error(S8 ) = |2 − 2.0002692| = 2.692 × 10−4 Error(S16 ) = |2 − 2.00001659| = 1.659 × 10−5 Error(S32 ) = |2 − 2.000001033| = 1.033 × 10−6 Error(S64 ) = |2 − 2.00000006453| = 6.453 × 10−8 The actual error does in fact decrease by about 1/16 each time N is doubled. For example, 0.004560/16 = 2.85 × 10−4 , which is roughly the same as 2.692 × 10−4 . 55. Use Simpson’s Rule to determine the average temperature in a museum over a 3-hour period, if the temperatures (in An airplane’s velocity is recorded at 5 min intervals during a 1-hour period with the following results, in mph: degrees Celsius), recorded at 15-min intervals, are 550, 575, 600, 580, 610, 640, 625, 595, 590, 620, 640, 640, 630 21, 21.3, 21.5, 21.8, 21.6, 21.2, 20.8, 20.6, 20.9, 21.2, 21.1, 21.3, 21.2 Use Simpson’s Rule to estimate the distance traveled during the hour. If T (t) represents the temperature at time t, then the average temperature Tave from t = 0 to t = 3 hours

SOLUTION

is given by Tave =

" 3 1 T (t) dt. 3−0 0

To use Simpson’s Rule to approximate this, let t = 1/4 (15 minute intervals). Then we have Tave =

1 1 1 1 [S12 ] = · · 21 + 4 · 21.3 + 2 · 21.5 + · · · + 4 · 21.3 + 21.2 ≈ 21.2111. 3 3 3 4

The average temperature is approximately 21.2◦ C. Times Scientists estimate the arrival times of tsunamis (seismic ocean waves) based on Further Tsunami InsightsArrival and Challenges √

" P the point of origin b and ocean depths. The speed s of a tsunami in miles per hour is approximately s = 15d, 57. Show = f (x) dinx feet. for all N and all endpoints a, b if f (x) = r x + s is a linear function (r, s constants). where that d is Tthe ocean depth N a (a) Let f (x) be the ocean depth x miles from P (in the direction of the coast). Argue using Riemann sums that the SOLUTION First, note thattsunami to travel M miles toward the coast is time T required for the " b " M2 r (b − ad2x) (r x + s) dTx = = + s(b − a). √ a 0 2 15 f (x) Now,(b) Use Simpson’s Rule to estimate T if M = 1,000 and the ocean depths (in feet), measured at 100-mile intervals * * ) ) starting from P, are N −1 N −1 −1 # # b−a r (b − a) b − a N# b−a TN (r x + s) = f (xi ) +9,000, f (b) 8,500, = a + 3,800, 2 b +s f (a) + 2 + 2 4,400, (2N ) 10,500, 7,000, a6,000, 3,200,i + 2,000 2N13,000, 11,500, 2N N 2N i=1 i=1 i=1 b − a (N − 1)N r (b2 − a 2 ) r (b − a) = (2N − 1)a + 2 + b + s(b − a) = + s(b − a). 2N N 2 2 b−a " ab + j x, y = f (x ), and 59. For N even, divide [a, b] into N subintervals of width x = . Set x j = j j 2 N then S2 = f (x) d x. In other words, show that Show that if f (x) = px + qx + r is a quadratic polynomial, a

b −a y2 j b+−4ya2j+1 + y2 j+2 3N d x = f (x) y0 + 4y1 + y2 6 a on the intervals [x2 j , x2 j+2 ], that is, S N = S20 + S22 + · · · + S2N −2 . (a) Show that S N is the sum ofthe approximations " x2 ja+2+ b 2j = f f (a), , and y = f (b). Hint: Show this first for f (x) = 1, x, x 2 and use linearity. where y0 = 58, (b) By Exercise S2y1 = 2 f (x) d x if 2f (x) is a quadratic polynomial. Use (a) to show that S N is exact for all 2j S2" = b

x2 j

N if f (x) is a quadratic polynomial. SOLUTION

S E C T I O N 8.2

Integration by Parts

415

(a) This result follows because the even-numbered interior endpoints overlap: (N# −2)/2

2j

S2 =

i=0

=

b−a (y0 + 4y1 + y2 ) + (y2 + 4y3 + y4 ) + · · · 6 b−a y0 + 4y1 + 2y2 + 4y3 + 2y4 + · · · + 4y N −1 + y N = S N . 6

(b) If f (x) is a quadratic polynomial, then by part (a) we have " x2 " x4 " xN " b f (x) d x + f (x) d x + · · · + f (x) d x = f (x) d x. S N = S20 + S22 + · · · + S2N −2 = x0

x2

x N −2

a

61. Use the Error Bound for S N to obtain another proof " that Simpson’s Rule is exact for all cubic polynomials. b

3 2 + cxvalue (4) (x) andany conclude, as in Exercise 59, fthat S N =is 0, exact forcan all that fS(x) Let = gives ax 3 +the bxexact + d,for with a x = d0,x be cubic polynomial. Then, so we 2 also a take K 4 = 0. This yields 4 cubic polynomials. Show by counterexample that S2 is not exact for integrals of x . 0 Error(S N ) ≤ = 0. 180N 4

Show SOLUTION

In other words, S N is exact for all cubic polynomials for all N . Calculate M10 and S10 for the integral

Sometimes, Simpson’s Rule Performs Poorly

8.2whose Integration by Parts value we know to be π (one-quarter of the area of the unit circle).

" 1 0

1 − x 2 d x,

4

(a) We usually expect S N to be more accurate than M N . Which of M10 and S10 is more accurate in this case? Preliminary Questions (b) How do you explain the result of part (a)? Hint: The Error Bounds are not valid because | f (x)| and | f (4) (x)| 1. Which derivative rule is used to derive the Integration by Parts formula? tend to ∞ as x → 1, but | f (4) (x)| goes to infinity faster. SOLUTION The Integration by Parts formula is derived from the Product Rule. 2. For each of the following integrals, state whether substitution or Integration by Parts should be used: " " " " 2 x cos(x 2 ) d x, x cos x d x, x 2 e x d x, xe x d x SOLUTION

"

(a)

"

(b) " (c)

"

(d)

x cos(x 2 ) d x: use the substitution u = x 2 . x cos x d x: use Integration by Parts. x 2 e x d x; use Integration by Parts. 2 xe x d x; use the substitution u = x 2 .

3. Why is u = cos x, v = x a poor choice for evaluating SOLUTION

original.

" x cos x d x?

Transforming v = x into v = 12 x 2 increases the power of x and makes the new integral harder than the

Exercises In Exercises 1–6, evaluate the integral using the Integration by Parts formula with the given choice of u and v . " 1.

x sin x d x;

SOLUTION

u = x, v = sin x

Using the given choice of u and v results in u=x

v = − cos x

u = 1

v = sin x

Using Integration by Parts, " " " x sin x d x = x(− cos x) − (1)(− cos x) d x = −x cos x + cos x d x = −x cos x + sin x + C. " xe2x d x;

u = x, v = e2x

416

CHAPTER 8


" (2x + 9)e x d x;

3.

SOLUTION

u = 2x + 9, v = e x

Using u = 2x + 9 and v = e x gives us u = 2x + 9

v = ex

u = 2

v = ex

Integration by Parts gives us " " (2x + 9)e x d x = (2x + 9)e x − 2e x d x = (2x + 9)e x − 2e x + C = e x (2x + 7) + C. " 5.

" x 3 ln x d x; u = ln x, v = x 3 x cos(4x) d x; u = x, v = cos(4x)

SOLUTION

Using u = ln x and v = x 3 gives us u = ln x

v = 14 x 4

u = 1x

v = x 3

Integration by Parts gives us " " 1 4 1 4 1 x − x x 3 ln x d x = (ln x) dx 4 x 4 " 1 1 1 4 x4 1 x 3 d x = x 4 ln x − x +C = (4 ln x − 1) + C. = x 4 ln x − 4 4 4 16 16 " In Exercises 7–32, use Integration by Parts to evaluate the integral. tan−1 x d x; u = tan−1 x, v = 1 " 7. (3x − 1)e−x d x SOLUTION

Let u = 3x − 1 and v = e−x . Then we have u = 3x − 1

v = −e−x

u = 3

v = e−x

Using Integration by Parts, we get " " (3x − 1)e−x d x = (3x − 1)(−e−x ) − (3)(−e−x ) d x = −e−x (3x − 1) + 3 " 9.

"

e−x d x = −e−x (3x − 1) − 3e−x + C = −e−x (3x + 2) + C.

" x x 2 e x d−x xe d x

SOLUTION

Let u = x 2 and v = e x . Then we have u = x2

v = ex

u = 2x

v = ex

Using Integration by Parts, we get "

" x 2 ex d x = x 2 ex − 2 " We must apply Integration by Parts again to evaluate "

xe x d x.

xe x d x. Taking u = x and v = e x , we get "

xe x d x = xe x −

(1)e x d x = xe x − e x + C.

Plugging this into the original equation gives us " x 2 e x d x = x 2 e x − 2 xe x − e x + C = e x (x 2 − 2x + 2) + C.

S E C T I O N 8.2

" 11.


" x cos22x d x x sin x d x

SOLUTION

Let u = x and v = cos 2x. Then we have u=x

v = 12 sin 2x

u = 1

v = cos 2x

Using Integration by Parts, we get " " 1 1 x cos 2x d x = x sin 2x − (1) sin 2x d x 2 2 " 1 1 1 1 sin 2x d x = x sin 2x + cos 2x + C. = x sin 2x − 2 2 2 4 " 13.

" x dx e−x sin x 2 sin(3x + 1) d x

SOLUTION

Let u = e−x and v = sin x. Then we have u = e−x

v = − cos x

u = −e−x

v = sin x

Using Integration by Parts, we get " " " −x −x −x −x e sin x d x = −e cos x − (−e )(− cos x) d x = −e cos x − e−x cos x d x. " We must apply Integration by Parts again to evaluate "

"

e−x cos x d x = e−x sin x −

e−x cos x d x. Using u = e−x and v = cos x, we get

(−e−x )(sin x) d x = e−x sin x +

"

e−x sin x d x.

Plugging this into the original equation, we get " " e−x sin x d x = −e−x cos x − e−x sin x + e−x sin x d x . " Solving this equation for

e−x sin x d x gives us "

" 15.

1 e−x sin x d x = − e−x (sin x + cos x) + C. 2

" x ln x4xd x e cos 3x d x

SOLUTION

Let u = ln x and v = x. Then we have u = ln x

v = 12 x 2

u = 1x

v = x

Using Integration by Parts, we get " " 1 1 2 1 x ln x d x = x 2 ln x − x dx 2 x 2 " 1 2 1 1 x2 1 1 2 x d x = x ln x − + C = x 2 (2 ln x − 1) + C. = x ln x − 2 2 2 2 2 4 " 17.

" x −9lnlnxx d x dx x2

417

418

CHAPTER 8

T E C H N I Q U E S O F I N T E G R AT I O N SOLUTION

Let u = ln x and v = x −9 . Then we have u = ln x

v = − 18 x −8

u = 1x

v = x −9

Using Integration by Parts, we get " " " ln x 1 1 1 1 x −9 ln x d x = − x −8 ln x − − x −8 d x = − 8 + x −9 d x 8 x 8 8 8x 1 x −8 1 1 ln x + C = − 8 ln x + + C. =− 8 + 8 −8 8 8x 8x " 19.

" x cos(2 − x) d x e x sin(2x) d x

SOLUTION

Let u = x and v = cos(2 − x). Then we have u=x

v = − sin(2 − x)

u = 1

v = cos(2 − x)

Using Integration by Parts, we get " " x cos(2 − x) d x = −x sin(2 − x) − (1)(− sin(2 − x)) d x " = −x sin(2 − x) + " 21.

sin(2 − x) d x = −x sin(2 − x) + cos(2 − x) + C.

" x 2x 2d x x ln x d x

SOLUTION

Let u = x and v = 2x . Then we have u=x

v=

u = 1

2x ln 2

v = 2x

Using Integration by Parts, we get x " x " " 2 x 2x 1 1 2 x 2x 2x 2x 1 x 2x d x = x dx = − 2x d x = − − (1) +C = x− + C. ln 2 ln 2 ln 2 ln 2 ln 2 ln 2 ln 2 ln 2 ln 2 " 23.

" (ln x)2 d x2 x sec x d x

SOLUTION

Let u = (ln x)2 and v = 1. Then we have u = (ln x)2

v=x

2 ln x x

v = 1

u =

Using Integration by Parts, we get " " " 2 (ln x)2 d x = (ln x)2 (x) − ln x x d x = x(ln x)2 − 2 ln x d x. x " We must apply Integration by Parts again to evaluate ln x d x. Using u = ln x and v = 1, we have "

" ln x d x = x ln x −

1 · x d x = x ln x − x

" d x = x ln x − x + C.

Plugging this into the original equation, we get " (ln x)2 d x = x(ln x)2 − 2 (x ln x − x) + C = x (ln x)2 − 2 ln x + 2 + C. "

cos−1 x d x

S E C T I O N 8.2

" 25.


sin−1 x d x

SOLUTION

Let u = sin−1 x and v = 1. Then we have u = sin−1 x 1 u = 1 − x2

v=x v = 1

Using Integration by Parts, we get " " We can evaluate

" 27.

sin−1 x d x = x sin−1 x −

"

x d x. 1 − x2

x d x by making the substitution w = 1 − x 2 . Then dw = −2x d x, and we have 1 − x2 " " " 1 1 −2x d x sin−1 x d x = x sin−1 x + = x sin−1 x + w−1/2 dw 2 2 2 1−x 1 = x sin−1 x + (2w1/2 ) + C = x sin−1 x + 1 − x 2 + C. 2

" x 5x d x−1 sec x d x

SOLUTION

Let u = x and v = 5x . Then we have 5x ln 5

u=x

v=

u = 1

v = 5x

Using Integration by Parts, we get x " " 5 5x x 5x d x = x − (1) dx = ln 5 ln 5 x 1 5 x 5x − = +C = ln 5 ln 5 ln 5 " 29.

" x 5x 1 5x d x − ln 5 ln 5 5x 1 x− + C. ln 5 ln 5

" x cosh 2x dxx (sin x)5 d x

SOLUTION

Let u = x and v = cosh 2x. Then we have u=x

v = 12 sinh 2x

u = 1

v = cosh 2x

Using Integration by Parts, we get " " 1 1 x cosh 2x d x = x sinh 2x − (1) sinh 2x d x 2 2 " 1 1 1 1 sinh 2x d x = x sinh 2x − cosh 2x + C. = x sinh 2x − 2 2 2 4 " 31.

" sinh−1 x−1d x tanh (4x) d x

SOLUTION

Using u = sinh−1 x and v = 1 gives us u = sinh−1 x 1 u = 1 + x2

v=x v = 1

Integration by Parts gives us "

sinh−1 x d x = x sinh−1 x −

"

1

1 + x2

x d x.

419

420

CHAPTER 8


For the integral on the right we’ll use the substitution w = 1 + x 2 , dw = 2x d x. Then we have " " √ 1 dw sinh−1 x d x = x sinh−1 x − √ = x sinh−1 x − w + C 2 w = x sinh−1 x − 1 + x 2 + C. " 33. Use the substitution u = x 1/2 and then Integration by Parts to evaluate (cos x)(cosh x) d x SOLUTION

"

√

e x d x.

Let w = x 1/2 . Then dw = 12 x −1/2 d x, or d x = 2 x 1/2 dw = 2w dw. Now, " " √ e x d x = 2 wew dw.

Using Integration by Parts with u = w and v = ew , we get " 2 wew dw = 2(wew − ew ) + C. Substituting back, we find "

√

√

√ e x d x = 2e x ( x − 1) + C.

" or both if necessary. In Exercises 35–44, evaluate using Integration by Parts, substitution, 3 x2 " Use substitution and then Integration by Parts to evaluate x e d x. 35.

x cos 4xd x

SOLUTION

Let u = x and v = cos 4x. Then we have u=x

v = 14 sin 4x

u = 1

v = cos 4x

Using Integration by Parts, we get " " 1 1 1 1 1 x cos 4x d x = x sin 4x − (1) sin 4x d x = x sin 4x − − cos 4x + C 4 4 4 4 4 =

1 1 x sin 4x + cos 4x + C. 4 16

"

"x dx √ ln(ln x) d x x +1 x SOLUTION Let u = x + 1. Then du = d x, x = u − 1, and " " " " x dx 1 (u − 1) du u = = du = (u 1/2 − u −1/2 ) du √ √ √ −√ u u u x +1

37.

= " 39.

2 3/2 2 u − 2u 1/2 + C = (x + 1)3/2 − 2(x + 1)1/2 + C. 3 3

" cos x ln(sin x) d x sin(ln x) d x

SOLUTION

Let w = sin x. Then dw = cos x d x, and " " cos x ln(sin x) d x = ln w dw.

Now use Integration by Parts with u = ln w and v = 1. Then u = 1/w and v = w, which gives us " " cos x ln(sin x) d x = ln w dw = w ln w − w + C = sin x ln(sin x) − sin x + C. " 41.

" √ sin 3x d2x x (x + 1)12 d x



First use substitution, with w = " sin

√

√

√ x and dw = d x/(2 x). This gives us

" x dx =

√ √ " (2 x) sin x d x = 2 w sin w dw. √ (2 x)

Now use Integration by Parts, with u = w and v = sin w. Then we have " " " √ sin x d x = 2 w sin w dw = 2 −w cos w − − cos w dw = 2(−w cos w + sin w) + C = 2 sin "

√

√ √ x − 2 x cos x + C.

ln(ln x) ln x d x " √ √x xex d x

43.

SOLUTION

Let w = ln x. Then dw = d x/x, and " " ln(ln x) ln x d x = w ln w dw. x

Now use Integration by Parts, with u = ln w and v = w. Then, 1 2 w 2

u = ln w

v=

u = w −1

v = w

and "

1 ln(ln x) ln x d x 1 = w 2 ln w − x 2 2 =

"

1 1 w dw = w 2 ln w − 2 2

w2 2

+C

1 1 1 (ln x)2 ln(ln x) − (ln x)2 + C = (ln x)2 [2 ln(ln x) − 1] + C. 2 4 4

" In Exercises 45–50, compute the definite integral. −1 x d x x tan " 2 45. xe9x d x 0 SOLUTION

47.

Let u = x and v = e9x . Then u = 1 and v = 19 e9x . Using Integration by Parts, 2 " 2 2 " 2 1 1 1 9x 1 xe9 x d x = xe9x − (1) e9x d x = xe − e9x 9 9 9 81 0 0 0 0 1 2 18 1 18 1 1 e − e (18e18 − e18 + 1) = (17e18 + 1). = − 0 − (1) = 9 81 81 81 18

" 4 "√ x 3 4 − x dx ln x d x 0 1

SOLUTION

Let u = 4 − x. Then x = 4 − u, du = −d x and " 4 √ " u=0 " u=0 x 4 − x dx = (4 − u)u 1/2 (−du) = − (4u 1/2 − u 3/2 ) du 0

u=4

= =

49.

" 4 "√π /4 x ln x d x x sin(2x) d x 1 0

√

u=0

(4u 1/2 − u 3/2 ) du =

u=4

2u 3/2 2u 5/2 (4) − 3 5

4 0

8 3/2 2 64 64 128 (4 ) − (45/2 ) = − = . 3 5 3 5 15

x. Then u = 1/x and v = 23 x 3/2 . Using Integration by Parts, 4 " 4 4 " 4 √ 2 2 3/2 2 2 2 1/2 x ln x d x = x 3/2 ln x − dx = ln x − · x 3/2 x x 3 3 3 3 1 1 3 1 1

SOLUTION

Let u = ln x and v =

" u=4

421

422

CHAPTER 8


= 1 (5) to evaluate 51. Use"Eq. tan−1 x d x

2 3/2 x 3

2 3/2 2 4 2 2 2 16 28 = ln x − ln 4 − 4 − (1) 0 − = ln 4 − . 3 1 3 3 3 3 3 9

" x 4 e x d x.

0

SOLUTION

"

" x 4 ex d x = x 4 ex − 4

" x 3 ex d x = x 4 ex − 4 x 3 ex − 3 x 2 ex d x "

= x 4 e x − 4x 3 e x + 12

" x 2 e x d x = x 4 e x − 4x 3 e x + 12 x 2 e x − 2 xe x d x "

= x 4 e x − 4x 3 e x + 12x 2 e x − 24

" xe x d x = x 4 e x − 4x 3 e x + 12x 2 e x − 24 xe x − e x d x

= x 4 e x − 4x 3 e x + 12x 2 e x − 24 xe x − e x + C. Thus, " x 4 e x d x = e x (x 4 − 4x 3 + 12x 2 − 24x + 24) + C. "

" (5). 53. Find a reduction formula for x n e−x d x similar to 4Eq. Use substitution and then Eq. (5) to evaluate x e7x d x. SOLUTION

Let u = x n and v = e−x . Then u = xn

v = −e−x

u = nx n−1

v = e−x

Using Integration by Parts, we get " " " x n e−x d x = −x n e−x − nx n−1 (−e−x ) d x = −x n e−x + n x n−1 e−x d x. " indicate a good method for evaluating the integral (but do not evaluate). " In Exercises 55–62, Your choices are algebraic −1 ln x d x? Evaluate x n ln x (specify d x for nu = −1. Which method should be used to evaluate manipulation, substitution and du), and Integration by Parts (specify u and v ). xIf it appears that the techniques you have learned thus far are not sufficient, state this. " √ 55. x ln x d x SOLUTION

" 57.

Use Integration by Parts, with u = ln x and v =

√

x.

" x3 √ x 2 − d xx dx 4 − x2

SOLUTION

2x Use substitution, followed by algebraic manipulation: Let u = 4 − x 2 . Then du = −2x d x, x 2 = 4 − u,

and "

"

x3 1 dx = − 2 2 4−x

"

1 (x 2 )(−2x d x) =− √ 2 u

"

1 (4 − u)(du) =− √ 2 u

"

" 2x + 3 dx dx 3x + 6 x2 + 4 − x2 SOLUTION Use substitution, with u = x 2 + 3x + 6 and du = (2x + 3) d x. " " 61. x sin(3x + 1 4) d x dx 2 x + 3x + 6 SOLUTION Use Integration by Parts, with u = x and v = sin(3x + 4). " " 63. Evaluate (sin−1 x)2 d x. Hint: First use Integration by Parts and then substitution. x cos 3x d x 59.

u 4 √ −√ u u

du.



423

First use integration by parts with v = 1 to get "

"

x sin−1 x d x . 1 − x2 Now use substitution on the integral on the right, with u = sin−1 x. Then du = d x/ 1 − x 2 and x = sin u, and we get (using Integration by Parts again) (sin−1 x)2 d x = x(sin−1 x)2 − 2

"

" x sin−1 x d x = u sin u du = −u cos u + sin u + C = − 1 − x 2 sin−1 x + x + C. 1 − x2 where cos u = 1 − sin2 u = 1 − x 2 . So the final answer is " (sin−1 x)2 d x = x(sin−1 x)2 + 2 1 − x 2 sin−1 x − 2x + C. " 4 65. Evaluate x"7 cos(x (ln x)2) ddxx. . Hint: First use substitution and then Integration by Parts. Evaluate x2 4 SOLUTION First, let w = x . Then dw = 4x 3 d x and " " 1 w cos x dw. x 7 cos(x 4 ) d x = 4 Now, use Integration by Parts with u = w and v = cos w. Then " " 1 1 1 1 1 x 7 cos(x 4 ) d x = w sin w − sin w dw = w sin w + cos w + C = x 4 sin(x 4 ) + cos(x 4 ) + C. 4 4 4 4 4 67. Find the volume of the solid obtained by revolving y = e x for 0 ≤ x ≤ 2 about the y-axis. Find f (x), assuming that SOLUTION By the Method of Cylindrical " solid is " Shells, the volume V of the x d x = f (x)e x − −1 x f"(x)e " 2 x e dx b (2π r )h d x = 2π xe x d x. V = a

0

Using Integration by Parts with u = x and v = e x , we find 2 V = 2π (xe x − e x ) = 2π (2e2 − e2 ) − (0 − 1) = 2π (e2 + 1). 0

69. Recall that the present value of an investment which pays out income continuously at a rate R(t) for T years is " T Find the volume of the solid obtained by revolving y = cos x for 0 ≤ x ≤ π2 about the y-axis. −r t R(t)e dt, where r is the interest rate. Find the present value if income is produced at a rate R(t) = 5,000 + 100t 0

dollars/year for 10 years. SOLUTION

The present value is given by

PV =

" T 0

R(t)e−r t dt =

" 10 0

(5000 + 100t)e−r t dt = 5000

" 10 0

e−r t dt + 100

" 10 0

Using Integration by Parts for the integral on the right, with u = t and v = e−r t , we find ) * " 10 −1 −r t 1 −r t 10 t −r t 10 P V = 5000 − e − e dt e + 100 − r r r 0 0 0 10 5000 −r t 10 100 1 e − te−r t + e−r t r r r 0 0 5000 −10r 100 1 1 −10r −10r =− − 1) − + e − 0+ (e 10e r r r r 5000 1000 100 5000 100 = e−10r − − − 2 + + 2 r r r r r = −

=

5000r + 100 − e−10r (6000r + 100) . r2

Prove the reduction formula

"

"

te−r t dt.

424

CHAPTER 8


" 71. Use Eq. (6) to calculate

(ln x)k d x for k = 2, 3.

SOLUTION

"

" (ln x)2 d x = x(ln x)2 − 2

"

ln x d x = x(ln x)2 − 2(x ln x − x) + C = x(ln x)2 − 2x ln x + 2x + C; "

(ln x)3 d x = x(ln x)3 − 3

(ln x)2 d x = x(ln x)3 − 3 x(ln x)2 − 2x ln x + 2x + C

= x(ln x)3 − 3x(ln x)2 + 6x ln x − 6x + C. " 1 x Prove xb thex reduction d x = b x formulas 73. Prove − 2 + C. ln b ln " b " x n cos x d1xand = xvn = sinbxx /−lnnb. Using x n−1 Integration sin x d x by Parts, we get SOLUTION Let u = x and v = b x . Then u = " " " " x xb x xb x n 1 1 n x bx 1 n−1 x bx d x = + C. − = cos bx x d x − x = −x− cos x ·+ n +xC x bsindxxd= ln b ln b ln b ln b ln b ln b (ln b)2

Further Insights and Challenges 75. Prove in two ways The Integration by Parts formula can be written " " a " a " f (x) d x = a f (a) − x f (x) d x u(x)v(x) d x = u(x)V (x)0− u (x)V (x) d x 0

8

" a V (x) satisfies V (x) = v(x). Firstwhere use Integration by Parts. Then assume f (x) is increasing. Use the substitution u = f (x) to prove that x f (x) d x 0 (a) to Show directly thatshaded the right-hand of Eq. (7)derive does not if V (x) is replaced by V (x) + C, where C is a is equal the area of the region inside Figure 1 and Eq.change (8) a second time. constant. " (b) Use u = tan−1 x and v = x in Eq. (7) to calculate y

f(a)

x tan−1y x= fd(x)x, but carry out the calculation twice: first with

V (x) = 12 x 2 and then with V (x) = 12 x 2 + 12 . Which choice of V (x) results in a simpler calculation? f(0) x 0

a

FIGURE 1 SOLUTION

Let u = f (x) and v = 1. Then Integration by Parts gives a " a " a " a f (x) d x = x f (x) − x f (x) d x = a f (a) − x f (x) d x. 0

0

0

0

Alternately, let u = f (x). Then du = f (x) d x, and if f (x) is either increasing or decreasing, it has an inverse function, and x = f −1 (u). Thus, " x=a x=0

x f (x) d x =

" f (a) f (0)

f −1 (u) du

which is precisely the area of the shaded region in Figure 1 (integrating along!the vertical axis). Since the area of the entire rectangle is a f (a), the difference between the areas of the two regions is 0a f (x) d x. " 1 that f (0) =−f (1) and that exists. Prove x a (1 x)b = d x,0 where a, bf are whole numbers. 77. Set Assume I (a, b) = 0 " " 1 (a) Use substitution to show that I (a, b) = I1(b,a). f (x) f (x) d x = − f (x)2 d x 1 0 0 . (b) Show that I (a, 0) = I (0, a) = a+1 Use this to prove that if f (0) = f (1) = 0 and f (x) = λ f (x) for some constant λ , then λ < 0. Can you think of a (c) Prove that for a ≥ 1 and b ≥ 0, function satisfying these conditions for some λ ? a I (a, b) = I (a − 1, b + 1) b+1 (d) Use (b) and (c) to calculate I (1, 1) and I (3, 2). a! b! . (e) Show that I (a, b) = (a + b + 1)! SOLUTION

S E C T I O N 8.2


(a) Let u = 1 − x. Then du = −d x and " 1 " u=0 (1 − u)a u b (−du) = u b (1 − u)a du = I (b, a). I (a, b) = u=1

0

(b) I (a, 0) = I (0, a) by part (a). Further, I (a, 0) =

" 1 0

x a (1 − x)0 d x =

" 1 0

xa dx =

1 . a+1

(c) Using Integration by Parts with u = (1 − x)b and v = x a gives 1 x a+1 b I (a, b) = (1 − x) + a + 1

0

" 1 b b x a+1 (1 − x)b−1 d x = I (a + 1, b − 1). a+1 0 a+1

The other equality arises from Integration by Parts with u = x a and v = (1 − x)b . (d) I (1, 1) =

1 1 1 1 1 I (1 − 1, 1 + 1) = I (0, 2) = · = 1+1 2 2 3 6

I (3, 2) =

1 1 1 1 1 1 I (4, 2) = · I (5, 0) = · = . 2 2 5 10 6 60

(e) We proceed as follows: I (a, b) =

a a−1 a I (a − 1, b + 1) = · I (a − 2, b + 2) b+1 b+1 b+2

.. . =

a a−1 1 · ··· I (0, b + a) b+1 b+2 b+a

=

1 a(a − 1) · · · (1) · (b + 1)(b + 2) · · · (b + a) b + a + 1

=

a! b! b! a! = . b! (b + 1)(b + 2) · · · (b + a)(b + a + 1) (a + b + 1)!

" " n cos(x 2 ) d x and J = n 2 ) d x. 79. LetShow In = by xdifferentiation n that if Pn (x) xis asin(x polynomial of degree n satisfying Pn (x) + Pn (x) = x n , then " (a) Find a reduction formula that expresses In in terms of Jn−2 . Hint: Write x n cos(x 2 ) as x n−1 (x cos(x 2 )). x n e x d x = Pn (x) e x + C Use the result of (a) to show that In can be evaluated explicitly if n is odd. (b) (c) Evaluate I3 .for n = 2, 3, 4. Find Pn (x) SOLUTION

(a) Integration by Parts with u = x n−1 and v = x cos(x 2 ) d x yields " 1 n−1 1 n−1 x n−2 sin(x 2 ) d x = x n−1 sin(x 2 ) − Jn−2 . In = x n−1 sin(x 2 ) − 2 2 2 2 (b) If n is odd, the reduction process will eventually lead to either " " x cos(x 2 ) d x or x sin(x 2 ) d x, both of which can be evaluated using the substitution u = x 2 . (c) Starting with the reduction formula from part (a), we find " 1 2 1 1 x sin(x 2 ) d x = x 2 sin(x 2 ) + cos(x 2 ) + C. I3 = x 2 sin(x 2 ) − 2 2 2 2

425

426

CHAPTER 8


8.3 Trigonometric Integrals Preliminary Questions

"

1. Describe the technique used to evaluate

sin5 x d x.

Because the sine function is raised to an odd power, rewrite sin5 x = sin x sin4 x = sin x(1 − cos2 x)2 and then substitute u = cos x. " 2. Describe a way of evaluating sin6 x d x. SOLUTION

Repeatedly use the reduction formula for powers of sin x. " 3. Are reduction formulas needed to evaluate sin7 x cos2 x d x? Why or why not?

SOLUTION

No, a reduction formula is not needed because the sine function is raised to an odd power. " 4. Describe a way of evaluating sin6 x cos2 x d x.

SOLUTION

SOLUTION Because both trigonometric functions are raised to even powers, write cos2 x = 1 − sin2 x and then apply the reduction formula for powers of the sine function.

5. Which integral requires more work to evaluate? " sin798 x cos x d x

" or

sin4 x cos4 x d x

Explain your answer. SOLUTION The first integral can be evaluated using the substitution u = sin x, whereas the second integral requires the use of reduction formulas. The second integral therefore requires more work to evaluate.

Exercises In Exercises 1–6, use the method for odd powers to evaluate the integral. " cos3 x d x.

1.

SOLUTION

Use the identity cos2 x = 1 − sin2 x to rewrite the integrand: " "

cos3 x d x = 1 − sin2 x cos x d x.

Now use the substitution u = sin x, du = cos x d x: " "

1 1 cos3 x d x = 1 − u 2 du = u − u 3 + C = sin x − sin3 x + C. 3 3 " 3.

" sin3 θ 5cos2 θ d θ sin x d x

SOLUTION

Write sin3 θ = sin2 θ sin θ = (1 − cos2 θ ) sin θ . Then " "

sin3 θ cos2 θ d θ = 1 − cos2 θ cos2 θ sin θ d θ .

Now use the substitution u = cos θ , du = − sin θ d θ : " "

"

sin3 θ cos2 θ d θ = − 1 − u 2 u 2 du = − u 2 − u 4 du 1 1 1 1 = − u 3 + u 5 + C = − cos3 θ + cos5 θ + C. 3 5 3 5 " 5.

" sin3 t cos3 t 5dt cos x sin x d x


Trigonometric Integrals

427

Write sin3 t = (1 − cos2 t) sin t dt. Then " " "

sin3 t cos3 t dt = (1 − cos2 t) cos3 t sin t dt = cos3 t − cos5 t sin t dt.

Now use the substitution u = cos t, du = − sin t dt: "

" 1 1 1 1 u 3 − u 5 du = − u 4 + u 6 + C = − cos4 t + cos6 t + C. sin3 t cos3 t dt = − 4 6 4 6 " the area of the shaded region in Figure 1. 7. Find sin2 x cos5 x d x y

1

y = cos3 x

p 2

p

x 3p 2

−1

FIGURE 1 Graph of y = cos3 x.

First evaluate the indefinite integral by writing cos3 x = (1 − sin2 x) cos x, and using the substitution u = sin x, du = cos x d x: "

"

" 1 1 cos3 x d x = 1 − sin2 x cos x d x = 1 − u 2 du = u − u 3 + C = sin x − sin3 x + C. 3 3 SOLUTION

The area is given by π /2 3π /2 1 1 sin x − sin3 x − sin x − sin3 x 3 3 0 π /2 π /2 0 π π 3π π π 1 3π 1 1 = sin − sin3 − 0 − sin − sin3 − sin − sin3 2 3 2 2 3 2 2 3 2

A=

" π /2

cos3 x d x −

" 3π /2

cos3 x d x =

1 1 1 = 1 − (1)3 − (−1) + (−1)3 + 1 − (1)3 = 2. 3 3 3 " In Exercises 9–12, evaluate the integral using the methods employed in Examples 3 and 4. 2 2 2 2 " Use the identity sin x = 1 − cos x to write sin x cos x d x as a sum of two integrals and then evaluate using 4 y dy 9. thecos reduction formula.

Using the reduction formula for cosm y, we get " " " 1 3 1 3 1 1 4 3 2 3 cos y d y = cos y sin y + dy cos y d y = cos y sin y + cos y sin y + 4 4 4 4 2 2

SOLUTION

= " 11.

1 3 3 cos3 y sin y + cos y sin y + y + C. 4 8 8

" 2 x dx sin4 x cos cos2 θ sin2 θ d θ

SOLUTION

Use the identity cos2 x = 1 − sin2 x to write: " " " "

4 2 4 2 4 sin x cos x d x = sin x 1 − sin x d x = sin x d x − sin6 x d x.

Using the reduction formula for sinm x: " " " 1 5 sin4 x d x sin4 x cos2 x d x = sin4 x d x − − sin5 x cos x + 6 6 " " 1 5 1 1 5 1 3 1 3 4 2 sin x d x = sin x cos x + sin x d x = sin x cos x + − sin x cos x + 6 6 6 6 4 4 " 1 1 1 = sin5 x cos x − sin2 x d x sin3 x cos x + 6 24 8 " 1 1 1 1 1 dx = sin5 x cos x − sin3 x cos x + − sin x cos x + 6 24 8 2 2

428

CHAPTER 8


=

1 5 1 1 1 sin x cos x − sin3 x cos x − sin x cos x + x + C. 6 24 16 16

" In Exercises 13–20, evaluate the integral using the reduction formulas on p. 438 as necessary. sin4 x cos6 x d x " 13. tan2 x sec2 x d x SOLUTION

" 15.

Use the substitution u = tan x, du = sec2 x d x. Then " " 1 1 tan2 x sec2 x d x = u 2 du = u 3 + C = tan3 x + C. 3 3

" tan4 θ 3sec2 θ 2d θ tan x sec x d x

SOLUTION

Use the substitution u = tan θ , du = sec2 θ d θ . Then " " 1 1 tan4 θ sec2 θ d θ = u 4 du = u 5 + C = tan5 θ + C. 5 5

"

" tan2 t dt tan2 x sec x d x SOLUTION Using the reduction formula for tanm t, we get " " tan2 t dt = tan t − dt = tan t − t + C.

17.

"

" sec3 x 3d x cot x d x SOLUTION Using the reduction formula for secm x, we get " " 1 1 1 1 sec x d x = tan x sec x + ln | sec x + tan x| + C. sec3 x d x = tan x sec x + 2 2 2 2

19.

" In Exercises 21–54, use the techniques and reduction formulas necessary to evaluate the integral. csc2 x d x " 21. cos5 x sin x d x SOLUTION

Use the substitution u = cos x, du = − sin x d x. Then " " 1 1 cos5 x sin x d x = − u 5 du = − u 6 + C = − cos6 x + C. 6 6

"

" dx cos4 (3x) cos3 2x sin 2x d x SOLUTION Use the substitution u = 3x, du = 3 d x, followed by the reduction formula for cosm x: " " " " 1 1 1 1 1 1 1 cos4 u du = cos3 u sin u + cos2 u du = cos3 u sin u + cos u sin u + du cos4 (3x) d x = 3 12 4 12 4 2 2 23.

= " 25.

1 1 1 3 1 1 cos3 u sin u + cos u sin u + u + C = cos3 (3x) sin(3x) + cos(3x) sin(3x) + x + C. 12 8 8 12 8 8

" ) sin4 (πθ ) d θ cos3 (πθ cos7 3x d x

SOLUTION

Use the substitution u = πθ , du = π d θ , and the identity cos2 u = 1 − sin2 u to write " " "

1 1 cos3 (πθ ) sin4 (πθ ) d θ = cos3 u sin4 u du = 1 − sin2 u sin4 u cos u du. π π

Now use the substitution w = sin u, dw = cos u du: "

"

" 1 5 1 1 1 7 1 − w2 w 4 dw = w4 − w 6 dw = cos3 (πθ ) sin4 (πθ ) d θ = w − w +C π π 5π 7π =

1 1 sin5 (πθ ) − sin7 (πθ ) + C. 5π 7π

S E C T I O N 8.3

" 27.


" dx sin4 (3x) cos498 y sin3 y d y

SOLUTION

Use the substitution u = 3x, du = 3 d x and the reduction formula for sinm x: " " " 1 1 1 sin4 u du = − sin3 u cos u + sin2 u du sin4 (3x) d x = 3 12 4 " 1 1 1 1 du = − sin3 u cos u + − sin u cos u + 12 4 2 2 1 1 1 sin3 u cos u − sin u cos u + u + C 12 8 8 1 1 3 = − sin3 (3x) cos(3x) − sin(3x) cos(3x) + x + C. 12 8 8

=−

" 29.

" sec 7t dt sin2 x cos6 x d x

SOLUTION

" 31.

" 2 x dx tan x sec csc3 x d x

SOLUTION

" 33.

Use the substitution u = 7t, du = 7 dt. Then " " 1 1 1 sec u du = ln | sec u + tan u| + C = ln | sec 7t + tan 7t| + C. sec 7t dt = 7 7 7

Use the substitution u = tan x, du = sec2 x d x. Then " " 1 1 tan x sec2 x d x = u du = u 2 + C = tan2 x + C. 2 2

" tan5 x 3sec4 x d3 x tan θ sec θ d θ

SOLUTION

Use the identity tan2 x = sec2 x − 1 to write "

" 2 sec2 x − 1 sec3 x(sec x tan x d x). tan5 x sec4 x d x =

Now use the substitution u = sec x, du = sec x tan x d x: " "

"

2 tan5 x sec4 x d x = u 2 − 1 u 3 du = u 7 − 2u 5 + u 3 du = " 35.

1 8 1 6 1 4 1 1 1 u − u + u + C = sec8 x − sec6 x + sec4 x + C. 8 3 4 8 3 4

" tan4 x 2sec x d4x tan x sec x d x

SOLUTION

Use the identity tan2 x = sec2 x − 1 to write " "

" " " 2 tan4 x sec x d x = sec2 x − 1 sec x d x = sec5 x d x − 2 sec3 x d x + sec x d x.

Now use the reduction formula for secm x: " " " " 1 3 sec3 x d x − 2 sec3 x d x + sec x d x tan4 x sec x d x = tan x sec3 x + 4 4 " " 5 1 sec3 x d x + sec x d x = tan x sec3 x − 4 4 " " 5 1 1 1 = tan x sec3 x − tan x sec x + sec x d x + sec x d x 4 4 2 2 " 5 3 1 sec x d x = tan x sec3 x − tan x sec x + 4 8 8 = " tan6 x sec4 x d x

5 1 3 tan x sec3 x − tan x sec x + ln | sec x + tan x| + C. 4 8 8

429

430

CHAPTER 8


" tan4 x sec3 x d x

37.

Use the identity tan2 x = sec2 x − 1 to write "

" " " " 2 sec2 x − 1 sec3 x d x = sec7 x d x − 2 sec5 x d x + sec3 x d x. tan4 x sec3 x d x =

SOLUTION

Now use the reduction formula for secm x: " " " " 1 5 4 3 5 5 tan x sec x + sec x d x − 2 sec5 x d x + sec3 x d x tan x sec x d x = 6 6 " " 7 1 5 5 sec x d x + sec3 x d x = tan x sec x − 6 6 " " 7 1 3 1 sec3 x d x + sec3 x d x = tan x sec5 x − tan x sec3 x + 6 6 4 4 " 7 1 1 sec3 x d x = tan x sec5 x − tan x sec3 x + 6 24 8 " 1 7 1 1 1 tan x sec3 x + tan x sec x + sec x d x = tan x sec5 x − 6 24 8 2 2 = " 39.

" sin 2x 2cos 2x 2d x tan x csc x d x

SOLUTION

" 41.

1 7 1 1 tan x sec5 x − tan x sec3 x + tan x sec x + ln | sec x + tan x| + C. 6 24 16 16

Use the substitution u = sin 2x, du = 2 cos 2x d x: " " " 1 1 1 1 sin 2x cos 2x d x = sin 2x(2 cos 2x d x) = u du = u 2 + C = sin2 2x + C. 2 2 4 4

" sin 2x cos 4x d x cos 4x cos 6x d x

SOLUTION

"

Use the formula for

" sin 2x cos 4x d x = −

sin mx cos nx d x:

cos(2 + 4)x cos(−2x) cos 6x 1 1 cos(2 − 4)x − +C =− − + C = cos 2x − cos 6x + C. 2(2 − 4) 2(2 + 4) −4 12 4 12

Here we’ve used the fact that cos x is an even function: cos(−x) = cos x. " " 3 tan (ln t) 43. 2 ) dt t tcos3 (tdt SOLUTION

Use the substitution u = ln t, du = 1t dt, followed by the reduction formula for tann x: "

tan3 (ln t) dt = t =

45.

" 2"π 0

2 dx sin cos2x(sin t) cos t dt

SOLUTION

Use the formula for " 2π 0

47.

" π /3 " π /2 3 sin x d3 x cos x d x 0 0

!

" tan3 u du =

1 tan2 u − 2

" tan u du

1 1 tan2 u − ln | sec u| + C = tan2 (ln t) − ln | sec(ln t)| + C. 2 2

sin2 x d x:

sin2 x d x =

x 2π sin 2x 2π sin 4π 0 sin 0 = − − − − = π. 2 4 2 4 2 4 0

S E C T I O N 8.3


431

Use the reduction formula for sinm x:

SOLUTION

" π /3 0

π /3

" 2 π /3 + sin x d x 3 0 0 ⎤ √ 2 π /3 1 2 1 5 3 1 2 =− − − 0⎦ − cos x −1 = . 2 2 3 8 3 2 24 0

1 sin3 x d x = − sin2 x cos x 3

⎡

= ⎣−

1 3

" π /2 dx " π /4 dx π /4 sin x cos x 0 SOLUTION Use the definition of csc x to simplify the integral:

49.

" π /2 " π /2 π /2 √ √ dx csc x d x = ln | csc x − cot x| = ln |1 − 0| − ln 2 − 1 = − ln 2 − 1 = sin x π /4 π /4 π /4 √ √ 1 ( 2 + 1) = ln( 2 + 1). = ln √ = ln √ √ 2−1 ( 2 − 1)( 2 + 1)

51.

" π /4 " π /3 5 tan x d x tan x d x 0 0

First use the reduction formula for tanm x to evaluate the indefinite integral: " " " 1 1 1 tan5 x d x = tan4 x − tan3 x d x = tan4 x − tan2 x − tan x d x 4 4 2

SOLUTION

=

1 1 tan4 x − tan2 x + ln | sec x| + C. 4 2

Now compute the definite integral: " π /4 0

π /4 1 1 tan4 x − tan2 x + ln | sec x| 4 2 0

√ 1 4 1 2 1 − 1 + ln 2 − (0 − 0 + ln 1) = 4 2

tan5 x d x =

=

53.

" π "sin π /4 3x cos44x d x sec x d x 0 −π /4

SOLUTION

" π 0

Use the formula for

!

√ 1 1 1 1 − + ln 2 − 0 = ln 2 − . 4 2 2 4

sin mx cos nx d x:

cos(3 − 4)x cos(−x) cos 7x π cos(3 + 4)x π = − − − 2(3 − 4) 2(3 + 4) −2 14 0 0 π 1 1 1 1 1 1 6 = cos x − cos 7x = (−1) − (−1) − (1) − (1) = − . 2 14 2 14 2 14 7 0

sin 3x cos 4x d x =

−

55. Use"the π identities for sin 2x and cos 2x listed on page 434 to verify that the first of the following formulas is equivalent to the second: sin x sin 3x d x " 0 1 sin4 x d x = (12x − 8 sin 2x + sin 4x) + C 32 " 1 3 3 sin4 x d x = − sin3 x cos x − sin x cos x + x + C 4 8 8 SOLUTION

First, observe sin 4x = 2 sin 2x cos 2x = 2 sin 2x(1 − 2 sin2 x) = 2 sin 2x − 4 sin 2x sin2 x = 2 sin 2x − 8 sin3 x cos x.

432

CHAPTER 8


Then 1 3 3 1 (12x − 8 sin 2x + sin 4x) + C = x − sin 2x − sin3 x cos x + C 32 8 16 4 3 3 1 = x − sin x cos x − sin3 x cos x + C. 8 8 4 " In Exercises 57–60, evaluate the identity cot2 x + 1 = csc2 x and methods similar to those for integrating 2 x cos3 using Evaluate sin x d x using the method described in the text and verify that your result is equivalent to the m n tan x sec x. " following result produced by a computer algebra system: 57. cot3 x csc x d x " 1 sin2 x cos3 x d x = (7 + 3 cos 2x) sin3 x + C SOLUTION Use the identity cot2 x = csc2 x − 1 to write 30 " cot3 x csc x d x =

"

csc2 x − 1 csc x cot x d x.

Now use the substitution u = csc x, du = − csc x cot x d x: " "

"

1 1 cot3 x csc x d x = − u 2 − 1 du = 1 − u 2 du = u − u 3 + C = csc x − csc3 x + C. 3 3 " 59.

" cot2 x 4csc2 x 2d x cot x csc x d x

SOLUTION

Use the substitution u = cot x, du = − csc2 x d x: " " "

1 1 cot2 x csc2 x d x = − cot2 x − csc2 x d x = − u 2 du = − u 3 + C = − cot3 x + C. 3 3

61. Find" the volume of the solid obtained by revolving y = sin x for 0 ≤ x ≤ π about the x-axis. cot2 x csc3 x d x SOLUTION Using the disk method, the volume is given by V =

" π 0

π (sin x)2 d x = π

" π 0

sin2 x d x = π

π2 π sin 2x π x − − 0 − (0) = . =π 2 4 2 2 0

" 63. Here is Integration another reduction forthe evaluating theformulas integral Eqs. J = (3) sin x cosn x d x when m and n are even. Use Use by Partsmethod to prove reductions andm (4). the identities sin2 x = 1 4

"

1 (1 − cos 2x), 2

cos2 x =

1 (1 + cos 2x) 2

(1 − cos 2x)m/2 (1 + cos 2x)n/2 d x. Then expand the right-hand side as a sum of integrals involving " smaller powers of sine and cosine in the variable 2x. Use this method to evaluate J = sin2 x cos2 x d x.

to write J =

Using the identities sin2 x = 12 (1 − cos 2x) and cos2 x = 12 (1 + cos 2x), we have " " 1 (1 − cos 2x)(1 + cos 2x) d x J = sin2 x cos2 x d x = 4 " " " 1

1 1 = 1 − cos2 2x d x = dx − cos2 2x d x. 4 4 4 " Now use the substitution u = 2x, du = 2 d x, and the formula for cos2 u du: SOLUTION

J= =

1 1 x− 4 8

" cos2 u du =

1 1 x− 4 8

u 1 + sin u cos u + C 2 2

1 1 1 1 1 x − (2x) − sin 2x cos 2x + C = x − sin 2x cos 2x + C. 4 16 16 8 16

" " 2 x d x. 65. Use the method of Exercise 63 to evaluate sin4 x cos Use the method of Exercise 63 to evaluate cos4 x d x.



433

Using the identities sin2 x = 12 (1 − cos 2x) and cos2 x = 12 (1 + cos 2x), we have " " 1 (1 − cos 2x)2 (1 + cos 2x) d x J = sin4 x cos2 x d x = 8 " 1

= 1 − 2 cos 2x + cos2 2x (1 + cos 2x) d x 8 " 1

1 − cos 2x − cos2 2x + cos3 2x d x. = 8

Now use the substitution u = 2x, du = 2 d x, together with the reduction formula for cosm x: " " " 1 1 1 1 cos u du − cos2 u du + cos3 u du J= x− 8 16 16 16 " 2 1 1 1 u 1 1 1 cos u du = x− sin u − + sin u cos u + cos2 u sin u + 8 16 16 2 2 16 3 3 1 1 1 1 1 1 x− sin 2x − (2x) − sin 2x cos 2x + cos2 2x sin 2x + sin 2x + C 8 16 32 32 48 24 1 1 1 1 = x− sin 2x − sin 2x cos 2x + cos2 2x sin 2x + C. 16 48 32 48 =

67. Prove the reduction formula Show that for n ≥ 2,

"

" k−1 x k x d x = tan " πtan "tan π /2 /2 k−2 x d x − n−1 sinn x d x k=− 1 sinn−2 x d x n 0 0 Hint: Use the identity tan2 x = (sec2 x − 1) to write tank x = (sec2 x − 1) tank−2 x. SOLUTION

Use the identity tan2 x = sec2 x − 1 to write " " " "

tank x d x = tank−2 x sec2 x − 1 d x = tank−2 x sec2 x d x − tank−2 x d x.

Now use the substitution u = tan x, du = sec2 x d x: " " " tank x d x = u k−2 du − tank−2 x d x =

" " 1 tank−1 x u k−1 − tank−2 x d x = − tank−2 x d x. k−1 k−1 " " 69. Use the substitution u = csc x − cot x to evaluate csc x d x (see Example 5). Evaluate I = sin2 x cos4 x d x using (15). Show: " u = csc x − cot x, SOLUTION Using the substitution 1 1 2 2 (a) I = sin3 x cos3 x + sin x cos x d x 2 6 2 du = − csc x cot x "+ csc x d x = csc x(csc x − cot x) d x, " 2 x cos2 x d x = 1 sin3 x cos x + 1 sin2 x d x (b) sin we have 4 4 " " " csc x(csc x − cot x) d x du csc x d x = = = ln |u| + C = ln | csc x − cot x| + C. csc x − cot x u 71. Let m, n be integers with m = ±n. Use formulas (25)–(27) in the table of trigonometric integrals to prove that Total Energy A 100-W " πlight bulb has resistance R "=π144 (ohms) when attached to household current, = 110 V, fcos = mx 60 cos Hz).nxThe is P = where the voltage varies as V =sinVmx 0 sin(2 sin πnxf t) d x(V =0 0, d x power = 0 supplied to the "bulb T 0 0 V 2 /R (joules per second) "and the total energy expended over a time period [0, T ] (in seconds) is U = P(t) dt. 2π

0

Compute U if the bulb remains sin on mx for cos 5 hours. nx d x = 0 0

These formulas, known as the orthogonality relations, play a basic role in the theory of Fourier Series (Figure 2). y

y

p

y = sin 2x sin 4x

x

p

x 2p

y = sin 3x cos 4x

FIGURE 2 By the orthogonality relations, the signed area under these graphs is zero.

434

CHAPTER 8

T E C H N I Q U E S O F I N T E G R AT I O N SOLUTION If m, n are integers, then m − n and m + n are integers, and therefore sin(m − n)π = sin(m + n)π = 0, since sin k π = 0 if k is an integer. Thus we have " π sin(m − n)x sin(m − n)π sin(m + n)x π sin(m + n)π sin mx sin nx d x = = − − − 0 = 0; 2(m − n) 2(m + n) 0 2(m − n) 2(m + n) 0 " π sin(m − n)x sin(m − n)π sin(m + n)x π sin(m + n)π cos mx cos nx d x = = + + − 0 = 0. 2(m − n) 2(m + n) 0 2(m − n) 2(m + n) 0

If k is an integer, then cos 2k π = 1. Using this fact, we have " 2π cos(m − n)x cos(m + n)x 2π sin mx cos nx d x = − − 2(m − n) 2(m + n) 0 0 cos(m − n)2π cos(m + n)2π 1 1 = − − − − − 2(m − n) 2(m + n) 2(m − n) 2(m + n) 1 1 1 1 = − − − − − = 0. 2(m − n) 2(m + n) 2(m − n) 2(m + n)


" π Use the trigonometric identity sin2 mx d x for m an arbitrary integer. 73. Evaluate 0

Use the substitution u =sin mx, d x.1 Then sin(m − n)x + sin(m + n)x mxdu cos=nxm = 2 " π " 1 x=π 2 1 u 1 mx sin 2u x=π sin 2mx π to prove Eq. (26) page sin2 in mxthe d xtable = of integrals sin on u du =438. = − − m x=0 m 2 4 m 2 4 0 x=0 0 π x π sin 2mx sin 2π m = − − (0). = 2 − 2 4m 4 0

SOLUTION

If m is an arbitrary integer, then sin 2m π = 0. Thus " π 0

sin2 mx d x =

π . 2

" π /2 Integration byxParts 75. Set Use Im = sinm d x. to prove (for m = 1) 0 " " 1 π m m−1 tan x secm−2 x m−2 m−2 and use Im = x d x (a) Show that I1 = 1, I2 = 2 2 sec Im−2 . x d xEq. = (28) to prove that+for m > 1, sec m m−1 m−1 (b) Show that I3 = 23 and I4 = 34 12 π2 . (c) Show more generally: I2m = I2m+1 =

2m − 1 2m − 3 1 π ··· · 2m 2m − 2 2 2 2m 2m − 2 2 ··· 2m + 1 2m − 1 3

(d) Conclude that

π 2·2 4·4 2m · 2m I2m = · ··· 2 1·3 3·5 (2m − 1)(2m + 1) I2m+1 SOLUTION

(a) I1 = I2 =

" π /2 0

" π /2 0

π /2 sin x d x = − cos x = 0 − (−1) = 1; sin2 x d x =

0

π /2 x 1 1 1 π π π . = − sin x cos x − (1)(0) − (0 − 0) = = 2 2 4 2 4 2 2 0

Using Eq. (28), we have Im =

" π /2 0

sinm x d x =

" m − 1 π /2 m−2 m−1 sin x dx = Im−2 . m m 0

S E C T I O N 8.4

Trigonometric Substitution

435

(b) Using the result from (a), we get 3−1 2 2 I1 = (1) = ; 3 3 3 4−1 3 1 π . I4 = I2 = 4 4 2 2 I3 =

(c) We’ll use induction to show these results. For I2m , the result is true for m = 1 and m = 2. Now assume the result is true for m = k − 1: I2(k−1) = I2k−2 =

2k − 3 2k − 5 1 π · ··· · 2k − 2 2k − 4 2 2

Using the relation Im = ((m − 1)/m)Im−2 , we have I2k =

2k − 1 2k − 1 = I · 2k 2k−2 2k

2k − 3 2k − 5 1 π · ··· · 2k − 2 2k − 4 2 2

.

For I2m+1 , the result is true for m = 1. Now assume the result is true for m = k − 1: I2(k−1)+1 = I2k−1 =

2k − 2 2k − 4 2 · ··· 2k − 1 2k − 3 3

Again using the relation Im = ((m − 1)/m)Im−2 , we have 2k + 1 − 1 2k 2k − 2 2k − 4 2 I2k+1 = I2k−1 = · ··· . 2k + 1 2k + 1 2k − 1 2k − 3 3 (d) First divide the two results from part (c) to obtain: I2m (2m − 1)(2m + 1) (2m − 3)(2m − 1) 1·3 π = · ··· · . I2m+1 2m · 2m (2m − 2)(2m − 2) 2·2 2 Solving for π /2, we get the desired result: 2·2 4·4 2m · 2m I π = · ··· · 2m . 2 1·3 3·5 (2m − 1)(2m + 1) I2m+1 This is a continuation of Exercise 75. (a) Prove that I2m+1 ≤ I2m ≤ I2m−1 . Hint: Observe that

8.4 Trigonometric Substitution Preliminary Questions

sin2m+1 x ≤ sin2m x ≤ sin2m−1 x

for

0 ≤ x ≤ π2

" I2m−1 1 (b) Show that = 1 + . 1. Explain why trigonometric substitution is not needed to evaluate x 9 − x 2 d x. I2m+1 2m

(c) Show that Because there is a factor of x in the integrand outside the radical and the derivative of 9 − x 2 is −2x, we 1 I2m this integral. may use the substitution u = 9 − x 2 , du = −2x d x1 to ≤1+ ≤ evaluate I2m+1 2m 2. State the trigonometric substitution appropriate to the given integral: " " I2m (a) (d) Prove 9 − x 2that d x lim (b) x 2 (x 2 − 16)3/2 d x = 1. m→∞ I2m+1 " " π 3/2 dthe by English mathematician John Wallis (1616–1703): deduce (c) (e)x 2Finally, (x 2 + 16) x infinite product for 2 discovered (d) (x 2 − 5)−2 d x SOLUTION

SOLUTION

(a) (b) (c) (d)

x x x x

= 3 sin θ = 4 sec θ = 4 tan θ √ = 5 sec θ

π 2 2 4 4 2m · 2m = lim · · · ··· m→∞ 1 3 3 5 2 (2m − 1)(2m + 1)

3. Which of the triangles in Figure 6 would be used together with the substitution x = 3 sin θ ? The substitution x = 3 sin θ implies sin θ = x3 . We therefore need a triangle whose opposite side has length x and whose hypotenuse has length 3. This describes the triangle in Figure 6(A). SOLUTION

436

CHAPTER 8


4. Express tan θ in terms of x for the angle in Figure 6(A).

3

3

x

q

9 − x2

q x

9 − x2 (A)

(B)

FIGURE 6 SOLUTION Tangent is the ratio of the length of the opposite side to the length of the adjacent side. For the triangle in Figure 6(A), it then follows that

x tan θ = . 9 − x2 5. Express sec θ in terms of x for the angle in Figure 6(B). SOLUTION Secant is the ratio of the length of the hypotenuse to the length of the adjacent side. For the triangle in Figure 6(B), it then follows that

sec θ =

3 . x

6. Express sin 2θ in terms of x, where x = sin θ . SOLUTION

1 − sin2 θ = 1 − x 2 . Thus, sin 2θ = 2 sin θ cos θ = 2x 1 − x 2 .

First note that if sin θ = x, then cos θ =

Exercises In Exercises 1–4, evaluate the integral by following the steps. " dx 1. I = 9 − x2 " (a) Show that the substitution x = 3 sin θ transforms I into d θ and evaluate I in terms of θ . (b) Evaluate I in terms of x. SOLUTION

(a) Let x = 3 sin θ . Then d x = 3 cos θ d θ , and

9 − x 2 = 9 − 9 sin2 θ = 3 1 − sin2 θ = 3 cos2 θ = 3 cos θ . Thus, " I =

dx = 9 − x2

"

3 cos θ d θ = 3 cos θ

(b) If x = 3 sin θ , then θ = sin−1 ( x3 ). Thus, I = θ + C = sin−1

x 3

" d θ = θ + C.

+ C.

" " dx 3. I = dx I = x 2 +9 2 " x x2 − 2 " I in terms of θ (refer to the table (a) Show that the substitution x = 3 tan√θ transforms I into sec θ d θ and evaluate 1 cos θ d θ and evaluate I in terms of (a) Show that the substitution x = 2 sec θ transforms the integral I into 2 of integrals in Section 8.3 if necessary). θ. (b) Show that if x = 3 tan θ , then sec θ = 13 x 2 + 9. (b) Use a right triangle to show that with the above substitution, sin θ = x 2 − 2/x. (c) Express I in terms of x. (c) Evaluate I in terms of x. SOLUTION

S E C T I O N 8.4


(a) If x = 3 tan θ , then d x = 3 sec2 θ d θ , and

x2 + 9 =

9 tan2 θ + 9 = 9(tan2 θ + 1) = 9 sec2 θ = 3 sec θ .

Thus, "

I =

"

dx x2 + 9

3 sec2 θ d θ = 3 sec θ

=

" sec θ d θ = ln | sec θ + tan θ | + C.

(b) Since x = 3 tan θ , we construct a right triangle with tan θ = x3 : x2 + 9

x

q 3

From this triangle we see that sec θ = 13 x 2 + 9. (c) Combining the results from parts (a) and (b), x2 + 9 + x 1 x 2 I = ln x + 9 + + C = ln +C 3 3 3 = ln | x 2 + 9 + x| − ln 3 + C = ln | x 2 + 9 + x| + C. In Exercises "5–10, use d x the indicated substitution to evaluate the integral. I = " 2 (x + 4)2 " 4 − x 2 d x, x = 2 sin θ 5. 1 cos2 θ d θ . (a) Show that the substitution x = 2 tan θ transforms the integral I into 8 SOLUTION Let x = 2 sin " θ . Then d x = 2 cos θ d θ , and 1 1 2 θ + sin θ cos θ to evaluate I "in terms of θ . (b) Use the formula " cos θ d θ =" 2 2 4 − xx2 d x = 4 − 4 sin22θ (2 cos θ d θ ) = 4 1 − sin2 θ cos θ d θ (c) Show that sin θ = and cos θ = . " " x2 + 4 x2 + 4 2 =4 cos θ cos θ d θ = 4 cos2 θ d θ (d) Express I in terms of x. =4

1 1 θ + sin θ cos θ + C = 2θ + 2 sin θ cos θ + C. 2 2

Since x = 2 sin θ , we construct a right triangle with sin θ = x2 : 2

x

q 4 − x2

From this triangle we see that cos θ = 12 4 − x 2 , so we have I = 2θ + 2 sin θ cos θ = 2 sin−1

x 2

+

+ C = 2 sin−1

4 − x2 +2 +C 2 2 2

x

x

1 x 4 − x 2 + C. 2

"

dx "1/2 x,2 x = 3 sec θ x x2 −9 d x, x = sin θ 0 1 − x2 SOLUTION Let x = 3 sec θ . Then d x = 3 sec θ tan θ d θ , and 7.

x2 − 9 =

9 sec2 θ − 9 = 3 sec2 θ − 1 = 3 tan2 θ = 3 tan θ .

Thus, "

dx

x x2 − 9

" =

1 (3 sec θ tan θ d θ ) = (3 sec θ )(3 tan θ ) 3

" dθ =

1 θ + C. 3

437

438

CHAPTER 8


Since x = 3 sec θ , θ = sec−1 ( x3 ), and

"

dx

=

x x2 − 9

x 1 sec−1 + C. 3 3

"

" 1 dx d x, x = 2 sec θ 3/2 (x 2 − 4) , x = 2 tan θ 1/2 x 2 x 2 + 4 SOLUTION Let x = 2 sec θ . Then d x = 2 sec θ tan θ d θ , and 9.

x 2 − 4 = 4 sec2 θ − 4 = 4(sec2 θ − 1) = 4 tan2 θ . This gives " I =

dx = (x 2 − 4)3/2

"

2 sec θ tan θ d θ = (4 tan2 θ )3/2

"

1 2 sec θ tan θ d θ = 4 8 tan3 θ

"

1 sec θ d θ = 4 tan2 θ

"

cos θ sin2 θ

dθ.

Now use substitution with u = sin θ and du = cos θ d θ . Then " 1 1 −1 I = + C. u −2 du = − u −1 + C = 4 4 4 sin θ Since x = 2 sec θ , we construct a right triangle with sec θ = x2 :

x

x2 − 4

q 2

From this triangle we see that sin θ =

x 2 − 4/x, so therefore

I =

4(

−1 x 2 − 4/x)

−x +C = + C. 4 x2 − 4

" 11. Is the 1substitution u = x 2 − 4 effective for evaluating the integral dx , x = 4 tan θ 0 (16 + x 2 )2 substitution. SOLUTION

"

x2 dx ? If not, evaluate using trigonometric x2 − 4

√ If u = x 2 − 4, then du = 2x d x, x 2 = u + 4, d x = du/2x = du/2 u + 4, and " I =

x2 dx = x2 − 4

"

(u + 4) √ u

du √ 2 u+4

=

1 2

"

u+4 u 2 + 4u

du

This substitution is clearly not effective for evaluating this integral. Instead, use the trigonometric substitution x = 2 sec θ . Then d x = 2 sec θ tan θ , x 2 − 4 = 4 sec2 θ − 4 = 2 tan θ , and we have "

" " x2 dx 4 sec2 θ (2 sec θ tan θ d θ ) = 4 sec3 θ d θ . = 2 tan θ x2 − 4 ! Now use the reduction formula for secm x d x from Section 8.8.3: " " tan θ sec θ 1 sec θ d θ = 2 tan θ sec θ + 2 ln | sec θ + tan θ | + C. + 4 sec3 θ d θ = 4 2 2 I =

Since x = 2 sec θ , we construct a right triangle with sec θ = x2 :

x q 2

x2 − 4

S E C T I O N 8.4


439

From this triangle we see that tan θ = 12 x 2 − 4. Therefore

1 x 1 x 1 2 1 I =2 x2 − 4 x − 4 + C = x x 2 − 4 + 2 ln + 2 ln + x + x 2 − 4 + C. 2 2 2 2 2 2 Finally, since 1 1 2 + ln |x + x 2 − 4|, ln (x + x − 4) = ln 2 2 and ln( 12 ) is a constant, we can “absorb” this constant into the constant of integration, so that I =

1 2 x x − 4 + 2 ln |x + x 2 − 4| + C. 2

" using " In Exercises 13–30, evaluate the integral trigonometric substitution. Refer to the table of trigonometric integrals x3 dx x dx as necessary. and in two ways: using trigonometric substitution and using the direct Evaluate both x2 − 4 x2 − 4 " x 2 d x u = x 2 − 4. substitution 13. 9 − x2 SOLUTION

Let x = 3 sin θ . Then d x = 3 cos θ d θ , 9 − x 2 = 9 − 9 sin2 θ = 9(1 − sin2 θ ) = 9 cos2 θ ,

and " I =

x2 dx = 9 − x2

"

9 sin2 θ (3 cos θ d θ ) =9 3 cos θ

" sin2 θ d θ = 9

1 1 θ − sin θ cos θ + C. 2 2

Since x = 3 sin θ , we construct a right triangle with sin θ = x3 : 3

x

q 9 − x2

9 − x 2 /3, and so

x 1 9 −1 x 9 x 9 − x2 9 I = sin − + C = sin−1 − x 9 − x 2 + C. 2 3 2 3 3 2 3 2

From this we see that cos θ =

" " 12 +d4x x 2 dx x2 − 9 SOLUTION First simplify the integral:

15.

I = Now let x =

√

3 tan θ . Then d x =

√

" " 12 + 4x 2 d x = 2 3 + x2 dx

3 sec2 θ d θ ,

3 + x 2 = 3 + 3 tan2 θ = 3(1 + tan2 θ ) = 3 sec2 θ , and

Since x =

" " "

√ 1 tan θ sec θ 2 3 2 + sec θ d θ I =2 3 sec θ 3 sec θ d θ = 6 sec θ d θ = 6 2 2 √

= 3 tan θ sec θ + 3 ln | sec θ + tan θ | + C. 3 tan θ , we construct a right triangle with tan θ = √x : 3

x2 + 3

x

q 3

440

CHAPTER 8


√ From this we see that sec θ = x 2 + 3/ 3. Therefore, 2 x2 + 3 x x +3 1 x I =3 √ + 3 ln √ + √ + C1 = x x 2 + 3 + 3 ln x 2 + 3 + x + 3 ln √ + C1 √ 3 3 3 3 3 = x x 2 + 3 + 3 ln x 2 + 3 + x + C, where C = 3 ln( √1 ) + C1 . 3 " " dt 17. dx 2 )3/2 (4 − t 2 x x 2 − 25 SOLUTION Let t = 2 sin θ . Then dt = 2 cos θ d θ , 4 − t 2 = 4 − 4 sin2 θ = 4(1 − sin2 θ ) = 4 cos2 θ , and

"

dt = (4 − t 2 )3/2

I =

"

2 cos θ d θ = (4 cos2 θ )3/2

"

1 2 cos θ d θ = 4 8 cos3 θ

"

1 dθ = 4 cos2 θ

" sec2 θ d θ =

1 tan θ + C. 4

Since t = 2 sin θ , we construct a right triangle with sin θ = 2t : 2

t

q 4 − t2

From this we see that tan θ = t/ 4 − t 2 , which gives us t 1 t I = + C. +C = 4 4 − t2 4 4 − t2 " 19.

"

dy dt

y 2 5 − y 2 t2 − 5

Let y =

SOLUTION

√

5 sin θ . Then d y =

√

5 cos θ d θ ,

5 − y 2 = 5 − 5 sin2 θ = 5(1 − sin2 θ ) = 5 cos2 θ , and " I = Since y =

√

dy = 2 y 5 − y2

"

√

5 cos θ d θ 1 = √ 5 (5 sin2 θ )( 5 cos θ )

"

dθ sin2 θ

=

1 5

" csc2 θ d θ =

5 sin θ , we construct a right triangle with sin θ = √y : 5

5

y

q 5 − y2

From this we see that cot θ =

" 21.

"

5 − y 2 /y, which gives us 1 − 5 − y2 5 − y2 I = +C =− + C. 5 y 5y

dz dt

z3 z2 − 4 t t2 − 4

SOLUTION

Let z = 2 sec θ . Then dz = 2 sec θ tan θ d θ , z 2 − 4 = 4 sec2 θ − 4 = 4(sec2 θ − 1) = 4 tan2 θ ,

1 (− cot θ ) + C. 5

S E C T I O N 8.4

and


" " " 1 1 dθ dz 2 sec θ tan θ d θ cos2 θ d θ = = = 8 8 (8 sec3 θ )(2 tan θ ) sec2 θ z3 z2 − 4 1 1 1 1 1 = θ + sin θ cos θ + C = θ+ sin θ cos θ + C. 8 2 2 16 16 "

I =

Since z = 2 sec θ , we construct a right triangle with sec θ = 2z : z

z2 − 4

q 2

From this we see that sin θ =

z 2 − 4/z and cos θ = 2/z. Then

z

z 1 1 z2 − 4 z2 − 4 1 2 −1 −1 + + I = + C. sec +C = sec 16 2 16 z z 16 2 8z 2

" " 2 x dx dx x2 + 251+ x 2 SOLUTION Let x = tan θ . Then d x = sec2 θ d θ , x 2 + 1 = tan2 θ + 1 = sec2 θ , and

23.

"

x2 dx = x2 + 1

I =

"

tan2 θ (sec2 θ d θ ) = sec θ

" tan2 θ sec θ d θ . "

Now use the identity tan2 θ = sec2 θ − 1 along with the reduction formula for

sec3 θ d θ :

" " " " "

tan θ sec θ 1 sec θ d θ − sec θ d θ sec2 θ − 1 sec θ d θ = sec3 θ d θ − sec θ d θ = + 2 2 " 1 1 1 1 sec θ d θ = tan θ sec θ − ln | sec θ + tan θ | + C. = tan θ sec θ − 2 2 2 2

I =

Since x = tan θ , we construct a right triangle with tan θ = x1 : x2 + 1

x

q 1

From this we see that sec θ =

x 2 + 1. Therefore, 1 1 I = x x 2 + 1 − ln x 2 + 1 + x + C. 2 2

"

" dx 2 (x 2 + x9)2d x (x 2 − 4)3/2 SOLUTION Let x = 3 tan θ . Then d x = 3 sec2 θ d θ ,

25.

x 2 + 9 = 9 tan2 θ + 9 = 9(tan2 θ + 1) = 9 sec2 θ , and " I =

dx = (x 2 + 9)2

"

3 sec2 θ d θ 1 = 27 81 sec4 θ

" cos2 θ d θ =

Since x = 3 tan θ , we construct a right triangle with tan θ = x3 : x2 + 9

x

q 3

1 1 1 θ + sin θ cos θ + C. 27 2 2

441

442

CHAPTER 8


From this we see that sin θ = x/ x 2 + 9 and cos θ = 3/ x 2 + 9. Thus

x

x 1 x x 1 3 1 −1 + C. tan + tan−1 + I = +C = 54 3 54 54 3 18(x 2 + 9) x2 + 9 x2 + 9 "

" dx d2x (x 2 − 4) (9x 2 + 4)2 SOLUTION Let x = 2 sec θ . Then d x = 2 sec θ tan θ d θ ,

27.

x 2 − 4 = 4 sec2 θ − 4 = 4(sec2 θ − 1) = 4 tan2 θ , and

" " 1 sec θ d θ dx 2 sec θ tan θ d θ = = 8 (x 2 − 4)2 16 tan4 θ tan3 θ " " " " 1 1 1 cos2 θ 1 − sin2 θ 3 θ dθ − 1 = d θ = d θ = csc csc θ d θ . 8 8 8 8 sin3 θ sin3 θ " Now use the reduction formula for csc3 θ d θ : "

I =

I =

" " " 1 1 1 1 1 cot θ csc θ + csc θ d θ − csc θ d θ = − cot θ csc θ − csc θ d θ − 8 2 2 8 16 16

=−

1 1 cot θ csc θ − ln | csc θ − cot θ | + C. 16 16

Since x = 2 sec θ , we construct a right triangle with sec θ = x2 :

x

x2 − 4

q 2

From this we see that cot θ = 2/ x 2 − 4 and csc θ = x/ x 2 − 4. Thus 2 1 x 1 2 x ln I =− − − +C 2 2 2 2 16 16 x −4 x −4 x −4 x − 4 1 x − 2 −x − ln = + C. 8(x 2 − 4) 16 x 2 − 4 " 29.

" x 3 9 −d xx 2 d x (x 2 + 1)3 Let x = 3 sin θ . Then d x = 3 cos θ d θ ,

SOLUTION

9 − x 2 = 9 − 9 sin2 θ = 9(1 − sin2 θ ) = 9 cos2 θ , and

" I =

x3

" 2 9 − x d x = (27 sin3 θ )(3 cos θ )(3 cos θ d θ )

"

" sin3 θ cos2 θ d θ = 243

= 243 " = 243

" cos2 θ sin θ d θ −

(1 − cos2 θ ) cos2 θ sin θ d θ

cos4 θ sin θ d θ .

Now use substitution, with u = cos θ and du = − sin θ d θ for both integrals: 1 1 I = 243 − cos3 θ + cos5 θ + C. 3 5 Since x = 3 sin θ , we construct a right triangle with sin θ = x3 :


S E C T I O N 8.4

3

x

q 9 − x2

From this we see that cos θ = 9 − x 2 /3. Thus ⎡ 3 5 ⎤ 2 2 9 − x 9 − x 1 1 ⎦ + C = −3(9 − x 2 )3/2 + 1 (9 − x 2 )5/2 + C. + I = 243 ⎣− 3 3 5 3 5 Alternately, let u = 9 − x 2 . Then " " √ 1 1 2 I = x3 9 − x2 dx = − (9 − u) u du = − 6u 3/2 − u 5/2 + C 2 2 5 =

1 1 5/2 − 3u 3/2 + C = (9 − x 2 )5/2 − 3(9 − x 2 )3/2 + C. u 5 5

31. Prove for a > 0: " the following x2 dx (x 2 + 1)3/2 SOLUTION

"

1 dx x = √ tan−1 √ + C a a x2 + a √ √ Let x = a u. Then, x 2 = au 2 , d x = a du, and "

dx 1 = √ a x2 + a

"

1 du 1 = √ tan−1 u + C = √ tan−1 a a u2 + 1

x √ a

+ C.

" dx 33. LetProve I = thefollowing for .a > 0: 2 x − 4x + 8 " d x4x + 8 =1 (x − 2) x 2 + 4. 1 −1 √x (a) Complete the square to show that x 2 − = √ tan +C + " 2a x 2 du a a (x 2 + a)2 +a . Evaluate I . (b) Use the substitution u = x − 2 to show that I = u 2 + 22 SOLUTION

(a) Completing the square, we get x 2 − 4x + 8 = x 2 − 4x + 4 + 4 = (x − 2)2 + 4. (b) Let u = x − 2. Then du = d x, and " " " dx dx du = . I = = 2 (x − 2) + 4 x 2 − 4x + 8 u2 + 4 Now let u = 2 tan θ . Then du = 2 sec2 θ d θ , u 2 + 4 = 4 tan2 θ + 4 = 4(tan2 θ + 1) = 4 sec2 θ , and " I =



Since u = 2 tan θ , we construct a right triangle with tan θ = u2 : u2 + 4

u

q 2

From this we see that sec θ = u 2 + 4/2. Thus 1 u2 + 4 u I = ln + + C1 = ln u 2 + 4 + u + ln + C1 = ln u 2 + 4 + u + C. 2 2 2

443

444

CHAPTER 8


Finally, we substitute back for x: 2 I = ln (x − 2) + 4 + x − 2 + C. " evaluate the integral by completing the square and using trigonometric substitution. In Exercises 35–40, dx Evaluate . First complete the square to write 12x − x 2 = 36 − (x − 6)2 . " dx 12x − x 2 35. x 2 + 4x + 13 SOLUTION

First complete the square: x 2 + 4x + 13 = x 2 + 4x + 4 + 9 = (x + 2)2 + 9.

Let u = x + 2. Then du = d x, and " I =

"

dx

=

x 2 + 4x + 13

"

dx (x + 2)2 + 9

=

du . u2 + 9

Now let u = 3 tan θ . Then du = 3 sec2 θ d θ , u 2 + 9 = 9 tan2 θ + 9 = 9(tan2 θ + 1) = 9 sec2 θ , and "


I =


Since u = 3 tan θ , we construct the following right triangle: u2 + 9

u

q 3

From this we see that sec θ = u 2 + 9/3. Thus 1 u2 + 9 u + + C1 = ln u 2 + 9 + u + ln + C1 I = ln 3 3 3 = ln (x + 2)2 + 9 + x + 2 + C = ln x 2 + 4x + 13 + x + 2 + C. "

" dx dx x + x2 2 + x − x2 SOLUTION First complete the square:

37.

1 1 x2 + x = x2 + x + − = 4

Let u = x + 12 . Then du = d x, and

"

I =

dx x + x2

" =

4

1 2 1 − . x+ 2 4

dx

= (x + 12 )2 − 14

"

du

. u 2 − 14

Now let u = 12 sec θ . Then du = 12 sec θ tan θ d θ , u2 −

1 1 1 1 1 = sec2 θ − = (sec2 θ − 1) = tan2 θ , 4 4 4 4 4

and I =

" 1 2 sec θ tan θ d θ 1 tan θ 2

" =

sec θ d θ = ln | sec θ + tan θ | + C.

Since u = 12 sec θ , we construct the following right triangle:

S E C T I O N 8.4

2u


4u2 − 1

q 1

From this we see that tan θ = 4u 2 − 1. Then

2 1 1 I = ln 2u + 4u 2 − 1 + C = ln 2 x + − 1 + C + 4 x+ 2 2 1 − 1 + C = ln 2x + 1 + 2 x 2 + x + C. = ln 2x + 1 + 4 x 2 + x + 4

39.

" " x 2 −24x + 3 d x x − 4x + 7 d x

SOLUTION

First complete the square: x 2 − 4x + 3 = x 2 − 4x + 4 − 1 = (x − 2)2 − 1.

Let u = x − 2. Then du = d x, and " " " I = x 2 − 4x + 3 d x = (x − 2)2 − 1 d x = u 2 − 1 du. Now let u = sec θ . Then du = sec θ tan θ d θ , u 2 − 1 = sec2 θ − 1 = tan2 θ , and " " "

I = tan2 θ (sec θ tan θ d θ ) = tan2 θ sec θ d θ = sec2 θ − 1 sec θ d θ "

"

=

sec3 θ d θ −

=

1 1 tan θ sec θ − 2 2

tan θ sec θ 1 + 2 2

sec θ d θ = "

"

" sec θ d θ − sec θ d θ

1 1 tan θ sec θ − ln | sec θ + tan θ | + C. 2 2

sec θ d θ =

Since u = sec θ , we construct the following right triangle: u

u2 − 1

q 1

From this we see that tan θ = u 2 − 1. Thus

1 1 1 1 I = u u 2 − 1 − ln u + u 2 − 1 + C = (x − 2) (x − 2)2 − 1 − ln x − 2 + (x − 2)2 − 1 + C 2 2 2 2 1 1 = (x − 2) x 2 − 4x + 3 − ln x − 2 + x 2 − 4x + 3 + C. 2 2 " 41. Evaluate

"

sec d x−1 x d x. Hint: First use Integration by Parts. (x 2 + 6x + 6)2 SOLUTION Let u = sec−1 x and v = 1. Then v = x, u = 1/x x 2 − 1, and " " " x dx I = sec−1 x d x = x sec−1 x − d x = x sec−1 x − . 2 x x −1 x2 − 1

To evaluate the integral on the right, let x = sec θ . Then d x = sec θ tan θ d θ , x 2 − 1 = sec2 θ − 1 = tan2 θ , and " " " sec θ tan θ d θ dx = sec θ d θ = ln | sec θ + tan θ | + C = ln x + x 2 − 1 + C. = 2 tan θ x −1 Thus, the final answer is

I = x sec−1 x − ln x + x 2 − 1 + C.

445

446

CHAPTER 8


" indicate In Exercises 43–52, a good method for evaluating the integral (but do not evaluate). Your choices are recognizing sin−1 x a basic Evaluate integration formula, algebraic manipulation, substitution (specify u and du), Integration by Parts (specify u and d x. Hint: First use Integration by Parts. x2 or trigonometric substitution (specify). If it appears that the techniques you have learned v ), a trigonometric method, thus far are not sufficient, state this. " dx 43. 12 − 6x − x 2 SOLUTION For this integral use a combination of three methods: Algebraic manipulation (complete the square), sub√ stitution (let u = x + 3), then trigonometric substitution (let u = 21 sin θ ). " " 45. x sec2 3x d x 3 sin x cos x d x SOLUTION

47.

Use Integration by Parts, with u = x and v = sec2 x.

" " 2x e 2d x 4x e +4x1 − 1 d x

SOLUTION First use substitution, with u = e2x . Then either use trigonometric substitution (with u = tan θ ) or recognize the formula for the inverse tangent: " du = tan−1 u + C. 1 + u2

"

" cot x csc d xx d x 9 − x2 SOLUTION Recognize the formula

49.

" cot x csc x d x = − csc x + C. " 51.

" dx ln 2) x d3x (x +

SOLUTION

Use the substitution u = x + 2, and then recognize the formula " 1 u −3 du = − 2 + C. 2u

" of the following integrals can be evaluated using the substitution u = 1 − x 2 and which require trigonometric 53. Which dx substitution? Determine the integral obtained after substitution in each case. " " (x + 1)(x + 2)3 3 (a) x 1 − x 2d x (b) x 2 1 − x 2d x " " x4 x (c) dx (d) dx 2 1−x 1 − x2 SOLUTION

(a) Use the substitution u = 1 − x 2 . Then du = −2x d x, x 2 = 1 − u, and so " " " 1 1 x3 1 − x2 dx = − x 2 1 − x 2 (−2x d x) = − (1 − u)u 1/2 du. 2 2 (b) Let x = sin θ . Then d x = cos θ d θ , 1 − x 2 = cos2 θ , and so " " " x 2 1 − x 2 d x = sin2 θ (cos θ ) cos θ d θ = sin2 θ cos2 θ d θ . (c) Let x = sin θ . Then d x = cos θ d θ , 1 − x 2 = cos2 θ , and so " " " x4 sin4 θ dx = cos θ d θ = sin4 θ d θ . cos θ 1 − x2 (d) Let u = 1 − x 2 . Then du = −2x d x, and we have " " " 1 1 x −2x d x du dx = − =− . 1/2 2 2 2 2 u 1−x 1−x Find the average height of a point on the semicircle y =

1 − x 2 for −1 ≤ x ≤ 1.

S E C T I O N 8.4


447

55. Find the volume of the solid obtained by revolving the graph of y = x 1 − x 2 over [0, 1] about the y-axis. Using the method of cylindrical shells, the volume is given by

SOLUTION

V = 2π

" 1 " 1 x x 1 − x 2 d x = 2π x 2 1 − x 2 d x. 0

0

To evaluate this integral, let x = sin θ . Then d x = cos θ d θ , 1 − x 2 = 1 − sin2 θ = cos2 θ , and

" I =

" "

" " x 2 1 − x 2 d x = sin2 θ cos2 θ d θ = 1 − cos2 θ cos2 θ d θ = cos2 θ d θ − cos4 θ d θ .

! Now use the reduction formula for cos4 θ d θ : * ) " " " 1 cos3 θ sin θ 1 3 2 2 cos θ d θ = − cos3 θ sin θ + cos2 θ d θ + I = cos θ d θ − 4 4 4 4 1 1 1 1 1 1 1 3 = − cos θ sin θ + θ + sin θ cos θ + C = − cos3 θ sin θ + θ + sin θ cos θ + C. 4 4 2 2 4 8 8 Since sin θ = x, we know that cos θ = 1 − x 2 . Then we have I =−

3/2 1 1 1 x + sin−1 x + x 1 − x 2 + C. 1 − x2 4 8 8

Now we can complete the volume: V = 2π

1 π2

3/2 1 −1 1 1 π − x 1 − x2 + sin x + x 1 − x 2 = 2π 0 + + 0 − (0) = . 4 8 8 16 8 0

57. Find the volume of revolution for the region in Exercise 56, but revolve around y = 3. Find the volume of the solid obtained by revolving the region between the graph of y 2 − x 2 = 1 and the line SOLUTION Using y = 2 about the the linewasher y = 2. method, the volume is given by " √3

" √3

2 2 2 2 2 3 − x + 1 − 1 dx V = √ π R − r d x = 2π − 3 " √

= 2π

0

0

3

9−6

x 2 + 1 + x 2 + 1 − 1 d x = 2π

" √3

9 − 6 x2 + 1 + x2 dx 0

√ 1 3 1 2 1 3 2 = 2π 9x − 6 x x + 1 + ln x + 1 + x + x 2 2 3 0 √ √ √ √ √ √ = 2π 9 3 − 3 3(2) − 3 ln 2 + 3 + 3 − (0) = 8π 3 − 6π ln 2 + 3 .

59. Havingwire ordered an 18-in. pizza field for yourself and P twolocated friends,atyou want to divide as A charged creates an electric at a point a distance D fromit up theusing wire vertical (Figure slices 7). The in Figure 8. Use EEq. of (7)the in field Exercise 63 below and a computer algebra system to find the value of x that divides the pizza perpendicular to the wire (in volts) is component ⊥ into equal parts. " x2 kλ D dx E⊥ = y 2 (x + D 2 )3/2 x1 where k = 8.99 × 109 N · m2 /C2 (Coulomb constant), λ is the charge density (coulombs per meter), and x1 , x 2 are shown in the figure. Suppose that λ = 6 × 10−4 C/m, and D = 3 m. Find E ⊥ if (a) x1 = 0 and x 2 = 30 m, and (b) x 1 = −15 m and x2 = 15 m. x −9

−x

x

9

FIGURE 8 Dividing a pizza into three equal parts.

448

CHAPTER 8

T E C H N I Q U E S O F I N T E G R AT I O N SOLUTION First find the value of x which divides evenly a pizza with a 1-inch radius. By proportionality, we can then take this answer and multiply by 9 to get the answer for the 18-inch pizza. The total area of a 1-inch radius pizza is π · 12 = π (in square inches). The three equal pieces will have an area of π /3. The center piece is further divided into 4 equal pieces, each of area π /12. From Example 1, we know that " x 1 1 1 − x 2 d x = sin−1 x + x 1 − x 2 . 2 2 0

Setting this expression equal to π /12 and solving for x using a computer algebra system, we find x = 0.265. For the 18-inch pizza, the value of x should be x = 9(0.265) = 2.385 inches.

Further Insights and Challenges 61. Hyperbolic Substitution Hyperbolic functions can be used instead of trigonometric substitution to treat integrals " dx " LetJn2 = 2 2 . Prove dx involving x ± a .(x Let+I 1) =n . x2 − 1 " 1 1 x = 1 − I intoJthe (a) Show that the substitution x = coshJn+1 t transforms dt = t + C. n +integral 2n 2n (x 2 + 1)n+1 (b) Show that I = cosh−1 x + C. Hint: Compute Jn using Integration by Parts with v = 1. Use this recursion relation to calculate J2 and J3 . (c) Trigonometric substitution with x = sec θ leads to I = ln |x + x 2 − 1| + C Show that the two answers coincide. SOLUTION

(a) Let x = cosh t. Then, d x = sinh t dt, and x 2 − 1 = cosh2 t − 1 = sinh2 t. Thus, " " " dx sinh t dt = dt = t + C. = sinh t x2 − 1 (b) Since x = cosh t, by definition t = cosh−1 x. Thus I = cosh−1 x + C. (c) To establish that cosh−1 x = ln |x + x 2 − 1|, first note that 1 d cosh−1 x = , dx x2 − 1 and

1 x 1 x + x2 − 1 1 d 2 ln |x + x − 1| = 1+ = = ; 2 2 2 2 2 dx x + x −1 x −1 x + x −1 x −1 x −1 in other words, cosh−1 x and ln |x + x 2 − 1| have the same √ derivative. The two functions can therefore differ by at most an additive constant; however, cosh−1 1 = 0 = ln |1 + 1 − 1|, so that constant must be zero and the two functions must be equal. 63. In Example "1, we proved the formula x 2 − 9 d x. " Let I = 1 1 1 − x 2 d x = sin−1 x + x 1 − x 2 "+ C 7 2 2 (a) Show that the substitution x = 3 cosh t transforms I into the integral 9 sinh2 t dt. Derive this formula using geometry rather than calculus by interpreting the integral as the area of part of the unit circle. (b) Evaluate the hyperbolic " a integral using the identity SOLUTION The integral 1 − x 2 d x is the area bounded by the unit circle, the x-axis, the y-axis, and the line 1 0 sinh2 t = (cosh 2t − 1) x = a. This area can be divided into two regions as follows:2 (c) Express the result in terms of x. Note that ysinh 2t = 2 sinh t cosh t. 1

II

q 0

I a

1

x

S E C T I O N 8.5

Region I is a triangle with base a and height π − cos−1 a = sin−1 a. Thus, 2

The Method of Partial Fractions

449

1 − a 2 . Region II is a sector of the unit circle with central angle θ =

a " a 1 1 1 1 x 1 − x 2 + sin−1 x . 1 − x 2 d x = a 1 − a 2 + sin−1 a = 2 2 2 2 0 0 "

dx in two ways and verify that the answers agree: first via trigonometric substitution and then x2 − 1 8.5using ThetheMethod identity of Partial Fractions 1 1 1 1 Preliminary Questions − = " 2 x −1 x +1 x2 − 1 √ 1. Suppose that f (x) d x = ln x + x + 1 + C. Can f (x) be a rational function? Explain. Compute

SOLUTION No, f (x) cannot be a rational function because the integral of a rational function cannot contain a term √ with a non-integer exponent such as x + 1.

2. Which of the following are proper rational functions? x (a) x −3 x 2 + 12 (c) (x + 2)(x + 1)(x − 3)

(b)

4 9−x

(d)

4x 3 − 7x (x − 3)(2x + 5)(9 − x)

SOLUTION

(a) No, this is not a proper rational function because the degree of the numerator is not less than the degree of the denominator. (b) Yes, this is a proper rational function. (c) Yes, this is a proper rational function. (d) No, this is not a proper rational function because the degree of the numerator is not less than the degree of the denominator. 3. Which of the following quadratic polynomials are irreducible? To check, complete the square if necessary. (b) x 2 − 5 (a) x 2 + 5 (c) x 2 + 4x + 6

(d) x 2 + 4x + 2

SOLUTION

(a) (b) (c) (d)

Square is already completed; irreducible. √ √ Square is already completed; factors as (x − 5)(x + 5). x 2 + 4x + 6 = (x + 2)2 + 2; irreducible. √ √ x 2 + 4x + 2 = (x + 2)2 − 2; factors as (x + 2 − 2)(x + 2 + 2).

4. Let P(x)/Q(x) be a proper rational function where Q(x) factors as a product of distinct linear factors (x − ai ). Then " P(x) d x Q(x) (choose correct answer): (a) is a sum of logarithmic terms Ai ln(x − ai ) for some constants Ai . (b) may contain a term involving the arctangent. SOLUTION

The correct answer is (a): the integral is a sum of logarithmic terms Ai ln(x − ai ) for some constants Ai .

Exercises 1. Match the rational function (a)–(d) with the corresponding partial fraction decomposition (i)–(iv). x 2 + 4x + 12 (a) (x + 2)(x 2 + 4) 2x 2 + 8x + 24 (b) (x + 2)2 (x 2 + 4) x 2 − 4x + 8 (c) (x − 1)2 (x − 2)2 x 4 − 4x + 8 (d) (x + 2)(x 2 + 4)

450

CHAPTER 8


4 4x − 4 − 2 x +2 x +4 8 4 5 + + + x − 1 (x − 1)2 (x − 2)2 −x + 2 2 + 2 + (x + 2)2 x +4 4 + 2 x +4

(i) x − 2 + −8 x −2 1 (iii) x +2 1 (iv) x +2 (ii)

SOLUTION

x 2 + 4x + 12 4 1 + 2 = . x +2 (x + 2)(x 2 + 4) x +4 1 −x + 2 2 2x 2 + 8x + 24 = . + 2 + (b) x + 2 (x + 2)2 (x + 2)2 (x 2 + 4) x +4 x 2 − 4x + 8 −8 8 4 5 (c) = + . + + x − 2 (x − 2)2 x − 1 (x − 1)2 (x − 1)2 (x − 2)2 x 4 − 4x + 8 4x − 4 4 (d) − 2 = x −2+ . 2 x +2 (x + 2)(x + 4) x +4 (a)

3. Clear denominators in the following partial fraction decomposition and determine the constant B (substitute a value Determine the constants A, B: of x or use the method of undetermined coefficients): 2x − 3 A B 1 = B + 3 3x 2 + 11x + 12 (x − 3)(x = − 4) x − 3 x − 4 − − x +1 x + 3 (x + 3)2 (x + 1)(x + 3)2 SOLUTION

Clearing denominators gives 3x 2 + 11x + 12 = (x + 3)2 − B(x + 1)(x + 3) − 3(x + 1).

Setting x = 0 then yields 12 = 9 − B(1)(3) − 3(1)

or

B = −2.

To use the method of undetermined coefficients, expand the right-hand side and gather like terms: 3x 2 + 11x + 12 = (1 − B)x 2 + (3 − 4B)x + (6 − 3B). Equating x 2 -coefficients on both sides, we find 3=1− B

or

B = −2.

In Exercises 5–8,constants use longindivision to fraction write f (x) as the sum of a polynomial and a proper rational function. Then Find the partial decomposition " the calculate f (x) d x. A Bx + C 2x + 4 = + 2 x −2 (x − 2)(x 2 + 4) x +4 x 5. f (x) = 3x − 9 SOLUTION Long division gives us x 1 1 = + . 3x − 9 3 x −3 Therefore the integral is "

x 1 dx = 3x − 9 3

"

" dx +

dx 1 = x + ln |x − 3| + C. x −3 3

x 3 +2x + 1 7. f (x) = x +2 f (x) = x − 2 x +3 SOLUTION Long division gives us x3 + x + 1 11 = x 2 + 2x + 5 + . x −2 x −2 Therefore the integral is " 3 " " 1 x +x +1 dx d x = (x 2 + 2x + 5) d x + 11 = x 3 + x 2 + 5x + 11 ln |x − 2| + C. x −2 x −2 3

S E C T I O N 8.5


In Exercises 9–46,3 evaluate the integral. x −1 " f (x) = d xx 2 − x 9. (x − 2)(x − 4) SOLUTION

The partial fraction decomposition has the form: A B 1 = + . (x − 2)(x − 4) x −2 x −4

Clearing denominators gives us 1 = A(x − 4) + B(x − 2). Setting x = 2 then yields 1 = A(2 − 4) + 0

or

1 A=− , 2

while setting x = 4 yields 1 = 0 + B(4 − 2)

B=

or

1 . 2

The result is: 1 − 12 1 = + 2 . (x − 2)(x − 4) x −2 x −4

Thus, "

1 dx =− (x − 2)(x − 4) 2

"

dx 1 + x −2 2

"

dx 1 1 = − ln |x − 2| + ln |x − 4| + C. x −4 2 2

"

" dx dx x(2x + 1) (x − 3)(x + 7) SOLUTION The partial fraction decomposition has the form:

11.

A B 1 = + . x(2x + 1) x 2x + 1 Clearing denominators gives us 1 = A(2x + 1) + Bx. Setting x = 0 then yields 1 = A(1) + 0 while setting x = − 12 yields

or

1 1=0+ B − 2

or

A = 1,

B = −2.

The result is: 1 1 −2 = + . x(2x + 1) x 2x + 1 Thus, "

dx = x(2x + 1)

"

dx − x

"

2 dx = ln |x| − ln |2x + 1| + C. 2x + 1

For the integral on the right, we have used the substitution u = 2x + 1, du = 2 d x. " "(2x − 1) d x 13. (3x + 5) d x x 2 −25x + 6 x − 4x − 5

451

452

CHAPTER 8


The partial fraction decomposition has the form: 2x − 1 x 2 − 5x + 6

=

2x − 1 A B = + . (x − 2)(x − 3) x −2 x −3

Clearing denominators gives us 2x − 1 = A(x − 3) + B(x − 2). Setting x = 2 then yields 3 = A(−1) + 0

or

A = −3,

5 = 0 + B(1)

or

B = 5.

while setting x = 3 yields

The result is: −3 5 2x − 1 = + . x −2 x −3 x 2 − 5x + 6 Thus, "

15.

(2x − 1) d x = −3 x 2 − 5x + 6

"

dx +5 x −2

"

dx = −3 ln |x − 2| + 5 ln |x − 3| + C. x −3

" " 2 (x + 3x −d x44) d x (x + + 5)(3x −+ 2)2) (x3)(x − 2)(x − 3)(x

SOLUTION

The partial fraction decomposition has the form: x 2 + 3x − 44 A B C = + + . (x + 3)(x + 5)(3x − 2) x +3 x + 5 3x − 2

Clearing denominators gives us x 2 + 3x − 44 = A(x + 5)(3x − 2) + B(x + 3)(3x − 2) + C(x + 3)(x + 5). Setting x = −3 then yields 9 − 9 − 44 = A(2)(−11) + 0 + 0

or

A = 2,

while setting x = −5 yields 25 − 15 − 44 = 0 + B(−2)(−17) + 0 and setting x = 23 yields 4 + 2 − 44 = 0 + 0 + C 9

11 3

17 3

or

B = −1,

or

C = −2.

The result is: x 2 + 3x − 44 2 −1 −2 = + + . (x + 3)(x + 5)(3x − 2) x +3 x + 5 3x − 2 Thus, "

(x 2 + 3x − 44) d x =2 (x + 3)(x + 5)(3x − 2)

"

dx − x +3

"

dx −2 x +5

"

2 dx = 2 ln |x + 3| − ln |x + 5| − ln |3x − 2| + C. 3x − 2 3

To evaluate the last integral, we have made the substitution u = 3x − 2, du = 3 d x. " " 2 (x + 11x) 3 d xd x 17. 2 (x − 1)(x + 1) 2 + x) (x + 1)(x SOLUTION The partial fraction decomposition has the form: A B C x 2 + 11x = . + + 2 x − 1 x + 1 (x − 1)(x + 1) (x + 1)2

S E C T I O N 8.5


Clearing denominators gives us x 2 + 11x = A(x + 1)2 + B(x − 1)(x + 1) + C(x − 1). Setting x = 1 then yields 12 = A(4) + 0 + 0

A = 3,

or

while setting x = −1 yields −10 = 0 + 0 + C(−2)

or

C = 5.

Plugging in these values results in x 2 + 11x = 3(x + 1)2 + B(x − 1)(x + 1) + 5(x − 1). The constant B can be determined by plugging in for x any value other than 1 or −1. If we plug in x = 0, we get 0 = 3 + B(−1)(1) + 5(−1)

B = −2.

or

The result is 3 x 2 + 11x −2 5 = . + + 2 x −1 x + 1 (x + 1)2 (x − 1)(x + 1) Thus, "

(x 2 + 11x) d x =3 (x − 1)(x + 1)2

"

dx −2 x −1

"

dx +5 x +1

"

5 dx = 3 ln |x − 1| − 2 ln |x + 1| − + C. 2 x +1 (x + 1)

"

dx " 2 − 21x) d x (x −(4x 1)2 (x − 2)2 (x − 3)2 (2x + 3) SOLUTION The partial fraction decomposition has the form:

19.

1 (x − 1)2 (x − 2)2

=

A C B D + . + + x − 1 (x − 1)2 x − 2 (x − 2)2

Clearing denominators gives us 1 = A(x − 1)(x − 2)2 + B(x − 2)2 + C(x − 2)(x − 1)2 + D(x − 1)2 . Setting x = 1 then yields 1 = B(1)

or

B = 1,

1 = D(1)

or

D = 1.


Plugging in these values gives us 1 = A(x − 1)(x − 2)2 + (x − 2)2 + C(x − 2)(x − 1)2 + (x − 1)2 . Setting x = 0 now yields 1 = A(−1)(4) + 4 + C(−2)(1) + 1

or

− 4 = −4 A − 2C,

while setting x = 3 yields 1 = A(2)(1) + 1 + C(1)(4) + 4

or

− 4 = 2 A + 4C.

Solving this system of two equations in two unknowns gives A = 2 and C = −2. The result is 1 2 −2 1 1 = + . + + x − 1 (x − 1)2 x − 2 (x − 2)2 (x − 1)2 (x − 2)2 Thus,

"

dx =2 (x − 1)2 (x − 2)2

"

dx + x −1

= 2 ln |x − 1| −

"

dx −2 (x − 1)2

"

dx + x −2

"

dx (x − 2)2

1 1 − 2 ln |x − 2| − + C. x −1 x −2

453

454

CHAPTER 8


"

" 48 d x dx x(x + 4)2 3 (x + 4) SOLUTION The partial fraction decomposition has the form:

21.

A B C 48 = + . + x x + 4 (x + 4)2 x(x + 4)2 Clearing denominators gives us 48 = A(x + 4)2 + Bx(x + 4) + C x. Setting x = −4 then yields 48 = 0 + 0 − 4C

C = −12,

or

while setting x = 0 yields 48 = A(16) + 0 + 0

A = 3.

or

Plugging in A = 3 and C = −12 gives us 48 = 3(x + 4)2 + Bx(x + 4) − 12x. Setting x = 1 now yields 48 = 3(25) + B(1)(5) − 12

B = −3.

or

The result is −3 −12 3 48 + = + . x x + 4 (x + 4)2 x(x + 4)2 Thus,

"

48 d x =3 x(x + 4)2

"

dx −3 x

"

dx − 12 x +4

"

12 dx = 3 ln |x| − 3 ln |x + 4| + + C. 2 x +4 (x + 4)

"

dx " 2 + x + 3) d x (x4) 2 (x − 1) (x − (x − 1)3 SOLUTION The partial fraction decomposition has the form:

23.

A 1 C B = + + . 2 2 x − 4 (x − 1) (x − 4) (x − 1) (x − 4) Clearing denominators, we get 1 = A(x − 4)(x − 1) + B(x − 1) + C(x − 4)2 . Setting x = 1 then yields 1 = 0 + 0 + C(9)

or

C=

1 , 9

1 = 0 + B(3) + 0

or

B=

1 . 3


Plugging in B = 13 and C = 19 , and setting x = 5, we find 1 1 1 = A(1)(4) + (4) + (1) 3 9

or

1 A=− . 9

The result is 1 1 − 19 1 9 3 = + + . x − 4 (x − 4)2 x −1 (x − 4)2 (x − 1)

Thus, "

1 dx =− 9 (x − 4)2 (x − 1)

"

dx 1 + x −4 3

"

1 dx + 9 (x − 4)2

"

dx 1 1 1 = − ln |x − 4| − + ln |x − 1| + C. x −1 9 3(x − 4) 9


S E C T I O N 8.5

455

"

3x + 6 " dx x 2 x x (x − 1)(x3−d 3) (3x + 7) SOLUTION The partial fraction decomposition has the form:

25.

3x + 6 A C B D = + 2 + + . x x −1 x −3 x 2 (x − 1)(x − 3) x Clearing denominators gives us 3x + 6 = Ax(x − 1)(x − 3) + B(x − 1)(x − 3) + C x 2 (x − 3) + Dx 2 (x − 1). Setting x = 0, then yields 6 = 0 + B(−1)(−3) + 0 + 0

B = 2,

or

while setting x = 1 yields 9 = 0 + 0 + C(1)(−2) + 0

or

15 = 0 + 0 + 0 + D(9)(2)

or

9 C =− , 2

and setting x = 3 yields D=

5 . 6

In order to find A, let’s look at the x 3 -coefficient on the right-hand side (which must equal 0, since there’s no x 3 term on the left): 0= A+C + D = A−

9 5 + , 2 6

A=

so

11 . 3

The result is 11 5 9 3x + 6 3 + 2 + −2 + 6 . = x x −1 x −3 x 2 (x − 1)(x − 3) x2

Thus, "

11 (3x + 6) d x = 2 3 x (x − 1)(x − 3) =

"

dx +2 x

"

9 dx − 2 2 x

"

dx 5 + x −1 6

"

dx x −3

2 9 5 11 ln |x| − − ln |x − 1| + ln |x − 3| + C. 3 x 2 6

" " 2 (3x −d2) x dx x −4 3 x(x − 1) SOLUTION First we use long division to write

27.

3x 2 − 2 46 = 3x + 12 + . x −4 x −4 Then the integral becomes "

(3x 2 − 2) d x = x −4

"

" (3x + 12) d x + 46

3 dx = x 2 + 12x + 46 ln |x − 4| + C. x −4 2

"

" d x2 2 +− x(x(x 1)x + 1) d x x2 + x SOLUTION The partial fraction decomposition has the form:

29.

A Bx + C 1 = + 2 . x x(x 2 + 1) x +1 Clearing denominators, we get 1 = A(x 2 + 1) + (Bx + C)x. Setting x = 0 then yields 1 = A(1) + 0

or

A = 1.

456

CHAPTER 8


This gives us 1 = x 2 + 1 + Bx 2 + C x = (B + 1)x 2 + C x + 1. Equating x 2 -coefficients, we find B+1=0

B = −1;

or

while equating x-coefficients yields C = 0. The result is −x 1 1 = + 2 . x x(x 2 + 1) x +1 Thus, "

dx = x(x 2 + 1)

"

dx − x

"

x dx . x2 + 1

For the integral on the right, use the substitution u = x 2 + 1, du = 2x d x. Then we have " " " du dx dx 1 1 = − = ln |x| − ln |x 2 + 1| + C. x 2 u 2 x(x 2 + 1) "

"x 2 d x 2 (3x − 4x + 5) d x x2 + 3 (x − 1)(x 2 + 1) SOLUTION First use long division to obtain

31.

x2 x2 + 3

=

x2 + 3 − 3 x2 + 3 −3 3 = 2 + 2 =1− 2 . 2 x +3 x +3 x +3 x +3

The integral becomes "

x2 dx = x2 + 3

"

" dx − 3

√ 1 x dx −1 √x −1 √ = x − 3 3 tan tan + C = x − + C. √ x2 + 3 3 3 3

" "

2 dxx dx 2 (x + 2 − 3 + 1) 2x1)(x SOLUTION The partial fraction decomposition has the form

33.

A x2 Bx + C = . + 2 x +1 (x + 1)(x 2 + 1) x +1 Clearing denominators, we get x 2 = A(x 2 + 1) + (Bx + C)(x + 1). Setting x = −1 then yields 1 = A(2) + 0

A=

or

1 . 2

This gives us x2 =

1 2 1 x + + Bx 2 + Bx + C x + C = 2 2

B+

1 1 x 2 + (B + C)x + C + . 2 2

Equating x 2 -coefficients, we find 1=B+

1 2

or

B=

1 , 2

while equating constant coefficients yields 0=C+

1 2

or

1 C=− . 2

The result is 1 1x − 1 x2 2 2 2. = + x +1 (x + 1)(x 2 + 1) x2 + 1


S E C T I O N 8.5

457

Thus, "

1 x2 dx = 2 2 (x + 1)(x + 1) =

"

dx 1 + x +1 2

"

1 (x − 1) d x = 2 2 x +1

"

dx 1 + x +1 2

"

1 x dx − 2 2 x +1

"

dx x2 + 1

1 1 1 ln |x + 1| + ln |x 2 + 1| − tan−1 x + C. 2 4 2

Here we used u = x 2 + 1, du = 2x d x for the second integral. " " x 2 d x2 35. 6x + 7x − 6 dx (3x +27)3 (x − 4)(x + 2) 1 SOLUTION This problem can be done without partial fraction decomposition. Let u = 3x + 7, so that x = 3 (u − 7), 1 2 2 x = 9 (u − 7) and du = 3 d x. Then "

" 1 u 2 − 14u + 49 1 −2 + 49u −3 du du = − 14u 27 u u3 1 49 1 49 14 = ln |u| + 14u −1 − u −2 + C = ln |3x + 7| + (3x + 7)−1 − (3x + 7)−2 + C. 27 2 27 27 54

1 x2 dx = 3 27 (3x + 7)

"

1 (u − 7)2 du = 3 27 u

"

"

" dx dx x 2 (x 2 + 25) 2 x(x + 25) SOLUTION The partial fraction decomposition has the form:

37.

1 x 2 (x 2 + 25)

=

A Cx + D B . + 2 + 2 x x x + 25

Clearing denominators, we get 1 = Ax(x 2 + 25) + B(x 2 + 25) + (C x + D)x 2 . Setting x = 0 then yields 1 = 0 + B(25) + 0

B=

or

1 . 25

This gives us 1 = Ax 3 + 25 Ax +

1 2 1 x + 1 + C x 3 + Dx 2 = ( A + C)x 3 + D + x 2 + 25 Ax + 1. 25 25

Equating x-coefficients yields 0 = 25 A

or

A = 0,

while equating x 3 -coefficients yields 0 = A+C =0+C

C = 0,

or

and equating x 2 -coefficients yields 0= D+

1 25

or

D=

1

−1

−1 . 25

The result is 1 = 252 + 2 25 . x 2 (x 2 + 25) x x + 25 Thus, "

" 39.

1 dx = 25 x 2 (x 2 + 25)

10 d x " 10 d x 2 (x + 1)(x 22+ 9) (x − 1) (x 2 + 9)

"

1 dx − 25 x2

"

1 dx 1 −1 x + C. = − − tan 25x 125 5 x 2 + 25

458

CHAPTER 8


The partial fraction decomposition has the form:

SOLUTION

Dx + E A Bx + C 10 + 2 = . + 2 2 2 x + 1 (x + 1)(x + 9) x +9 (x + 9)2 Clearing denominators gives us 10 = A(x 2 + 9)2 + (Bx + C)(x + 1)(x 2 + 9) + (Dx + E)(x + 1). Setting x = −1 then yields 10 = A(100) + 0 + 0

A=

or

1 . 10

Expanding the right-hand side, we find 18 81 1 x 4 + (B + C)x 3 + 9B + C + D + x 2 (9B + 9C + D + E)x + 9C + E + . 10 = B + 10 10 10 Equating x 4 -coefficients yields 1 =0 10

B+

or

B=−

1 , 10

C=

1 , 10

while equating x 3 -coefficients yields −

1 +C =0 10

or

and equating x 2 -coefficients yields −

9 1 18 + +D+ =0 10 10 10

D = −1.

or

Finally, equating constant coefficients, we find 10 =

81 9 +E+ 10 10

E = 1.

or

The result is 1 1 1 10 10 + − 10 x + 10 + −x + 1 . = x +1 (x + 1)(x 2 + 9)2 x2 + 9 (x 2 + 9)2

Thus, "

1 10 d x = 2 2 10 (x + 1)(x + 9)

"

dx 1 − x + 1 10

"

1 x dx + 2 10 x +9

"

dx − 2 x +9

"

x dx + 2 (x + 9)2

"

dx (x 2 + 9)2

For the second and fourth integrals, use the substitution u = x 2 + 9, du = 2x d x. Then we have " "

1 1 10 d x dx 1 2 + 9| + 1 tan−1 x + + = . ln |x + 1| − ln |x 10 20 30 3 (x + 1)(x 2 + 9)2 2(x 2 + 9) (x 2 + 9)2 For the last integral, use the trigonometric substitution x = 3 tan θ ,

d x = 3 sec2 θ d θ ,

x 2 + 9 = tan2 θ + 9 = 9 sec2 θ .

Then, "

dx = (x 2 + 9)2

"

3 sec2 θ d θ 1 = 27 (9 sec2 θ )2

"

dθ 1 = 27 sec2 θ

"

1 1 θ + sin θ cos θ + C. 27 2 2

1 cos2 θ d θ =

Now we construct a right triangle with tan θ = x3 : x2 + 9

x

q 3

.

S E C T I O N 8.5


459

From this we see that sin θ = x/ x 2 + 9 and cos θ = 3/ x 2 + 9. Thus "

x

1 x 1 3 1 x dx −1 −1 x + + + C. = + C = tan tan 2 2 2 2 2 54 3 54 54 3 (x + 9) 18(x + 9) x +9 x +9 Collecting all the terms, we obtain "

1 1 1 10 d x 2 + 9| + 1 tan−1 x + = ln |x + 1| − ln |x 2 2 2 10 20 30 3 (x + 1)(x + 9) 2(x + 9)

x x 1 + +C + tan−1 54 3 18(x 2 + 9)

x x +9 1 7 1 + + C. = ln |x + 1| − ln |x 2 + 9| + tan−1 10 20 135 3 18(x 2 + 9) "

" 100x d x d x2 (x − 3)(x + 1)2 x(x 2 + 8)2 SOLUTION The partial fraction decomposition has the form:

41.

Dx + E A 100x Bx + C + 2 = . + 2 2 2 x − 3 (x − 3)(x + 1) x +1 (x + 1)2 Clearing denominators, we get 100x = A(x 2 + 1)2 + (Bx + C)(x − 3)(x 2 + 1) + (Dx + E)(x − 3). Setting x = 3 then yields 300 = A(100) + 0 + 0

or

A = 3.

Expanding the right-hand side, we find 100x = (B + 3)x 4 + (C − 3B)x 3 + (B − 3C + D + 6)x 2 + (C − 3B − 3D + E)x + (3 − 3C − 3E). Equating coefficients of like powers of x then yields B +3=0 C − 3B = 0 B − 3C + D + 6 = 0 C − 3B − 3D + E = 100 3 − 3C − 3E = 0 The solution to this system of equations is B = −3,

C = −9,

D = −30,

E = 10.

Therefore 3 100x −3x − 9 −30x + 10 + = , + 2 x −3 (x − 3)(x 2 + 1)2 x +1 (x 2 + 1)2 and

"

" (−30x + 10) d x (−3x − 9) d x + x2 + 1 (x 2 + 1)2 " " " " " x dx dx x dx dx dx −3 − 9 − 30 =3 + 10 . x −3 x2 + 1 x2 + 1 (x 2 + 1)2 (x 2 + 1)2

100x d x =3 (x − 3)(x 2 + 1)2

"

dx + x −3

"

For the second and fourth integrals, use the substitution u = x 2 + 1, du = 2x d x. Then we have " " 3 dx 100x d x 2 + 1| − 9 tan−1 x + 15 + 10 = 3 ln |x − 3| − . ln |x 2 2 2 2 2 (x − 3)(x + 1) x +1 (x + 1)2 For the last integral, use the trigonometric substitution x = tan θ , d x = sec2 θ d θ . Then x 2 + 1 = tan2 θ + 1 = sec2 θ , and " " " dx 1 1 sec2 θ d θ = = cos2 θ = θ + sin θ cos θ + C. 2 2 (x 2 + 1)2 sec4 θ We construct the following right triangle with tan θ = x:

460

CHAPTER 8


1 + x2

x

1

From this we see that sin θ = x/ 1 + x 2 and cos θ = 1/ 1 + x 2 . Thus " 1 1 x x 1 1 dx −1 + C. = tan x + + C = tan−1 x + 2 + 1) 2 2 2 2 2 (x 2 + 1)2 2(x 1+x 1+x Collecting all the terms, we obtain " 1 3 15 x 100x d x 2 −1 −1 + 10 = 3 ln |x − 3| − ln |x + 1| − 9 tan x + 2 tan x + +C 2 2 (x − 3)(x 2 + 1)2 x +1 2(x 2 + 1) = 3 ln |x − 3| −

5x + 15 3 + C. ln |x 2 + 1| − 4 tan−1 x + 2 2 x +1

"

9 dx " dx (x + 1)(x 2 −22x + 6) (x + 2)(x + 4x + 10) SOLUTION The partial fraction decomposition has the form:

43.

Bx + C A 9 + 2 = . x +1 (x + 1)(x 2 − 2x + 6) x − 2x + 6 Clearing denominators gives us 9 = A(x 2 − 2x + 6) + (Bx + C)(x + 1). Setting x = −1 then yields 9 = A(9) + 0

or

A = 1.

Expanding the right-hand side gives us 9 = (1 + B)x 2 + (−2 + B + C)x + (6 + C). Equating x 2 -coefficients yields 0=1+ B

or

B = −1,

while equating constant coefficients yields 9=6+C

or

C = 3.

The result is 9 −x + 3 1 + 2 = . x +1 (x + 1)(x 2 − 2x + 6) x − 2x + 6 Thus, "

9 dx = (x + 1)(x 2 − 2x + 6)

"

dx + x +1

"

(−x + 3) d x . x 2 − 2x + 6

To evaluate the integral on the right, we first write " " " " (−x + 3) d x (x − 1 − 2) d x (x − 1) d x dx = − = − + 2 . x 2 − 2x + 6 x 2 − 2x + 6 x 2 − 2x + 6 x 2 − 2x + 6 For the first integral, use the substitution u = x 2 − 2x + 6, du = (2x − 2) d x. Then " " (x − 1) d x (2x − 2) d x 1 1 − = − = − ln |x 2 − 2x + 6| + C. 2 2 x 2 − 2x + 6 x 2 − 2x + 6 For the second integral, we first complete the square: " " " dx dx dx = 2 = 2 . 2 x 2 − 2x + 6 (x 2 − 2x + 1) + 5 (x − 1)2 + 5

S E C T I O N 8.5


Now let u = x − 1, du = d x. Then " " dx 2 u −1 du 1 −1 √ −1 x√ 2 tan + C = tan + C. √ √ = 2 = 2 (x − 1)2 + 5 u2 + 5 5 5 5 5 Collecting all the terms, we have " 9 dx −1 1 2 − 2x + 6| + √2 tan−1 x√ + C. ln |x = ln |x + 1| − 2 (x + 1)(x 2 − 2x + 6) 5 5 45.

" " 2 (x + 3) 25ddxx (x 2x(x + 2x 3)2+ 5)2 2 ++2x

SOLUTION

The partial fraction decomposition has the form: x2 + 3 (x 2 + 2x + 3)2

Cx + D Ax + B + = 2 . x + 2x + 3 (x 2 + 2x + 3)2

Clearing denominators gives us x 2 + 3 = ( Ax + B)(x 2 + 2x + 3) + C x + D. Expanding the right-hand side, we get x 2 + 3 = Ax 3 + (2 A + B)x 2 + (3 A + 2B + C)x + (3B + D). Equating coefficients of like powers of x then yields A=0 2A + B = 1 3 A + 2B + C = 0 3B + D = 3 The solution to this system of equations is A = 0,

B = 1,

C = −2,

D = 0.

Therefore x2 + 3 (x 2 + 2x + 3)2

1 −2x = 2 , + 2 x + 2x + 3 (x + 2x + 3)2

and "

(x 2 + 3) d x = (x 2 + 2x + 3)2

"

dx − x 2 + 2x + 3

"

2x d x . (x 2 + 2x + 3)2

The first integral can be evaluated by completing the square: " " " dx dx dx = = . x 2 + 2x + 3 x 2 + 2x + 1 + 2 (x + 1)2 + 2 Now use the substitution u = x + 1, du = d x. Then " " 1 dx +1 du −1 x√ = = tan + C. √ x 2 + 2x + 3 u2 + 2 2 2 For the second integral, let u = x 2 + 2x " " 2x d x = (x 2 + 2x + 3)2 " =

+ 3. We want du = (2x + 2) d x to appear in the numerator, so we write " " (2x + 2 − 2) d x (2x + 2) d x dx = − 2 2 2 2 2 2 (x + 2x + 3) (x + 2x + 3) (x + 2x + 3)2 " " 1 du dx dx −2 =− −2 2 2 2 2 u u (x + 2x + 3) (x + 2x + 3)2 " −1 dx = 2 . −2 2 x + 2x + 3 (x + 3x + 3)2

Finally, for this last integral, complete the square, then substitute u = x + 1, du = d x: " " " dx dx du = = . (x 2 + 2x + 3)2 ((x + 1)2 + 2)2 (u 2 + 2)2

461

462

CHAPTER 8


√ √ Now use the trigonometric substitution u = 2 tan θ . Then du = 2 sec2 θ d θ , and u 2 + 2 = 2 tan2 θ + 2 = 2 sec2 θ . Thus √ " √ √ √ " √ " 2 sec2 θ d θ 2 du 2 θ d θ = 2 1 θ + 1 sin θ cos θ = 2 θ + 2 sin θ cos θ + C. cos = = 4 4 2 2 8 8 (u 2 + 2)2 4 sec4 θ √ We construct a right triangle with tan θ = u/ 2: u2 + 2

u

q 2

√ From this we see that sin θ = u/ u 2 + 2 and cos θ = 2/ u 2 + 2. Therefore √ √ √ " 2 2 2 u u du −1 = tan + +C √ 8 8 (u 2 + 2)2 2 u2 + 2 u2 + 2 √ √ 2 2 u +1 u x +1 −1 x√ + C = + C. = tan−1 √ tan + + 2 2 8 8 4(u + 2) 4(x + 2x + 3) 2 2 Collecting all the terms, we have "

* √ ) (x 2 + 3) d x 1 2 −1 x +1 x +1 x +1 −1 −1 +C = √ tan √ √ tan − + −2 8 (x 2 + 2x + 3)2 x 2 + 2x + 3 4(x 2 + 2x + 3) 2 2 2 √ 1 2 x +1 2 + (x + 1) = √ + +C tan−1 √ + 4 2(x 2 + 2x + 3) 2 2 √ 3 2 x +1 x +3 = tan−1 + C. + √ 2 4 2(x + 2x + 3) 2

" dx " in two ways: using partial fractions and trigonometric substitution. Verify that the two answers 47. Evaluated x 2 x −1 4 x −1 agree. SOLUTION

The partial fraction decomposition has the form: 1 1 A B = = + . (x − 1)(x + 1) x −1 x +1 x2 − 1

Clearing denominators gives us 1 = A(x + 1) + B(x − 1). Setting x = 1, we get 1 = A(2) or A = 12 ; while setting x = −1, we get 1 = B(−2) or B = − 12 . The result is 1 − 12 1 2 + . = x −1 x +1 x2 − 1

Thus,

"

1 dx = 2 x2 − 1

"

1 dx − x −1 2

"

1 1 dx = ln |x − 1| − ln |x + 1| + C. x +1 2 2

Using trigonometric substitution, let x = sec θ . Then d x = tan θ sec θ d θ , and x 2 − 1 = sec2 θ − 1 = tan2 θ . Thus " " " " sec θ d θ cos θ d θ dx tan θ sec θ d θ = = = tan θ sin θ cos θ x2 − 1 tan2 θ " = csc θ d θ = ln | csc θ − cot θ | + C. Now we construct a right triangle with sec θ = x: x q 1

x2 − 1

S E C T I O N 8.5


463

From this we see that csc θ = x/ x 2 − 1 and cot θ = 1/ x 2 − 1. Thus " x −1 dx 1 x = ln − + C = ln + C. x2 − 1 x2 − 1 x2 − 1 x2 − 1 To check that these two answers agree, we write √ x − 1 √x − 1 x −1 x − 1 1 1 x − 1 1 ln |x − 1| − ln |x + 1| = ·√ = ln = ln √ . = ln x +1 x2 − 1 x + 1 2 2 2 x + 1 x − 1 " √ √ x dx 2 = 10x(xu − 49. Evaluate (sometimes a rationalizing substitution). . Hint:(x Use substitution = 30)x and find the called volume of the solid obtained by revolving the Graph the equation − the 40)y x −1 region between the√graph and the x-axis√for 0 ≤ x ≤ 30 around the x-axis. SOLUTION Let u = x. Then du = (1/2 x) d x = (1/2u) d x. Thus " √

"

" " u(2u du) u 2 du (u 2 − 1 + 1) du = 2 = 2 2 2 u −1 u −1 u2 − 1 " " " " 1 2 du 2 du u2 − 1 + du = 2 du + = 2u + . =2 u2 − 1 u2 − 1 u2 − 1 u2 − 1

x dx = x −1

The partial fraction decomposition of the remaining integral has the form: A B 2 2 = + . = (u − 1)(u + 1) u−1 u+1 u2 − 1 Clearing denominators gives us 2 = A(u + 1) + B(u − 1). Setting u = 1 yields 2 = A(2) + 0 or A = 1, while setting u = −1 yields 2 = 0 + B(−2) or B = −1. The result is 2 u2 − 1

=

−1 1 + . u−1 u+1

Thus, "

2 du = u2 − 1

"

du − u−1

"

du = ln |u − 1| − ln |u + 1| + C. u+1

The final answer is " √

√ √ √ x dx = 2u + ln |u − 1| − ln |u + 1| + C = 2 x + ln | x − 1| − ln | x + 1| + C. x −1

" evaluate the integral using the appropriate method or combination of methods covered thus far in In Exercises 51–66, x dx . the text.Evaluate 1/2 − x 1/3 x " dx 51. 2 x 4 − x2 SOLUTION

Use the trigonometric substitution x = 2 sin θ . Then d x = 2 cos θ d θ , 4 − x 2 = 4 − 4 sin2 θ = 4(1 − sin2 θ ) = 4 cos2 θ ,

and

"

dx = x2 4 − x2

"

2 cos θ d θ (4 sin2 θ )(2 cos θ )

=

1 4

"

1 csc2 θ d θ = − cot θ + C. 4

Now construct a right triangle with sin θ = x/2:

2 q 4 − x2

x

464

CHAPTER 8


From this we see that cot θ =

4 − x 2 /x. Thus

"

" 53.

1 dx =− 2 2 4 x 4−x

4 − x2 4 − x2 +C =− + C. x 4x

" dx 2 dx x(xcos − 21)4x

SOLUTION

Using partial fractions, we first write A B C 1 = + . + 2 x x − 1 x(x − 1) (x − 1)2

Clearing denominators gives us 1 = A(x − 1)2 + Bx(x − 1) + C x. Setting x = 0 yields 1 = A(1) + 0 + 0

A = 1,

or

while setting x = 1 yields 1 =0+0+C

or

C = 1,

1 = 1 + 2B + 2

or

B = −1.

and setting x = 2 yields

The result is 1 1 −1 1 = + . + x x − 1 (x − 1)2 x(x − 1)2 Thus, "

" 55.

dx = x(x − 1)2

"

dx − x

"

dx + x −1

"

1 dx = ln |x| − ln |x − 1| − + C. 2 x −1 (x − 1)

" dx (x 2x+sec 9)22x d x

SOLUTION

Use the trigonometric substitution x = 3 tan θ . Then d x = 3 sec2 θ d θ , x 2 + 9 = 9 tan2 θ + 9 = 9(tan2 θ + 1) = 9 sec2 θ ,

and "

dx = (x 2 + 9)2

"

3 3 sec2 θ d θ = 81 (9 sec2 θ )2

"

1 sec2 θ d θ = 27 sec4 θ

" cos2 θ d θ =

1 27

1 1 θ + sin θ cos θ + C. 2 2

Now construct a right triangle with tan θ = x/3: x2 + 9

x

q 3

From this we see that sin θ = x/ x 2 + 9 and cos θ = 3/ x 2 + 9. Thus "

x

1 x 1 x 3 1 dx −1 −1 x + + = tan tan + C = + C. 2 2 + 9) 2 2 54 3 54 54 3 18(x x +9 x +9 x2 + 9 " 57.

" tan5 x sec−1x d x θ sec θ d θ



Use the trigonometric identity tan2 x = sec2 x − 1 to write " "

2 tan5 x sec x d x = sec2 x − 1 tan x sec x d x.

Now use the substitution u = sec x, du = sec x tan x d x: " "

" tan5 x sec x d x = (u 2 − 1)2 du = u 4 − 2u 2 + 1 du =

59.

1 5 2 3 1 2 u − u + u + C = sec5 x − sec3 x + sec x + C. 5 3 5 3

" " 2 x dxx d x 3/2 (x 2(x −21) − 1)3/2

SOLUTION

Use the trigonometric substitution x = sec θ . Then d x = sec θ tan θ d θ , x 2 − 1 = sec2 θ − 1 = tan2 θ ,

and "

" " (sec2 θ ) sec θ tan θ d θ sec3 θ d θ (tan2 θ + 1) sec θ d θ = = 2 3/2 2 (tan θ ) tan θ tan2 θ " " " " tan2 θ sec θ d θ sec θ d θ = + = sec θ d θ + csc θ cot θ d θ tan2 θ tan2 θ

x2 dx = 2 (x − 1)3/2

"

= ln | sec θ + tan θ | − csc θ + C. Now construct a right triangle with sec θ = x: x

x2 − 1

q 1

From this we see that tan θ =

x 2 − 1 and csc θ = x/ x 2 − 1. So the final answer is " x x2 dx = ln x + x 2 − 1 − + C. 2 3/2 (x − 1) x2 − 1

"

" dx dx x(x 2 −21) x(x + x) SOLUTION Using partial fractions, we first write

61.

1 x(x 2 − 1)

=

1 A B C = + + . x(x − 1)(x + 1) x x −1 x +1

Clearing denominators gives us 1 = A(x − 1)(x + 1) + Bx(x + 1) + C x(x − 1). Setting x = 0 then yields 1 = A(−1) + 0 + 0

or

A = −1,

1 = 0 + B(2) + 0

or

B=

1 , 2

1 = 0 + 0 + C(2)

or

C=

1 . 2

1

1


and setting x = −1 yields

The result is −1 1 = + 2 + 2 . x x −1 x +1 x(x 2 − 1)

465

466

CHAPTER 8


Thus,

"

dx =− x(x 2 − 1)

"

1 dx + x 2

"

dx 1 + x −1 2

"

dx 1 1 = − ln |x| + ln |x − 1| + ln |x + 1| + C. x +1 2 2

" √ " x dx (x + 1)d x x 3 + 12 (x + 4x + 8)2 3 SOLUTION Use the substitution u = x 3/2 , du = 2 x 1/2 d x. Then x 3 = (x 3/2 )2 = u 2 , so we have " " √ 2 2 2 x dx du = = tan−1 u + C = tan−1 (x 3/2 ) + C. 3 3 3 x3 + 1 u2 + 1

63.

" 65.

" dx tan3 −1 3θ d θ x 4 (x + 1)

SOLUTION

Use the substitution u = x 3 , du = 3x 2 d x. Then d x = du/3x 2 , and we have " " " " 1 1 du du du dx = = = . 3 3 x 4 (x 3 + 1) x 4 (3x 2 )(u + 1) x 6 (u + 1) u 2 (u + 1)

Now we can use partial fractions: 1 C B A . = + 2 + u u+1 u 2 (u + 1) u Clearing denominators gives us 1 = Au(u + 1) + B(u + 1) + Cu 2 . Setting u = 0 then yields 1 = 0 + B(1) + 0

or

B = 1,

1 = 0 + 0 + C(1)

or

C = 1,

while setting u = −1 yields

and setting u = 1 yields 1 = 2A + 2 + 1

or

A = −1.

The result is 1 1 −1 1 + 2 + . = u u+1 u 2 (u + 1) u Thus, 1 3

"

" " " 1 du 1 1 du du du = + − + = − ln |u| − + ln |u + 1| + C. 3 u u+1 3 u u 2 (u + 1) u2

The final answer is

"

dx x 4 (x 3 + 1)

1 1 1 = − lnx 3 − 3 + lnx 3 + 1 + C. 3 3 3x

67. Show the substitution θ = 2 tan−1 t (Figure 2) yields the formulas " that x 1/2 d x 1 − t2 2t 2 dt x 1/3 + 1 , sin θ = , dθ = cos θ = 1 + t2 1 + t2 1 + t2

10

This substitution transforms the integral of any rational function of cos θ and " sin θ into an integral of a rational function dθ of t (which can then be evaluated using partial fractions). Use it to evaluate . cos θ + (3/4) sin θ 1 + t2 q /2 1

FIGURE 2

t



467

If θ = 2 tan−1 t, then d θ = 2 dt/(1 + t 2 ). We also have that cos( θ2 ) = 1/ 1 + t 2 and sin( θ2 ) =

t/ 1 + t 2 . To find cos θ , we use the double angle identity cos θ = 1 − 2 sin2 ( θ2 ). This gives us

2

t

cos θ = 1 − 2 1 + t2

=1−

2t 2 1 + t 2 − 2t 2 1 − t2 = = . 1 + t2 1 + t2 1 + t2

To find sin θ , we use the double angle identity sin θ = 2 sin( θ2 ) cos( θ2 ). This gives us 1 2t t sin θ = 2 = . 2 2 1 + t2 1+t 1+t With these formulas, we have "

dθ = cos θ + (3/4) sin θ

"

"

2 dt

2

1+t

= 1−t 2 + 3 2t 2 2 4 1+t

8 dt = 4(1 − t 2 ) + 3(2t)

1+t

"

8 dt = 4 + 6t − 4t 2

"

The partial fraction decomposition has the form A B 4 = + . 2 2 − t 1 + 2t 2 + 3t − 2t Clearing denominators gives us 4 = A(1 + 2t) + B(2 − t). Setting t = 2 then yields 4 = A(5) + 0 while setting t = − 12 yields 4=0+ B

or

5 2

or

A=

4 , 5

B=

8 . 5

The result is 4

8

4 = 5 + 5 . 2−t 1 + 2t 2 + 3t − 2t 2 Thus, "

4 4 dt = 2 5 2 + 3t − 2t

"

dt 8 + 2−t 5

"

dt 4 4 = − ln |2 − t| + ln |1 + 2t| + C. 1 + 2t 5 5

The original substitution was θ = 2 tan−1 t, which means that t = tan( θ2 ). The final answer is then " dθ 4 θ 4 θ = − ln ln 2 − tan + 1 + 2 tan + C. 3 5 2 5 2 cos θ + sin θ 4

Further and Challenges UseInsights the substitution of Exercise 67 to evaluate 69. Prove the general formula "

"

dθ . cos θ + sin θ

1 x −a dx = ln +C (x − a)(x − b) a−b x −b

where a, b are constants such that a = b. SOLUTION

The partial fraction decomposition has the form: A B 1 = + . (x − a)(x − b) x −a x −b

Clearing denominators, we get 1 = A(x − b) + B(x − a).

4 dt . 2 + 3t − 2t 2

468

CHAPTER 8


Setting x = a then yields 1 = A(a − b) + 0

or

A=

1 , a−b

1 = 0 + B(b − a)

or

B=

1 . b−a

while setting x = b yields

The result is 1

1

1 = a−b + b−a . (x − a)(x − b) x −a x −b Thus,

"

1 dx = (x − a)(x − b) a−b

"

dx 1 + x −a b−a

"

dx 1 1 = ln |x − a| + ln |x − b| + C x −b a−b b−a x − a 1 1 1 + C. ln |x − a| − ln |x − b| + C = ln = a−b a−b a − b x − b

P(x) 71. Suppose that Q(x) = (x − a)(x − shows b), where be a proper rational function so that The method of partial fractions thata = b, and let Q(x) " 1 d xP(x) 1 A − Bln x + 1 + C == ln x − 1+ 2 2 x2 − 1 Q(x) (x − a) (x − b) P(a) system Mathematica P(b) The computer algebra evaluates this integral as − tanh−1 x, where tanh−1 x is the inverse (a) Show that A = and B = . hyperbolic tangent function. Can Q you Q (a) (b)reconcile the two answers? (b) Use this result to find the partial fraction decomposition for P(x) = 3x − 2 and Q(x) = x 2 − 4x − 12. SOLUTION

(a) Clearing denominators gives us P(x) = A(x − b) + B(x − a). Setting x = a then yields P(a) = A(a − b) + 0

or

A=

P(a) , a−b

P(b) = 0 + B(b − a)

or

B=

P(b) . b−a

while setting x = b yields

Now use the product rule to differentiate Q(x): Q (x) = (x − a)(1) + (1)(x − b) = x − a + x − b = 2x − a − b; therefore, Q (a) = 2a − a − b = a − b Q (b) = 2b − a − b = b − a Substituting these into the above results, we find A=

P(a) Q (a)

and

B=

P(b) . Q (b)

(b) The partial fraction decomposition has the form: 3x − 2 P(x) 3x − 2 A B = = 2 = + ; Q(x) (x − 6)(x + 2) x −6 x +2 x − 4x − 12 A=

P(6) 3(6) − 2 16 = = = 2; Q (6) 2(6) − 4 8

B=

3(−2) − 2 −8 P(−2) = = = 1. Q (−2) 2(−2) − 4 −8

The result is 2 3x − 2 1 = + . x −6 x +2 x 2 − 4x − 12 Suppose that Q(x) = (x − a1 )(x − a2 ) · · · (x − an ) where the roots a j are all distinct. Let

P(x) be a proper Q(x)

S E C T I O N 8.6

Improper Integrals

469

8.6 Improper Integrals Preliminary Questions 1. State whether the integral converges or diverges: " ∞ (a) x −3 d x 1 " ∞ x −2/3 d x (c)

" 1 (b)

x −3 d x

0

" 1 (d)

1

x −2/3 d x

0

SOLUTION

(a) The integral is improper because one of the limits of integration is infinite. Because the power of x in the integrand is less than −1, this integral converges. (b) The integral is improper because the integrand is undefined at x = 0. Because the power of x in the integrand is less than −1, this integral diverges. (c) The integral is improper because one of the limits of integration is infinite. Because the power of x in the integrand is greater than −1, this integral diverges. (d) The integral is improper because the integrand is undefined at x = 0. Because the power of x in the integrand is greater than −1, this integral converges. " π /2 cot x d x an improper integral? Explain. 2. Is 0

Because the integrand cot x is undefined at x = 0, this is an improper integral. " b 1 d x an improper integral. 3. Find a value of b > 0 that makes 2 x −4 0

SOLUTION

Any value of b satisfying |b| ≥ 2 will make this an improper integral. " ∞ dx converges? 4. Which comparison would show that x + ex 0

SOLUTION

SOLUTION

Note that, for x > 0, 1 1 < x = e−x . x + ex e

Moreover

" ∞

e−x d x

0

converges. Therefore, " ∞ 0

1 dx x + ex

converges by the comparison test.

" ∞ −x e 5. Explain why it is not possible to draw any conclusions about the convergence of d x by comparing with x 1 " ∞ dx the integral . x 1 SOLUTION

For 1 ≤ x < ∞, 1 e−x < , x x

but

" ∞ dx x 1

diverges. Knowing that an integral is smaller than a divergent integral does not allow us to draw any conclusions using the comparison test.

470

CHAPTER 8


Exercises 1. Which of the following integrals is improper? Explain your answer, but do not evaluate the integral. " 2 " ∞ dx dx (a) d x (b) x 0.2 0 x 1/3 1 " ∞ " 1 (c) e−x d x (d) e−x −1

0

" π /2 (e)

" ∞ sec x d x

(f)

0

" 1

" 1 sin x d x

(g)

(h)

0

0

" ∞ (i)

sin x d x 0

" 3 ln x d x

(j)

1

dx 3 − x2

ln x d x 0

SOLUTION

Improper. The function x −1/3 is infinite at 0. Improper. Infinite interval of integration. Improper. Infinite interval of integration. Proper. The function e−x is continuous on the finite interval [0, 1]. Improper. The function sec x is infinite at π2 . Improper. Infinite interval of integration. Proper. The function sin x is continuous on the finite interval [0, 1]. (h) Proper. The function 1/ 3 − x 2 is continuous on the finite interval [0, 1]. (i) Improper. Infinite interval of integration. (j) Improper. The function ln x is infinite at 0. " ∞ −2/3 3. Prove that Let f (x) = xx−4/3 .d x diverges by showing that 1" R " R (a) Evaluate f (x) d x. 1 x −2/3 d x = ∞ lim " ∞ R→∞ 1 (b) Evaluate f (x) d x by computing the limit 1 SOLUTION First compute the proper integral: " R R " R f (x) d x

lim 1/3 − 3 = 3 R 1/3 − 1 . = 3R x −2/3 d x = 3x 1/3R→∞ 1

(a) (b) (c) (d) (e) (f) (g)

1

1

Then show divergence: " ∞ 1

x −2/3 d x = lim

" R

R→∞ 1

x −2/3 d x = lim 3 R 1/3 − 1 = ∞. R→∞

" 3 In Exercises 5–46, determine whether the improper integral converges and, if so, evaluate it. dx Determine if converges by computing " ∞ dx 0 (3 − x)3/2 5. x 19/20 " R 1 dx lim SOLUTION First evaluate the integral over the finite interval [1, R] for3/2 R > 1: R→3− 0 (3 − x) R " R dx 1/20 = 20R 1/20 − 20. = 20x 19/20 x 1 1 Now compute the limit as R → ∞: " ∞ 1

The integral does not converge. " 4 " ∞0.0001t d x dt 7. e −∞ x 20/19 1

" R

dx = lim = lim 20R 1/20 − 20 = ∞. 19/20 19/20 R→∞ 1 x R→∞ x dx


Improper Integrals

471

First evaluate the integral over the finite interval [R, 4] for R < 4: " 4 R

4

e(0.0001)t e(0.0001)t dt = 0.0001

Now compute the limit as R → −∞: " 4 e(0.0001)t dt = lim −∞

" 4

R→−∞ R

= 10,000 e0.0004 − e(0.0001)R .

R

e(0.0001)t dt =

lim

R→−∞

10,000 e0.0004 − e(0.0001)R

= 10,000 e0.0004 − 0 = 10,000e0.0004 .

" 5 " ∞ dx dt 0 x 20/19 20 t SOLUTION The function x −20/19 is infinite at the endpoint 0, so we’ll first evaluate the integral on the finite interval [R, 5] for 0 < R < 5: 5 " 5

1 1 dx −1/19 = −19 5−1/19 − R −1/19 = 19 = −19x − . R 1/19 51/19 R x 20/19 R 9.

Now compute the limit as R → 0+ : " 5 " 5 1 1 dx dx = lim = lim 19 − = ∞; R 1/19 51/19 R→0+ R x 20/19 R→0+ 0 x 20/19 thus, the integral does not converge. " 4 " 5d x 11. √ dx 4−x 0 0 x 19/20 √ SOLUTION The function 1/ 4 − x is infinite at x = 4, but is left-continuous at x = 4, so we’ll first evaluate the integral on the interval [0, R] for 0 < R < 4: " R R √ √ √ √ dx = −2 4 − x = −2 4 − R − (−2) 4 = 4 − 2 4 − R. √ 0 4 − x 0 Now compute the limit as R → 4− : " 4 " R

√ dx dx = lim = lim 4 − 2 4 − R = 4 − 0 = 4. √ √ − − 4−x R→4 4−x R→4 0 0 " ∞ " x6−3 d x dx 2 5 (x − 5)3/2 SOLUTION First evaluate the integral on the finite interval [2, R] for 2 < R: 13.

" R 2

R

x −2 x −3 d x = −2

2

=

1 −1 −1 1 = − − . 2 2 8 2R 2(2 ) 2R 2

Now compute the limit as R → ∞: " R " ∞ 1 1 1 x −3 d x = lim x −3 d x = lim = . − 2 8 8 R→∞ R→∞ 2R 2 2 " ∞ " ∞ dx dx −3 (x + 4)3/2 3 0 (x + 1) SOLUTION First evaluate the integral on the finite interval [−3, R] for R > −3: 15.

R 2 −2 dx −2 −1/2 = −2(x + 4) = √ − √ =2− √ . 3/2 R+4 R+4 1 −3 (x + 4) −3

" R

Now compute the limit as R → ∞: " R " ∞ dx dx 2 = lim = lim = 2 − 0 = 2. 2 − √ R→∞ −3 (x + 4)3/2 R→∞ R+4 −3 (x + 4)3/2

472

CHAPTER 8


17.

" 1 " d∞ x −2x d x 0 x 0.2e 2

The function x −0.2 is infinite at x = 0 and right-continuous at x = 0, so we’ll first evaluate the integral on the interval [R, 1] for 0 < R < 1: 1 " 1

x 0.8 dx 0.8 . = = 1.25 1 − R 0.8 R x 0.2 SOLUTION

R

Now compute the limit as R → 0+ : " 1 " 1

dx dx 0.8 = 1.25(1 − 0) = 1.25. = lim = lim 1.25 1 − R R→0+ R x 0.2 R→0+ 0 x 0.2 19.

" ∞ " ∞ e−3x−1/3 dx x dx 4 2

SOLUTION

First evaluate the integral on the finite interval [4, R] for R > 4: " R 4

R

e−3x e−3x d x = −3

=−

4

1

1 −3R − e−12 = e e−12 − e−3R . 3 3

Now compute the limit as R → ∞: " ∞ " R 1

1 1 −12 e e−12 − 0 = 12 . e−3x d x = lim e−3x d x = lim − e−3R = 3 3 R→∞ R→∞ 3e 4 4 21.

" 0 " ∞3x e d3xx e dx −∞ 4

SOLUTION

First evaluate the integral on the finite interval [R, 0] for R < 0: " 0 R

0

e3x e3x d x = 3

=

R

1 e3R − . 3 3

Now compute the limit as R → −∞: " 0 −∞

e3x d x =

" 0 lim

R→−∞ R

e3x d x =

lim

R→−∞

1 e3R − 3 3

=

1 1 −0= . 3 3

" ∞ " 1 dx dx 2 (x + 3)4 0 x 3.7 SOLUTION First evaluate the integral on the finite interval [2, R] for R > 2: 23.

" R 2

R (x + 3)−3 −1 1 dx 1 = − 3 . = −3 3 (R + 3)3 (x + 3)4 5 2

Now compute the limit as R → ∞: " R " ∞ 1 1 dx dx 1 1 1 1 = lim = lim − − 0 = . = 3 53 375 R→∞ 2 (x + 3)4 R→−∞ 3 53 (R + 3)3 2 (x + 3)4 " 3 " 2d x √ dx 3−x 2 1 √ 1 (x − 1) SOLUTION The function f (x) = 1/ 3 − x is infinite at x = 3 and is left continuous at x = 3, so we first evaluate the integral on the interval [1, R] for 1 < R < 3: R " R √ √ √ dx √ = −2 3 − x = −2 3 − R + 2 2. 3−x 1 1 25.

Now compute the limit as R → 3− : " R " 3 √ √ dx dx = lim = 0 + 2 2 = 2 2. √ √ R→3− 1 3−x 3−x 1

S E C T I O N 8.6

Improper Integrals

473

" ∞ " 4d x dx 0 1+x −2 (x + 2)1/3 SOLUTION First evaluate the integral on the finite interval [0, R] for R > 0: 27.

" R R dx = ln |1 + x|0 = ln |1 + R| − ln 1 = ln |1 + R|. 1 + x 0 Now compute the limit as R → ∞: " ∞ 0

" R dx dx = lim = lim ln |1 + R| = ∞; 1+x R→∞ 0 1 + x R→∞

thus, the integral does not converge. " ∞ " 0 dx 29. 2 )22 0 (1 + xex−x dx −∞ SOLUTION First evaluate the indefinite integral using the trigonometric substitution x = tan θ . Then d x = sec2 θ d θ , 1 + x 2 = 1 + tan2 θ = sec2 θ , and we have " " " 1 1 dx sec2 θ d θ = = cos2 θ d θ = θ + sin θ cos θ + C. 2 2 (1 + x 2 )2 sec4 θ Now construct a right triangle with tan θ = x: 1 + x2

x

q 1

From this we see that sin θ = x/ 1 + x 2 and cos θ = 1/ 1 + x 2 . Thus " 1 dx x 1 1 x 1 −1 + C. = tan x + + C = tan−1 x + 2 2 2 2 2 2 2 (1 + x ) 2(1 + x 2 ) 1+x 1+x Finally, " ∞ 0

" R 1 dx dx R −1 R + tan − 0 = lim = lim 2 R→∞ 0 (1 + x 2 )2 R→∞ (1 + x 2 )2 2(1 + R 2 ) 0 1/R π π 1 1 π + = lim tan−1 R + = +0= . = 2 2 2 0+2 4 4 R→∞ 2 2/R + 2

" ∞ " e6−x cos x d x x dx 0 √ x −3 3 SOLUTION First evaluate the indefinite integral using Integration by Parts, with u = e−x , v = cos x. Then u = −e−x , v = sin x, and " " " e−x cos x d x = e−x sin x − sin x(−e−x ) d x = e−x sin x + e−x sin x d x. 31.

Now use Integration by Parts again, with u = e−x , v = sin x. Then u = −e−x , v = − cos x, and " " e−x cos x d x = e−x sin x + −e−x cos x − e−x cos x d x . Solving this equation for

! −x e cos x d x, we find " 1 e−x cos x d x = e−x (sin x − cos x) + C. 2

Thus, " R 0

and

e−x cos x d x =

R 1 −x sin R − cos R sin 0 − cos 0 1 sin R − cos R − + , e (sin x − cos x) = = R R 2 2 2 2e 2e 0

" ∞ 0

1 sin R − cos R + 2 R→∞ 2e R

e−x cos x d x = lim

=0+

1 1 = . 2 2

474

CHAPTER 8


33.

" 1 " 0 dx 2 −x 0 1 −x xe2 d x −∞

The function (1 − x 2 )−1/2 is infinite at x = 1, and is left-continuous at x = 1, so we’ll first evaluate the integral on the interval [0, R] for 0 < R < 1: SOLUTION

" R 0

R dx = sin−1 x = sin−1 R − sin−1 0 = sin−1 R. 0 1 − x2

Thus, " 1

dx π = lim sin−1 R = sin−1 1 = . − 2 2 R→1 0 1−x

" ∞ √ x√ " e1 dxx 35. √ e dx x√ 1 x 0 √ 1 SOLUTION Let u = x, du = 2 x −1/2 d x. Then " √x " √ " √ e dx dx =2 e x = 2 eu du = 2eu + C = 2e x + C, √ √ x 2 x and " ∞ √x " R √x

√ √ R e dx e dx = lim = lim 2e x = lim 2e R − 2 = ∞. √ √ 0 x x R→∞ 0 R→∞ R→∞ 0 The integral does not converge. " ∞ x dx " ∞ 37. x 2 xe + −2x 1 dx 0 1

SOLUTION

First, evaluate the integral on the finite interval [0, R] for R > 0: " R 0

R x 1 2 ) = 1 ln(1 + R 2 ). ln(1 + x d x = 2 2 2 x +1 0

Now compute the limit as R → ∞: " ∞ 0

" R x x 1 d x = lim d x = lim ln(1 + R 2 ) = ∞. 2 2 R→∞ 0 x + 1 R→∞ 2 x +1

The integral does not converge. " ∞ " sin π /2x d x 39. 0 sec θ d θ 0

SOLUTION

First evaluate the integral on the finite interval [0, R] for R > 0: " R 0

R sin x d x = − cos x = − cos R + cos 0 = 1 − cos R. 0

Thus, " R 0

sin x d x = lim (1 − cos R) = 1 − lim cos R. R→∞

R→∞

This limit does not exist, since the value of cos R oscillates between 1 and −1 as R approaches infinity. Hence the integral does not converge. " π /2 " π /2 41. tan x sec x d x tan x d x 0 0

SOLUTION

The function tan x sec x is infinite and left-continuous at x = π2 so we’ll first evaluate the integral on [0, R]

for 0 < R < π2 :

" R 0

R tan x sec x d x = sec x = sec R − 1. 0

S E C T I O N 8.6

Improper Integrals

475

Thus, " π /2 0

tan x sec x d x =

lim (sec R − 1) = ∞.

R→ π2 −

The integral does not converge. " 1 " 1 x ln x d x 43. ln x d x 0 0

The function x ln x is undefined at x = 0, so we’ll first evaluate the integral on [R, 1] for 0 < R < 1. Use Integration by Parts with u = ln x and v = x. Then u = 1/x, v = x 2 /2, and 1 1 " " 1 1 1 1 1 2 1 x ln x d x = x 2 ln x − x dx = x ln x − x 2 2 2 R 2 4 R R R 1 1 1 1 1 1 2 1 = ln 1 − − R ln R − R 2 = R 2 − R 2 ln R − . 2 4 2 4 4 2 4 SOLUTION

Thus, " 1 0

ln x d x = lim

R→0+

1 1 2 1 2 R − R ln R − 4 2 4

1 2 1 = − − lim R ln R. 4 R→0+ 2

To evaluate the limit, rewrite the function as a quotient and apply L’Hôpital’s Rule: " 1

1 1 1 1 ln R R2 1 1 R x ln x d x = − − lim = − = − − lim − lim =− −0=− . 1 1 + + + 4 R→0 2 2 4 R→0 −4 3 4 R→0 −4 4 4 0 R

R

" 1 "ln2x dx 0 x2 1 x ln x SOLUTION Use Integration by Parts, with u = ln x and v = x −2 . Then u = 1/x, v = −x −1 , and " " 1 1 1 ln x dx d x = − = − ln x − + C. ln x + 2 2 x x x x x 45.

The function is infinite and right-continuous at x = 0, so we’ll first evaluate the integral on [R, 1] for 0 < R < 1: " 1 1 1 1 1 1 1 1 1 1 ln x d x = − = − ln x − ln 1 − − − ln R − = ln R + − 1. 2 x x R 1 1 R R R R a x Thus, " 1 1 ln x 1 ln R + 1 d x = lim ln R + − 1 = −1 + lim = −∞. R R R→0+ R R→0+ 0 x2 The integral does not converge. " ∞ dx " 47. Let I ∞ = ln x . x − 2)(x − 3) 42 d(x x R > 4, (a) Show 1that for " R 4

R − 3 dx − ln 1 = ln (x − 2)(x − 3) R − 2 2

(b) Then show that I = ln 2. SOLUTION

(a) The partial fraction decomposition takes the form 1 A B = + . (x − 2)(x − 3) x −2 x −3 Clearing denominators gives us 1 = A(x − 3) + B(x − 2). Setting x = 2 then yields A = −1, while setting x = 3 yields B = 1. Thus, " " " x − 3 dx dx dx + C, = − = ln |x − 3| − ln |x − 2| + C = ln (x − 2)(x − 3) x −3 x −2 x − 2

476

CHAPTER 8


and, for R > 4, " R 4

x − 3 R dx = ln R − 3 − ln 1 . = ln (x − 2)(x − 3) x −2 4 R − 2 2

(b) Using the result from part (a), R − 3 − ln 1 = ln 1 − ln 1 = ln 2. ln R − 2 2 2 R→∞

I = lim

" 1 dx " ∞ d x and, if so, evaluate. 49. Determine if I = converges . Evaluate the integral I =+ 5) x(2x 0 x(2x + 5) 1 SOLUTION The partial fraction decomposition takes the form A B 1 = + . x(2x + 5) x 2x + 5 Clearing denominators gives us 1 = A(2x + 5) + Bx. Setting x = 0 then yields A = 15 , while setting x = − 52 yields B = − 25 . Thus, "

1 dx = x(2x + 5) 5

"

2 dx − x 5

1 1 1 x dx = ln |x| − ln |2x + 5| + C = ln + C, 2x + 5 5 5 5 2x + 5

"

and, for 0 < R < 1, dx 1 x 1 1 1 1 R = ln ln − ln = . 5 2x + 5 R 5 7 5 2R + 5 R x(2x + 5)

" 1

Thus, 1 1 1 R ln − ln = ∞. 7 5 2R + 5 R→0+ 5

I = lim The integral does not converge.

" ∞if the doubly infinite improper integral converges and, if so, evaluate it. Use definition (2). In Exercises 51–54, determine dx converges and, if so, evaluate. Determine if I = " ∞ dx 2 (x + 3)(x + 1)2 51. −∞ 1 + x 2 SOLUTION

First note that "

dx = tan−1 x + C. 1 + x2

Thus, " R

dx dx −1 R − tan−1 0 = π ; tan = lim = lim 2 R→∞ 0 1 + x 2 R→∞ 0 1 + x2 " 0 " 0

π dx dx tan−1 0 − tan−1 R = ; = lim = lim 2 2 2 R→−∞ R 1 + x R→−∞ −∞ 1 + x " ∞

and

" ∞

dx π π = + = π. 2 2 2 ∞ 1+x

53.

" ∞ " ∞ −x 2 xe −|x| dx e dx −∞ −∞

SOLUTION

First note that "

2 2 1 xe−x d x = − e−x + C. 2

S E C T I O N 8.6

Improper Integrals

Thus, 1 1 −R 2 1 − e = ; 2 2 2 R→∞ R→∞ 0 0 " 0 " 0 2 2 2 1 1 1 xe−x d x = lim xe−x d x = lim − + e−R = − ; 2 2 2 R→−∞ R R→−∞ −∞ " ∞

" R

xe−x d x = lim 2

and

" ∞ −∞

" ∞ 55. For which values d x of a does

" ∞

2

xe−x d x = 2

1 1 − = 0. 2 2

eax d x converge?

0

−∞ (x 2 + 1)3/2

SOLUTION

xe−x d x = lim

First evaluate the integral on the finite interval [0, R] for R > 0: R

" R

1 eax d x = eax a

0

= 0

1 aR e −1 . a

Thus, " ∞ 0

1 aR e −1 . R→∞ a

eax d x = lim

If a > 0, then ea R → ∞ as R → ∞. If a < 0, then ea R → 0 as R → ∞, and " ∞ 0

1 1 aR e −1 =− . a R→∞ a

eax d x = lim

The integral converges for a < 0. 1 " 1 under the graph of f (x) = 57. Sketch the region for −∞ < x < ∞ and show that its area is π . dx 1 + x 2 if p ≥ 1. converges if p < 1 and diverges Show that p 0 x SOLUTION The graph is shown below. y 1 0.8 0.6 0.4 0.2 −4

−2

x 2

4

Since (1 + x 2 )−1 is an even function, we can first compute the area under the graph for x > 0: " R 0

dx −1 x R = tan−1 R − tan−1 0 = tan−1 R. = tan 0 1 + x2

Thus, " ∞ 0

dx π = lim tan−1 R = . 2 R→∞ 1 + x2

By symmetry, we have " ∞

" 0 " ∞ dx dx dx π π = + = + = π. 2 2 2 2 2 −∞ 1 + x −∞ 1 + x 0 1+x

" ∞ " ∞ dx " converges by comparing with x −3 d x. ∞ 59. Show that dx 1 1 that converges. Show that 1 x 3 + 4 ≤ 2 for all x and use this to prove 1 4 1 x +"1 ∞ x x4 + 1 SOLUTION The integral x −3 d x converges because 3 > 1. Since x 3 + 4 ≥ x 3 , it follows that 1

1 1 ≤ 3. x3 + 4 x

477

478

CHAPTER 8


Therefore, by the comparison test, " ∞ 1

61. " ∞

dx converges. x3 + 4

" ∞Then use the Comparison Test and (3) to show that −x 2 ≤ e−x for x ≥ 1 (Figure 11). ∞0≤ Show"that d xe 2x −3 d x. Show that converges by comparing with 2 3−4 −x x 2 2 e d x converges.

0 y 1

y = e −x

2

y = e −|x| x

− 4 −3 −2 −1

1

2

3

4

2 FIGURE 11 Comparison of y = e−|x| and y = e−x .

SOLUTION

For x ≥ 1, x 2 ≥ x, so −x 2 ≤ −x and e−x ≤ e−x . Now " ∞ " ∞ 2 e−x d x converges, so e−x d x converges 2

1

by the comparison test. Finally, because e

1 −x 2

is continuous on [0, 1], " ∞ 2 e−x d x converges. 0

" ∞ " ∞ " 1∞− sin x 63. Show that 2 d x converges. e−|x| d x (Figure 11). Prove that xe2−x d x converges by comparing with 1 " ∞ −∞ −∞ 1 − sin x 2 SOLUTION Let f (x) = . Since f (x) ≤ 2 and 2x −2 d x = 2, it follows that 2 x x 1 " ∞ 1 − sin x d x converges x2 1 by the comparison test. In Exercises 65–74, use the Comparison x aTest to determine whether or not the integral converges. = ∞ (by Exercise 56 in Section 7.7). Let a > 0. Recall that lim " ∞ x→∞ ln x 1 that xdax > 2 ln x for all x sufficiently large. 65. (a) Show 1 x 5 + 2 −x a < x −2√for all x sufficiently large. (b) Show that e " 5 SOLUTION Since x∞ +−x 2 a≥ x 5 = x 5/2 , it follows that (c) Show that e d x converges. 1 1 1 ≤ 5/2 . 5 x x +2 " ∞ The integral d x/x 5/2 converges because 52 > 1. Therefore, by the comparison test: 1

" ∞ 1

dx x5 + 2

also converges.

" ∞ " ∞d x √ dx x −31 3 1/2 2x + 4) 1 (x +√ √ SOLUTION Since x ≥ x − 1, we have (for x > 1) 67.

1 1 √ ≤ √ . x x −1

S E C T I O N 8.6

" ∞

√ d x/ x =

Improper Integrals

479

" ∞

d x/x 1/2 diverges because 12 < 1. Since the function x −1/2 is continuous (and there" ∞ d x/x 1/2 diverges. Therefore, by the comparison test, fore finite) on [1, 3], we also know that

The integral

1

1

3

" ∞

√

3

dx also diverges. x −1

" ∞ " e5−(x+x −1 ) d x dx 1 0 x 1/3 + x 3 1 1 SOLUTION For all x ≥ 1, x > 0 so x + x ≥ x. Then and − x + x −1 ≤ −x " ∞ e−x d x converges by direct computation: The integral 69.

−1 e−(x+x ) ≤ e−x .

1

" ∞ 1

e−x d x = lim

" R

R→∞ 1

R e−x d x = lim −e−x = lim −e−R + e−1 = 0 + e−1 = e−1 . R→∞

1

R→∞

Therefore, by the comparison test, " ∞

−1 e−(x+x ) also converges.

1

" 1 x "e 1 dx x sin 0 x2 √ dx x 0 SOLUTION For 0 < x < 1, e x > 1, and therefore 71.

ex 1 < 2. 2 x x " 1 The integral 0

d x/x 2 diverges since 2 > 1. Therefore, by the comparison test, " 1 x e also diverges. 0 x2

" 1 " ∞1 √1 d x 0 x4 + 4 x x dx x +e 1 √ √ SOLUTION For 0 < x < 1, x 4 + x ≥ x, and 73.

1 x4 + " 1 The integral 0

1 √ ≤ √ . x x

√ (1/ x) d x converges, since p = 12 < 1. Therefore, by the comparison test, " 1

dx √ also converges. x

0 x4 +

75. An "investment ∞ ln x pays a dividend of $250/year continuously forever. If the interest rate is 7%, what is the present value of the entire income d x stream generated by the investment? 1 sinh x SOLUTION The present value of the income stream after T years is " T 0

T

250e−0.07t 250e−0.07t dt = −0.07

0

=

−250 −0.07T 250

e 1 − e−0.07T . −1 = 0.07 0.07

Therefore the present value of the entire income stream is " ∞ " T 250 250 250

1 − e−0.07T = 250e−0.07t = lim 250e−0.07t = lim (1 − 0) = = $3,571.43. 0.07 0.07 T →∞ 0 T →∞ 0.07 0 An investment is expected to earn profits at a rate of 10,000e0.01t dollars/year forever. Find the present value of the income stream if the interest rate is 4%.

480

CHAPTER 8


77. Compute the present value of an investment that generates income at a rate of 5,000te0.01t dollars/year forever, assuming an interest rate of 6%. SOLUTION

The present value of the income stream after T years is " T " T

5000te0.01t e−0.06t dt = 5000 te−0.05t dt 0

0

Compute the indefinite integral using Integration by Parts, with u = t and v = e−0.05t . Then u = 1, v = (−1/0.05)e−0.05t , and " " −t −0.05t 1 20 −0.05t + +C e e−0.05t dt = −20te−0.05t + e te−0.05t dt = 0.05 0.05 −0.05 = e−0.05t (−20t − 400) + C. Thus, " T 5000 0

T te−0.05t dt = 5000e−0.05t (−20t − 400)0 = 5000e−0.05T (−20T − 400) − 5000(−400) = 2,000,000 − 5000e−0.05T (20T + 400).

Use L’Hôpital’s Rule to compute the limit: 5000(20) 5000(20T + 400) lim 2,000,000 − = 2,000,000 − 0 = $2,000,000. = 2,000,000 − lim T →∞ T →∞ 0.05e0.05T e0.05T for 1 ≤ x < ∞ 79. Let S be the solid obtained by rotating the region below the graph of y = x −1 about the x-axis Find the volume of the solid obtained by rotating the region below the graph of y = e−x about the x-axis for (Figure 12). 0 ≤ x < ∞. (a) Use the Disk Method (Section 6.3) to compute the volume of S. Note that the volume is finite even though S is an infinite region. (b) It can be shown that the surface area of S is " ∞ A = 2π x −1 1 + x −4 d x. 1

Show that A is infinite. If S were a container, you could fill its interior with a finite amount of paint, but you could not paint its surface with a finite amount of paint. y

y = x −1

x

FIGURE 12 SOLUTION

(a) The volume is given by V =

" ∞ 1

π

2 1 d x. x

First compute the volume over a finite interval: R 2 " R x −1 1 −1 −1 1 −2 π dx = π x dx = π − =π 1− . =π x −1 R 1 R 1 1 1

" R

Thus, V = lim

" ∞

R→∞ 1

1 π x −2 d x = lim π 1 − = π. R R→∞

S E C T I O N 8.6

Improper Integrals

481

(b) For x > 1, we have √ 1 1 x4 + 1 x2 1 x4 + 1 x4 1 1+ 4 = = ≥ = = . 4 3 3 3 x x x x x x x x The integral

" ∞ 1 d x diverges, since p = 1 ≥ 1. Therefore, by the comparison test, x 1 " ∞ 1 1 1 + 4 d x also diverges. x x 1

Finally, A = 2π

" ∞ 1 1 1 + 4 dx x x 1

diverges. 81. When a capacitor of capacitance C is charged by a source of voltage V , the power expended at time t is Compute the volume of the solid obtained by rotating the region below the graph of y = (x 2 + 1)−1 about the x-axis for −∞ < x < ∞. V 2 −t/RC − e−2t/RC ), (e P(t) = R where R is the resistance in the circuit. The total energy stored in the capacitor is " ∞ W = P(t) dt 0

Show that W = 12 C V 2 . SOLUTION

The total energy contained after the capacitor is fully charged is W =

" V 2 ∞ −t/RC e − e−2t/RC dt. R 0

The energy after a finite amount of time (t = T ) is T " V2 RC −2t/RC V 2 T −t/RC −2t/RC −t/RC e dt = −e + −RCe e R 0 R 2 0 1 1 2 −T /RC −2T /RC = V C −e + e − −1 + 2 2 1 1 − e−T /RC + e−2T /RC . = CV 2 2 2 Thus, W = lim C V 2 T →∞

1 1 − e−T /RC + e−2T /RC 2 2

= CV 2

1 1 − 0 + 0 = C V 2. 2 2

When a radioactive substance decays, thethat fraction at time is end f (t)of = ae−kt , where >0 Conservation of Energy can be used to show whenofaatoms mass present m oscillates at tthe withkspring " spring ∞ constant k, the period of oscillation is is the decay constant. It can be shown that the average life of an atom (until it decays) is A = − t f (t) dt. Use 0 " ∞ " √2E/k √ dx Integration by Parts to show that A = f (t)Tdt=and A. What is the average decay time of Radon-222, whose 4 compute m 0 0 2E − kx 2 half-life is 3.825 days? √ where E is the total energy of the mass. Show that this is an improper integral with value T = 2π m/k. SOLUTION Let u = t, v = f (t). Then u = 1, v = f (t), and ∞ " ∞ " ∞ A=− t f (t) dt = −t f (t) + f (t) dt. 83.

0

0

0

Since f (t) = e−kt , we have R ∞ −R −1 −kt −t f (t) 0 = lim −te = lim −Re−Rt + 0 = lim Rt = lim = 0. Rt R→∞ R→∞ R→∞ R→∞ e Re 0

482

CHAPTER 8


Here we used L’Hôpital’s Rule to compute the limit. Thus " ∞ " ∞ A= f (t) dt = e−kt dt. 0

0

Now, " R 0

R

1 e−kt dt = − e−kt k

=−

0

1

1 −k R e 1 − e−k R , −1 = k k

so 1 1

1 1 − e−k R = (1 − 0) = . k k R→∞ k

A = lim

Because k has units of (time)−1 , A does in fact have the appropriate units of time. To find the average decay time of Radon-222, we need to determine the decay constant k, given the half-life of 3.825 days. Recall that k=

ln 2 tn

where tn is the half-life. Thus, A=

tn 3.825 1 = = ≈ 5.518 days. k ln 2 ln 2

" ∞ α x dof According for a gas at α temperature T , that the fraction of gas Let Jnto=Maxwell’s x n e−Law x, velocity where n distribution ≥ 1 is an integer and > 0. Prove Jn = (n/ α )Jmolecules n−1 and 0 with speed between v and v + v is approximately f (v) v, where n! J0 = 1/α . Use this to compute J4 . Show that Jn = n+1

.m 3/2 2 f (v) = 4απ v 2 e−mv /(2kT ) 1 2 π kT SOLUTION Using Integration by Parts, with u = x n and v = e−α x , we get u = nx n−1 , v = − α e−α x , and " " 8kT 1/2 1 n −α x constant). n n − α x n−1 − α x (m is the molecular mass of the gas and k Boltzmann’s Show that the average speed v is , x e x dx = − x e + e d x. πm α α where " ∞ Thus, v = "v f (v) dv " ∞ n −R n 1 n −α x R 0n ∞ n−1 −α x n − α x x e d x = lim − x e + x e d x = lim + 0 + Jn−1 . Jn = αR α α α R→∞ R→∞ α e 0 0 0 85.

Use L’Hôpital’s Rule repeatedly to compute the limit: −Rn −n R n−1 −n(n − 1)R n−2 −n(n − 1)(n − 2) · · · (3)(2)(1) = lim = lim = · · · = lim = 0. α R 2 α R 3 α R R→∞ α e R→∞ α e R→∞ R→∞ α e α n+1 eα R lim

Finally, Jn = 0 +

n n Jn−1 = Jn−1 . α α

J0 can be computed directly: J0 =

" ∞ 0

R 1 1 −α R 1 1 e e−α x d x = lim − e−α x = lim − − 1 = − (0 − 1) = . α α α α R→∞ 0 R→∞ R→∞ 0

e−α x d x = lim

" R

With this starting point, we can work up to J4 : J1 =

1 1 J = α 0 α

J2 =

2 2 J1 = α α

J3 =

3 3 J = α 2 α

J4 =

4 4 J = α 3 α

1 1 = 2; α α 2! 1 2 = 3 = 2+1 ; 2 α α α 3! 2 6 = 4 = 3+1 ; α3 α α 4! 6 24 = 5 = 4+1 . α4 α α

We can use induction to prove the formula for Jn . If Jn−1 =

(n − 1)! , αn

Improper Integrals

S E C T I O N 8.6

483

then we have Jn =

n n (n − 1)! n! Jn−1 = · = n+1 . n α α α α

n amount of electromagnetic energy with frequency between ν and There is a function F(ν ) such that xthe 87. for xT =is0 proportional and f (0) = 0. Let a > 0byand n > 1. Define fbody (x) =at temperature ν+ ν radiated a so-called black to F(ν ) ν . The total radiated energy is " ∞ eax − 1 Rule to to Planck’s show thatRadiation f (x) is continuous at x = 0. F(νL’Hˆ ) d νo. pital’s According Law, E = (a) Use " ∞ 0 (b) Show that f (x) d x converges. Hint: Show 2x n e−ax if x is large enough. Then use the Com that f (x) ≤ 3 0 8π h ν parison Test and Exercise 85. F(ν ) = eh ν /kT − 1 c3

where c, h, k are physical constants. To derive this law, Planck introduced the quantum hypothesis in 1900, which thus marked the birth of quantum mechanics. Show that E is finite (use Exercise 86). SOLUTION

The total radiated energy E is given by E=

" ∞ 0

" 8π h ∞ ν3 d ν. h ν /kT 3 −1 c 0 e

F(ν ) d ν =

Let α = h/kT . Then E=

" 8π h ∞ ν 3 d ν. c3 0 eαν − 1

Because α > 0 and 8π h/c3 is a constant, we know E is finite by Exercise 86. A probability (x) defined for x ≥ 0 such that f (x) ≥ 0 and " ∞density function on "[0,N∞) is a function "f ∞ "89. ∞ −x 2 d x and J = −x 2 d x. Although e−x 2 has no elementary antiderivative, it is known that Let J = e e f (x) d x = 1. The mean value ofN f (x) is the quantity μ = x f (x) d x. For k > 0, find a constant C such that 0 0 0 0 √ −kx π /2. Let Tdensity N th trapezoidal =a probability N be theand Ce J is compute μ . approximation to J N . Calculate T4 and show that T4 approximates J to three decimal places. −kx SOLUTION For Ce to be a probability distribution, it is required that " R " ∞ C −kx R C −kx −kx Ce d x = lim Ce d x = lim = 1. − e = k k R→∞ 0 R→∞ 0 0 This requires C = k, so that f (x) = ke−kx . To compute " ∞ " ∞ −kx μ= x(ke ) dx = k xe−kx d x 0

0

we must use integration by parts with u = x , v = e−kx , hence du = d x and v = − 1k e−kx :

μ=

" ∞ 0

x f (x) d x =

= lim

R→∞

" ∞ 0

−Re−Rk −

1 k

" R

kxe−kx d x = lim k R→∞ 0

xe−kx d x = lim R→∞

" R

R −xe−kx + 0 0

e−kx d x

R 1 −Rk 1 1 − 1 = 0 − (−1) = . e−kx e = lim −Re−Rk − 0 k k k R→∞

In Exercises 90–93, the Laplace transform of a function f (x) is the function L f (s) of the variable s defined by the improper integral (if it converges): " ∞ L f (s) = f (x)e−sx d x 0

Laplace transforms are widely used in physics and engineering.

α 91. Show thatthat if fif(x)f (x) = sin α x, then C L fa(s) = 2 then2 .L f (s) = C/s for s > 0. Show =C where constant, s +α SOLUTION

If f (x) = sin α x, then the Laplace transform of f (x) is " ∞ e−sx sin α x d x L f (s) = 0

484

CHAPTER 8


First evaluate the indefinite integral using Integration by Parts, with u = sin α x and v = e−sx . Then u = α cos α x, v = − 1s e−sx , and " " 1 −sx α −sx e−sx cos α x d x. e sin α x d x = − e sin α x + s s Use Integration by Parts again, with u = cos α x, v = e−sx . Then u = −α sin α x, v = − 1s e−sx , and " " 1 α e−sx sin α x d x. e−sx cos α x d x = − e−sx cos α x − s s ! Substituting this into the first equation and solving for e−sx sin α x d x, we get " 1 α α2 e−sx sin α x d x e−sx sin α x d x = − e−sx sin α x − 2 e−sx cos α x − 2 s s s

" −e−sx 1s sin α x + α2 cos α x −e−sx (s sin α x + α cos α x) s

e−sx sin α x d x = = 2 s2 + α2 1 + α2 "

s

Thus, " R

1 e−sx sin α x d x = 2 s + α2 0

0+α s sin α R + α cos α R − −1 −es R

s sin α R + α cos α R 1 α − . = 2 es R s + α2

Finally we take the limit, noting the fact that, for all values of R, |s sin α R + α cos α R| ≤ s + |α | s sin α R + α cos α R 1 α 1 α − (α − 0) = 2 . = 2 L f (s) = lim 2 s R 2 R→∞ s + α 2 e s +α s + α2 93. Compute L f (s), where f (x) = cos ααxx and s > 0. Compute L f (s), where f (x) = e and s > α . SOLUTION If f (x) = cos α x, then the Laplace transform of f (x) is " ∞ L f (x) = e−sx cos α x d x 0

First evaluate the indefinite integral using Integration by Parts, with u = cos α x and v − e−sx . Then u = −α sin α x, v = − 1s e−sx , and " " α 1 e−sx cos α x d x = − e−sx cos α x − e−sx sin α x d x. s s Use Integration by Parts again, with u = sin α x d x and v = −e−sx . Then u = α cos α x, v = − 1s e−sx , and " " 1 α e−sx cos α x d x. e−sx sin α x d x = − e−sx sin α x + s s ! Substituting this into the first equation and solving for e−sx cos α x d x, we get " " 1 1 α α e−sx cos α d x e−sx cos α x d x = − e−sx cos α x − − e−sx sin α x + s s s s " 2 1 α α = − e−sx cos α x + 2 e−sx sin α x − 2 e−sx cos α x d x s s s

" e−sx α2 sin α x − 1s cos α x e−sx (α sin α x − s cos α x) s e−sx cos α x d x = = 2 s2 + α2 1+ α s2

Thus, " R

1 e−sx cos α x d x = 2 s + α2 0

0−s α sin α R − s cos α R − . 1 es R

Finally we take the limit, noting the fact that, for all values of R, |α sin α R − s cos α R| ≤ |α | + s s 1 α sin α R − s cos α R 1 (s + 0) = 2 . s + = 2 L f (s) = lim 2 R→∞ s + α 2 es R s + α2 s + α2

Improper Integrals

S E C T I O N 8.6

485


" 1 " 1/2 95. Let I = x p ln x d x. dx Let p be converges if and only if p > 1. 0 an integer. Show that x(ln x) p 0 (a) Show that I diverges for p = −1. (b) Show that if p = −1, then " x p+1 1 p x ln x d x = ln x − +C p+1 p+1 (c) Use L’Hôpital’s Rule to show that I converges if p > −1 and diverges if p < −1. SOLUTION

(a) If p = −1, then I =

" 1 0

x −1 ln x d x =

" 1 ln x d x. x 0

Let u = ln x, du = (1/x) d x. Then "

ln x dx = x

" u du =

u2 1 + C = (ln x)2 + C. 2 2

Thus, " 1 1 1 1 ln x d x = (ln 1)2 − (ln R)2 = − (ln R)2 , x 2 2 2 R and 1 I = lim − (ln R)2 = ∞. R→0+ 2 The integral diverges for p = −1. (b) If p = 1, then use Integration by Parts, with u = ln x and v = x p . Then u = 1/x, v = x p+1 / p + 1, and x p ln x d x =

1 x p+1 ln x − p+1 p+1

"

1

" x p+1 1 x p dx ln x − x p+1 p+1 x p+1 1 x p+1 1 x p+1 ln x − +C = ln x − + C. = p+1 p+1 p+1 p+1 p+1

"

x p+1

dx =

(c) Let p < −1. Then I = lim

" 1

R→0+ R

= lim

R→0+

* 1 1 R p+1 1 ln 1 − − ln R − p+1 p+1 p+1 p+1 R p+1 ln R + . ( p + 1)2

) x p ln x =

lim

R→0+

R p+1 −1 − p+1 ( p + 1)2

Since p < −1, p + 1 < 0, and we have ln R 1 −1 − + = ∞. I = lim ( p + 1)R − p−1 ( p + 1)2 R − p−1 R→0+ ( p + 1)2 The integral diverges for p < −1. On the other hand, if p > −1, then p + 1 > 0, and I =

1 1 −1 −1 −1 lim R p+1 ln R + + lim R p+1 = +0= . p + 1 R→0+ ( p + 1)2 ( p + 1)2 R→0+ ( p + 1)2 ( p + 1)2

In Exercises 96–98, an improper integral I =

" ∞ a

f (x) d x is called absolutely convergent if

It can be shown that if I is absolutely convergent, then it is convergent. " ∞ 2 " ∞ e−xsincos 97. Show that x x d x is absolutely convergent. d x is absolutely convergent. Show that 1 2 x 1

" ∞ a

| f (x)| d x converges.

486

CHAPTER 8


" ∞ By the result of Exercise 61, we know that

SOLUTION

e−x d x is convergent. Then 2

" ∞

0

gent. Because | cos x| ≤ 1 for all x, we have 2 2 2 −x 2 cos x = | cos x| e−x ≤ e−x = e−x . e

e−x d x is also conver2

1

Therefore, by the comparison test, we have " ∞ −x 2 cos x d x also converges. e " ∞ Since

1

e−x cos x d x converges absolutely, it itself converges. 2

1

99. The gamma function, which plays " ∞ an important role in advanced applications, is defined for n ≥ 1 by sin x " ∞ f (0) = 1. Then f (x) is continuous and I is not improper at Let f (x) = f (x) d x. We define and I = x 0 t n−1 e−t dt (n) = x = 0. 0 (a) Show that (a) Show that the integral defining (n) converges for n ≥ 1 (it actually converges for all n > 0). Hint: Show that n−1 " R t e−t < t −2 for t sufficiently large. " R sin x cos x cos x R d x = − − dx (b) Show that (n + 1) = n(n) using Integration by Parts. x x 1 x2 1 1 (c) Show that (n + 1) = n! if n ≥ 1 is an integer. Hint: Use (a) repeatedly. Thus, (n) provides a way of defining " ∞ n-factorial when n is not an cosinteger. x d x converges. Conclude that the limit as R → ∞ of the integral in (a) exists and is finite. (b) Show that 2 x SOLUTION 1 (c)repeated Show that (a) By use Iofconverges. L’Hôpital’s Rule, we can compute the following limit: It is known that I = π2 . However,t I is not absolutely convergent. The convergence depends on cancellation, as shown e et et in Figure 13. lim n+1 = lim = · · · = lim = ∞. t→∞ t t→∞ (n + 1)t n t→∞ (n + 1)! This implies that, for t sufficiently large, we have et ≥ t n+1 ; therefore et

" ∞ The integral 1

t n+1 ≥ = t2 t n−1 t n−1

t n−1 e−t ≤ t −2 .

or

t −2 dt converges because p = 2 > 1. Therefore, by the comparison test, " ∞

t n−1 e−t dt also converges,

M

where M is the value above which the above comparisons hold. Finally, because the function t n−1 e−t is continuous for all t, we know that " ∞ (n) = t n−1 e−t dt converges for all n ≥ 1. 0

(b) Using Integration by Parts, with u = t n and v − e−t , we have u = nt n−1 , v = −e−t , and " ∞ " ∞ ∞ (n + 1) = t n e−t dt = −t n e−t 0 + n t n−1 e−t dt 0

= lim

R→∞

−R n eR

0

− 0 + n(n) = 0 + n(n) = n(n).

Here, we’ve computed the limit as in part (a) with repeated use of L’Hôpital’s Rule. (c) By the result of part (b), we have (n + 1) = n(n) = n(n − 1)(n − 1) = n(n − 1)(n − 2)(n − 2) = · · · = n! (1). If n = 1, then (1) =

" ∞ 0

R

e−t dt = lim −e−t = lim 1 − e−R = 1. R→∞

0

R→∞

Thus (n + 1) = n! (1) = n! n! Use the results of Exercise 99 to show that the Laplace transform (see Exercises 90–93 above) of x n is n+1 . s


487

CHAPTER REVIEW EXERCISES " 5 1. Estimate 2

SOLUTION

x 2 = 5. Then

f (x) d x by computing T2 , M3 , T6 , and S6 for a function f (x) taking on the values in the table below: x

2

2.5

3

3.5

4

4.5

5

f (x)

1 2

2

1

0

− 32

−4

−2

To calculate T2 , divide [2, 5] into two subintervals of length x = 32 with endpoints x 0 = 2, x1 = 3.5, T2 =

9 1 3 1 · ( f (2) + 2 f (3.5) + f (5)) = 0.75 + 2 · 0 + (−2) = − . 2 2 2 8

To calculate M3 , divide [2, 5] into three subintervals of length x = 1 with midpoints c1 = 2.5, c2 = 3.5, c3 = 4.5. Then M3 = 1 · ( f (2.5) + f (3.5) + f (4.5)) = 2 + 0 − 4 = −2. 1 To calculate T6 , divide [2, 5] into 6 subintervals of length 5−2 6 = 2 with endpoints x 0 = 2, x 1 = 2.5, x 2 = 3, x 3 = 3.5, x 4 = 4 , x 5 = 4.5, x 6 = 5. Then

1 1 · ( f (2) + 2 f (2.5) + 2 f (3) + 2 f (3.5) + 2 f (4) + 2 f (4.5) + f (5)) 2 2 1 1 3 13 = +2·2+2·1+2·0+2· − + 2(−4) + (−2) = − . 4 2 2 8

T6 =

1 Finally, to calculate S6 , divide [2, 5] into 6 subintervals of length x = 5−2 6 = 2 with endpoints x 0 = 2, x 1 = 2.5, x2 = 3, x 3 = 3.5, x4 = 4 , x 5 = 4.5, x 6 = 5. Then

1 1 · ( f (2) + 4 f (2.5) + 2 f (3) + 4 f (3.5) + 2 f (4) + 4 f (4.5) + f (5)) 3 2 1 1 3 7 = +4·2+2·1+4·0+2· − + 4(−4) + (−2) = − . 6 2 2 4

S6 =

3. The rainfall rate (in inches per hour) was measured hourly during a 10-hour thunderstorm with the following results: State whether the approximation M N or TN is larger or smaller than the integral. " π " 2π 0, 0.41, 0.49, 0.32, 0.3, 0.23, 0.09, 0.08, 0.05, 0.11, 0.12 (a) sin x d x (b) sin x d x π 0 Use Simpson’s " 8 Rule to estimate the total rainfall during the 10-hour "period. 5 dx (d) ln rainfall x d x during the 10-hour period is approxi(c) SOLUTION xWe have 10 subintervals of length x = 1. Thus, the total 2 1 2 mately 1 · 1 [0 + 4 · 0.41 + 2 · 0.49 + 4 · 0.32 + 2 · 0.3 + 4 · 0.23 + +2 · 0.09 + 4 · 0.08 + 2 · 0.05 + 4 · 0.11 + 0.12] 3 = 2.19 inches.

S10 =

In Exercises 4–9, compute the given approximation to the integral. 5.

" 4 " 1 3 6t −x+2 1 dt, T3 e d x, M5 2 0

2 8 10 Divide the interval [2, 4] into 3 subintervals of length x = 4−2 3 = 3 , with endpoints 2, 3 , 3 , 4. Then, 1 8 10 T3 = x f (2) + 2 f +2f + f (4) 2 3 3 ⎛ ⎞ 3 3 1 2 ⎝ 8 10 = · 6 · 23 + 1 + 2 6 · +1+2 6· + 1 + 6 · 43 + 1⎠ = 25.976514. 2 3 3 3

SOLUTION

7.

" 4 " πd/2x √ , T6 1 x 3 + 1 sin θ d θ , π /4

M4

488

CHAPTER 8


Then T6 =

1 x 2

1 3 5 7 Divide the interval [1, 4] into 6 subintervals of length x = 4−1 6 = 2 with endpoints 1, 2 , 2, 2 , 3, 2 , 4.

f (1) + 2 f ⎛

3 5 7 + 2 f (2) + 2 f + 2 f (3) + 2 f + f (4) 2 2 2

1 1 1⎜ 1 + 2 3 = · ⎝ 3 2 2 1 +1 3 2

9.

+1

1 1 + 2 3 +2 3 2 +1 5 2

⎞

1 1 ⎟ 1 +2 + 3 +2 3 ⎠ = 0.358016. 2 3 +1 ( 72 ) + 1 4 + 1 +1

" 9 " 1 2 cos(x−x)2d x, S8 e d x, S4 5 0

1 11 13 15 Divide the interval [5, 9] into 8 subintervals of length x = 9−5 8 = 2 with endpoints 5, 2 , 6, 2 , 7, 2 , 8, 17 2 , 9. Then 1 11 13 15 17 S8 = x f (5) + 4 f + 2 f (6) + 4 f + 2 f (7) + 4 f + 2 f (8) + 4 f + f (9) 3 2 2 2 2

1 1 cos(52 ) + 4 cos(5.52 ) + 2 cos(62 ) + 4 cos(6.52 ) = · 3 2 + 2 cos(72 ) + 4 cos(7.52 ) + 2 cos(82 ) + 4 cos(8.52 ) + cos(92 ) SOLUTION

= 0.608711. 11. Suppose that the second derivative of the function A(h) in Exercise 10 satisfies | A (h)| ≤ 1.5. Use the Error Bound The table below gives the area A(h) of a horizontal cross section of a pond at depth h. Use the Trapezoidal Rule to find the maximum possible error in your estimate of the volume V of the pond. to estimate the volume V of the pond (Figure 1). SOLUTION The Error Bound for the Trapezoidal Rule states that h (ft) A(h) (acres) h (ft) A(h) (acres) K 2 (b − a)3 , 0.8 0 Error (T 2.8 10 N) ≤ 12N 2 2 2.4 12 0.6 14 We estimated 0.2 the volume of the pond by T9 ; hence all x ∈ [a, b]. where K 2 is a number such that f (x) 4≤ K 2 for 1.8 0.1 A (h) ≤ 1.5 acres/ft2 we may take N = 9. The interval of depth is [0, 18] 6hence b −1.5 a = 18 − 016= 18. Since 8 1.2 18 0 K 2 = 1.5, to find that the error cannot exceed 1.5 · 183 K 2 (b − a)3 = = 9 acre · ft = 392,040 ft3 , 12N 2 12 · 92 where we have used the fact that 1 acre = 43560 ft2 . 13.

a bound Let f (x) = sin(x 3 ). Find " 3 for the error Find a bound for the error M16 − x 3 d x . " π /2 1 f (x) d x . T24 − 0

Hint: Find a bound K 2 for | f (x)| by plotting f (x) with a graphing utility. SOLUTION

Using the error bound for T24 we obtain:

3 " π /2 K 2 π2 − 0 K2π3 − f (x) d x = , ≤ T 24 55, 296 12 · 242 0 where K 2 is a number such that f (x) < k2 for all x ∈ 0, π2 . We compute the first and second derivative of f (x) = sin(x 3 ): f (x) = 3x 2 cos(x 3 )

f (x) = 6x cos(x 3 ) + 3x 2 · 3x 2 − sin(x 3 ) = 6x cos(x 3 ) − 9x 4 sin(x 3 ) The graph of f (x) = 6x cos(x 3 ) − 9x 4 sin(x 3 ) on the interval 0, π2 shows that f (x) ≤ 30 on this interval. We may choose K 2 = 30 and find " π /2 5π 3 30π 3 f (x) d x ≤ = = 0.0168220. T24 − 55,296 9216 0


489

y 30 20 10 x 2

−10 −20 −30

3

4

5

" 5 a value N such 15. FindFind a value of Nofsuch thatthat S N approximates x −1/4 d x with an error of at most 10−2 (but do not calculate S N ). 2 " π /4 −4(x). We differentiate f (x) = x −1/4 four SOLUTION To use the error bound we must M find the fourth f (4) tanderivative x d x ≤ 10 N − 0 times to obtain: 1 5 −9/4 45 585 −17/4 f (x) = − x −5/4 , f (x) = , f (x) = − x −13/4 , f (4) (x) = . x x 4 16 64 256 For 2 ≤ x ≤ 5 we have: (4) f (x) =

585 585 ≤ = 0.120099. 17/4 256x 256 · 217/4

Using the error bound with b = 5, a = 2 and K 4 = 0.120099 we have: Error (S N ) ≤

0.162134 0.120099(5 − 2)5 = . 4 180N N4

We must choose a value of N such that: 0.162134 ≤ 10−2 N4 N 4 ≥ 16.2134 N ≥ 2.00664 The smallest even value of N that is needed to obtain the required precision is N = 4. In Exercises 17–25, compute(a)–(e) the integral using the suggested(i)–(v) method. Match the integrals with their antiderivatives on the basis of the general form (do not evaluate the " integrals). " 3 θ sin8 θ d θ [write cos3 θ as cos θ (1 − sin2 θ )] " 17. cos x dx (2x + 9) d x (a) (b) 2 −4 x2 + 4 " x We " SOLUTION use the identity cos2 θ = 1 − sin2 θ to rewrite the integral: dx (c) sin3 x cos"2 x d x (d) " "

x 16x 2 − 1 " cos3 θ sin8 θ d θ = cos2 θ sin8 θ cos θ d θ = 1 − sin2 θ sin8 θ cos θ d θ . 16 d x (e) 4)2 Now, we usex(x the−substitution u = sin θ , du = cos θ d θ : (i)" sec−1 4x + C "

"

u9 u 11 sin9 θ sin11 θ 4 2 8 3 8 cos|x|θ − sinlog θ |x d θ−=4| − 1 − u + uC du = u 8 − u 10 du = − +C = − + C. (ii) log 9 11 9 11 x −4 1 " (iii) (3 cos5 x − 3 cos3 x sin2 x − 7 cos3 x) + C "303 sec 19. tan4xθ d θ (trigonometric identity, reduction formula) 9 xeθ−12x x ln(x (Integration by Parts) 2 + 4) + C (iv) tan−1 d+

2 2 2 SOLUTION 2 We use the identity 1 + tan2 θ = sec2 θ to write tan4 θ = sec2 θ − 1 and to rewrite the integral as (v) x − 4 + C "

"

sec3 θ tan4 θ d θ

"

2

sec3 θ 1 − sec2 θ d θ = sec3 θ 1 − 2sec2 θ + sec4 θ d θ " =

" sec7 θ d θ − 2

" sec5 θ d θ +

Now we use the reduction formula " " tan θ secm−2 θ m−2 secm−2 θ d θ . secm θ d θ = + m−1 m−1 We have " sec5 θ d θ =

tan θ sec3 θ 3 + 4 4

" sec3 θ d θ + C,

sec3 θ d θ .

490

CHAPTER 8


and "

tan θ sec5 θ 5 sec7 θ d θ = + 6

=

"

tan θ sec5 θ 5 sec5 θ d θ = +

6

6

5 tan θ sec5 θ 5 + tan θ sec3 θ + 6 24 8

6

tan θ sec3 θ 3 + 4 4

" sec3 θ d θ

" sec3 θ d θ + C.

Therefore, " tan θ sec5 θ 5 5 + tan θ sec3 θ + sec3 θ d θ 6 24 8 " " 3 tan θ sec3 θ 3 sec θ d θ + sec3 θ d θ + −2 4 4

" sec3 θ tan4 θ d θ =

tan θ sec5 θ 7 tan θ sec3 θ 1 − + 6 24 8

=

" sec3 θ d θ .

We again use the reduction formula to compute " " tan θ sec θ tan θ sec θ 1 1 sec3 θ d θ = sec θ d θ = + + ln | sec θ + tan θ | + C. 2 2 2 2 Finally, " sec3 θ tan4 θ d θ =

tan θ sec5 θ 7 tan θ sec3 θ tan θ sec θ 1 − + + ln | sec θ + tan θ | + C. 6 24 16 16

"

dx " 4x + 4d x (trigonometric substitution) 2 d x (partial fractions) x(x − 1)3/2 (x − 5)(x + 3) SOLUTION Substitute x = sec θ , d x = sec θ tan θ d θ . Then,

21.

and

"

3/2

x2 − 1

3/2

3/2

= sec2 θ − 1 = tan2 θ = tan3 θ , "

dx

sec θ tan θ d θ = sec θ tan3 θ

3/2 = x x2 − 1

"

dθ = tan2 θ

" cot2 θ d θ .

Using a reduction formula we find that: " cot2 θ d θ = − cot θ − θ + C so

"

dx

3/2 x x2 − 1

= − cot θ − θ + C.

We now must return to the original variable x. We use the relation x = sec θ and the figure to obtain: " 1 dx − sec−1 x + C. 3/2 = − 2 x −1 x x2 − 1

x

x2 − 1

q 1

"

" dx d x (rewrite integrand) −1 (substitution) x + x3/2 x + x 1/2 SOLUTION We rewrite the integrand as follows: "

23.

dx = x + x −1

"

x dx . x2 + 1

+C


Now, we substitute u = x 2 + 1. Then du = 2x d x and "

" 25.

dx = x + x −1

" 1 2 du

=

u

1 2

"

du 1 1

= ln |u| + C = ln 1 + x 2 + C. u 2 2

" dx (complete square, substitution, partial fractions) x 2 x+−2 4xtan −−1 5 x d x (Integration by Parts)

SOLUTION

The partial fraction decomposition takes the form 1 A B = + . x −1 x +5 x 2 + 4x − 5

Clearing denominators gives us 1 = A(x + 5) + B(x − 1). Setting x = 1 then yields A = 16 , while setting x = −5 yields B = − 16 . Therefore, "

1 dx = 6 x 2 + 4x − 5

"

dx 1 − x −1 6

"

dx 1 1 1 x − 1 = ln |x − 1| − ln |x + 5| + C = ln + C. x +5 6 6 6 x + 5

In Exercises 26–69, compute the integral using the appropriate method or combination of methods. " " 2 x x 2 e4x ddxx 27. 9 − x2 SOLUTION

Substitute x = 3 sin θ , d x = 3 cos θ d θ . Then

9 − x 2 = 9 − 9sin2 θ = 9 1 − sin2 θ = 9cos2 θ = 3 cos θ ,

and "

" 9sin2 θ · 3 cos θ d θ = 9 sin2 θ d θ 3 cos θ θ sin 2θ 9θ 9 sin θ cos θ =9 − +C = − + C. 2 4 2 2

x2 dx = 9 − x2

"

We now must return to the original variable x. Since x = 3 sin θ , we have t = sin−1 3x . Using the figure we obtain "

x 9 x 9 − · · d x = sin−1 2 3 2 3 9 − x2 x2

9 − x2 9 −1 x x 9 − x 2 − + C = sin + C. 3 2 3 2

3

x

9 − x2

" 29.

" sec2 θ 9tan4 θ d3θ cos 6θ sin 6θ d θ

SOLUTION

We substitute u = tan θ , du = sec2 θ d θ to obtain "

" sec2 θ tan4 θ d θ =

u 4 du =

tan5 θ u5 +C = + C. 5 5

"

" dt (6x + 4)d x (t 2 − 1)2 2 x −1 SOLUTION Substitute t = sin θ , dt = cos θ d θ . Then

31.

2

2

2 2

t 2 − 1 = 1 − t 2 = 1 − sin2 θ = cos2 θ = cos4 θ ,

491

492

CHAPTER 8


and

"

"

dt

2 =

t2 − 1

cos θ d θ = cos4 θ

"

dθ = cos3 θ

" sec3 θ d θ .

We use a reduction formula to compute the resulting integral: " " " tan θ sec θ tan θ sec θ dt 1 1 3 + sec θ d θ = + ln | sec θ + tan θ | + C. 2 = sec θ d θ = 2 2 2 2 2 t −1

1

t

q 1 − t2

We now must return to the original variable t. Using the relation t = sin θ and the accompanying figure, " 1 1 1 t 1 t |1 + t| dt 1 t 1 · + ln + + ln + C. +C = 2 = 2 · 2 2 2 2 2 2 2 2 2 1 − t 1−t 1−t 1−t 1−t 1 − t2 t −1 " 33.

" sin 2θdsin θ 2 θdθ cos4 θ We use the trigonometric identity sin 2θ = 2 sin θ cos θ to rewrite the integral: " " " 2 2 sin 2θ sin θ d θ = 2 sin θ cos θ sin θ d θ = 2sin3 θ cos θ d θ .

SOLUTION

Now, we substitute u = sin θ . Then du = cos θ d θ and " " u4 sin4 θ 2 sin 2θ sin θ d θ = 2 u 3 du = +C = + C. 2 2 " 35.

" (ln(x + 1))2 d x ln(9 − 2x) d x First, substitute w = x + 1, dw = d x. Then " " (ln(x + 1))2 d x = (ln w)2 dw.

SOLUTION

Now, we use Integration by Parts with u = (ln w)2 and v = 1. We find u = 2 lnww , v = w, and " " (ln w)2 dw = w(ln w)2 − 2 ln w dw. 1 , v = w, and We use Integration by Parts again, this time with u = ln w and v = 1. We find u = w " " ln w d x = w ln w − dw = w ln w − w + C.

Thus,

" (ln w)2 dw = w(ln w)2 − 2w ln w + 2w + C,

and

" (ln(x + 1))2 d x = (x + 1) [ln(x + 1)]2 − 2(x + 1) ln(x + 1) + 2(x + 1) + C. "

37.

" cos4 (9x − 2) d x sin5 θ d θ We substitute u = 9x − 2, du = 9 d x and then use a reduction formula to evaluate the resulting integral.

SOLUTION

We obtain: "

1 cos4 (9x − 2) d x =

9

"

1 cos4 u du =

9

cos3 u sin u 3 + 4 4

" cos2 u du


"

"

cos3 u sin u 1 + 36 12

=

cos3 (9x − 2) sin(9x − 2) 9x − 2 sin(18x − 4) + + + C. 36 24 48

cos2 u du =

cos3 u sin u 1 + 36 12

=

u sin 2u + 2 4

+C

" sin 2x sec2 x d x sin 3x cos 5x d x

39.

We use the trigonometric identity sin 2x = 2 cos x sin x to rewrite the integrand:

SOLUTION

sin 2xsec2 x = 2 sin x cos xsec2 x =

2 sin x 2 sin x cos x = = 2 tan x. cos x cos2 x

Hence, "

" sin 2xsec2 x d x =

" 41.

2 tan x d x = 2 ln | sec x| + C.

" (sec√ x + tan x)22 d x tan x sec x d x We rewrite the integrand as

SOLUTION

(sec x + tan x)2 = sec2 x + 2 sec x tan x + tan2 x = 2 sec x tan x + 2 sec2 x − 1. Therefore, "

" (sec x + tan x)2 d x = 2

" 43.

" sec x tan x d x + 2

" sec2 x d x −

d x = 2 sec x + 2 tan x − x + C.

" θ cot2 5d θ 3 sin2 θ cos θ d θ

SOLUTION

We substitute u = θ2 . Then du = 12 d θ and " " θ cot2 d θ = 2 cot2 u du. 2

Now, we use a reduction formula to compute " " θ θ cot2 d θ = 2 cot2 u du = 2(− cot u − u) + C = −2 cot − θ + C. 2 2 "

dt " dt (t − 3)2 (t + 4) (t − 3)(t + 4) SOLUTION The partial fraction decomposition has the form

45.

1 (t − 3)2 (t + 4)

=

B C A . + + t + 4 t − 3 (t − 3)2

Clearing denominators gives us 1 = A(t − 3)2 + B(t − 3)(t + 4) + C(t + 4). 1 . Lastly, setting t = 0 yields Setting t = 3 then yields C = 17 , while setting t = −4 yields A = 49

1 = 9 A − 12B + 4C

or

B=−

1 . 49

Hence, "

dt 1 = 2 49 (t − 3) (t + 4) =

" 47.

dx " 2 x x 2x −+4 9 d x

"

dt 1 − t + 4 49

"

dt 1 + t −3 7

"

dt (t − 3)2

1 1 1 −1 1 t + 4 1 1 ln |t + 4| − ln |t − 3| + · +C = ln − · + C. 49 49 7 t −3 49 t −3 7 t −3

493

494

CHAPTER 8


Substitute x = 2 sec θ , d x = 2 sec θ tan θ d θ . Then

SOLUTION

and

x2 − 4 = "

4sec2 θ − 4 = 4 sec2 θ − 1 = 4tan2 θ = 2 tan θ , "

dx

=

x x2 − 4

"

1 2 sec θ tan θ d θ = 2 sec θ · 2 tan θ 2

dθ =

1 θ + C. 2

Now, return to the original variable x. Since x = 2 sec θ , we have sec θ = x2 or θ = sec−1 2x . Thus, " 1 dx x = sec−1 + C. 2 2 2 x x −4 "

dx " dx 3/2 x + ax 1/2 x + x 2/3 SOLUTION Let u = x 1/2 or x = u 2 . Then d x = 2u du and " " " 2u du du dx = = 2 . x 3/2 + ax 1/2 u 3 + au u2 + a

49.

If a > 0, then "

dx =2 x 3/2 + ax 1/2

"

2 du = √ tan−1 a u2 + a

If a = 0, then

"

Finally, if a < 0, then

"

u √ a

2 + C = √ tan−1 a

x + C. a

dx 2 = − √ + C. x x 3/2

du = u2 + a

" u2 −

du √

−a

2 ,

and the partial fraction decomposition takes the form

u2 −

1 √

−a

2 =

u−

A √

−a

+

u+

B √

−a

.

Clearing denominators gives us √ √ −a) + B(u − −a). √ , while setting u = − −a yields B = − √1 1 = A(u +

Setting u =

√

−a then yields A = √1 "

2 −a

dx =2 3/2 x + ax 1/2

2 −a

"

1 du = √ 2 −a u +a

"

1 du −√ √ u − −a −a

"

. Hence, du √ u + −a

√ √ 1 1 = √ ln |u − −a| − √ ln u + −a + C −a −a √ √ √ u − −a x − −a 1 1 ln ln √ = √ √ + C = √ √ + C. −a u + −a −a x + −a In summary,

⎧ ⎪ √2 tan−1 x + C ⎪ a ⎪ a ⎪ " ⎨ √ √ dx x− 1 = √ ln √ √−a + C −a x+ −a ⎪ x 3/2 + ax 1/2 ⎪ ⎪ ⎪ ⎩− √2 + C x

51.

" "2 (x − x)ddxx 3 (x(x+−2)b) 2+4

a>0 a 0) converges if p > 1 and diverges if p ≤ 1. Here, p = 23 < 1, hence the xp a integral diverges. " ∞ " 4d x 73. dx x 12/5 9 0 x 2/3 SOLUTION

" ∞ " R 5 −7/5 R dx dx 5 −7/5 5 −7/5 = lim = lim − = lim + x R · 9 − 7 7 R→∞ 9 x 12/5 R→∞ 7 R→∞ x 12/5 9 9 =0+

75.

5 5 5 −7/5 = = . ·9 2/5 7 7·9·9 63 · 92/5

" 9 " 04x e d xd x

−∞ −∞ x 2 + 1

SOLUTION

" 9 −∞

77.

e4x d x =

" 9 lim

R→−∞ R

e4x d x =

e36 1 4x 9 1 36 1 4R lim e = lim e − e = . 4 4 R→−∞ 4 R→−∞ 4 R

" ∞ " π /2 d x 2)(2x 1 (x +cot θ d θ+ 3) 0

SOLUTION

First, evaluate the indefinite integral. The following partial fraction decomposition has the form 1 1 2 =− + . (x + 2)(2x + 3) x + 2 2x + 3

Clearing denominators gives us 1 = A(2x + 3) + B(x + 2). Setting x = −2 then yields A = −1, while setting x = − 32 yields B = 2. Hence, "

dx =− (x + 2)(2x + 3)

"

dx +2 x +2

"

2x + 3 dx + C. = − ln |x + 2| + ln |2x + 3| + C = ln 2x + 3 x +2


499

Now, for R > 1, " R 1

and

" ∞ 1

79.

2x + 3 R dx = ln 2R + 3 − ln 5 , = ln (x + 2)(2x + 3) x + 2 1 R+2 3

5 3 6 dx 2R + 3 = lim ln − ln = ln 2 + ln = ln . (x + 2)(2x + 3) R+2 3 5 5 R→∞

" 5 " ∞ −1/3 (5 − x) dx (5 + x)−1/3 d x 2 0

SOLUTION

" 5 2

(5 − x)−1/3 d x = lim

" R

R→5− 2

= lim − R→5−

(5 − x)−1/3 d x =

R 3 lim − (5 − x)2/3 R→5− 2 2

35/3 3

3

(5 − R)2/3 − 32/3 = − 0 − 32/3 = . 2 2 2

In Exercises 80–85, use the Comparison Test to determine if the improper integral converges or diverges. " ∞ " ∞2 81. (sin x)e d x −x d x 8

x2 − 4 The following inequality holds for all x,

0 ≤ sin2 x e−x ≤ e−x .

8

SOLUTION

We use direct computation to show that the improper integral of e−x over the interval [8, ∞) converges: R " ∞ " R

e−x d x = lim e−x d x = lim −e−x = lim −e−R + e−8 = 0 + e−8 = e−8 . R→∞ 8

8

R→∞

8

" ∞ Therefore, by the Comparison Test, the improper integral 8

R→∞

(sin2 x)e−x d x also converges.

" ∞ " ∞ dx dx x 1/3 4+ x 2/3 2 1 x + cos x 3 SOLUTION If x ≥ 1, then x 1/3 ≥ 1; therefore,

x 1/3 + x 2/3 = x 1/3 1 + x 1/3 ≤ x 1/3 x 1/3 + x 1/3 = x 1/3 · 2x 1/3 = 2x 2/3 .

83.

Hence, 1 1 ≥ 2/3 . x 1/3 + x 2/3 2x

" ∞ " ∞ dx dx diverges; hence also diverges. Therefore, by the Comparison Test, the improper The integral 2/3 1 2x 2/3 " ∞ 1 x dx also diverges. integral x 1/3 + x 2/3 1 " ∞ " e1−x 3 d x 85. dx 0

0 x 1/3 + x 2/3

SOLUTION

" ∞

3 3 For x > 1, e x ≥ x; hence e x ≥ x 3 , therefore 0 ≤ e−x ≤ x −3 . Since

" ∞ dx converges, the integral x3 1

e−x d x also converges by the Comparison Test. We write 3

1

" ∞ 0

e−x d x = 3

" 1 0

e−x d x + 3

" ∞

e−x d x. 3

1

The " ∞first integral on the right hand side has a finite value and the second integral converges. We conclude that the integral 3 e−x d x converges. 0

Calculate the volume of the infinite solid obtained by rotating the region under y = (x 2 + 1)−2 for 0 ≤ x < ∞ about the y-axis.

500

CHAPTER 8


87. (a) (b) (c)

Let R be the region under the graph of y = (x + 1)−1 for 0 ≤ x < ∞. Which of the following quantities is finite? The area of R The volume of the solid obtained by rotating R about the x-axis The volume of the solid obtained by rotating R about the y-axis

SOLUTION

(a) The area of R is R " R " ∞ dx dx = lim = lim ln |x + 1| = lim ln(R + 1) − ln 1 = ∞. x +1 R→∞ 0 x + 1 R→∞ R→∞ 0 0 Hence, the area of R is not finite. (b) Using the Disk Method, the volume of the solid obtained by rotating R about the x-axis is

π

" ∞ 0

" R 1 R dx 1 = π lim = π lim − = π lim + 1 = π. − R+1 R→∞ 0 (x + 1)2 R→∞ x + 1 0 R→∞ (x + 1)2 dx

Hence, the volume of the solid obtained by rotating R about the x-axis is finite. (c) Using the Shell Method, the volume of the solid obtained by rotating R about the y-axis is " ∞

2π

0

" R x x dx d x = 2π lim . x +1 x +1 R→∞ 0

Now, R " R " R " R x dx (x + 1) − 1 1 = dx = d x = (x − ln(x + 1)) 1− x +1 x +1 0 x +1 0 0 0 = R − (ln(R + 1) − ln 1) = R − ln(R + 1). Thus, " R ln(R + 1) x dx = 2π lim (R − ln(R + 1)) = 2π lim R 1 − = ∞. R R→∞ 0 x + 1 R→∞ R→∞

2π lim

Hence, the volume of the solid obtained by rotating R about the y-axis is not finite. 89. According to kinetic " ∞ theory, the molecules of ordinary matter are in constant random motion. The probability that a 2 1 e−E/(kT )2 thextemperature molecule has kinetic in adsmall intervalfor [E,allE n+>E] is approximately Show that energy x n e−x x converges 0. Hint: First observekTthat x n e−x , where < x n eT−xisfor > 1. Then 0 (in kelvins) and k Boltzmann’s constant. Compute the average kinetic energy E in terms of k and T , where show that x n e−x < x −2 for x sufficiently large. " ∞ 1 E= Ee−E/kT d E kT 0 By definition,

SOLUTION

" ∞ 0

Ee−E/kT d E = lim

" R

R→∞ 0

Ee−E/kT d E.

We compute the definite integral using Integration by Parts with u = E, v = e−E/kT . Then u = 1, v = −kT e−E/kT and " R " R R R Ee−E/kT d E = −kT e−E/kT E + kT e−E/kT d E = −kT e−R/kT R − (kT )2 e−E/kT E=0

0

E=0

0

= −kT Re−R/kT − k 2 T 2 e−R/kT − k 2 T 2 e0 = k 2 T 2 − kT Re−R/kT − k 2 T 2 e−R/kT .

We now let R → ∞, obtaining: " ∞ 0

Ee−E/RT d E = lim

" R

R→∞ 0

Ee−E/RT d E = lim

R→∞

k 2 T 2 − kT Re−R/kT − k 2 T 2 e−R/kT

= k 2 T 2 − kT lim Re−R/kT − 0 = k 2 T 2 − kT lim Re−R/kT . R→∞

R→∞

We compute the remaining limit using L’Hôpital’s Rule:

R→∞

R

dR

1 = lim d d R = 0. = lim 1 R/kT R/kT R→∞ e R/kT R→∞ R→∞ e e dR kT

lim Re−R/kT = lim


501

Thus, " ∞ 0

and E=

Ee−E/RT d E = k 2 T 2 ,

" ∞ 1 1 Ee−E/kT d E = · k 2 T 2 = kT. kT 0 kT

forss > > 0. α .See Exercises 90–93 in Section 8.6 91. Compute the the Laplace transform L f (s) f (x)f (x) = x 2=eαxx for Compute Laplace transform L f of (s)the of function the function for the definition of L ftransform (s). SOLUTION The Laplace is the following integral: " ∞ " ∞ " R

x 2 eα x e−sx d x = x 2 e(α −s)x d x = lim x 2 e(α −s)x d x. L x 2 eα x (s) = 0

R→∞ 0

0

1 e(α −s)x We compute the definite integral using Integration by Parts with u = x 2 , v = e(α −s)x . Then u = 2x, v = α −s and R " R " R 1 1 x 2 e(α −s)x d x = − 2x · x 2 e(α −s)x e(α −s)x d x α−s α−s 0 0 x=0 " R 2 1 xe(α −s)x d x. R 2 e(α −s)R − = α−s α −s 0

We compute the resulting integral using Integration by Parts again, this time with u = x and v = e(α −s)x . Then u = 1, 1 e(α −s)x and v = α −s " R 0

R R " R 1 1 x 1 (α −s)x − e(α −s)x d x = e e(α −s)x e(α −s)x − 2 α−s α −s 0 α −s (α − s) x=0 x=0

R (α −s)R 1 1 R 1 = − − e(α −s)R + e e(α −s)R . e(α −s)R − e0 = α−s α −s (α − s)2 (α − s)2 (α − s)2

xe(α −s)x d x = x ·

Thus, " R 0

x 2 e(α −s)x d x = =

and

L x 2 eα x (s) =

2 (s − α )3

−

1 2 R 2 e(α −s)R − α −s α −s

1 R (α −s)R − e(α −s)R + e 2 2 α −s (α − s) (α − s) 1

1 2 2 2R + e(α −s)R − e(α −s)R , R 2 e(α −s)R − 3 3 α −s (α − s) (α − s) (α − s)2

1 2 2 lim R 2 e−(s−α )R − lim e−(s−α )R − lim Re−(s−α )R . 3 s − α R→∞ R→∞ (s − α ) (s − α )2 R→∞

Now, since s > α , lim e−(s−α )R = 0. We use L’Hôpital’s Rule to compute the other two limits: R→∞

lim Re−(s−α )R = lim

R

lim R 2 e−(s−α )R = lim

R2

R→∞

R→∞

R→∞ e(s−α )R R→∞ e(s−α )R

1 = 0; R→∞ (s − α )e(s−α )R

= lim = lim

2R

R→∞ (s − α )e(s−α )R

2

= lim

R→∞ (s − α )2 e(s−α )R

Finally,

L x 2 eα x (s) =

2 (s − α )3

−0−0−0=

2 (s − α )3

.

" 2 /2 " 93. Let Jn = x n e−x x n d x d x. Let In = . x 2 + 12 /2 + C. (a) Show that J1 = −e−x n−1 x 2 n−1 −x /2 = . − (n In−2 (a) Prove that I n (b) Prove Jn = −x en − 1 + − 1)J n−2 . 0 ≤ Jn5 .≤ 5. (c) Use (a) and to compute J3 and (b) Use (a) (b) to calculate In for (c) Show that, in general, SOLUTION I2n+1 =

x 2n x 2n−2 x2 1 − + · · · + (−1)n−1 + (−1)n ln(x 2 + 1) + C 2n 2n − 2 2 2

= 0.

502

CHAPTER 8

T E C H N I Q U E S O F I N T E G R AT I O N 2

(a) Let u = − x2 . Then du = −x d x and " " 2 2 J1 = xe−x /2 d x = − eu du = −eu + C = −e−x /2 + C. 2 (b) Using Integration by Parts with u = x n−1 and v = xe−x /2 , we find " 2 2 2 Jn = −x n−1 e−x /2 + (n − 1) x n−2 e−x /2 d x = −x n−1 e−x /2 + (n − 1)Jn−2 .

(c) Using the results from parts (a) and (b), 2 2 J3 = −x 3−1 e−x /2 + (3 − 1)J3−2 = −x 2 e−x /2 + 2J1 2 2 2 = −x 2 e−x /2 − 2e−x /2 + C = −e−x /2 (x 2 + 2) + C

and then 2 2 J5 = −x 5−1 e−x /2 + (5 − 1)J5−2 = −x 4 e−x /2 + 4J3 2 2 2 = −x 4 e−x /2 − 4e−x /2 (x 2 + 2) + C = −e−x /2 (x 4 + 4x 2 + 8) + C.

FURTHER APPLICATIONS 9 OF THE INTEGRAL AND TAYLOR POLYNOMIALS 9.1 Arc Length and Surface Area Preliminary Questions 1. Which integral represents the length of the curve y = cos x between 0 and π ? " π " π 1 + cos2 x d x, 1 + sin2 x d x 0

0

Let y = cos x. Then y = − sin x, and 1 + (y )2 = 1 + sin2 x. Thus, the length of the curve y = cos x between 0 and π is " π 1 + sin2 x d x. SOLUTION

0

2. How do the arc lengths of the curves y = f (x) and y = f (x) + C over an interval [a, b] differ (C is a constant)? Explain geometrically and then justify using the arc length formula. SOLUTION The graph of y = f (x) + C is a vertical translation of the graph of y = f (x); hence, the two graphs should have the same arc length. We can explicitly establish this as follows: 2 " b " b d ( f (x) + C) d x = 1+ 1 + [ f (x)]2 d x = length of y = f (x). length of y = f (x) + C = dx a a

Exercises 1. Express the arc length of the curve y = x 4 between x = 2 and x = 6 as an integral (but do not evaluate). SOLUTION

Let y = x 4 . Then y = 4x 3 and s=

" 6 " 6 1 + (4x 3 )2 d x = 1 + 16x 6 d x. 2

2

2 1 2 1 3 π as an integral (but )2 = −2 . arc length = tan x ≤ Show do not evaluate). + x −1y for 1 ≤x xfor≤02.≤Hint: that 1 + (y + x x curve x 3. FindExpress the arcthe length of y =of the 4 12 4 SOLUTION

1 3 x + x −1 . Then y = 12 2 x2 2 −2 +1= −x (y ) + 1 = 4 Let y =

x 2 −2 x , and 4 x4 x4 1 1 − + x −4 + 1 = + + x −4 = 16 2 16 2

x2 + x −2 4

2

Thus, 1 2 " 22 " 2 2 " 2 2 x2 1 1 x 3 1 + (y )2 d x = dx = + 2 + 2 dx s= 4 x x 1 1 1 4 " 2 2 x x2 1 1 + 2 d x since + 2 >0 = 4 4 x x 1 13 1 2 x3 = − . = 12 x 1 12 4 over In Exercises 5–10, calculate the arc xlength 1 the given interval. Find the arc length of y = + 2 over [1, 4]. Hint: Show that 1 + (y )2 is a perfect square. 2 2x

.

504

CHAPTER 9

F U RT H E R A P P L I C AT I O N S O F T H E I N T E G R A L A N D TAY L O R P O LY N O M I A L S

5. y = 3x + 1, SOLUTION

[0, 3]

Let y = 3x + 1. Then y = 3, and s =

" 3 √ √ 1 + 9 d x = 3 10. 0

7. y = x 3/2 ,

[1, 2] [1, 3] y = 9 − 3x, 1/2 , and SOLUTION Let y = x 3/2 . Then y = 3 2x 3/2 " 2 √ 1 √ 9 8 13 8 9 3/2 2 11 3/2 = 1 + x dx = = − 1+ x 22 22 − 13 13 . s= 4 27 4 27 2 4 27 1 1 1 ln x, [1, 2e] 9. y = 14 x 21 −3/2 y = 3 x 2 − x 1/2 , [2, 8] 1 1 SOLUTION Let y = 4 x 2 − 2 ln x. Then

y =

x 1 − , 2 2x

and 1 + (y )2 = 1 +

x x 1 2 1 1 1 2 x2 = . − + + 2 = + 2 2x 4 2 4x 2 2x

Hence, " 2e " 2e s= 1 + (y )2 d x = 1

" 2e x

1

" 2e x 1 2 x + 1 dx dx = + 2 2 2x 2x 1

1 x 1 d x since + > 0 on [1, 2e] 2x 2 2x 1 2e ln 2 1 1 x2 + ln x = e2 + + . = 4 2 2 4 1

=

2

+

In Exercises 11–14,x), approximate y = ln(cos [0, π ] the arc length of the curve over the interval using the Trapezoidal Rule TN , the Midpoint Rule M N , or Simpson’s Rule4 S N as indicated. 11. y = 14 x 4 , [1, 2], SOLUTION

T5

Let y = 14 x 4 . Then 1 + (y )2 = 1 + (x 3 )2 = 1 + x 6 .

Therefore, the arc length over [1, 2] is " 2 1 + x 6 d x. 1

Now, let f (x) =

1 + x 6 . With n = 5, x =

1 2−1 = 5 5

and

6 7 8 9 5 {xi }i=0 = 1, , , , , 2 . 5 5 5 5

Using the Trapezoidal Rule, * ) " 2 4 # x 6 1 + x dx ≈ f (xi ) + f (x 5 ) = 3.957736. f (x 0 ) + 2 2 1 i=1 The arc length is approximately 3.957736 units. 13. y = x −1 , [1, 2], πS8 y = sin x, [0, 2 ], M8 SOLUTION Let y = x −1 . Then y = −x −2 and 1 1 + (y )2 = 1 + 4 . x

S E C T I O N 9.1

Arc Length and Surface Area

505

Therefore, the arc length over [1, 2] is

Now, let f (x) =

" 2 1 1 + 4 d x. x 1 1 + 14 . With n = 8, x

x =

2−1 1 = 8 8

9 5 11 3 13 7 15 8 {xi }i=0 = 1, , , , , , , , 2 . 8 4 8 2 8 4 8

and

Using Simpson’s Rule, * ) " 2 4 3 # # 1 x 1 + 4 dx ≈ f (x 2i−1 ) + 2 f (x 2i ) + f (x 8 ) = 1.132123. f (x 0 ) + 4 3 x 1 i=1 i=1 The arc length is approximately 1.132123 units. 15. Calculate the2 length of the astroid x 2/3 + y 2/3 = 1 (Figure 11). y = e−x , [0, 2], S8 y 1

x

−1

1

−1

FIGURE 11 Graph of x 2/3 + y 2/3 = 1. SOLUTION We will calculate the arc length of the portion of the asteroid in the first quadrant and then multiply by 4. By implicit differentiation

2 −1/3 2 −1/3 + y y = 0, x 3 3 so x −1/3 y 1/3 y = − −1/3 = − 1/3 . y x Thus y 2/3 x 2/3 + y 2/3 1 = 2/3 , 1 + (y )2 = 1 + 2/3 = x x 2/3 x and s=

" 1

1

0 x 1/3

dx =

3 . 2

The total arc length is therefore 4 · 32 = 6. 17. Find the arc length of the curve shown in Figure 12. Show that the arc length of the astroid x 2/3 + y 2/3 = a 2/3 (for a > 0) is proportional to a. y

0.5 x 1

2

3

FIGURE 12 Graph of 9y 2 = x(x − 3)2 . SOLUTION

Using implicit differentiation, 18yy = x(2)(x − 3) + (x − 3)2 = 3(x − 3)(x − 1)

Hence, (y )2 =

(x − 3)2 (x − 1)2 (x − 3)2 (x − 1)2 (x − 1)2 (x − 3)2 (x − 1)2 = = = 4x 36y 2 4(9y 2 ) 4x(x − 3)2

506

CHAPTER 9


and 1 + (y )2 =

(x − 1)2 + 4x (x + 1)2 = . 4x 4x

Finally, s=

" 3 0

" 3 (x + 1)2 |x + 1| dx = √ dx 4x 2 x 0

" 3 x +1 √ d x since x + 1 > 0 on [0, 3] = 0 2 x 3 " 3 √ 1 1/2 1 −1/2 1 3/2 = + x + x 1/2 = 2 3. x x dx = 2 2 3 0 0 19. Let f (x) = mx + r be a linear function (Figure 13). Use the Pythagorean Theorem to verify that the arc length over Find the value of a such that the arc length of the catenary y = cosh x for −a ≤ x ≤ a equals 10. [a, b] is equal to " b 1 + f (x)2 d x a

y

m(b − a) b−a

r

x a

b

FIGURE 13 SOLUTION

Let h denote the length of the hypotenuse. Then, by Pythagoras’ Theorem, h 2 = (b − a)2 + m 2 (b − a)2 = (b − a)2 (1 + m 2 ),

or

h = (b − a) 1 + m 2

since b > a. Moreover, ( f (x))2 = m 2 , so " b s= 1 + m 2 d x = (b − a) 1 + m 2 = h. a

" r dx 21. Show that the circumference C of the circle of radius r is C ="2 1 . Use a substitution to show that dx 2 /r 2improper integral). Evaluate, thus −r 1 − x (an Show that the circumference of the unit circle is equal to 2 −1 1 − x 2 C is proportional to r . verifying that the circumference is 2π . SOLUTION Let y = r 2 − x 2 denote the upper half of a circle of radius r centered at the origin. Then x2 r2 1 1 + (y )2 = 1 + 2 = 2 = , 2 2 2 r −x r −x 1 − x2 r

and the circumference of the circle is given by C =2

" r

dx . −r 1 − x 2 /r 2

Using the substitution u = x/r , du = d x/r , we find C = 2r

" 1

du = kr, −1 1 − u 2

where k=2

" 1

du −1 1 − u 2

is the circumference of the unit circle. Thus, the circumference is proportional to r .

S E C T I O N 9.1


507

√ √ Express length = [0, x over [0, 1]Use as atrigonometric definite integral. Then useEvaluate the substitution 23. Calculate the the arc arc length of yof=g(x) x 2 over a]. Hint: substitution. for a = u1.= x to show that this arc length is equal to the arc length of x 2 over [0, 1] (but do not evaluate the integrals). Explain this result graphically. √ SOLUTION Let g(x) = x. Then 1 + 4x 1 + g (x)2 = 4x

With the substitution u =

√

and s =

" 1 1 + 4x 0

4x

" 1√ 1 + 4x √ d x. dx = 2 x 0

1 x, du = √ d x, this becomes 2 x s=

" 1 1 + 4u 2 du. 0

Now, let f (x) = x 2 . Then 1 + f (x)2 = 1 + 4x 2 , and s=

" 1 1 + 4x 2 d x. 0

√ Thus, the two arc lengths are equal. This is explained graphically by the fact that for x ≥ 0, x 2 and x are inverses of each other. This means that the two graphs are symmetric with respect to the line y = x. Moreover, the graphs of x 2 and √ x intersect at x = 0 and at x = 1. Thus, it is clear that the arc length of the two graphs on [0, 1] are equal. √ 25. Find the arc length of y = e x over [0, a]. Hint: Try the substitution u = 1 + e2x followed by partial fractions. Use a graphing utility to sketch the graphs of f (x) = x and g(x) = ln x over [1, 4]. Then, by applying the )2 = 1 +show Comparison arc length that the the arc arc length length over of y [0, = a] f (x) SOLUTION LetTest y =toe xthe . Then 1 + (yintegrals, e2x , and is is: (a) Less than the arc length of y = g(x) over [1, " a4]. 1 + e2x d x. (b) Greater than the arc length of y = g(x) over [4, 8]. 0

Now, let u = 1 + e2x . Then du =

2e2x u2 − 1 1 · dx dx = 2 u 1 + e2x

and the arc length is " a " x=a " x=a " x=a u u2 1 2x 1 + e dx = u· 2 1+ 2 du = du = du u −1 u −1 0 x=0 x=0 u 2 − 1 x=0 x=a " x=a 1 1 1 1 1 1 1+ = − du = u + ln(u − 1) − ln(u + 1) 2 u − 1 2 u + 1 2 2 x=0 x=0 ) * a 1 1 + e2x − 1 = 1 + e2x + ln 2x 2 1+e +1 0 √ 1 + e2a − 1 √ 1 1 1+ 2 − 2 + ln √ = 1 + e2a + ln 2 2 2−1 1 + e2a + 1 √ 1 1 + e2a − 1 √ − 2 + ln(1 + 2). = 1 + e2a + ln 2a 2 1+e +1 27. Use Eq. (4) to compute the arc length of y = ln(sin x) for π4 ≤ x ≤ π2 . Show that the arc length of y = ln( f (x)) for a ≤ x ≤ b is SOLUTION With f (x) = sin x, Eq. (4) yields " b f (x)2 + f (x)2 dx π /2 " π /2 " π /2 2 2 f (x) sin x + cos xa s= csc x d x = ln (csc x − cot x) dx = sin x π /4 π /4 π /4 √ √ 1 = ln( 2 + 1). = ln 1 − ln( 2 − 1) = ln √ 2−1 x of y = f (x) over [a, b] is at most 29. Show that if 0 ≤ f (x) ≤ 1 for√all x, then the arc length e +1 for f (x)Use = x, equals − a). Eq.the (4)arc to length compute the arc2(b length of y = ln x over [1, 3]. e −1

√

2(b − a). Show that

508

CHAPTER 9

F U RT H E R A P P L I C AT I O N S O F T H E I N T E G R A L A N D TAY L O R P O LY N O M I A L S SOLUTION

If 0 ≤ f (x) ≤ 1 for all x, then s=

" b " b √ √ 1 + f (x)2 d x ≤ 1 + 1 d x = 2(b − a). a

a

If f (x) = x, then f (x) = 1 and s=

" b √ √ 1 + 1 d x = 2(b − a). a

31. Approximate the arc length of one-quarter of the unit circle (which we know is π ) by computing the length of the Letapproximation a, r > 0. Showwith thatNthe length of(Figure the curve x r + y r = ar for 0 ≤ x ≤ 2a is proportional to a. polygonal =arc 4 segments 14). y

x 0.25 0.5 0.75 1

FIGURE 14 One-quarter of the unit circle SOLUTION

With y =

1 − x 2 , the five points along the curve are P0 (0, 1), P1 (1/4,

Then

√

15/4), P2 (1/2,

√

3/2), P3 (3/4,

1 2 √ 2 21 4 − 15 3 P0 P1 = + 16 4 1 √ 2 √ 2 21 2 3 − 15 3 P1 P2 = + 16 4 1 √ 2 √ 2 21 2 3− 7 3 P2 P3 = + 16 4 7 1 P3 P4 = + 16 16

√

7/4), P4 (1, 0)

≈ .252009

≈ .270091

≈ .323042 ≈ .707108

and the total approximate distance is 1.552250 whereas π /2 ≈ 1.570796. In Exercises 32–39, compute the surface area of revolution about the x-axis over the interval. 33. y = 4x + 3, [2, 4] y = 4x + 3, [0, 1] SOLUTION Let y = 4x + 3. Then 1 + (y )2 = 17 and S A = 2π

" 4 2

4 √

√ √ (4x + 3) 17 d x = 2π 17 2x 2 + 3x = 60π 17. 2

)3/2 , [0, 8] 35. y = (4 −1x 2/3 y = 12 x 3 + x −1 , [1, 4] SOLUTION Let y = (4 − x 2/3 )3/2 . Then y = −x −1/3 (4 − x 2/3 )1/2 , and 1 + (y )2 = 1 +

4 − x 2/3 4 = 2/3 . 2/3 x x

Therefore, S A = 2π

" 8 0

(4 − x 2/3 )3/2

2 x 1/3

d x.

S E C T I O N 9.1


509

Using the substitution u = 4 − x 2/3 , du = − 23 x −1/3 d x, we find S A = 2π

" 0 4

u 3/2 (−3) du = 6π

12 5/2 4 384π 3/2 u du = πu = . 5 5 0 0

" 4

37. y = e x , −x [0, 1] y = e , [0, 1] SOLUTION Let y = e x . Then y = e x and S A = 2π

" 1 0

e x 1 + e2x d x.

Using the substitution e x = tan θ , e x d x = sec2 θ d θ , we find that "

" 1 1 e x 1 + e2x d x = sec3 θ d θ = sec θ tan θ + ln | sec θ + tan θ | + C 2 2 1 1 = e x 1 + e2x + ln | 1 + e2x + e x | + C. 2 2

Finally, SA =

1 π e x 1 + e2x + π ln | 1 + e2x + e x | 0

√ √ = π e 1 + e2 + π ln( 1 + e2 + e) − π 2 − π ln( 2 + 1) √ 1 + e2 + e 2 √ . = π e 1 + e − π 2 + π ln 2+1 39. y = sin x, [0, π ] y = x 2 , [0, 4] SOLUTION Let y = sin x. Then y = cos x, and S A = 2π

" π 0

sin x 1 + cos2 x d x.

Using the substitution cos x = tan θ , − sin x d x = sec2 θ d θ , we find that " " 1 1 sin x 1 + cos2 x d x = − sec3 θ d θ = − sec θ tan θ − ln | sec θ + tan θ | + C 2 2 1 1 = − cos x 1 + cos2 x − ln | 1 + cos2 x + cos x| + C. 2 2 Finally, 1 S A = 2π − cos x 1 + cos2 x − 2 √ 1 √ 1 = 2π 2 − ln( 2 − 1) + 2 2

π 1 ln | 1 + cos2 x + cos x| 2 0

√ √ √ √ 1 1 2 + ln( 2 + 1) = 2π 2 + ln( 2 + 1) . 2 2

2 by rotating the top half of the unit circle x 2 + y 2 = r 2 41. Prove that surface of aobtained sphere ofbyradius r isy 4=π rcosh Find thethe area of thearea surface rotating x over [− ln 2, ln 2] around the x-axis. about the x-axis. x SOLUTION Let y = r 2 − x 2 . Then y = − , and 2 r − x2

x2 r2 = 2 . 1 + (y )2 = 1 + 2 2 r −x r − x2 Finally, S A = 2π

" r " r r r2 − x2 d x = 2π r d x = 2π r (2r ) = 4π r 2 . −r −r r2 − x2

In Exercises 42–45, use a computer algebra system to find the exact or approximate surface area of the solid generated by rotating the curve about the x-axis. y = 14 x 2 − 12 ln x for 1 ≤ x ≤ e,

exact area

510

CHAPTER 9


43. y = x 2 for 0 ≤ x ≤ 4, SOLUTION

exact area

Let y = x 2 . Then y = 2x, 1 + (y )2 = 1 + 4x 2 , and S A = 2π

" 4

x 2 1 + 4x 2 d x.

0

Using a computer algebra system to evaluate the definite integral, we find √ √ π 129 65 π− ln(8 + 65). SA = 4 32 45. y = tan x for 0 ≤ x ≤ π4 , approximate area y = x 3 for 0 ≤ x ≤ 4, approximate area SOLUTION Let y = tan x. Then y = sec2 x, 1 + (y )2 = 1 + sec4 x, and S A = 2π

" π /4 0

tan x 1 + sec4 x d x.

Using a computer algebra system to approximate the value of the definite integral, we find S A ≈ 3.83908. A merchant intends to produce specialty carpets in the shape of the region in Figure 15, bounded by the 47. Find the surface area of the torus obtained by rotating the circle x 2 + (y − b)2 = r 2 about the x-axis. axes and graph of y = 1 − x n (units in yards). Assume that material costs $50/yd2 and that it costs 50L dollars to cut the carpet, where L is the length of the curved side of the carpet. The carpet can be sold for 150 A dollars, where A is the carpet’s area. Using numerical integration with a computer algebra system, find the whole number n for which the merchant’s profits are maximal. y 1 y = 1 − xn 0.5

A

x 0.5

1

FIGURE 15 SOLUTION

The area of the carpet is A=

" 1 0

(1 − x n )d x = x −

x n+1 1 1 n =1− = , n + 1 0 n+1 n+1

while the length of the curved side of the carpet is " 1 " 1 1 + (nx n−1 )2 d x = 1 + n 2 x 2n−2 d x. L= 0

0

Using these formulas, we find that the merchant’s profit is given by 150 A − (50 A + 50L) = 100 A − 50L =

" 1 100n 1 + n 2 x 2n−2 d x. − 50 n+1 0

Using a CAS, we find that the merchant’s profit is maximized (approximately $3.31 per carpet) when n = 13. The table below lists the profit for 1 ≤ n ≤ 15. n 1 2 3 4 5 6 7 8

Profit −20.71067810 −7.28047621 −2.39328273 −0.01147138 1.30534545 2.08684099 2.57017349 2.87535925

n

Profit

9 10 11 12 13 14 15

3.06855532 3.18862208 3.25953632 3.29668137 3.31024566 3.30715476 3.29222024

Prove that the portion of a sphere of radius R seen by an observer located at a distance d above the North Pole 2π d R 2


S E C T I O N 9.1

511

49. Suppose that the observer in Exercise 48 moves off to infinity, that is, d → ∞. What do you expect the limiting value of the observed area to be? Check your guess by calculating the limit using the formula for the area in the previous exercise. We would assume the observed surface area would approach 2π R 2 which is the surface area of a hemisphere of radius R. To verify this, observe: SOLUTION

2π R 2 d 2π R 2 = lim = 2π R 2 . d→∞ R + d d→∞ 1

lim S A = lim

d→∞


x 2 y 2 51. FindShow the surface of the ellipsoid obtained by cone rotating the ellipse + h is π r= 1r 2about x-axis. that thearea surface area of a right circular of radius r and aheight + h 2the . Hint: Rotate a line b y = mx about the x-axis for 0 ≤ x ≤ h, where m is determined suitably by the radius r . SOLUTION Taking advantage of symmetry, we can find the surface area of the ellipsoid by doubling the surface area obtained by rotating the portion of the ellipse in the first quadrant about the x-axis. The equation for the portion of the ellipse in the first quadrant is y=

b 2 a − x 2. a

Thus, a 4 + (b2 − a 2 )x 2 b2 x 2 = , 1 + (y )2 = 1 + 2 2 2 a (a − x ) a 2 (a 2 − x 2 ) and " a " a b a 4 + (b2 − a 2 )x 2 2 2 S A = 4π a −x d x = 4π b 0 a 0 a a2 − x 2

1+

2 b − a2 x 2 d x. a4

We now consider two cases. If b2 > a 2 , then we make the substitution a2 b2 − a 2 x = tan θ , d x = sec2 θ d θ , a2 b2 − a 2 and find that x=a " x=a a2 a2 S A = 4π b sec3 θ d θ = 2π b (sec θ tan θ + ln | sec θ + tan θ |) b2 − a 2 x=0 b2 − a 2 x=0 ⎞ ⎛ 2 2 2 2 2 2 − a 2 a b a b − a − a b 2 + 2π b 2+ 1+ ⎠ = ⎝2π bx 1 + ln x x x a4 a4 a2 0 b2 − a 2 a2 = 2π b2 + 2π b ln b2 − a 2

b + a

b2 − a 2 . a

On the other hand, if a 2 > b2 , then we make the substitution a2 a 2 − b2 x = sin θ , d x = cos θ d θ , 2 a a 2 − b2 and find that x=a " x=a a2 a2 S A = 4π b cos2 θ d θ = 2π b (θ + sin θ cos θ ) 2 2 2 2 x=0 a −b a −b x=0 ⎡ ⎤ 2 a − b2 a2 a 2 − b2 ⎦ a = ⎣2π bx 1 − sin−1 x x 2 + 2π b 4 a a2 a 2 − b2 0 a2 = 2π b2 + 2π b sin−1 a 2 − b2

a 2 − b2 . a

Observe that in both cases, as a approaches b, the value of the surface area of the ellipsoid approaches 4π b2 , the surface area of a sphere of radius b. Show that if the arc length of f (x) over [0, a] is proportional to a, then f (x) must be a linear function

512

CHAPTER 9


53.

let η =

Let L be the arc length of the upper half of the ellipse with equation y =

b a 2 − x 2 (Figure 17) and a

b2 1 − 2 . Use substitution to show that a

" π /2 1 − η2 sin2 θ d θ L=a −π /2

Use a computer algebra system to approximate L for a = 2, b = 1. y 1

x

−2

2

FIGURE 17 Graph of the ellipse y = 12 SOLUTION

Let y =

4 − x 2.

b 2 a − x 2 . Then a 1 + (y )2 =

and s=

" a −a

b2 x 2 + a 2 (a 2 − x 2 ) a 2 (a 2 − x 2 )

b2 x 2 + a 2 (a 2 − x 2 ) d x. a 2 (a 2 − x 2 )

With the substitution x = a sin t, d x = a cos t dt, a 2 − x 2 = a 2 cos2 t and " π /2 " π /2 a 2 b2 sin2 t + a 2 a 2 cos2 t b2 sin2 t dt = a cos t + cos2 t dt s=a a 2 (a 2 cos2 t) a2 π /2 −π /2 Because

b2 b2 1 − 2 , η2 = 1 − 2 a a

η= we then have

1 − η2 sin2 t = 1 −

b2 1− 2 a

b2 b2 sin2 t = 1 − sin2 t + 2 sin2 t = cos2 t + 2 sin2 t a a

which is the same as the expression under the square root above. Substituting, we get " π /2 s=a 1 − η2 sin2 t dt −π /2

When a = 2 and b = 1, η2 = 34 . Using a computer algebra system to approximate the value of the definite integral, we find s ≈ 4.84422.

f (x) an increasing function and of letthe g(x) be its inverse. Argue basismass of arc lengthρthat 55. LetLet M be thebetotal mass of a metal rodon in [a, the b] shape curve y= f (x) over [a,onb]the whose density (x) the following varies as aequality functionholds: of x. Use Riemann sums to justify the formula " b " f (b) " b

1 + f (x)2 d x = 1 + g (y)2 d y 5 ρ (x) 1 + f (x)2 d x M = a f (a) a

Then use the substitution u = f (x) to prove Eq. (5). SOLUTION Since the graphs of f (x) and g(x) are symmetric with respect to the line y = x, the arc length of the curves will be equal on the respective domains. Since the domain of g is the range of f , on f (a) to f (b), g(x) will have the same arc length as f (x) on a to b. If g(x) = f −1 (x) and u = f (x), then x = g(u) and du = f (x) d x. But g (u) =

1 1 1 = ⇒ f (x) = f (g(u)) f (x) g (u)

Now substituting u = f (x), " b " f (b) s= 1 + f (x)2 d x = a

f (a)

1+

2 " f (b) 1 (u) du = g g (u)2 + 1 du g (u) f (a)

S E C T I O N 9.2

Fluid Pressure and Force

513

9.2 Fluid Pressure and Force Preliminary Questions 1. How is pressure defined? SOLUTION

Pressure is defined as force per unit area.

2. Fluid pressure is proportional to depth. What is the factor of proportionality? SOLUTION

The factor of proportionality is the weight density of the fluid, w = ρ g, where ρ is the mass density of the

fluid. 3. When fluid force acts on the side of a submerged object, in which direction does it act? SOLUTION

Fluid force acts in the direction perpendicular to the side of the submerged object.

4. Why is fluid pressure on a surface calculated using thin horizontal strips rather than thin vertical strips? SOLUTION

Pressure depends only on depth and does not change horizontally at a given depth.

5. If a thin plate is submerged horizontally, then the fluid force on one side of the plate is equal to pressure times area. Is this true if the plate is submerged vertically? SOLUTION When a plate is submerged vertically, the pressure is not constant along the plate, so the fluid force is not equal to the pressure times the area.

Exercises 1. A box of height 6 ft and square base of side 3 ft is submerged in a pool of water. The top of the box is 2 ft below the surface of the water. (a) Calculate the fluid force on the top and bottom of the box. (b) Write a Riemann sum that approximates the fluid force on a side of the box by dividing the side into N horizontal strips of thickness y = 6/N . (c) To which integral does the Riemann sum converge? (d) Compute the fluid force on a side of the box. SOLUTION

(a) At a depth of 2 feet, the pressure on the top of the box is (62.5)(2) = 125 lb/ft2 ; therefore, the force on the top of the box is (125)(32 ) = 1125 lb. At a depth of 8 feet, the pressure on the bottom of the box is (62.5)(8) = 500 lb/ft2 ; therefore, the force on the bottom of the box is (500)(32 ) = 4500 lb. (b) Let y j denote the depth of the jth strip, for j = 1, 2, 3, . . . , N ; the pressure at this depth is 62.5y j . Because the strip has thickness y feet and width 3 feet, the area of the strip is 3y, and the force exerted by the fluid on the strip is 187.5y j y. Summing over the strips, we find that the fluid force on the side of the box is approximately F≈

N #

187.5y j y.

j=1

(c) As N → ∞, the Riemann sum

N #

187.5y j y converges to the definite integral 187.5

" 8

j=1

y d y. 2

(d) Using the result from part (c), the fluid force on the side of the box is 8 y 2 F = 187.5 y d y = 187.5 = 187.5(32 − 2) = 5625 lb. 2 2 2 " 8

3. Repeat Exercise 2, but assume that the top of the triangle is located 3 ft below the surface of the water. A plate in the shape of an isosceles triangle with base 1 ft and height 2 ft is submerged vertically in a tank of SOLUTION water so that its vertex touches the surface of the water (Figure 7). 1 y. f (y) y−3 (a) Show that the width of the triangle at depth y − is 3f (y) = 2 so f (y) = (a) Examine the figure below. By similar triangles, = . 2 1 2 (b) Consider a thin strip of thickness y at depth y. Explain why the fluid force on a side of this strip is approximately equal to w 12 y 2 y, where w = 62.5 lb/ft3 . (c) Write an approximation for the total fluid force F on a side of the plate as a Riemann sum and indicate the integral to which it converges. (d) Calculate F.

514

CHAPTER 9


(b) The pressure at a depth of y feet is wy lb/ft2 , and the area of the strip is approximately f (y) y = 12 (y − 3)y ft2 . Therefore, the fluid force on this strip is approximately 1 1 (y − 3)y = wy(y − 3)y. wy 2 2 (c) F ≈

N # j=1

w

y 2j − 3y j 2

y. As N → ∞, the Riemann sum converges to the definite integral " w 5 2 (y − 3y) d y. 2 3

(d) Using the result of part (c), 5 " w y3 62.5 w 5 2 3y 2 125 75 27 1625 F= − − lb. (y − 3y) d y = − 9− = = 2 3 2 3 2 2 3 2 2 6 3 5. Let F be the fluid force (in Newtons) on a side of a semicircular plate of radius r meters, submerged vertically in plate R inisFigure 8, bounded by the parabola y =9). x 2 and y = 1, is submerged vertically in water. Let F water soThe thatthin its diameter level with the water’s surface (Figure be the fluid force on one side of R. √ x (a) Show that the width of R at height y is f (y) =r 2 y and the fluid force on a side of a horizontal strip of thickness y 1/2 y at height y is approximately 2wy (1 − y)y. r " 1 2 r 2 − y2 (b) Write a Riemann sum that approximates F and use it to explain why F = 2w y 1/2 (1 − y) d y. 0

(c) Calculate F. FIGURE 9

(a) Show that the width of the plate at depth y is 2 r 2 − y 2 . (b) Calculate F using Eq. (2). SOLUTION

(a) Place the origin at the center of the semicircle and point the positive y-axis downward. The equation for the edge of 2 + y 2 = r 2 . At a depth of y, the plate extends from the point (− r 2 − y 2 , y) on the left the semicircular plate is then x to the point ( r 2 − y 2 , y) on the right. The width of the plate at depth y is then

r 2 − y2 − − r 2 − y2 = 2 r 2 − y2. (b) With w = 9800 N/m3 , F = 2w

r " r 19600 2 19600r 3 y r 2 − y2 d y = − (r − y 2 )3/2 = N. 3 3 0 0

7. A semicircular plate of radius r , oriented as in Figure 9, is submerged in water so that its diameter is located at a Calculate the force on one side of a circular plate with radius 2 ft, submerged vertically in a tank of water so that depth of m feet. Calculate the force on one side of the plate in terms of m and r . the top of the circle is tangent to the water surface. SOLUTION Place the origin at the center of the semicircular plate with the positive y-axis pointing downward. The water surface is then at y = −m. Moreover, at location y, the width of the plate is 2 r 2 − y 2 and the depth is y + m. Thus, " r

F = 2w (y + m) r 2 − y 2 d y. 0

S E C T I O N 9.2


515

Now, r " r 1 1 y r 2 − y 2 d y = − (r 2 − y 2 )3/2 = r 3 . 3 3 0 0 Geometrically, " r r 2 − y2 d y 0

2 represents the area of one quarter of a circle of radius r , and thus has the value π4r . Bringing these results together, we find that

F=

2 3 1 125 3 125 wr + m π wr 2 = r + m π r 2 lb. 3 2 3 4

9. Calculate the total force (in Newtons) on a side of the plate in Figure 11(A), submerged in water. Figure 10 shows the wall of a dam on a water reservoir. Use the Trapezoidal Rule and the width and depth SOLUTION The width the to plate variesthe linearly fromon 4 meters measurements in the of figure estimate total force the wall.at a depth of 3 meters to 7 meters at a depth of 5 meters. Thus, at depth y, the width of the plate is 3 3 1 4 + (y − 3) = y − . 2 2 2 Finally, the force on a side of the plate is F =w

" 5 3 1 1 3 1 2 5 y y− dy = w y − y = 45w = 441000 N. 2 2 2 4 3 3

11. The plate in Figure 12 is submerged in water with its top level with the surface of the water. The left and right edges Calculate the total force (in Newtons) on a side of the plate in Figure 11(B), submerged in a fluid of mass density of the plate are the 3curves y = x 1/3 and y = −x 1/3 . Find the fluid force on a side of the plate. ρ = 800 kg/m . y 2

Water level y = x 1/3

y = −x 1/3

x

−8

8

FIGURE 12

At height y, the plate extends from the point (−y 3 , y) on the left to the point (y 3 , y) on the right; thus, the width of the plate is f (y) = y 3 − (−y 3 ) = 2y 3 . Because the water surface is at height y = 2, the horizontal strip at height y is at a depth of 2 − y. Consequently, SOLUTION

F =w

" 2 0

(2 − y)(2y 3 ) d y = 2w

1 4 1 5 2 16w y − y = . 2 5 5 0

If distances are in feet, then w = 62.5 lb/ft3 and F = 200 lb; if distances are in meters, then w = 9800 N/m3 and F = 31360 N. 13. In the notation of Exercise 12, calculate the fluid force on a side of the plateπR if it is oriented as in Figure 13(A). R be in the shape of the under y substitution. = sin x for 0 ≤ x ≤ 2 in Figure 13(A). Find the fluid force You mayLet need to the useplate Integration by Parts andregion trigonometric on a side of R if it is rotated counterclockwise by 90◦ and submerged in a fluid of density 140 lb/ft3 with its top SOLUTION Place the origin at the lower left corner of the plate. Because the fluid surface is at height y = 1, the edge level with the surface of the fluid as in (B). horizontal strip at height y is at a depth of 1 − y. Moreover, this strip has a width of

π − sin−1 y = cos−1 y. 2 Thus, F =w

" 1 0

(1 − y) cos−1 y d y.

Starting with integration by parts, we find " 1 0

1 " 1 y − 1 y2 1 2 −1 2 y − y cos y + dy 2 0 1 − y2 0 " " 1 " 1 y − 1 y2 1 1 1 y y2 2 = cos−1 1 + dy = dy − d y. 2 2 2 2 0 0 0 1− y 1− y 1 − y2

(1 − y) cos−1 y d y =

516

CHAPTER 9


Now, " 1 0

y 1 − y2

dy = −

1

1 − y 2 = 1. 0

For the remaining integral, we use the trigonometric substitution y = sin θ , d y = cos θ d θ and find y=1 " " 1 y=1 2 1 y2 1 1 dy = sin θ d θ = (θ − sin θ cos θ ) 2 2 0 2 4 y=0 1− y y=0 1

1 π = sin−1 y − y 1 − y 2 = . 4 8 0 Finally,

π w ≈ 85.0 lb. F = 1− 8 15. Calculate the fluid force on one side of the “infinite” plate B in Figure 14. Let A be the region under the graph of y = ln x for 1 ≤ x ≤ e (Figure 14). Calculate the fluid force on one side of a plate in the shape of region A if the waterysurface is at y = 1. y = ln x 1 A x B

1

e

FIGURE 14 SOLUTION Because the fluid surface is at height y = 1, the horizontal strip at height y is at a depth of 1 − y. Moreover, this strip has a width of e y . Thus, " 0 (1 − y)e y d y. F =w

−∞

Using integration by parts, we find " 0 −∞

0 (1 − y)e y d y = (1 − y)e y + e y −∞ = 2.

Thus, F = 2w. If distances are in feet, then w = 62.5 lb/ft3 and F = 125 lb; if distances are in meters, then w = 9800 N/m3 and F = 19600 N. 17. Repeat Exercise 16, but assume that the top edge of the plate lies at a depth of 6 m. A square plate of side 3 m is submerged in water at an incline of 30◦ with the horizontal. Its top edge is located SOLUTION Because plateCalculate is 3 meters a side, at aone horizontal angle of 30◦ , and has its top edge at the surface of thethe water. the on fluid force is(insubmerged Newtons) on side of the plate. located at a depth of 6 meters, the bottom edge of the plate is located at a depth of 6 + 3 sin 30◦ = 15 2 meters. Let y denote the depth at any point of the plate. The width of each horizontal strip of the plate is then y = 2y, sin 30◦ and F = 2w

" 15/2 6

3y d y =

243 w = 595350 N. 4

19. Calculate the fluid force on one side of the plate (an isosceles triangle) shown in Figure 15(B). Figure 15(A) shows a ramp inclined at 30◦ leading into a swimming pool. Calculate the fluid force on the ramp. Water surface

Water surface 4 30˚

y f(y)

10 Vertical change Δy

6 3

60˚ (B)

(A)

FIGURE 15



517

3 y and width A horizontal strip at depth y has length f (y) = 10

y 2 = √ y. sin 60◦ 3 Thus, √ " 10 200 3 3 y2 d y = w w. 5 3 0

√ F=

If distances are in feet, then w = 62.5 lb/ft3 and F ≈ 7216.88 lb; if distances are in meters, then w = 9800 N/m3 and F ≈ 1,131,606.5 N. 21. Calculate the fluid pressure on one of the slanted sides of the trough in Figure 16, filled with corn syrup as in Exercise The trough in Figure 16 is filled with corn syrup, whose density is 90 lb/ft3 . Calculate the force on the front side 20. of the trough.

b h

d a

FIGURE 16 SOLUTION

The diagram above displays a side view of the trough. From this diagram, we see that sin θ =

h 2

b−a 2

. + h2

Thus,

2 b−a " h + h 2 dh 2 90 2 w b−a 2 F= d · y dy = + h2. = 45dh sin θ 0 h 2 2 Figure 17 shows an object whose face is an equilateral triangle with 5-ft sides. The object is 2 ft thick and is Further Insights and Challenges submerged in water with its vertex 3 ft below the water surface. Calculate the fluid force on both a triangular face 23. The of therectangular trough in Figure is object. an equilateral triangle of side 3. Assume that the trough is filled with water to and aend slanted edge of18the height y. Calculate the fluid force on each side of the trough as a function of the level y and the length l of the trough.

3

y

FIGURE 18 SOLUTION Because we want the answer to be a function of y, to avoid confusion, we will use h to denote the variable in the vertical direction. Now, place the origin at the lower vertex of the trough and orient the positive h-axis pointing upward. First, consider the faces at the front and back ends of the trough. A horizontal strip at height h has a length of 2h √ and is at a depth of y − h. Thus, 3 √ y " y 2h 2 3 3 y F =w (y − h) √ d y = w √ h 2 − √ h 3 = wy . 9 3 3 3 3 0 0

518

CHAPTER 9


For the slanted sides, we note that each side makes an angle of 60◦ with the horizontal. If we let denote the length of the trough, then √ " 3 2w y (y − h) d y = wy 2 . F= √ 3 3 0 25. Prove that the force on the side of a rectangular plate of area A submerged vertically in a fluid is equal to p0 A, where A rectangular plate of side is submerged vertically in a fluid of density w, with its top edge at depth h. Show p0 is the fluid pressure at the center point of the rectangle. that if the depth is increased by an amount h, then the force on a side of the plate increases by w Ah, where A is SOLUTION denote the length of the vertical side of the rectangle, x denote the length of the horizontal side of the the area ofLet the plate. rectangle, and suppose the top edge of the rectangle is at depth y = m. The pressure at the center of the rectangle is then p0 = w m + , 2 and the force on the side of the rectangular plate is F=

" +m m

wx wx ( + m)2 − m 2 = ( + 2m) = Aw + m = Ap0 . 2 2 2

wx y d y =

If the density of a fluid varies with depth, then the pressure at depth y is a function p(y) (which need not equal wy as in the case of constant density). Use Riemann sums to argue that the total force F on the flat side of a " b 9.3 Center of Mass f (y) p(y) d y, where f (y) is the width of the side at depth y. submerged object submerged vertically is F = a


1. What are the x- and y-moments of a lamina whose center of mass is located at the origin? SOLUTION

Because the center of mass is located at the origin, it follows that Mx = M y = 0.

2. A thin plate has mass 3. What is the x-moment of the plate if its center of mass has coordinates (2, 7)? SOLUTION The x-moment of the plate is the product of the mass of the plate and the y-coordinate of the center of mass. Thus, Mx = 3(7) = 21.

3. The center of mass of a lamina of total mass 5 has coordinates (2, 1). What are the lamina’s x- and y-moments? SOLUTION The x-moment of the plate is the product of the mass of the plate and the y-coordinate of the center of mass, whereas the y-moment is the product of the mass of the plate and the x-coordinate of the center of mass. Thus, Mx = 5(1) = 5, and M y = 5(2) = 10.

4. Explain how the Symmetry Principle is used to conclude that the centroid of a rectangle is the center of the rectangle. SOLUTION Because a rectangle is symmetric with respect to both the vertical line and the horizontal line through the center of the rectangle, the Symmetry Principle guarantees that the centroid of the rectangle must lie along both of these lines. The only point in common to both lines of symmetry is the center of the rectangle, so the centroid of the rectangle must be the center of the rectangle.

Exercises 1. Four particles are located at points (1, 1) (1, 2) (4, 0) (3, 1) (a) Find the moments Mx and M y and the center of mass of the system, assuming that the particles have equal mass m. (b) Find the center of mass of the system, assuming the particles have mass 3, 2, 5, and 7, respectively. SOLUTION

(a) Because each particle has mass m, Mx = m(1) + m(2) + m(0) + m(1) = 4m; M y = m(1) + m(1) + m(4) + m(3) = 9m; and the total mass of the system is 4m. Thus, the coordinates of the center of mass are M y Mx 9m 4m 9 , = , = ,1 . M M 4m 4m 4

S E C T I O N 9.3

Center of Mass

519

(b) With the indicated masses of the particles, Mx = 3(1) + 2(2) + 5(0) + 7(1) = 14; M y = 3(1) + 2(1) + 5(4) + 7(3) = 46; and the total mass of the system is 17. Thus, the coordinates of the center of mass are M y Mx 46 14 , = , . M M 17 17 3. Point masses of equal size are placed at the vertices of the triangle with coordinates (a, 0), (b, 0), and (0, c). Show Find the center of mass for the system of particles of mass 4, 2, 5, 1 located at (1, 2), (−3, 2), (2, −1), (4, 0). that the center of mass of the system of masses has coordinates ( 13 (a + b), 13 c). SOLUTION

Let each particle have mass m. The total mass of the system is then 3m. and the moments are Mx = 0(m) + 0(m) + c(m) = cm; and M y = a(m) + b(m) + 0(m) = (a + b)m.

Thus, the coordinates of the center of mass are M y Mx (a + b)m cm a+b c , = , = , . M M 3m 3m 3 3 2 5. Sketch lamina of constant density ρ = 3 g/cm the graph of y = x 2 for Pointthe masses of Smass m 1 , m 2 , and m 3 are placed at theoccupying points (−1,the 0),region (3, 0),beneath and (0, 4). 0 ≤ x ≤ 3. (a) Suppose that m 1 = 6. Show that there is a unique value of m 2 such that the center of mass lies on the y-axis. (a) Use formulas (1) and (2) to compute Mx and M y . (b) Suppose that m 1 = 6 and m 2 = 4. Find the value of m 3 such that yCM = 2. (b) Find the area and the center of mass of S. SOLUTION

A sketch of the lamina is shown below y 8 6 4 2 x 0

0.5 1 1.5 2 2.5 3

(a) Using (2), Mx = 3

" 9 0

y(3 −

√

y) d y =

9y 2 729 6 5/2 9 − y = 10 . 2 5 0

Using (1), My = 3

" 3 0

x(x 2 ) d x =

3x 4 3 243 = . 4 0 4

(b) The area of the lamina is x 3 3 2 A= x dx = = 9 cm2 . 3 0 0 " 3

With a constant density of ρ = 3 g/cm2 , the mass of the lamina is M = 27 grams, and the coordinates of the center of mass are M y Mx 243/4 729/10 9 27 , = , = , . M M 27 27 4 10 7. Find the moments and center of mass of the lamina of uniform density ρ occupying the region underneath y = x 32 for 0 ≤ Use x ≤ Eqs. 2. (1) and (3) to find the moments and center of mass of the lamina S of constant density ρ = 2 g/cm occupying the region between y = x 2 and y = 9x over [0, 3]. Sketch S, indicating the location of the center of SOLUTION With uniform density ρ, mass. Mx =

" 2 1 64ρ ρ (x 3 )2 d x = 2 0 7

and

My = ρ

" 2 0

x(x 3 ) d x =

32ρ . 5

520

CHAPTER 9


The mass of the lamina is M =ρ so the coordinates of the center of mass are

9. (a) (b)

" 2 0

x 3 d x = 4ρ,

M y Mx , M M

=

8 16 , . 5 7

Let T be the triangular lamina in Figure 17. the graph of y = 1 − x 2 for 0 ≤ x ≤ 1 in two ways, Calculate Mx (assuming ρ = 1) for the region underneath Show that the horizontal cut at height y has length 4 − 23 y and use Eq. (2) to compute Mx (with ρ = 1). first using Eq. (2) and then using Eq. (3). Use the Symmetry Principle to show that M y = 0 and find the center of mass. y 6

x

−2

2

FIGURE 17 Isosceles triangle. SOLUTION

(a) The equation of the line from (2, 0) to (0, 6) is y = −3x + 6, so x =2−

1 y. 3

The length of the horizontal cut at height y is then 1 2 2 2 − y = 4 − y, 3 3 and Mx =

" 6 2 y 4 − y d y = 24. 3 0

(b) Because the triangular lamina is symmetric with respect to the y-axis, xcm = 0, which implies that M y = 0. The total mass of the lamina is " 2 (−3x + 6) d x = 12, M =2 0

so ycm = 24/12. Finally, the coordinates of the center of mass are (0, 2). In Exercises 10–17, find the centroid of the region lying underneath the graph of the function over the given interval. √ 11. f (x) = x, [4, 9] f (x) = 6 − 2x, [0, 3] SOLUTION The moments of the region are Mx =

" 1 9 65 x dx = 2 4 4

My =

and

" 9 √ 422 . x x dx = 5 4

The area of the region is A= so the coordinates of the centroid are

13. f (x) = 9 − x32 , [0, 3] f (x) = x , [0, 1]

" 9 √ 4

M y Mx , A A

x dx =

=

38 , 3

633 195 , . 95 152


Center of Mass

521

The moments of the region are Mx =

" 1 3 324 (9 − x 2 )2 d x = 2 0 5

My =

and

" 3 0

x(9 − x 2 ) d x =

81 . 4


" 3

(9 − x 2 ) d x = 18,

0

M y Mx , A A

=

9 18 , . 8 5

15. f (x) = e−x , [0, 4] f (x) = (1 + x 2 )−1/2 , [0, 1] SOLUTION The moments of the region are Mx =

" 1 4 −2x 1

e dx = 1 − e−8 2 0 4

My =

and

" 4 0

4 xe−x d x = −e−x (x + 1) = 1 − 5e−4 . 0

The area of the region is A=

" 4 0

so the coordinates of the centroid are

M y Mx , A A

e−x d x = 1 − e−4 ,

=

1 − 5e−4 1 − e−8 . , 1 − e−4 4(1 − e−4 )

17. f (x) = sin x, [0, π ] f (x) = ln x, [1, 2] SOLUTION The moments of the region are π " 1 π 2 1 π sin x d x = (x − sin x cos x) = ; and 2 0 4 4 0 π " π x sin x d x = (−x cos x + sin x) = π . My =

Mx =

0

0


" π 0

M y Mx , A A

sin x d x = 2, =

π π , . 2 8

19. Sketch the region between y = x + 4 and y = 2 − x for 0 ≤ x ≤ 2. Using symmetry, explain why the centroid of Calculate the moments and center of mass of the lamina occupying the region between the curves y = x and the region 2lies on the line y = 3. Verify this by computing the moments and the centroid. y = x for 0 ≤ x ≤ 1. SOLUTION A sketch of the region is shown below.

5 4 3 2 1 0.5

1

1.5

2

522

CHAPTER 9


The region is clearly symmetric about the line y = 3, so we expect the centroid of the region to lie along this line. We find " 1 2

(x + 4)2 − (2 − x)2 d x = 24; Mx = 2 0 " 2 28 x ((x + 4) − (2 − x)) d x = My = ; and 3 0 " 2 A= ((x + 4) − (2 − x)) d x = 8. 0

Thus, the coordinates of the centroid are 6 , 3 . 7

In Exercises 20–25, find the centroid of the region lying between the graphs of the functions over the given interval. √ 21. y = x 2 , y = √ x, [0, 1] y = x, y = x, [0, 1] SOLUTION The moments of the region are Mx =

" 1 1 3 (x − x 4 ) d x = 2 0 20

and

My =

" 1

√ 3 x( x − x 2 ) d x = . 20 0

The area of the region is " 1 √ 1 ( x − x 2) d x = , 3 0

A= so the coordinates of the centroid are

9 9 , . 20 20

Note: This makes sense, since the functions are inverses of each other. This makes the region symmetric with respect to the line y = x. Thus, by the symmetry principle, the center of mass must lie on that line. 23. y = e x , y = 1, [0, 1] y = x −1 , y = 2 − x, [1, 2] SOLUTION The moments of the region are Mx =

" 1 1 2x e2 − 3 (e − 1) d x = 2 0 4

and

My =

" 1 0

x(e x − 1) d x =

xe x − e x −

1 2 1 1 x = . 2 2 0


" 1 0

(e x − 1) d x = e − 2,

1 e2 − 3 , . 2(e − 2) 4(e − 2)

25. y = sin x, y = cos x, [0, π /4] y = ln x, y = x − 1, [1, 3] SOLUTION The moments of the region are " " 1 π /4 1 π /4 1 (cos2 x − sin2 x) d x = cos 2x d x = ; and 2 0 2 0 4 √ π /4 " π /4 π 2 x(cos x − sin x) d x = [(x − 1) sin x + (x + 1) cos x] = − 1. My = 4 0 0 Mx =


" π /4 0

(cos x − sin x) d x =

√

2 − 1,

√ π 2−4 1 √ , √ . 4( 2 − 1) 4( 2 − 1)

S E C T I O N 9.3

Center of Mass

523

27. Sketch the region enclosed by y = 0, y = (x + 1)3 , and y = (1 − x)32 and find its centroid. Sketch the region enclosed by y = 0, y = x + 1, and y = (x − 1) and find its centroid. SOLUTION A sketch of the region is shown below. y 1

x

−1

The moments of the region are 1 Mx = 2

" 0 −1

1

(x + 1)6 d x +

" 1 0

(1 − x)6 d x

=

1 ; and 7

M y = 0 by the Symmetry Principle. The area of the region is A=

" 0 −1

(x + 1)3 d x +

so the coordinates of the centroid are 0, 27 .

" 1 0

(1 − x)3 d x =

1 , 2

In Exercises 28–32, find the centroid of the region.

x 2 y 2 2 y = 2 1 for arbitrary a, b > 0

x+ 29. Top half of the ellipse +b =1 Top half of the ellipsea 2 4 SOLUTION The equation of the top half of the ellipse is y=

b2 −

b2 x 2 a2

Thus, ⎛ ⎞2 " 1 a ⎝ 2 b2 x 2 ⎠ 2ab2 b − 2 dx = . Mx = 2 −a 3 a By the Symmetry Principle, M y = 0. The area of the region is one-half the area of an ellipse with axes of length a and b; i.e., 12 π ab. Finally, the coordinates of the centroid are 4b . 0, 3π 31. Quarter of the unit circle lying in the first quadrant Semicircle of radius r with center at the origin SOLUTION By the Symmetry Principle, the center of mass must lie on the line y = x in the first quadrant. Therefore, we need only calculate one of the moments of the region. With y = 1 − x 2 , we find " 1 1 My = x 1 − x2 dx = . 3 0 The area of the region is one-quarter of the area of a unit circle; i.e., 14 π . Thus, the coordinates of the centroid are 4 4 , . 3π 3π 33. Find the centroid for the shaded region of the semicircle of radius r in Figure 18. What is the centroid when r = 1 Triangular plate with vertices (−c, 0), (0, c), (a, b), where a, b, c > 0, and b < c and h = 12 ? Hint: Use geometry rather than integration to show that the area of the region is r 2 sin−1 ( 1 − h 2 /r 2 ) − h r 2 − h2. y

r

h x

FIGURE 18

524

CHAPTER 9


From the symmetry of the region, it is obvious that the centroid lies along the y-axis. To determine the y-coordinate of the centroid, we must calculate the moment about the x-axis and the area of the region. Now, the length of the horizontal cut of the semicircle at height y is

r 2 − y2 − − r 2 − y2 = 2 r 2 − y2. SOLUTION

Therefore, taking ρ = 1, we find Mx = 2

" r 2 y r 2 − y 2 d y = (r 2 − h 2 )3/2 . 3 h

Observe that the region is comprised ofa sector of the circle with the triangle between the two radii removed. The angle of the sector is 2θ , where θ = sin−1 1 − h 2 /r 2 , so the area of the sector is 12 r 2 (2θ ) = r 2 sin−1 1 − h 2 /r 2 . The triangle has base 2 r 2 − h 2 and height h, so the area is h r 2 − h 2 . Therefore, YC M =

2 (r 2 − h 2 )3/2 Mx . = 3 A r 2 sin−1 1 − h 2 /r 2 − h r 2 − h 2

When r = 1 and h = 1/2, we find √ 2 (3/4)3/2 3 3 3 √ √ = YC M = √ . 4π − 3 3 sin−1 23 − 43 In Exercises 34–36, use the additivity of moments to find the COM of the region. 35. An ice cream cone consisting of a semicircle on top of an equilateral triangle of side 6 (Figure 20) Isosceles triangle of height 2 on top of a rectangle of base 4 and height 3 (Figure 19) y

6 x

−3

3

FIGURE 20 SOLUTION The region is symmetric with respect to the y-axis, so M y = 0 by the Symmetry Principle. The moment about the x-axis for the triangle is

2 triangle Mx = √

" 3√3

3 0

y 2 d y = 54.

√ √ For the semicircle, first note that the center is (0, 3 3), so the equation is x 2 + (y − 3 3)2 = 9, and Mxsemi = 2

" 3+3√3 √ y 9 − (y − 3 3)2 d y. √ 3 3

√ Using the substitution w = y − 3 3, dw = d y, we find " 3 √ Mxsemi = 2 (w + 3 3) 9 − w 2 dw 0

√ " 3 √ " 3 27π 3 w 9 − w 2 dw + 6 3 9 − w 2 dw = 18 + =2 , 2 0 0 ! where we have used the fact that 03 9 − w 2 dw represents the area of one-quarter of a circle of radius 3. The total moment about the x-axis is then √ 27π 3 triangle semi Mx = Mx + Mx = 72 + . 2 √ Because the area of the region is 9 3 + 92π , the coordinates of the center of mass are √ 16 + 3π 3 . 0, √ π+2 3

S E C T I O N 9.3

Center of Mass

525

37. Let S be the lamina of mass density ρ = 1 obtained by removing a circle of radius r from the circle of radius 2r Three-quarters of the unit circle (remove the part in the fourth quadrant) big shown in Figure 21. Let MxS and M yS denote the moments of S. Similarly, let M y and M ysmall be the y-moments of the larger and smaller circles. y

2r r x

FIGURE 21

(a) Use the Symmetry Principle to show that MxS = 0. big

(b) Show that M yS = M y

− M ysmall using the additivity of moments.

big

(c) Find M y and M ysmall using the fact that the COM of a circle is its center. Then compute M yS using (b). (d) Determine the COM of S. SOLUTION

(a) Because S is symmetric with respect to the x-axis, MxS = 0. (b) Because the small circle together with the region S comprise the big circle, by the additivity of moments, big

M yS + M ysmall = M y . big

Thus M yS = M y

− M ysmall . big

big

(c) The center of the big circle is the origin, so x cm = 0; consequently, M y small = −r ; consequently small circle is (−r, 0), so xcm

= 0. On the other hand, the center of the

small · Asmall = −r · π r 2 = −π r 3 . M ysmall = x cm

By the result of part (b), it follows that M yS = 0 − (−π r 3 ) = π r 3 . (d) The area of the region S is 4π r 2 − π r 2 = 3π r 2 . The coordinates of the center of mass of the region S are then

r πr 3 , 0 = ,0 . 3 3π r 2

Find the COM of the laminas in Figure 22 obtained by removing squares of side 2 from a square of side 8. Further Insights and Challenges 39. A median of a triangle is a segment joining a vertex to the midpoint of the opposite side. Show that the centroid of a triangle lies on each of its medians, at a distance two-thirds down from the vertex. Then use this fact to prove that the three medians intersect at a single point. Hint: Simplify the calculation by assuming that one vertex lies at the origin and another on the x-axis. SOLUTION Orient the triangle by placing one vertex at (0, 0) and the long side of the triangle along the x-axis. Label cx ac the vertices (0, 0), (a, 0), (b, c). Thus, the equations of the short sides are y = cx b and y = b−a − b−a . Now,

" " 1 b 1 a cx − ac 2 ac2 (cx/b)2 d x + dx = ; 2 0 2 b b−a 6 " b " a cx − ac ac(a + b) x(cx/b) d x + x dx = ; and My = b − a 6 0 b ac M= . 2 a+b c , . To show that the centroid lies on each median, let y1 be the median from (b, c), y2 so the center of mass is 3 3 the median from (0, 0) and y3 the median from (a, 0). We find a+b 2c c y1 (x) = (x − a/2), so y1 = ; 2b − a 3 3 c c a+b y2 (x) = x, so y2 = ; a+b 3 3 Mx =

526

CHAPTER 9


y3 (x) =

c (x − a), b − 2a

so

y3

a+b 3

=

c . 3

This shows that the center of mass lies on each median. We now show that the center of mass is 23 of the way from each vertex. For y1 , note that x = b gives the vertex and x = a2 gives the midpoint of the opposite side, so two-thirds of this distance is a+b 2 a −b = , x =b+ 3 2 3 a+b the x-coordinate of the center of mass. Likewise, for y2 , two-thirds of the distance from x = 0 to x = a+b 2 is 3 , and for y3 , the two-thirds point is 2 b a+b x =a+ −a = . 3 2 3

A similar method shows that the y-coordinate is also two-thirds of the way along the median. Thus, since the centroid lies on all three medians, we can conclude that all three medians meet at a single point, namely the centroid. 41. Find the COM of a system of two weights of masses m 1 and m 2 connected by a lever of length d whose mass density Let P be the COM of a system of two weights with masses m and m 2 separated by a distance d. Prove ρ is uniform. Hint: The moment of the system is the sum of the moments of1 the weights and the lever. Archimedes’s Law of the (weightless) Lever: P is the point on line between the two weights such that 1 m 1 = 2 m 2 , SOLUTION A distance be the cross-sectional area where j Let is the from mass j to P.of the rod. Place the rod with m 1 at the origin and rod lying on the positive 1 x-axis. The y-moment of the rod is M y = 2 ρ Ad 2 , the y-moment of the mass m 2 is M y = m 2 d, and the total mass of the system is M = m 1 + m 2 + ρ Ad. Therefore, the x-coordinate of the center of mass is m 2 d + 12 ρ Ad 2

m 1 + m 2 + ρ Ad

.

43. Prove directly that Eqs. (2) and (3) are equivalent in the following situation. Let f (x) be a positive decreasing Let under the=graph of f Show (x) over thatthe interval [−a, a], where f (x) ≥ 0. function onSymmetry [0, b] suchPrinciple that f (b) = 0.R Setbedthe = region f (0) and g(y) f −1 (y). Assume that R is symmetrical with respect to the y-axis. " " d (a) Explain why f (x) is even, that is, f1(x) b= f (−x). f (x)2 d x = yg(y) d y (b) Show that x f (x) is an odd function.2 0 0 Useapply (b) tothe prove that M y = Hint:(c) First substitution y 0. = f (x) to the integral on the left and observe that d x = g (y) d y. Then apply (d) Prove that the COM of R lies on the y-axis (a similar argument applies to symmetry with respect to the x-axis). Integration by Parts. SOLUTION

f (x) ≥ 0 and f (x) < 0 shows that f has an inverse g on [a, b]. Because f (b) = 0, f (0) = d, and

f −1 (x) = g(x), it follows that g(d) = 0 and g(0) = b. If we let x = g(y), then d x = g (y) d y. Thus, with y = f (x), " " " 1 b 2 1 0 2 1 b f (x)2 d x = y dx = y g (y) d y. 2 0 2 0 2 d Using Integration by Parts with u = y 2 and v = g (y) d y, we find * ) 0 " " 0 " d " 0 1 2 1 1 0 2 0 − d 2 g(d) − y g (y) d y = yg(y) d y = yg(y) d y = yg(y) d y. y g(y) − 2 2 d 2 2 d d 0 d

Let R be a lamina of uniform density submerged in a fluid of density w (Figure 23). Prove the following law: The fluid force on one side of R is equal to the area of R times the fluid pressure on the centroid. Hint: Let g(y) be 9.4theTaylor Polynomials horizontal width of R at depth y. Express both the fluid pressure [Eq. (2) in Section 9.2] and y-coordinate of the centroid in terms of g(y).

Preliminary Questions 1. What is T3 (x) centered at a = 3 for a function f (x) such that f (3) = 9, f (3) = 8, f (3) = 4, and f (3) = 12. SOLUTION

In general, with a = 3, f (3) f (3) (x − 3)2 + (x − 3)3 . T3 (x) = f (3) + f (3)(x − 3) + 2 6

Using the information provided, we find T3 (x) = 9 + 8(x − 3) + 2(x − 3)2 + 2(x − 3)3 .

S E C T I O N 9.4

Taylor Polynomials

527

2. The dashed graphs in Figure 3 are Taylor polynomials for a function f (x). Which of the two is a MacLaurin polynomial? y

y y = f(x) 1

2

3

-1

y = f(x) x

1

3

-1

x

2

(A)

(B)

FIGURE 3 SOLUTION A MacLaurin polynomial always gives the value of f (0) exactly. This is true for the Taylor polynomial sketched in (B); thus, this is the MacLaurin polynomial.

3. For which value of x does the MacLaurin polynomial Tn (x) satisfy Tn (x) = f (x), no matter what f (x) is? SOLUTION

A MacLaurin polynomial always gives the value of f (0) exactly.

4. Let Tn (x) be the MacLaurin polynomial of a function f (x) satisfying | f (4) (x)| ≤ 1 for all x. Which of the following statements follow from the Error Bound? (a) |T4 (2) − f (2)| ≤ 24 /24 (b) |T3 (2) − f (2)| ≤ 23 /6 (c) |T3 (2) − f (2)| ≤ 1/3 SOLUTION

For a function f (x) satisfying | f (4) (x)| ≤ 1 for all x, |T3 (2) − f (2)| ≤

1 (4) 23 23 | f (x)|24 ≤ < . 24 12 6

Thus, (b) is the correct answer.

Exercises In Exercises 1–14, calculate the Taylor polynomials T2 (x) and T3 (x) centered at x = a for the given function and value of a. 1. f (x) = sin x, SOLUTION

a=0

First, we calculate and evaluate the needed derivatives: f (x) = sin x

f (a) = 0

f (x) = cos x

f (a) = 1

f (x) = − sin x

f (a) = 0

f (x) = − cos x

f (a) = −1

Now, T2 (x) = f (a) + f (a)(x − a) +

f (a) 0 (x − a)2 = 0 + 1(x − 0) + (x − 0)2 = x; and 2 2

f (a) f (a) (x − a)2 + (x − a)3 T3 (x) = f (a) + f (a)(x − a) + 2 6 0 −1 1 = 0 + 1(x − 0) + (x − 0)2 + (x − 0)3 = x − x 3 . 2 6 6 1 3. f (x)f (x) = = sin ,x, a a==0 π /2 1+x SOLUTION

First, we calculate and evaluate the needed derivatives: f (x) =

1 1+x

f (a) = 1

f (x) =

−1 (1 + x)2

f (a) = −1

528

CHAPTER 9


f (x) =

2 (1 + x)3

f (a) = 2

f (x) =

−6 (1 + x)4

f (a) = −6

Now, f (a) T2 (x) = f (a) + f (a)(x − a) + (x − a)2 2 2 = 1 + (−1)(x − 0) + (x − 0)2 = 1 − x + x 2 ; and 2 f (a) f (a) T3 (x) = f (a) + f (a)(x − a) + (x − a)2 + (x − a)3 2 6 2 −6 = 1 + (−1)(x − 0) + (x − 0)2 + (x − 0)3 = 1 − x + x 2 − x 3 . 2 6 5. f (x) = tan x, a = 0 1 f (x) = , a=1 SOLUTION First, 1 +we x calculate and evaluate the needed derivatives: f (x) = tan x

f (a) = 0

f (x) = sec2 x

f (a) = 1

f (x) = 2 sec2 x tan x

f (a) = 0

f (x) = 2 sec4 x + 4 sec2 x tan2 x

f (a) = 2

Now, f (a) 0 T2 (x) = f (a) + f (a)(x − a) + (x − a)2 = 0 + 1(x − 0) + (x − 0)2 = x; and 2 2 f (a) f (a) (x − a)2 + (x − a)3 T3 (x) = f (a) + f (a)(x − a) + 2 6 0 2 1 = 0 + 1(x − 0) + (x − 0)2 + (x − 0)3 = x + x 3 . 2 6 3 1 7. f (x)f (x) = = tan2x, , aa = =0π4 1+x SOLUTION

First, we calculate and evaluate the needed derivatives: f (x) =

1 1 + x2

f (a) = 1

f (x) =

−2x (x 2 + 1)2

f (a) = 0

f (x) =

2(3x 2 − 1) x2 + 1

f (a) = −2

f (x) =

−24x(x 2 − 1) (x 2 + 1)4

f (a) = 0

Now, f (a) −2 T2 (x) = f (a) + f (a)(x − a) + (x − a)2 = 1 + 0(x − 0) + (x − 0)2 = 1 − x 2 ; and 2 2 f (a) f (a) T3 (x) = f (a) + f (a)(x − a) + (x − a)2 + (x − a)3 2 6 −2 0 = 1 + 0(x − 0) + (x − 0)2 + (x − 0)3 = 1 − x 2 . 2 6 9. f (x) = e x , a = 0 1 , a = −1 f (x) = SOLUTION First, 1 +we x 2 calculate and evaluate the needed derivatives: f (x) = e x

f (a) = 1

S E C T I O N 9.4

f (x) = e x

f (a) = 1

f (x) = e x

f (a) = 1

f (x) = e x

f (a) = 1

Taylor Polynomials

529

Now, f (a) T2 (x) = f (a) + f (a)(x − a) + (x − a)2 2 1 1 = 1 + 1(x − 0) + (x − 0)2 = 1 + x + x 2 ; and 2 2 f (a) f (a) T3 (x) = f (a) + f (a)(x − a) + (x − a)2 + (x − a)3 2 6 1 1 1 1 = 1 + 1(x − 0) + (x − 0)2 + (x − 0)3 = 1 + x + x 2 + x 3 . 2 6 2 6 11. f (x) = e−x x+ e−2x , a = 0 f (x) = e , a = ln 2 SOLUTION First, we calculate and evaluate the needed derivatives: f (x) = e−x + e−2x

f (a) = 2

f (x) = −e−x − 2e−2x

f (a) = −3

f (x) = e−x + 4e−2x

f (a) = 5

f (x) = −e−x − 8e−2x

f (a) = −9

Now, f (a) T2 (x) = f (a) + f (a)(x − a) + (x − a)2 2 5 5 = 2 + (−3)(x − 0) + (x − 0)2 = 2 − 3x + x 2 ; and 2 2 f (a) f (a) T3 (x) = f (a) + f (a)(x − a) + (x − a)2 + (x − a)3 2 6 5 −9 5 3 = 2 + (−3)(x − 0) + (x − 0)2 + (x − 0)3 = 2 − 3x + x 2 − x 3 . 2 6 2 2 13. f (x) = ln(x + 1), a = 0 f (x) = x 2 e−x , a = 1 SOLUTION First, we calculate and evaluate the needed derivatives: f (x) = ln(x + 1)

f (a) = 0

f (x) =

1 x +1

f (a) = 1

f (x) =

−1 (x + 1)2

f (a) = −1

f (x) =

2 (x + 1)3

f (a) = 2

Now, f (a) −1 1 T2 (x) = f (a) + f (a)(x − a) + (x − a)2 = 0 + 1(x − 0) + (x − 0)2 = x − x 2 ; and 2 2 2 f (a) f (a) T3 (x) = f (a) + f (a)(x − a) + (x − a)2 + (x − a)3 2 6 −1 2 1 1 = 0 + 1(x − 0) + (x − 0)2 + (x − 0)3 = x − x 2 + x 3 . 2 6 2 3 In Exercises f (x)15–18, = coshcompute x, a =T02 (x) at x = a and use a calculator to compute the error | f (x) − T2 (x)| at the given value of x. 15. y = e x ,

x = −0.5,

a=0

530

CHAPTER 9


Let f (x) = e x . Then f (x) = e x , f (x) = e x , f (a) = 1, f (a) = 1 and f (a) = 1. Therefore 1 1 T2 (x) = 1 + 1(x − 0) + (x − 0)2 = 1 + x + x 2 , 2 2

and 1 T2 (−0.5) = 1 + (−0.5) + (−0.5)2 = 0.625. 2 Using a calculator, we find 1 f (−0.5) = √ = 0.606531, e so |T2 (−0.5) − f (−0.5)| = 0.0185. 17. y = x −3/2 , x = 0.3,π a = 1 y = cos x, x = 12 , a = 0 3 15 3 SOLUTION Let f (x) = x −3/2 . Then f (x) = − 2 x −5/2 , f (x) = 4 x −7/2 , f (a) = 1, f (a) = − 2 and f (a) = 15 . Therefore 4 3 3 15/4 15 T2 (x) = 1 + − (x − 1) + (x − 1)2 = 1 − (x − 1) + (x − 1)2 , 2 2 2 8 and 15 3 T2 (0.3) = 1 − (−0.7) + (−0.7)2 = 2.96875. 2 8 Using a calculator, f (0.3) = 6.08581, so |T2 (0.3) − f (0.3)| = 3.11706. 19. Show that the nth MacLaurin polynomial for f (x) = e x is y = esin x , x = 1.5, a = π2 x2 xn x + ··· + Tn (x) = 1 + + 1! 2! n! SOLUTION

With f (x) = e x , it follows that f (n) (x) = e x and f (n) (0) = 1 for all n. Thus, xn 1 1 x2 + ··· + . Tn (x) = 1 + 1(x − 0) + (x − 0)2 + · · · + (x − 0)n = 1 + x + 2 n! 2 n!

In Exercises 21–26, find Tn (x) at x = a for all n. 1 Show that the nth Taylor polynomial for at a = 1 is x +1 1 21. f (x) = , a=0 1−x 1 (x − 1)n x − 1 (x − 1)2 1 Tn (x) = − + + · · · + (−1)n n+1 SOLUTION Let f (x) = 1−x . Then 2 4 8 2 1 f (0) = 1 = 0! f (x) = 1−x f (x) =

1 (1 − x)2

f (0) = 1 = 1!

f (x) =

2 (1 − x)3

f (0) = 2 = 2!

f (x) =

6 (1 − x)4

f (0) = 6 = 3!

.. . f (n) (x) =

.. . n! (1 − x)n+1

f (n) (0) = n!

Therefore, 2 6 n! Tn (x) = 1 + 1(x − 0) + (x − 0)2 + (x − 0)3 + · · · + (x − 0)n 2 6 n! = 1 + x + x2 + x3 + · · · + xn.

S E C T I O N 9.4

23. f (x) = e x , f (x) = SOLUTION Letx

Taylor Polynomials

531

a=1 1 (n) x (n) , a x= 4 f (x) − 1 = e . Then f (x) = e and f (1) = e for all n. Therefore, e e Tn (x) = e + e(x − 1) + (x − 1)2 + · · · + (x − 1)n . 2! n!

25. f (x) = x 5/2√, a = 2 f (x) = x, a = 1 SOLUTION Let f (x) = x 5/2 . Then f (x) = x 5/2

f (2) = 25/2

5 3/2 x 2 15 1/2 f (x) = x 4 15 −1/2 f (x) = x 8 f (x) =

As we continue taking derivatives, a pattern emerges: 15 1 − x −3/2 f (4) (x) = 8 2 15 1 3 f (5) (x) = − − x −5/2 8 2 2 15 1 3 5 f (6) (x) = − − − x −7/2 8 2 2 2

5 3/2 2 2 15 1/2 f (2) = 2 4 15 −1/2 f (x) = 2 8 f (2) =

15 1 −3/2 − 2 8 2 15 1 3 −5/2 f (5) (2) = − − 2 8 2 2 15 1 3 5 −7/2 f (6) (2) = − − − 2 8 2 2 2 f (4) (2) =

.. .

.. .

Thus, for n ≥ 4, f (n) (2) =

1 15 (−1)n−3 n−3 (1 · 3 · 5 · · · · · (2n − 7)) 2−(2n−5)/2 . 8 2

To express 1 · 3 · · · · · (2n − 7) in closed form, we note that the product of the even numbers 2 · 4 · 6 · 2n − 8 is 2n−4 (n − 4)! so that the product of the odd numbers is (2n − 7)! 1 · 3 · · · · · (2n − 7) = n−4 2 (n − 4)! Therefore, for n ≥ 4, f n (2) = 15(−1)n−3

(2n − 7)! (13−6n)/2 2 . (n − 4)!

Finally, √ √ 15(x − 2)2 (2n − 7)! (13−6n)/2 5(x − 2)3 (x − 2)n . Tn (x) = 4 2 + 5 2(x − 2) + 2 √ + √ + · · · + 15(−1)n−3 n!(n − 4)! 4 2 16 2 x 27. y= f (x) Plot = cos x, e atogether = π /4 with the MacLaurin polynomials Tn (x) for n = 1, 3, 5 and then for n = 2, 4, 6 on the interval [−3, 3]. What difference do you notice between the even and odd MacLaurin polynomials?

The odd polynomials are concave down on the left and lie beneath the graph of y = e x ; the even polynomials are concave up on the left and lie above the graph of y = e x . SOLUTION

y

y

n odd

−2

n even

−1

15

15

10

10

5

5 x 1

2

−2

−1

x 1

2

29. Use the Error Bound to find the maximum possible size of |cos 0.3 − T5 (0.3)|, where T5 (x) is the MacLaurin poly1 nomial. Verify withtogether a calculator. Plot your f (x) result = with the Taylor polynomials Tn (x) at a = 1 for 1 ≤ n ≤ 4 on the interval [−2, 8] 1+x (be sure to limit the upper plot range). (a) Over which interval does T4 appear to give a close approximation to f (x)?

532

CHAPTER 9


Using the Error Bound, we have | cos 0.3 − T5 (0.3)| ≤

K |0.3|6 , 6!

where K is a number such that | f (6) (x)| ≤ K for x between 0 and 0.3. With f (x) = cos x, f (6) (x) = − cos x, so | f (6) (x)| ≤ 1 for all x. We can therefore take K = 1 and then | cos 0.3 − T5 (0.3)| ≤

|0.3|6 = 1.0125 × 10−6 . 6!

Now, with f (x) = cos x, T5 (x) = 1 −

1 2 1 4 x + x , 2 24

and 1 1 T5 (0.3) = 1 − (0.3)2 + (0.3)4 = 0.9553375. 2 24 Using a calculator, we find cos 0.3 ≈ 0.955336489, so | cos 0.3 − T5 (0.3)| = 1.010874 × 10−6 . 31. Let Tn be the MacLaurin polynomial of f (x) = e x and let c > 0. Show that we may take K = ec in the Error 0.1 possible −5 size of Calculate T4 (x) at Then a = use 1 for (x) =Bound x 11/2toand use theaError Bound to that find |T the maximum Bound for |Tn (c) − f (c)|. thef Error determine number n such n (0.1) − e | ≤ 10 . (5) (x)| on [1, c] with c > 1 is f (5) (c). Show that the max | f |T4 (1.2) − f (1.2)|. Hint: SOLUTION Let f (x) = e x . We know that f (n+1) (x) = e x is always increasing, and, because c > 0, we can take K = ec in the error bound. Thus, c n+1 Tn (x) − e x ≤ e · x (n + 1)!

e0.1 · (0.1)n+1 . Tn (0.1) − e0.1 ≤ (n + 1)!

so

By trial and error, we find e0.1 · (0.1)3 ≈ 1.842 × 10−4 > 10−5 T2 (0.1) − e0.1 ≤ 3! but e0.1 · (0.1)4 ≈ 4.605 × 10−6 < 10−5 . T3 (0.1) − e0.1 ≤ 4! Thus, n = 3. 33. Calculate T3 (x)√at a = 0 for f (x) = tan−1 x. Then compute T3 ( 12 ) and use the Error Bound to find a bound for Let f (x) = 1 + x and let Tn (x) be the Taylor polynomial centered at a = 8. |T3 ( 12 ) − tan−1 ( 12 )|. Refer to the graph in Figure 4 to find an acceptable value of K . (a) Find T3 (x) and calculate T3 (8.02). √ − 9.02|. (b) Use the Error Bound to find a bound for |T3 (8.02) y 5 4 3 2 1 x −1

1

2

3

FIGURE 4 Graph of f (4) (x) = −24x(x 2 − 1)/(x 2 + 1)4 , where f (x) = tan−1 x. SOLUTION

Let f (x) = tan−1 x. Then f (x) = tan−1 x

f (0) = 0

f (x) =

1 1 + x2

f (0) = 1

f (x) =

−2x (1 + x 2 )2

f (0) = 0

f (x) =

(1 + x 2 )2 (−2) − (−2x)(2)(1 + x 2 )(2x) (1 + x 2 )4

f (0) = −2

S E C T I O N 9.4

Taylor Polynomials

533

and 0 −2 x3 (x − 0)3 = x − . T3 (x) = 0 + 1(x − 0) + (x − 0)2 + 2 6 3 Since f (4) (x) ≤ 5 for x ≥ 0, we may take K = 5 in the error bound; then, 4 T3 1 − tan−1 1 ≤ 5(1/2) = 5 . 2 2 4! 384 35. Show that the MacLaurin polynomials for f (x) = sin x are Calculate T2 (x) for f (x) = sech x at a = 0. Then compute T2 ( 12 ) and use the Error Bound to find a bound for 3 5 2n−1 n−1 x x + x − T2n−1 (x) = T2n = x − 3! 1 5! · ·1· + (−1) (2n − 1)! . T2 − f 2 2 Use the Error Bound with n = 4 to show that Hint: Plot f (x) to find an acceptable value of K .x 3 |x|5 (for all x) sin x − x − ≤ 6 120 SOLUTION

Let f (x) = sin x. Then f (x) = sin x

f (0) = 0

f (x) = cos x

f (0) = 1

f (x) = − sin x

f (0) = 0

f (x) = − cos x

f (0) = −1

f (4) (x) = sin x

f (4) (0) = 0

f (5) (x) = cos x

f (5) (0) = 1

.. .

.. .

Consequently, T2n−1 (x) = x −

x3 x 2n−1 x5 + + · · · + (−1)n+1 3! 5! (2n − 1)!

and T2n (x) = x − With n = 4 and K = 1,

x3 x 2n−1 x5 + + · · · + (−1)n+1 + 0 = T2n−1 (x). 3! 5! (2n − 1)! |x|5 x 3 |x|5 ≤ . sin x − x − ≤K 6 5! 5!

37. Find n such that |Tn (1.3) − ln(1.3)| ≤ 10−4 , where Tn is the Taylor polynomial for f (x) = ln x at a = 1. Find n such that |Tn (0.1) − cos(0.1)| ≤ 10−7 , where T−2 MacLaurin polynomial for f (x) = cos x (see n is the −1 −3 , f (4) (x) = −6x −4 , etc. In general, SOLUTION LetCalculate f (x) = ln|Tx.(0.1) Then−f cos(0.1)| (x) = x for, this f (x) = −x , f verify (x) =the 2x bound Example 5). value of n and on the error. n f (n) (x) = (−1)n+1 (n − 1)!x −n . Now, | f (n+1) (x)| is decreasing on the interval [1, 1.3], so | f (n+1) (x)| ≤ | f (n+1) (1)| = n! for all x ∈ [1, 1.3]. We can therefore take K = n! in the error bound, and |Tn (1.3) − ln(1.3)| ≤ n!

|1.3 − 1|n+1 (0.3)n+1 = . (n + 1)! n+1

With n = 5, (0.3)6 = 1.215 × 10−4 > 10−4 , 6 but with n = 6, (0.3)7 = 3.124 × 10−5 < 10−4 . 7 Therefore, the error is guaranteed to be below 10−4 for n = 6.

534

CHAPTER 9


√ √ 10−6 , where Tn (x) is the Taylor polynomial for f (x) = x x at a = 1. 39. Find n such that |Tn (1.3) − 1.3| ≤ −6 Find n such that |Tn (1) − e| ≤ 10 , where Tn is the MacLaurin polynomial for f (x) = e . SOLUTION Using the Error Bound, we have |Tn (1.3) −

√

1.3| ≤ K

|1.3 − 1|n+1 |0.3|n+1 =K , (n + 1)! (n + 1)!

√ where K is a number such that | f (n+1) (x)| ≤ K for x between 1 and 1.3. For f (x) = x, | f (n) (x)| is decreasing for x > 1, hence the maximum value of | f (n+1) (x)| occurs at x = 1. We may therefore take 1 · 3 · 5 · · · (2n + 1) 2n+1 1 · 3 · 5 · · · (2n + 1) 2 · 4 · 6 · · · (2n + 2) (2n + 2)! = · . = 2 · 4 · 6 · · · (2n + 2) 2n+1 (n + 1)!22n+2

K = | f (n+1) (1)| =

Then |Tn (1.3) −

√

1.3| ≤

(2n + 2)! |.3|n+1 (2n + 2)! · (0.075)n+1 . = 2n+2 (n + 1)! (n + 1)!2 [(n + 1)!]2

With n = 9 (20)! (0.075)10 = 1.040 × 10−6 > 10−6 , [(10)!]2 but with n = 10 (22)! (0.075)11 = 2.979 × 10−7 < 10−6 . [(11)!]2 Hence, n = 10 will guarantee the desired accuracy. 41. Let n ≥ 1. Show that if |x| is small, then Let Tn (x) be the Taylor polynomial for f (x) = ln x at a = 1 and let c > 1. (k+1) 1 − n Example (x)+on1)[1, (see 4). (a) Show that the maximum of f (k+1)(x 1/nc]≈is1 f+ x +(1) x2 2 n+1 n 2n |c − 1| (b) Prove |Tn (c) − ln c| ≤ . 1 Use this approximation with n = 6nto+estimate 1.51/6 . (c) Find n such that |Tn (1.5) − ln 1.5| ≤ 10−2 . SOLUTION Let f (x) = (x + 1)1/n . Then f (x) = (x + 1)1/n

f (0) = 1

1 (x + 1)1/n−1 n 1 1 − 1 (x + 1)1/n−2 f (x) = n n f (x) =

f (0) =

1 n

f (0) =

1 n

1 −1 n

and T2 (x) = 1 +

1 (x) + n

1 1 − n n2

2 x 1−n x x 2. =1+ + 2 n 2n 2

With n = 6 and x = 0.5, 1.51/6 ≈ T2 (0.5) =

307 ≈ 1.065972. 288

43. Find the fourth MacLaurin polynomial for f (x) = sin x cos x xby multiplying the fourth MacLaurin polynomials for thatfthe polynomial for f (x) = e sin x is equal to the product of the third MacLaurin f (x) =Verify sin x and (x)third = cosMacLaurin x. polynomials of e x and sin x (after discarding terms of degree greater than 3 in the product). 3 SOLUTION The fourth MacLaurin polynomial for sin x is x − x6 , and the fourth MacLaurin polynomial for cos x is 2

4

x . Multiplying these two polynomials, and then discarding terms of degree greater than 4, we find that the 1 − x2 + 24 fourth MacLaurin polynomial for f (x) = sin x cos x is

T4 (x) = x −

2x 3 . 3

1 1 2 ).2You 45. FindFind the the MacLaurin polynomials of Tn (x)2for by fsubstituting for xmay in the of to the sum (see (x) = cos(x−x useMacLaurin the fact thatpolynomials Tn (x) is equal MacLaurin polynomials 1 − x 1+x of the21). terms up to degree n obtained by substituting x 2 for x in the nth MacLaurin polynomial of cos x. Exercise


Taylor Polynomials

535

1 are of the form The MacLaurin polynomials for 1−x

Tn (x) = 1 + x + x 2 + · · · + x n . Accordingly, the MacLaurin polynomials for

1 are of the form 1+x 2

T2n (x) = 1 − x 2 + x 4 − x 6 + · · · + (−x 2 )n . 47. Let f (x) = 3x 3 + 2x 2 − x − 4. Calculate T j (x) for j = 1, 2, 3, 4, 5 at both a = 0 and 8b −aa = 1. Show that for the length θ seventeenth-century Dutch scientist Christian Huygens used the approximation θ ≈ T3 (x) =The f (x) in both cases. 3 3 2 of a circular of the unit+circle, is the length of the chord AC of angle θ and b is length of the chord AB SOLUTION Letarc f (x) = 3x 2x −where x − 4.a Then of angle θ /2 (Figure 5). 3 2 f (x) = 3x x− 4 and show f (0) = −4the Huygens f (1) = 0 θ /2) and + b 2x = 2−sin( θ /4), that approximation amounts to the (a) Prove that a = 2 sin( approximation 2 f (x) = 9x + 4x − 1 f (0) = −1 f (1) = 12 16 θ 2 θ =4 f (1) = 22 f (x) = 18x + 4 θ≈ sin f−(0)sin 3 4 3 2 f (0) = 18 f (1) = 18 f (x) = 18 (b) Compute the fifth MacLaurin polynomial of the function on the right. (x)show = 0 that the error in the Huygens f (4) (0) =0 f (4) 0 0.00022|θ |5 . f (4)to (c) Use the Error Bound approximation is (1) less= than f (5) (x) = 0

f (5) (0) = 0

f (5) (1) = 0

At a = 0, T1 (x) = −4 − x; T2 (x) = −4 − x + 2x 2 ; T3 (x) = −4 − x + 2x 2 + 3x 3 = f (x); T4 (x) = T3 (x); and T5 (x) = T3 (x). At a = 1, T1 (x) = 12(x − 1); T2 (x) = 12(x − 1) + 11(x − 1)2 ; T3 (x) = 12(x − 1) + 11(x − 1)2 + 3(x − 1)3 = −4 − x + 2x 2 + 3x 3 = f (x); T4 (x) = T3 (x); and T5 (x) = T3 (x). e x + e−x − 2 nthbeTaylor polynomial atoxpital’s = aRule for atopolynomial f (x)Consider of degreeL n. Let polynomials Tn (x) be thecan 49. Taylor used instead of L’Hˆ evaluate limits. = Based lim on the result. of 1 − cos x x→0 Bound. Exercise 47, guess the value of | f (x) − Tn (x)|. Prove that your guess is correct using the Error (a) Show that the second MacLaurin polynomial for f (x) = e x + e−x − 2 is T2 (x) = x 2 . Use the Error Bound with n = 2 to show that e x + e−x − 2 = x 2 + g1 (x) where lim

x→0

g1 (x) = 0. Similarly, prove that x2 1 − cos x =

g2 (x) = 0. x→0 x 2

1 2 x + g2 (x) 2

where lim

g (x) 1+ 12 x (b) Evaluate L by using (a) to show that L = lim . g (x) x→0 1 + 22 2 x (c) Evaluate L again using L’Hôpital’s Rule.

536

CHAPTER 9


(a) Let f (x) = e x + e−x − 2. Then f (x) = e x − e−x , f (x) = e x + e−x , f (0) = 0, f (x) = 0 and f (0) = 2. Thus, T2 (x) = x 2 . Let g1 (x) = f (x) − T2 (x) = e x + e−x − 2 − x 2 . By the Error Bound, for |x| < 1, |g1 (x)| ≤

K 1 |x|3 , 3!

where K 1 = max{ f (x) : |x| ≤ 1}, so K |x| |g1 (x)| ≤ 1 →0 2 3! x as x → 0. Likewise, 1 − cos x has MacLaurin polynomial

x2 T2 (x) = 1 − 1 − 2

=

x2 , 2

so, for g2 (x) = 1 − cos x − 12 x 2 , K |x| |g2 (x)| ≤ 2 →0 3! x2 as x → 0. Here, we may take K 2 = 1. (b) By substitution x 2 + g1 (x) 1 + g1 (x)/x 2 = lim = 2. x→0 x 2 /2 + g2 (x) x→0 1/2 + g2 (x)/x 2

L = lim (c) Using L’Hôpital’s Rule

e x + e−x − 2 e x − e−x e x + e−x = lim = lim = 2. sin x x→0 1 − cos x x→0 x→0 cos x

L = lim

51. A light wave of wavelength λ travels from A to B by passing through an aperture (see Figure 6 for the notation). Use the method of Exercise 49 to evaluate The aperture (circular region) is located in a plane that is perpendicular to AB. Let f (r ) = d + h , that is, f (r ) is 1 −Fresnel cos x zones, used to determine sin x − x the optical disturbance at B, are the the distance AC + C B as a function of r . The and lim . lim ) = AB + concentric bands bounded by the circlesx→0 of radiusx 2Rn such that f (Rn x→0 x 3 n λ /2 = d + h + n λ /2. (a) Show that f (r ) = d 2 + r 2 + h 2 + r 2 , and use the MacLaurin polynomial of order 2 to show 1 1 1 2 f (r ) ≈ d + h + + r 2 d h √ (b) Deduce that Rn ≈ n λ L, where L = (d −1 + h −1 )−1 . (c) Estimate the radii R1 and R100 for blue light (λ = 475 × 10−7 cm) if d = h = 100 cm.

C d' A

r

d

h'

O h

R1 R2

B

R3

FIGURE 6 The Fresnel zones are the regions between the circles of radius Rn . SOLUTION

(a) From the diagram, we see that AC = Moreover,

d 2 + r 2 and C B = h 2 + r 2 . Therefore, f (r ) = d 2 + r 2 + h 2 + r 2 .

r r f (r ) = + , 2 2 2 d +r h + r2

f (r ) =

f (0) = d + h, f (0) = 0 and f (0) = d −1 + h −1 . Thus, f (r ) ≈ T2 (r ) = d + h +

1 2

d2 h2 + , (d 2 + r 2 )3/2 (h 2 + r 2 )3/2

1 1 2 + r . d h

Taylor Polynomials

S E C T I O N 9.4

537

(b) Solving 1 f (Rn ) ≈ d + h + 2

1 1 + d h

Rn2 = d + h +

nλ 2

yields Rn =

√ n λ (d −1 + h −1 )−1 = n λ L,

where L = (d −1 + h −1 )−1 . (c) With d = h = 100 cm, L = 50 cm. Taking λ = 475 × 10−7 cm, it follows that √ R1 ≈ λ L = 0.04873 cm; and √ R100 ≈ 100λ L = 0.4873 cm.

Further Insights and Challenges 53. UseShow Taylor’s to show that if f (n+1)of (t)f (x) ≥ 0=forarcsin all t,x then that Theorem the nth MacLaurin polynomial for nthe oddnth is MacLaurin polynomial Tn (x) satisfies Tn (x) ≤ f (x) for all x ≥ 0. 1 x3 1 · 3 x5 1 · 3 · 5 x7 1 · 3 · 5 · · · (n − 2) x n SOLUTION From Taylor’s Theorem, + + + ··· + Tn (x) = x + 2 3 2·4 5 2 · 4"· 6 7 2 · 4 · 6 · · · (n − 1) n 1 x n (n+1) Rn (x) = f (x) − Tn (x) = (x − u) f (u) du. n! 0 If f (n+1) (t) ≥ 0 for all t then " 1 x (x − u)n f (n+1) (u) du ≥ 0 n! 0 since (x − u)n ≥ 0 for 0 ≤ u ≤ x. Thus, f (x) − Tn (x) ≥ 0, or f (x) ≥ Tn (x). 55. This exercise is intended to reinforce the proof of Taylor’s Theorem. Use Exercise 53 to show"that x for x ≥ 0 and all n, (a) Show that f (x) = T0 (x) + f (u) du. xn x2 a +···+ ex ≥ 1 + x + (b) Use Integration by Parts to prove the formula 2! n! " x " x x on the same coordinate axes. Does this inequality remain true for x < 0? Sketch the graphs of e , T1 (x), and T2 (x) (x − u) f (2) (u) du = − f (a)(x − a) + f (u) du a

a

(c) Prove the case n = 2 of Taylor’s Theorem: f (x) = T1 (x) +

" x a

(x − u) f (2) (u) du.

SOLUTION

(a) T0 (x) +

" x a

f (u) du = T0 (x) + f (x) − f (a) (from FTC2) = f (a) + f (x) − f (a) = f (x).

(b) Using Integration by Parts with w = x − u and v = f (u) du, x " x " x (x − u) f (u) du = f (u)(x − u) + f (u) du a

a

a

= f (x)(x − x) − f (a)(x − a) + = − f (a)(x − a) +

" x a

" x a

f (u) du

f (u) du.

(c) T1 (x) +

" x a

(x − u) f (u) du = f (a) + f (a)(x − a) + − f (a)(x − a) + = f (a) + f (x) − f (a) = f (x).

" x a

f (u) du

538

CHAPTER 9


" 1/2 2 ) obtain Integrals Using Taylor Taylor can be used numerical 57. UseApproximating the fourth MacLaurin polynomial and the Polynomials method of Exercise 56 polynomials to approximate sin(xto d x. Find a bound 0 approximations to integrals. for the error. (a) Let L > 0. Show that if two functions f (x) and g(x) satisfy | f (x) − g(x)| < L for all x ∈ [a, b], then SOLUTION The fourth MacLaurin polynomial for sin(x 2 ) is x 2 . Thus, by Exercise 56, " b " 1/2 " 1/2 f (x) − g(x) d x < L(b −1a) asin(x 2 ) d x ≈ x2 dx = . 24 0 0 (b) Show that |T3 (x) − sin x| ≤ ( 12 )5 /5! for all x ∈ [0, 12 ]. On [0, 1/2], " 1/2 " 1/2 (c) Evaluate T3 (x) d x as an approximation to sin x d x.6 Use (a) and (b) to find a bound for the size of (1/2) 0 0 sin(x 2 ) − T2 (x) , ≤ the error. 6 so the error in the approximation is bounded by (1/2)6 6

1 − 0 ≈ 1.302 × 10−3 . 2

59. Prove by induction that for all k, The Second Derivative Test forlocal extrema fails when f (a) = 0. This exercise shows us how to extend the test under the following assumption: dj (x − a)k k(k − 1) · · · (k − j + 1)(x − a)k− j = j d x f (a) =k! f (a) = f (a) = 0 butk! f (4) (a) > 0

(4)k(a)(x j +(x1−f a) − a)4 . 1 for k = j (a) Show that T4 (x) = fd(a) 24 = j for x k! = a. x=a (b) Show that T4 (x) > df x(a) 0 for k = j (c) By the Error Bound, there is a constant C such that |T4 (x) − f (x)| ≤ C|x − a|5 for all x in an interval containing agrees (x)f at = ax to order n. close to but not equal to a. Use a. thisUse to prove that this and (b)Tnto(x) show thatwith f (x)f > (a)x for sufficiently SOLUTION The first clearly true forAj similar = 0. Suppose theshows formula arbitrary j. Then (d) Conclude that formula f (a) is aislocal minimum. argument thatis ftrue (a) for is aan local maximum if f (4) (a) < 0. (x − a)k d dj d k(k − 1) · · · (k − j + 1)(x − a)k− j d j+1 (x − a)k = = k! dx dx j k! dx k! d x j+1 =

k(k − 1) · · · (k − j + 1)(k − ( j + 1) + 1)(x − a)k−( j+1) k!

as desired. Note that if k = j, then the numerator is k!, the denominator is k! and the value of the derivative is 1; otherwise, the value of the derivative is 0 at x = a. In other words, (x − a)k dj 1 for k = j = j k! dx 0 for k = j x=a Applying this latter formula, it follows that n j (k) (a) # f d dj k (x − a) Tn (a) = = f ( j) (a) j j k! dx d x x=a x=a k=0 as required. The following equation arises in the description of Bose–Einstein condensation (the quantum theory of gases cooled to near absolute zero):

CHAPTER REVIEW EXERCISES

2 " 4 A ∞ x 2 e−x d x A = √ 0 In Exercises 1–4, calculate the arc length over the given interval. π 0 1 − Ae−x 2

x −3to derive an approximate expression for A0 in terms of A for |A| small. x5 It is necessary + , [3, 5] 1. y = 2 10 6 x 2 e−x (a) Show that the second MacLaurin polynomial for the function f ( A) = (where A is the variable and 2 x5 x −3 1 − Ae−x SOLUTION Let y = + . Then x is treated as a constant) is6 10 2 2 2 2 2 T22 ( A) = x 2xe4−x x+−4 x 2 e−2x A +xx82 e−3x 1 A x −8 =1+ − − + 1 + (y ) ="1 + 2 4 2 4 4A ∞ 2 (b) Use the approximation A0 ≈ √ T2 ( A) d x to show 2 π 08 1 x −8 x4 x −4 x + + = 1= 2 2 +1 2 3 . 4A ≈ 2 A +4 √ A + √ A 0 3 3 2 2 You may use the formula (valid for λ > 0)


539

Because 12 (x 4 + x −4 ) > 0 on [3, 5], the arc length is " 5 " 5 4 −4 5 −3 5 x x 2918074 x x + − . s= 1 + (y )2 d x = dx = = 2 2 10 6 10125 3 3 3

3. y = 4x − 2, [−2, 2] y = e x/2 + e−x/2 , [0, 2] SOLUTION Let y = 4x − 2. Then

√ 2 1 + y = 1 + 42 = 17. Hence, s=

" 2 √ −2

√ 17 d x = 4 17.

√ √ √ √ 5. Show that2/3 the arc length of y = 2 x over [0, a] is equal to a(a + 1) + ln( a + a + 1). Hint: Apply the , [1, 2] y = x 2 substitution x = tan θ to the arc length integral. √ 1 SOLUTION Let y = 2 x. Then y = √ , and x

2 1 x +1 1 √ 1+ y = 1+ = x + 1. = √ x x x Thus, s=

" a 1 √ 1 + x d x. √ x 0

We make the substitution x = tan2 θ , d x = 2 tan θ sec2 θ d θ . Then " x=a " x=a 1 sec θ · 2 tan θ sec2 θ d θ = 2 sec3 θ d θ . s= x=0 tan θ x=0 We use a reduction formula to obtain x=a √ √ √ √ a tan θ sec θ 1 s=2 = ( x 1 + x + ln | 1 + x + x|) + ln | sec θ + tan θ | 2 2 x=0 0

√ √ √ √ √ √ = a 1 + a + ln | 1 + a + a| = a(a + 1) + ln a+ a+1 . In Exercises 7–10, calculate the surface area of the solid obtained by rotating the curve over the given interval about the Find a numerical approximation to the arc length of y = tan x over [0, π /4]. x-axis. 7. y = x + 1, SOLUTION

[0, 4]

Let y = x + 1. Then y = 1, and

√ √ y 1 + y 2 = (x + 1) 1 + 1 = 2(x + 1).

Thus, S A = 2π

" 4√ 0

√ 2(x + 1) d x = 2 2π

x2 +x 2

4 √ = 24 2π . 0

2 1 9. y = x 3/2 2 − x 1/2 2 , [1, 2] y 3= x 3/4 2− x 5/4 , [0, 1] 3 5 2 1 SOLUTION Let y = x 3/2 − x 1/2 . Then 3 2 y =

√

1 x− √ , 4 x

and 2 2 2 √ √ 1 1 1 1 1 1 x− √ =1+ x − + x+ √ . =x+ + = 1 + y = 1 + 2 16x 2 16x 4 x 4 x

540

CHAPTER 9


Because

√

x + √1 ≥ 0, the surface area is x

√ " b " 2 2 √ 1 x 2 3/2 x 2π y 1 + y d x = 2π − x+ √ dx 3 2 4 x a 1 " 2 2 2x 3 67 1 1 x2 1 2 2 1 π. x + x− x− d x = 2π − − x = = 2π 3 6 2 8 9 6 8 36 1 1 11. Compute the total surface area of the coin obtained by rotating the region in Figure 1 about the x-axis. The top and 1 bottom parts y = ofx 2the , region [0, 2] are semicircles with a radius of 1 mm. 2 y 1 mm

4 mm x

FIGURE 1 SOLUTION

The generating half circle of the edge is y = 2 +

1 − x 2 . Then,

−2x −x y = = , 2 2 1−x 1 − x2 and 1 + (y )2 = 1 +

x2 1 = . 1 − x2 1 − x2

The surface area of the edge of the coin is " 1 " 1

2 1 y 1 + y d x = 2π dx 2π 2 + 1 − x2 −1 −1 1 − x2 " " 1 1 dx 1 − x2 + dx = 2π 2 −1 1 − x 2 −1 1 − x 2 " 1 1 dx = 2π 2 arcsin x|−1 + −1

= 2π (2π + 2) = 4π 2 + 4π . We now add the surface area of the two sides of the disk, which are circles of radius 2. Hence the surface area of the coin is:

4π 2 + 4π + 2π · 22 = 4π 2 + 12π . 13. Calculate the fluid force on the side of a right triangle of height 3 ft and base 2 ft submerged in water vertically, with Calculate the fluid force on the side of a right triangle of height 3 ft and base 2 ft submerged in water vertically, its upper vertex located at a depth of 4 ft. with its upper vertex at the surface of the water. SOLUTION We need to find an expression for the horizontal width f (y) at depth y.

4 y y–4 3 f( y)

2


541

By similar triangles we have: 2 f (y) = y−4 3

f (y) =

so

2(y − 4) . 3

Hence, the force on the side of the triangle is 7 " 2w 7 2 2w y 3 2 y f (y) d y = F =w − 2y = 18w. y − 4y d y = 3 4 3 3 4 4 " 7

The density of water is w = 62.5 lb/ft2 , hence, F = 18 · 62.5 = 1125 lb. 15. Calculate the moments and COM of the lamina occupying the region under y = x(4 − x) for 0 ≤ x ≤ 4, assuming 2. a horizontal oil tank is an ellipse (Figure 2) with equation (x/4)2 + (y/3)2 = 1 (length in feet). = of 5 lb/ft a densityThe of ρend Assume that the tank is filled with oil of density 55 lb/ft3 . SOLUTION Because the lamina is symmetric with respect to the vertical line x = 2, by the symmetry principle, we total force F on the end of the tank when the tank is full. know(a) thatCalculate xcm = 2.the Now, (b) Would you expect the total force on the lower half of the tank than, or equal to 12 F? to be greater than, less " 4 " 4 4 5 16 (or 1 expectation. 256 ρ the force Explain. Then compute on the5 lower 2half exactly and confirm refute) your f (x)2 d x = x (4 − x)2 d x = Mx = x 3 − 2x 4 + x 5 = . 2 0 2 0 2 3 5 3 0 Moreover, the mass of the lamina is 4 " 4 " 4 1 160 f (x) d x = 5 x(4 − x) d x = 5 2x 2 − x 3 = . M=ρ 3 3 0 0 0 Thus, the coordinates of the center of mass are 2,

256/3 160/3

8 = 2, . 5

17. Find the centroid of the region between the −1 semicircle y = 1 − x 2 and the top half of the ellipse y = 12 1 − x 2 Sketch the region between y = 4(x + 1) and y = 1 for 0 ≤ x ≤ 3 and find its centroid. (Figure 3). SOLUTION

calculate

Since the region is symmetric with respect to the y-axis, the centroid lies on the y-axis. To find ycm we ⎡ 2 ⎤ " 2 2 1 1 ⎣ 1 − x ⎦ dx Mx = 1 − x2 − 2 −1 2 =

1 " 3 1 1 1 3

1 1 − x2 dx = x − x 3 = . 2 −1 4 8 3 2 −1

The area of the lamina is π2 − π4 = π4 , so the coordinates of the centroid are 2 1/2 = 0, . 0, π /4 π 2 − 2 and on the right by a semicircle 19. FindAthe centroid the shaded region region in Figure 4 bounded the left by = 2y Calculate plate in the of shape of the shaded in Figure 3 is on submerged inxwater. the fluid pressure on a side of radius Hint:ifUse symmetry andisadditivity of the1.plate the water surface y = 1. of moments. y x = ± 1 − y/2

semicircle of radius 1 x 1

FIGURE 4 SOLUTION The region is symmetric with respect to the x-axis, hence the centroid lies on the x-axis; that is, ycm = 0. To compute the area and the moment with respect to the y-axis, we treat the left side and the right side of the region separately. Starting with the left side, we find " 0 " 0 x x M yleft = 2 x + 1 dx and Aleft = 2 + 1 d x. 2 −2 −2 2

542

CHAPTER 9


In each integral we make the substitution u = x2 + 1, du = 12 d x, and find M yleft = 8

" 1 0

(u − 1)u 1/2 du = 8

" 1

2 5/2 2 3/2 1 32 3/2 1/2 u du = 8 −u − u u = − 15 5 3 0 0

and Aleft = 4

" 1 0

u 1/2 du =

8 3/2 1 8 u = . 3 3 0

On the right side of the region right =2 My

1 " 1 2 2 2 3/2 2 x 1 − x d x = − (1 − x ) = , 3 3 0 0

and Aright = π2 (because the right side of the region is one-half of a circle of radius 1). Thus, 32 2 22 + =− ; 15 3 15 16 + 3π 8 π A = Aleft + Aright = + = ; 3 2 6 right

M y = M yleft + M y

and the coordinates of the centroid are

=−

44 −22/15 ,0 . ,0 = − (16 + 3π )/6 80 + 15π

In Exercises 20–25, find the Taylor polynomial at x = a for the given function. 21. f (x) = 3(x + 2)3 − 5(x + 2), T3 (x), a = −2 f (x) = x 3 , T3 (x), a = 1 SOLUTION T3 (x) is the Taylor polynomial of f consisting of powers of (x + 2) up to three. Since f (x) is already in this form we conclude that T3 (x) = f (x). 23. f (x)f (x) = (3x 2)1/3 , T T(x), 3 (x), a a==1 2 = x+ln(x), 4 SOLUTION We start by computing the first three derivatives of f (x) = (3x + 2)1/3 : 1 (3x + 2)−2/3 · 3 = (3x + 2)−2/3 3 2 f (x) = − (3x + 2)−5/3 · 3 = −2(3x + 2)−5/3 3 10 f (x) = (3x + 2)−8/3 · 3 = 10(3x + 2)−8/3 3 f (x) =

Evaluating the function and its derivatives at x = 2, we find f (2) = 2, f (2) =

1 1 5 , f (2) = − , f (2) = . 4 16 128

Therefore, f (2) f (2) (x − 2)2 + (x − 2)3 T3 (x) = f (2) + f (2)(x − 2) + 2! 3! 1 5/128 −1/16 = 2 + (x − 2) + (x − 2)2 + (x − 2)3 4 2! 3! 1 1 5 (x − 2)3 . = 2 + (x − 2) − (x − 2)2 − 4 32 768 25. f (x) = ln(cos x),2 T3 (x), a = 0 f (x) = xe−x , T4 (x), a = 0 SOLUTION We start by computing the first three derivatives of f (x) = ln(cos x): f (x) = −

sin x = − tan x cos x

f (x) = −sec2 x f (x) = −2 sec2 x tan x


543

Evaluating the function and its derivatives at x = 0, we find f (0) = 0, f (0) = 0, f (0) = −1, f (0) = 0. Therefore, f (0) 2 f (0) 3 0 0 x2 1 x + x = 0 + x − x2 + x3 = − . T3 (x) = f (0) + f (0)x + 2! 3! 1! 2! 3! 2 √ 27. Use the fifth MacLaurin polynomial of f (x) = e x to approximate e. Use a calculator to determine the error. 3x Find the nth MacLaurin polynomial for f (x) = e . SOLUTION Let f (x) = e x . Then f (n) (x) = e x and f (n) (0) = 1 for all n. Hence, f (0) 2 f (0) 3 f (4) (0) 4 f (5) (0) 5 x + x + x + x T5 (x) = f (0) + f (0)x + 2! 3! 4! 5! =1+x +

x2 x3 x4 x5 + + + . 2! 3! 4! 5!

For x = 12 we have 1 1 =1+ + T5 2 2 =1+ Using a calculator, we find that

√

2 1 2

2!

3 +

1 2

3!

4 +

1 2

4!

5 +

1 2

5!

1 1 1 1 1 + + + + = 1.648697917 2 8 48 384 3840

e = 1.648721271. The error in the Taylor polynomial approximation is |1.648697917 − 1.648721271| = 2.335 × 10−5 .

√ 29. Let T4 (x) be the Taylor polynomial for f (x) = x at a = 16. Use the Error Bound to find the maximum possible −1 x at a = 1 to approximate f (1.1). Use a calculator to determine Use the third Taylor polynomial of f (x) = tan size of | f (17) − T4 (17)|. the error. SOLUTION Using the Error Bound, we have | f (17) − T4 (17)| ≤ K

(17 − 16)5 K = , 5! 5!

√ where K is a number such that f (5) (x) ≤ K for all 16 ≤ x ≤ 17. Starting from f (x) = x we find f (x) =

1 −1/2 1 3 15 x , f (x) = − x −3/2 , f (x) = x −5/2 , f (4) (x) = − x −7/2 , 2 4 8 16

and f (5) (x) =

105 −9/2 . x 32

For 16 ≤ x ≤ 17, (5) f (x) =

105 105 105 ≤ = . 9/2 8388608 32x 9/2 32 · 16

Therefore, we may take K =

105 . 8388608

Finally, | f (17) − T4 (17)| ≤

105 1 · ≈ 1.044 · 10−7 . 8388608 5!

31. Let T4 (x) be the Taylor polynomial for f (x) = x ln x at x = 1 computed in Exercise 22. Use the Error Bound to Find nforsuch that |T − e| < 10−8 , where Tn (x) is the nth MacLaurin polynomial for f (x) = e x . find a bound | f (1.2) −nT(1) 4 (1.2)|.

544

CHAPTER 9


Using the Error Bound, we have | f (1.2) − T4 (1.2)| ≤ K

(0.2)5 (1.2 − 1)5 = K, 5! 120

where K is a number such that f (5) x ≤ K for all 1 ≤ x ≤ 1.2. Starting from f (x) = x ln x, we find f (x) = ln x + x

1 1 1 2 = ln x + 1, f (x) = , f (x) = − 2 , f (4) (x) = 3 , x x x x

and f (5) (x) =

−6 . x4

For 1 ≤ x ≤ 1.2, 6 6 (5) f (x) = 4 ≤ 4 = 6. x 1 Hence we may take K = 6 to obtain: | f (1.2) − T4 (1.2)| ≤

(0.2)5 6 = 1.6 × 10−5 . 120

1 33. Show thebenth polynomial for ffor (x)f = (x) x. = 1 + x + x 2 + · · · + x n . Conclude by Tnsinh theMacLaurin nth MacLaurin polynomial (x) = sin is x+ Letthat Tn (x) 1−x (a) Show that T5 (x) = T6 (x) = T7 (x) = T8 (x). Hint: Tn (x) is the sum1of the MacLaurin polynomials of sin x and substituting sinh x. x/4 for x so that the nth MacLaurin polynomial for f (x) = 1 + x/4 is (b) Show that | f n (x)| ≤ 1 + cosh x for all n. Hint: Note that | sinh x| ≤ | cosh x| for all x. 2.6 (x)9 = 1 + 1 x + 1 x 2 + · · · + 1 x n (c) Show that |T8 (x) − f (x)| ≤ Tn|x| for −1 4≤ x ≤41. 2 4n 9! 1 What is the nth MacLaurin polynomial for g(x) = ? Hint: g(x) = f (−x). 1+x SOLUTION

Let f (x) = (1 − x)−1 . Then, f (x) = (1 − x)−2 , f (x) = 2(1 − x)−3 , f (x) = 3!(1 − x)−4 , and, in

general, f (n) (x) = n!(1 − x)−(n+1) . Therefore, f (n) (0) = n! and Tn (x) = 1 +

1! n! 2! x + x2 + · · · + xn = 1 + x + x2 + · · · + xn. 1! 2! n!

Upon substituting x/4 for x, we find that the nth MacLaurin polynomial for f (x) = Tn (x) = 1 +

1 is 1 − x/4

1 1 1 x + 2 x2 + · · · + n xn. 4 4 4

Substituting −x for x, the nth MacLaurin polynomial for g(x) =

1 is 1+x

Tn (x) = 1 − x + x 2 − x 3 + − · · · + (−x)n . 5 Let Tn (x) be the MacLaurin polynomial of f (x) = . 4 + 3x − x 2 1/4 1 + (a) Show that f (x) = 5 . 1 − x/4 1 + x (b) Let ak be the coefficient of x k in Tn (x) for k ≤ n. Use the result of Exercise 33 to show that 1 ak = k+1 + (−1)k 4

INTRODUCTION TO 10 DIFFERENTIAL EQUATIONS 10.1 Solving Differential Equations Preliminary Questions 1. Determine the order of the following differential equations: (b) (y )3 + x = 1 (a) x 5 y = 1 (c) y + x 4 y = 2 SOLUTION

(a) The highest order derivative that appears in this equation is a first derivative, so this is a first order equation. (b) The highest order derivative that appears in this equation is a first derivative, so this is a first order equation. (c) The highest order derivative that appears in this equation is a third derivative, so this is a third order equation. 2. Is y = sin x a linear differential equation? SOLUTION

Yes.

3. Give an example of a nonlinear differential equation of the form y = f (y). SOLUTION

One possibility is y = y 2 .

4. Can a nonlinear differential equation be separable? If so, give an example. SOLUTION

Yes. An example is y = y 2 .

5. Give an example of a linear, nonseparable differential equation. SOLUTION

One example is y + y = x.

Exercises 1. Which of the following differential equations are first-order? (b) y = y 2 (a) y = x 2 (c) (y )3 + yy = sin x (d) x 2 y − e x y = sin y y (e) y + 3y = x

(f) yy + x + y = 0

SOLUTION

(a) (b) (c) (d) (e) (f)

The highest order derivative that appears in this equation is a first derivative, so this is a first order equation. The highest order derivative that appears in this equation is a second derivative, so this is not a first order equation. The highest order derivative that appears in this equation is a first derivative, so this is a first order equation. The highest order derivative that appears in this equation is a first derivative, so this is a first order equation. The highest order derivative that appears in this equation is a second derivative, so this is not a first order equation. The highest order derivative that appears in this equation is a first derivative, so this is a first order equation.

In Exercises 3–9, verify that theingiven function a solution of the differential equation. Which of the equations Exercise 1 areislinear? 3. y − 8x = 0, SOLUTION

y = 4x 2

Let y = 4x 2 . Then y = 8x and y − 8x = 8x − 8x = 0.

5. yy + 4x = 0, y = 12−8x − 4x 2 y + 8y = 0, y = 4e

546

C H A P T E R 10

I N T R O D U C T I O N T O D I F F E R E N T I A L E Q U AT I O N S SOLUTION

Let y =

12 − 4x 2 . Then −4x y = , 12 − 4x 2

and yy + 4x =

−4x 12 − 4x 2 + 4x = −4x + 4x = 0. 12 − 4x 2

2 −1/2 7. (x 2 − 1)y + x y = 0, y = 4(x 2 − 1) y + 4x y = 0, y 2= 3e−2x SOLUTION Let y = 4(x − 1)−1/2 . Then y = −4x(x 2 − 1)−3/2 , and

(x 2 − 1)y + x y = (x 2 − 1)(−4x)(x 2 − 1)−3/2 + 4x(x 2 − 1)−1/2 = −4x(x 2 − 1)−1/2 + 4x(x 2 − 1)−1/2 = 0. 9. y − 2y + 5y = 0, y = e x sin 2x y − 2x y + 8y = 0, y = 4x 4 − 12x 2 + 3 SOLUTION Let y = e x sin 2x. Then y = 2e x cos 2x + e x sin 2x, y = −4e x sin 2x + 2e x cos 2x + 2e x cos 2x + e x sin 2x = −3e x sin 2x + 4e x cos 2x, and y − 2y + 5y = −3e x sin 2x + 4e x cos 2x − 4e x cos 2x − 2e x sin 2x + 5e x sin 2x = (−3e x − 2e x + 5e x ) sin 2x + (4e x − 4e x ) cos 2x = 0. 11. Consider the differential equation y = e y cos x. Which of the following equations are separable? Write those that are separable in the form y = f (x)g(y) (but (a) Write it as e−y d y = cos x d x. " " do not solve). −y d y = 2 cos x d x. (b) Integrate both sides of e (a) x y + y = 0 (b) 1 − x 2 y = e y sin x (c) Show = xy2 = +− y 2ln |C − sin x| is the general solution. (c) y that (d) Find the particular solution satisfying y(0) = 0.

(d) y = 9 − y 2

SOLUTION

(a) As y = dd xy , we regroup: dy = e y cos x dx e−y d y = cos x d x (b) Integrating both sides of the resulting equation yields −e−y + C1 = sin x + C2 e−y = − sin x + C (Note that a linear sum of arbitrary constants is another arbitrary constant.) (c) Solving the last equation for y yields −y = ln |− sin x + C| y = − ln |C − sin x| (d) Setting y(0) = 0 in the last equation yields 0 = − ln(C) so that C = 1. Therefore, y(x) = − ln |1 − sin x|. In Exercises 13–28, solve using separation of variables. Verify that x 2 y + e y = 0 is separable by rewriting it in the form y = f (x)g(y). x y 2 that e−y d y = −x −2 d x. 13. (a) y =Show (b) Show that e−y = −x −1 + C. (c) Verify that y = − lnC − x −1 is the general solution.


Rewrite dy = x y2 dx

dy = x d x. y2

as

Integrating both sides of this equation yields "

dy = y2 −

" x dx

1 1 = x 2 + C. y 2

Thus, 1

y=−1

,

2 2x +C

where C is an arbitrary constant. 15. y = 9y 1 y = Rewrite xy SOLUTION 2 dy = 9y dx

dy = 9 d x. y

as


dy = y

" 9 dx

ln |y| = 9x + C. Solving for y, we find |y| = e9x+C = eC e9x y = ±eC e9x = Ae9x , where A = ±eC is an arbitrary constant. dy =0 17. 2 y +=6y 2(4+−4 y) dx SOLUTION

Rewrite 2

dy + (6y + 4) = 0 dx

as dy = −(3y + 2) dx

dy = −d x. 3y + 2

and then


dy =− 3y + 2

" dx

ln |3y + 2| = −x + C. Solving for y, we find |3y + 2| = e−x+C = eC e−x 3y + 2 = ±eC e−x 1 2 2 y = ± eC e−x − = Ae−x − , 3 3 3 where A = ± 13 eC is an arbitrary constant. 19.

dy d−y te y =√0 dt =2 y dt

Solving Differential Equations

547

548

C H A P T E R 10


Rewrite dy = te y dt

e−y d y = t dt.

as


e−y d y = −e−y =

" t dt 1 2 t + C. 2

Solving for y, we find 1 −y = ln − t 2 + C 2 1 2 y = − ln C − t , 2 where C is an arbitrary constant. y 2 (1 − x 2 ) 21. y = 1 − x 2 y = x y SOLUTION Rewrite dy = y 2 (1 − x 2 ) dx

dy = (1 − x 2 ) d x. y2

as


dy = y2

" (1 − x 2 ) d x

−y −1 = x −

1 3 x + C. 3

Solving for y, we find y −1 =

1 3 x −x +C 3 1

y= 1 , 3 3x − x + C where C is an arbitrary constant. dx + 1)= x = x 2 + 1 23. (t 2 yy dt SOLUTION

Rewrite (t 2 + 1)

dx = (x 2 + 1) dt

1 1 dx = 2 dt. x2 + 1 t +1

as


1 dx = x2 + 1

"

1 dt t2 + 1

tan−1 x = tan−1 t + C. Solving for x, we find

x = tan tan−1 t + C . We can simplify this expression by applying the sum formula for the tangent function: x=

tan(tan−1 t) + tan C t + tan C t+A = = , 1 − t tan C 1 − At 1 − tan(tan−1 t) tan C

where A = tan C is an arbitrary constant. 25. y = x sec2y (1 + x )y = x 3 y



Rewrite dy = x sec y dx

cos y d y = x d x.

as


" cos y d y = sin y =

x dx 1 2 x + C. 2

Solving for y, we find

y = sin−1

1 2 x +C , 2

where C is an arbitrary constant. dy d=y y tan t dt = tan y dt SOLUTION Rewrite

27.

dy = y tan t dt

as

1 d y = tan t dt. y


1 dy = y

" tan t dt

ln |y| = ln |sec t| + C. Solving for y, we find |y| = eln |sec t|+C = eC |sec t| y = ±eC sec t = A sec t, where A = ±eC is an arbitrary constant. In Exercises d x 29–41, solve the initial value problem. = t tan x 29. y +dt2y = 0, y(ln 2) = 3 SOLUTION

First, we find the general solution of the differential equation. Rewrite dy + 2y = 0 dx

as

1 d y = −2 d x, y

and then integrate to obtain ln |y| = −2x + C. Thus, y = Ae−2x , where A = ±eC is an arbitrary constant. The initial condition y(ln 2) = 3 allows us to determine the value of A. 3 = Ae−2(ln 2) ;

1 3= A ; 4

Finally, y = 12e−2x . −y , y(0) = −1 xe2y 31. yy y= − + 4 = 0, y(1) = 4 2

so 12 = A.

549

550

C H A P T E R 10


First, we find the general solution of the differential equation. Rewrite y

2 dy = xe−y dx

2

ye y d y = x d x,

as

and then integrate to obtain 1 y2 1 e = x 2 + C. 2 2 Thus,

y = ± ln(x 2 + A), where A = 2C is an arbitrary constant. The initial condition y(0) = −1 allows us to determine the value of A. Since y(0) < 0, we have y = − ln(x 2 + A), and −1 = − ln( A); 1 = ln( A); so e = A. Finally,

y = − ln(x 2 + e). 33. y = (x − 1)(y − 2), y(0) = 3 dy = x −3 , y(2) = 0 y2 SOLUTION d x First, we find the general solution of the differential equation. Rewrite dy = (x − 1)(y − 2) dx

1 d y = (x − 1) d x, y−2

as

and then integrate to obtain ln |y − 2| =

1 2 x − x + C. 2

Thus, 2 y = Ae(1/2)x −x + 2,

where A = ±eC is an arbitrary constant. The initial condition y(0) = 3 allows us to determine the value of A. 3 = Ae0 + 2 = A + 2 so

A = 1.

Finally, 2 y = e(1/2)x −x + 2.

dy = ye−td,y y(0) = 1 dt (1 − t) − y = 0, y(2) = −4 dt SOLUTION First, we find the general solution of the differential equation. Rewrite

35.

dy = ye−t dt

1 d y = e−t dt, y

as

and then integrate to obtain ln |y| = −e−t + C. Thus, −t

y = Ae−e , where A = ±eC is an arbitrary constant. The initial condition y(0) = 1 allows us to determine the value of A. 1 = Ae−1

so

A = e.

Finally, y = (e)e−e dy = te−y , dt

y(1) = 0

−t

−t

= e1−e .

S E C T I O N 10.1

dy 37. t 2 − t = 1 + y + t y, dt SOLUTION


y(1) = 0

First, we find the general solution of the differential equation. Rewrite t2

dy = 1 + t + y + t y = (1 + t)(1 + y) dt

as 1 1+t d y = 2 dt, 1+y t and then integrate to obtain ln |1 + y| = −t −1 + ln |t| + C. Thus, t y = A 1/t − 1, e where A = ±eC is an arbitrary constant. The initial condition y(1) = 0 allows us to determine the value of A. 1 0= A − 1 so A = e. e Finally, et y = 1/t − 1. e 1 − x 2 y 2= y 2 , 2 y(0) = 1 1 − x y = y + 1, y(0) = 0 SOLUTION First, we find the general solution of the differential equation. Rewrite

39.

dy 1 − x2 = y2 dx

dy dx = , y2 1 − x2

as

and then integrate to obtain −

1 = sin−1 x + C. y

Thus, y=−

1 sin−1 x + C

,

where C is an arbitrary constant. The initial condition y(0) = 1 allows us to determine the value of C. 1=−

1 sin−1 0 + C

=−

1 C

so C = −1.

Finally, y=−

1 sin−1 x − 1

=

1 1 − sin−1 x

.

41. y = y 2 sin x, y(0) = 3 π y = tan y, y(ln 2) = 2 SOLUTION First, we find the general solution of the differential equation. Rewrite dy = y 2 sin x dx

as

y −2 d y = sin x d x,

and then integrate to obtain −y −1 = − cos x + C. Thus, y=

1 , A + cos x

551

552

C H A P T E R 10

I N T R O D U C T I O N T O D I F F E R E N T I A L E Q U AT I O N S

where A = −C is an arbitrary constant. The initial condition y(0) = 3 allows us to determine the value of A. 3=

1 ; A+1

A+1=

1 3

so

A=

1 2 −1=− . 3 3

Finally, y=

1 . cos x − (2/3)

43. Find all values of a such that y = eax isa a solution of Find all values of a such that y = x is a solution of y +2y −−2 8y = 0 y − 6x y = 0 SOLUTION Let y = eax . Then y = aeax

and

y = a 2 eax .

Substituting into the differential equation, we find y + 2y − 8y = eax (a 2 + 2a − 8). Because eax is never zero, y + 2y − 8y = 0 if only if a 2 + 2a − 8 = (a + 4)(a − 2) = 0. Hence, y = eax is a solution of the differential equation y + 2y − 8y = 0 provided a = −4 or a = 2. In Exercises 45–48, use Eq. (4) and Torricelli’s Law [Eq. (5)]. 2 Show that if y(t) is a solution of t (y − 1)y = 2y, then t y = Ce y for some constant C [we cannot solve for y(t) explicitly]. Find t such that y(t) has = 2,height assuming = 1.of area 30 ft2 . Water leaks through a hole in the 45. A cylindrical tank filled with water 10 ftthat andy(1) a base bottom of area 13 ft2 . How long does it take (a) for half of the water to leak out and (b) for the tank to empty? Because the tank has a constant cross-sectional area of 30 ft2 and the hole has an area of 13 ft2 , the differential equation for the height of the water in the tank is SOLUTION

1v v dy = 3 = . dt 30 90

By Torricelli’s Law, √ v = − 2gy = −8 y, using g = 32 ft/s2 . Thus, 4√ dy y. =− dt 45 Separating variables and then integrating yields 4 dt 45 4 2y 1/2 = − t + C 45

y −1/2 d y = −

Solving for y, we find 2 2 y(t) = C − t . 45 Since the tank is originally full, we have the initial condition y(0) = 10, whence y(t) =

√

10 −

2 2 t . 45

When half of the water is out of the tank, y = 5, so we solve: 5=

√

10 −

2 2 t 45

for t, finding t=

√ 45 √ ( 10 − 5) ≈ 20.84 sec. 2

√ 10 = C. Therefore,

S E C T I O N 10.1


553

When all of the water is out of the tank, y = 0, so √ 2 10 − t = 0 and 45

t=

45 √ 10 ≈ 71.15 sec. 2

47. The tank in Figure 7(B) is a cylinder of radius 10 ft and length 40 ft. Assume that the tank is half-filled with water A conical tank filled with water has height 12 ft [Figure 7(A)]. Assume that the top is a circle of radius 4 ft and and that water leaks through a hole in the bottom of area B = 32 in.2 . Determine the water level y(t) and the time te when that water leaks through a hole in the bottom of area 2 in. . Let y(t) be the water level at time t. the tank is empty. SOLUTION When the water is at height y over the bottom, the top cross section is a rectangle with length 40 ft, and with width x satisfying the equation: (a) Show that the cross-sectional area of the tank at height y is A(y) = π9 y 2 . 2 (y − 10)2for = y(t). 100. Use the initial condition y(0) = 12. (b) Find the differential equation satisfied(x/2) by y(t)+and solve How does take for the tank to empty? Thus,(c)x = 2 long 20y − y 2 ,itand

A(y) = 40x = 80 20y − y 2 . 1 ft2 and v = −√2gy = −8√ y, it follows that With B = 3 in2 = 48 1 8√ y dy 1 48 =− . =− √ 2 dt 480 20 − y 80 20y − y

Separating variables and integrating then yields: 1 20 − y d y = − dt 480 2 t − (20 − y)3/2 = − +C 3 480 When t = 0, y = 10, so C = − 23 103/2 , and t 2 2 − 103/2 − (20 − y)3/2 = − 3 480 3 2/3 t + 103/2 y(t) = 20 − . 320 The tank is empty when y = 0. Thus, te satisfies the equation 20 −

2/3 te 3/2 + 10 = 0. 320

It follows that te = 320(203/2 − 103/2 ) ≈ 18502.4 seconds. 49. Figure 8 shows a circuit consisting of a resistor of R ohms, a capacitor of C farads, and a battery of voltage V . When 2 filled with water has height h ft and a on base area A leaks through a hole to in the the circuitAis cylindrical completed,tank the amount of charge q(t) (in coulombs) theofplates offtthe. Water capacitor varies according the 2 bottom equation of area B(tftin. seconds) differential √ (a) Show that the time required for the tank to empty is proportional to A h/B. dq 1 R to V+h −1/2 q ,=where V V is the volume of the tank. (b) Show that the emptying time is proportional dt C (c) Two tanks have the same volume and same-sized hole, but different heights and bases. Which tank empties first: (a) Solve for or q(t). the taller shorter tank? (b) Show that lim q(t) = C V . t→∞

(c) Find q(t), assuming that q(0) = 0. Show that the capacitor charges to approximately 63% of its final value C V after a time period of length τ = RC (τ is called the time constant of the capacitor). R

V

C

FIGURE 8 An RC circuit. SOLUTION

554

C H A P T E R 10


(a) Upon rearranging the terms of the differential equation, we have q − CV dq =− . dt RC Separating the variables and integrating both sides, we obtain dt dq =− q − CV RC "

dq =− q − CV

"

dt RC

and ln |q − C V | = −

t + k, RC

where k is an arbitrary constant. Solving for q(t) yields 1

q(t) = C V + K e− RC t , where K = ±ek . (b) Using the result from part (a), we calculate

1 1 lim q(t) = lim C V + K e− RC t = C V + K lim e− RC t = C V + K · 0 = C V . t→∞

t→∞

t→∞

(c) Using the result from part (a), the condition q(0) = 0 determines K = −C V . Thus,

1 q(t) = C V 1 − e− RC t , and q(τ ) = q(RC) = C V (1 − e−1 ) ≈ 0.632C V . dV Oneinhypothesis growth rateR of V 0.01 of a cell is that to the cell’s Assume the circuitfor of the Figure 8 that =the 100volume , C = F, and V = 10 is V. proportional How many seconds doessurface it take dt for the charge on the capacitor plates to reach half of its limiting value? 3 2 area A. Since V has cubic units such as cm and A has square units such as cm , we may assume roughly that A ∝ V 2/3 , dV = kV 2/3 for some constant k. If this hypothesis is correct, which dependence of volume on time would and hence dt we expect to see (again, roughly speaking) in the laboratory? (a) Linear (b) Quadratic (c) Cubic

51.

SOLUTION

Rewrite dV = kV 2/3 dt

as

V −2/3 dv = k dt,

and then integrate both sides to obtain 3V 1/3 = kt + C V = (kt/3 + C)3 . Thus, we expect to see V increasing roughly like the cube of time. g , but let f (x) = e2x and find a function g(x) such that ( f g) = f g . Do the 53. In general, ( f g) not equal We might alsoisguess that to thef rate at which a snowball melts is proportional to its surface area. What is the samedifferential for f (x) =equation x. satisfied by the volume V of a spherical snowball at time t? Suppose the snowball has radius = fhalf g , of 4 cm and If that volume SOLUTION ( fitg)loses weitshave

after 10 min. According to this model, when will the snowball disappear?

f (x)g(x) + g (x) f (x) = f (x)g (x) g (x)( f (x) − f (x)) = −g(x) f (x) f (x) g (x) = g(x) f (x) − f (x) Now, let f (x) = e2x . Then f (x) = 2e2x and g (x) 2e2x = 2. = 2x g(x) 2e − e2x

S E C T I O N 10.1


555

Integrating and solving for g(x), we find dg = 2 dx dg ln |g| = 2x + C g(x) = Ae2x , where A = ±eC is an arbitrary constant. If f (x) = x, then f (x) = 1, and 1 g (x) = . g(x) 1−x Thus, 1 dg = dx g 1−x ln |g| = − ln |1 − x| + C g(x) =

A , 1−x

where A = ±eC is an arbitrary constant. 55. If a bucket of water spins about a vertical axis with constant angular velocity ω (in radians per second), the water A boy standing at point B on a dock holds a rope of length attached to a boat at point A [Figure 9(A)]. As climbs up the side of the bucket until it reaches an equilibrium position (Figure 10). Two forces act on a particle located the boy walks along the dock, holding the rope taut, the boat moves along a curve called a tractrix (from the Latin at a distance x from the vertical axis: the gravitational force −mg acting downward and the force of the bucket on the tractus meaning “to pull”). The segment from a point P on the curve to the x-axis along the tangent line has constant length . Let y = f (x) be the equation of the tractrix.y y (a) Show that y 2 + (y/y )2 = 2 and conclude y = − . Why must we choose the negative square root? 2 − y2 y = f (x) (b) Prove that the tractrix is the graph of m 2 x + 2 − y 2 x = ln − 2 − y 2 mg y x x

FIGURE 10

particle (transmitted indirectly through the liquid) in the direction perpendicular to the surface of the water. These two forces must combine to supply a centripetal force m ω 2 x, and this occurs if the diagonal of the rectangle in Figure 10 is normal to the water’s surface (that is, perpendicular to the tangent line). Prove that if y = f (x) is the equation of the curve obtained by taking a vertical cross section through the axis, then −1/y = −g/(ω 2 x). Show that y = f (x) is a parabola. At any point along the surface of the water, the slope of the tangent line is given by the value of y at that point; hence, the slope of the line perpendicular to the surface of the water is given by −1/y . The slope of the resultant force generated by the gravitational force and the centrifugal force is SOLUTION

g −mg =− 2 . mω2x ω x Therefore, the curve obtained by taking a vertical cross-section of the water surface is determined by the equation 1 g − =− 2 y ω x

or

y =

ω2 x. g

Performing one integration yields y = f (x) =

ω2 2 x + C, g

where C is a constant of integration. Thus, y = f (x) is a parabola. 57. Find the family of curves satisfying y = x/y and sketch several members of the family. Then find the differential 3y x equationShow for the orthogonal familyequations (see Exercise and add some members of this orthogonal = − solution, that the differential y =56), find andits y general define orthogonal families of curves; that is, the x 3y family to your plot. graphs of solutions to the first equation intersect the graphs of the solutions to the second equation in right angles (Figure 11). Find these curves explicitly.

556

C H A P T E R 10


Separation of variables and integration applied to y = x/y gives y dy = x dx 1 1 2 y = x2 + C 2 2 y = ± x2 + C

If y(x) is a curve of the family orthogonal to these, it must have tangent lines of slope −y/x at every point (x, y). This gives y = −y/x Separation of variables and integration give dx dy =− y x ln |y| = − ln |x| + C A x

y=

Several solution curves of both differential equations appear below:

"

–1

1 –1

59. Let v(t) be the velocity of an object of mass m in free fall near the earth’s surface. If we assume that air resistance is A 50-kg model rocket lifts off by expelling fuel at a rate dv of k = 4.752 kg/s for 10 s. The fuel leaves the end of v satisfies the differential equation m m(t) constant k > t.0.From the law of = be −gthe + kv proportional to with v 2 , then the rocket an exhaust velocity of b = 100 m/s. Let massfor ofsome the rocket at time dt conservation of momentum, we find the following differential equation for the rocket’s velocity v(t) (in meters per (a) Set α = (g/k)1/2 and rewrite the differential equation as second): dv k 2 2 dm = − (α − v ) m(t)v dt (t) =m−9.8m(t) + b dt Then solve using separation of variables with initial condition v(0) = 0. (a) Show that m(t) = 50 − 4.75t kg. (b) Show that the terminal velocity lim v(t) is equal to −α . (b) Solve for v(t) and compute t→∞ the rocket’s velocity at rocket burnout (after 10 s). SOLUTION

(a) Let α = (g/k)1/2 . Then k 2 k g g k dv α − v2 = − + v2 = − − v2 = − dt m m m k m Separating variables and integrating yields "

dv k =− m α 2 − v2

" dt = −

k t +C m

We now use partial fraction decomposition for the remaining integral to obtain "

" α + v 1 1 1 1 dv + dv = = ln 2α α+v α −v 2α α−v α 2 − v2 Therefore, α + v 1 = − k t + C. ln 2α α − v m The initial condition v(0) = 0 allows us to determine the value of C: α + 0 1 = − k (0) + C ln 2α α − 0 m

S E C T I O N 10.1

C= Finally, solving for v, we find

v(t) = −α


557

1 ln 1 = 0. 2α √

1 − e−2( gk/m)t

√

1 + e−2( gk/m)t

.

√

(b) As t → ∞, e−2( gk/m)t → 0, so

v(t) → −α

1−0 1+0

= −α .

Further Insights and Challenges 61. In most cases of interest, the general solution of a differential equation of order n depends on n arbitrary constants. In Section 6.2, are we exceptions. computed the volume V of a solid as the integral of cross-sectional area. Explain this This exercise shows there formula in terms of differential equations. Let V (y) be the volume of the solid up to height y and let A(y) be the (a) Show that (y )2 + y 2 = 0 is a first-order equation with only one solution y = 0. cross-sectional area at height y as in Figure 12. (b) Show that (y )2 + y 2 + 1 = 0 is a first-order equation with no solutions. (a) Explain the following approximation for small y: SOLUTION

(yy+ y) − Vy(y) = (a) (y )2 + y 2 ≥ 0 and equals zero if and onlyV if 0 and = 0≈ A(y) y 2 2 2 2 " b (b) (y ) + y + 1 ≥ 1 > 0 for all y and y, so (y ) + y + 1 = 0 has no solution dV A(y) d y. = A(y). Then derive the formula V = (b) Use Eq. (9) to justify the differential equation 63. A spherical tank of radius R is half-filled with water. d ySuppose that water leaks through a hole ina the bottom of area 2 λ x 2 P(x) = water x + level ax +atbtime witht (seconds). a, b constants. Show that y = Ce (for any constant C) is a solution of B ft (a) . LetLet y(t) be the √ and only y + ay +dby y if P(λ ) = 0. −8B y = 0 if (a) Show that = . (b) Show that C1 e3x−+y 2C)2 e−x is a solution of y − 2y − 3y = 0 for any constants C1 , C2 . dt y = π (2Ry (b) Show that for some constant C, π 2 3/2 1 5/2 Ry − y =C −t 4B 3 5 (c) Use the initial condition y(0) = R to compute C and show that C = te , the time at which the tank is empty. (d) Show that te is proportional to R 5/2 and inversely proportional to B. SOLUTION

(a) At height y above the bottom of the tank, the cross section is a circle of radius

r = R 2 − (R − y)2 = 2Ry − y 2 . The cross-sectional area function is then A(y) = π (2Ry − y 2 ). The differential equation for the height of the water in the tank is then √ 8B y dy . =− dt π (2Ry − y 2 ) (b) Rewrite the differential equation as π

2Ry 1/2 − y 3/2 d y = − dt, 8B and then integrate both sides to obtain

π 4B

2 3/2 1 5/2 − y Ry = C − t, 3 5

where C is an arbitrary constant. (c) The initial condition y(0) = R allows us to determine the value of C: π 2 5/2 1 5/2 7π 5/2 C= − R R R . = 4B 3 5 60B Moreover, note that y = 0 when t = C, C = te , the time at which the tank is empty. (d) From part (c), te =

7π 5/2 R , 60B

from which it is clear that te is proportional to R 5/2 and inversely proportional to B.

558

C H A P T E R 10


10.2 Graphical and Numerical Methods Preliminary Questions

1. What is the slope of the segment in the slope field for y· = t y + 1 at the point (2, 3)? · SOLUTION The slope of the segment in the slope field for y = t y + 1 at the point (2, 3) is (2)(3) + 1 = 7. 2. What is the equation of the isocline of slope c = 1 for y· = y 2 − t?

√ The isocline of slope c = 1 has equation y 2 − t = 1, or y = ± 1 + t. 3. True or false? In the slope field for y· = ln y, the slopes at points on a vertical line t = C are all equal.

SOLUTION

SOLUTION

This statement is true because the right-hand side of the differential equation does not explicitly depend on

the variable t.

4. What about the slope field for y· = ln t? Are the slopes at points on a vertical line t = C all equal?

No, because the right-hand side of this differential equation does explicitly depend on the variable t. 5. Let y(t) be the solution to y· = F(t, y) with y(1) = 3. How many iterations of Euler’s Method are required to approximate y(3) if the time step is h = 0.1? SOLUTION

SOLUTION The initial condition is specified at t = 1 and we want to obtain an approximation to the value of the solution at t = 3. With a time step of h = 0.1,

3−1 = 20 0.1 iterations of Euler’s method are required.

Exercises

1. Figure 8 shows the slope field for y· = sin y sin t. Sketch the graphs of the solutions with initial conditions y(0) = 1 and y(0) = −1. Show that y(t) = 0 is a solution and add its graph to the plot. y 3 2 1 t

0 −1 −2 −3

−3 −2 −1

0

1

2

3

·

FIGURE 8 Slope field for y = sin y sin t. SOLUTION

The sketches of the solutions appear below. y 3 2 1 t

0 −1 −2 −3

−3 −2 −1

0

1

2

3

If y(t) = 0, then y = 0; moreover, sin 0 sin t = 0. Thus, y(t) = 0 is a solution of y· = sin y sin t. 3. Show that f (t) = 12 (t − 12 ) is a solution to y· = t − 2y. Sketch the four solutions with y(0) = ±0.5, ±1 on the slope Figure 9 shows the slope field of y· = y 2 − t 2 . Sketch the integral curve passing through the point (0, −1), the field in Figure 10. The slope field suggests that every solution approaches f (t) as t → ∞. Confirm this by showing that curve through (0, 0), and the curve through (0, 2). Is y(t) = 0 a solution? y = f (t) + Ce−2t is the general solution. y 1

y = 1 (t − 1 ) 2

0.5 t

0 −0.5 −1

−1

−0.5

0

0.5

1

1.5

·

2

FIGURE 10 Slope field for y = t − 2y.

2


Graphical and Numerical Methods

559

Let y = f (t) = 12 (t − 12 ). Then y· = 12 and 1 1 y· + 2y = + t − = t, 2 2

so f (t) = 12 (t − 12 ) is a solution to y· = t − 2y. The slope field with the four required solutions is shown below.

Now, let y = f (t) + Ce−2t = 12 (t − 12 ) + Ce−2t . Then 1 y· = − 2Ce−2t , 2 and

·y + 2y = 1 − 2Ce−2t + t − 1 + 2Ce−2t = t. 2 2

Thus, y = f (t) + Ce−2t is the general solution to the equation y· = t − 2y. 5. Show that the isoclines of y· = 1/y ·are horizontal lines. Sketch the slope field for −2 ≤ t, y ≤ 2 and plot the Consider the differential equation y = y(0) t − y. solutions with initial conditions y(0) = 0 and = 1. (a) Sketch the slope field of the differential equation y· = t − y in the range −1 ≤ 1 1 t ≤ 3, −1 ≤ y ≤ 3. As an aid, SOLUTION The isocline of slope c is defined by y = c. This is equivalent to y = c , which is a horizontal line. The observe that the isocline of slope c is the line t − y = c, so the segments have slope c at points on the line y = t − c. slope field and the solutions are shown below. (b) Show that y = t − 1 + Ce−t is a solution for all C. Since lim e−t = 0, these solutions approach the particular t→∞

solution y = t − 1 as t → ∞. Explain how this behavior is reflected in your slope field.

7. Sketch the slope field of y· =· t y for −2 ≤ t, y ≤ 2. Based on the sketch, determine lim y(t), where y(t) is a Show that the isoclines of y = t are vertical lines. Sketch the slope field for −2 ≤ t, t→∞ y ≤ 2 and plot the integral y(t) y(0) < 0? solution with y(0) > 0. What limand curves passing through (0,is−1) (0,if1). t→∞ · SOLUTION The slope field for y = t y is shown below. y 2 1 t

0 −1 −2

−2

−1

0

1

2

With y(0) > 0, the slope field indicates that y is an always increasing, always concave up function; consequently, limt→∞ y = ∞. On the other hand, when y(0) < 0, the slope field indicates that y is an always decreasing, always concave down function; consequently, limt→∞ y = −∞.

560

C H A P T E R 10


9. OneMatch of thethe slope fields in Figures andslope (B) isfield the in slope field 11(A)–(F). for y· = t 2 . The other is for y· = y 2 . Identify which differential equation12(A) with its Figures is which. ·In each case, sketch the solutions with initial conditions y(0) = 1, y(0) = 0, and y(0) = −1. (i) y = −1 y y y (ii) y· = 3 3 t 2 2 (iii) y· = t 2 y · 2 1 1 (iv) y = t y t t 0 0 · 2 2 (v) y = t + y −1 −1 · (vi) y = t −2

−3

−2

−3 −2 −1

−3 0

1

2

3

−3 −2 −1

FIGURE 12(A)

0

1

2

3

FIGURE 12(B)

√ For y = t 2 , y only depends on t. The isoclines of any slope c will be the two vertical lines t = ± c. This indicates that the slope field will be the one given in Figure 12(A). The solutions are sketched below: SOLUTION

y 2 1 t

0 −1 −2

−2

−1

0

1

2

√ For y = y 2 , y only depends on y. The isoclines of any slope c will be the two horizontal lines y = ± c. This indicates that the slope field will be the one given in Figure 12(B). The solutions are sketched below: y 2 1 t

0 −1 −2

−2

−1

0

1

2

11. Sketch the slope field of y· =· t 2 − y in the region −3 ≤ t ≤ 3, −3 ≤ y ≤ 3 and sketch the solutions satisfying the1,slope field= of−1. y = t/y in the region −2 ≤ t ≤ 2, −2 ≤ y ≤ 2. y(1)(a) = 0,Sketch y(1) = and y(1) (b) Check that y = ± t 2 +·C is 2the general solution. SOLUTION The slope field for y = t − y, together with the required solution curves, is shown below. (c) Sketch the solutions on the slope field with initial conditions y(0) = 1 and y(0) = −1. y 3 2 1 t

0 −1 −2 −3

−3 −2 −1

0

1

2

3

Let y(t) be the solution to y· = te−y satisfying y(0) = 0.· Let F(t, y) = t 2 − y and let y(t) be the solution of y = F(t, y) satisfying y(2) = 3. Let h = 0.1 be the time Use Euler’s Method with time step h = 0.1 to approximate y(0.1), y(0.2), . . . , y(1). step in Euler’s Method and set y0 = y(2) = 3. Use separation of variables to find y(t) exactly. (a) Calculate y1 = y0 + h F(2, 3). Compute the error in the approximations to y(0.1), y(0.5), and y(1). (b) Calculate y2 = y1 + h F(2.1, y1 ). SOLUTION (c) Calculate y3 = y2 + h F(2.2, y2 ) and continue computing y4 , y5 , and y6 . (d) Find approximations to y(2.2) and y(2.5).

13. (a) (b) (c)

S E C T I O N 10.2

Graphical and Numerical Methods

561

(a) With y0 = 0, t0 = 0, h = 0.1, and F(t, y) = te−y , we compute n

tn

yn

0

0

0

1

0.1

y0 + h F(t0 , y0 ) = 0

2

0.2

y1 + h F(t1 , y1 ) = 0.01

3

0.3

y2 + h F(t2 , y2 ) = 0.029801

4

0.4

y3 + h F(t3 , y3 ) = 0.058920

5

0.5

y4 + h F(t4 , y4 ) = 0.096631

6

0.6

y5 + h F(t5 , y5 ) = 0.142026

7

0.7

y6 + h F(t6 , y6 ) = 0.194082

8

0.8

y7 + h F(t7 , y7 ) = 0.251733

9

0.9

y8 + h F(t8 , y8 ) = 0.313929

10

1.0

y9 + h F(t9 , y9 ) = 0.379681

(b) Rewrite dy = te−y dt

as

e y d y = t dt,

and then integrate both sides to obtain ey =

1 2 t + C. 2

Thus, 1 y = ln t 2 + C . 2 Applying the initial condition y(0) = 0 yields 0 = ln |C|, so C = 1. The exact solution to the initial value problem is 1 2 then y = ln 2 t + 1 . (c) The three errors requested are computed here: |y(0.1) − y1 | = |0.00498754 − 0| = 0.00498754; |y(0.5) − y5 | = |0.117783 − 0.0966314| = 0.021152; |y(1) − y10 | = |0.405465 − 0.379681| = 0.025784. In Exercises 14–19, use Euler’s Method with h = 0.1 to approximate the given value of y(t). 15. y(1); y· = y,· y(0) = 0 y(0.5); y = y 2 , y(0) = 0 SOLUTION Let F(t, y) = y. With t0 = 0, y0 = 0, and h = 0.1, we compute y(0.0) = y0

=0

y(0.1) ≈ y1 = y0 + 0.1(y0 ) =0 y(0.2) ≈ y2 = y1 + 0.1(y1 ) =0 y(0.3) ≈ y3 = y2 + 0.1(y2 ) =0 y(0.4) ≈ y4 = y3 + 0.1(y3 ) =0 y(0.5) ≈ y5 = y4 + 0.1(y4 ) =0 y(0.6) ≈ y6 = y5 + 0.1(y5 ) =0 y(0.7) ≈ y7 = y6 + 0.1(y6 ) =0 y(0.8) ≈ y8 = y7 + 0.1(y7 ) =0 y(0.9) ≈ y9 = y8 + 0.1(y8 ) =0 y(1.0) ≈ y10 = y9 + 0.1(y9 )=0 17. y(1.5); y· =· t sin y, y(1) = 2 y(0.7); y = −yt, y(0) = 1

562

C H A P T E R 10


Let F(t, y) = t sin y. With t0 = 1, y0 = 2 and h = 0.1, we compute =2

y(1.0) = y0

y(1.1) ≈ y0 + 0.1(t0 sin(y0 )) = 2.09093 y(1.2) ≈ y1 + 0.1(t1 sin(y1 )) = 2.18638 y(1.3) ≈ y2 + 0.1(t2 sin(y2 )) = 2.28435 y(1.4) ≈ y3 + 0.1(t3 sin(y3 )) = 2.38264 y(1.5) ≈ y4 + 0.1(t4 sin(y4 )) = 2.47898 19. y(0.5); y· =· t − y, y(0) = 1 y(2.6); y = t/y, y(0) = 2 SOLUTION Let F(t, y) = t − y. With t0 = 0, y0 = 1, and h = 0.1, we compute =1

y(0.0) = y0

y(0.1) ≈ y1 = y0 + 0.1(t0 − y0 ) = 0.9 y(0.2) ≈ y2 = y1 + 0.1(t1 − y1 ) = 0.82 y(0.3) ≈ y3 = y2 + 0.1(t2 − y2 ) = 0.758 y(0.4) ≈ y4 = y3 + 0.1(t3 − y3 ) = 0.7122 y(0.5) ≈ y5 = y4 + 0.1(t4 − y4 ) = 0.68098

Further Insights and Challenges " t In Exercises 21–22, use a modification of Euler’s Method, Euler’s Midpoint Method, that gives a significant improvement · = f (t) with initial condition y(a) = 0 is y(t) = If f (t) is continuous [a,initial b], then the solution y f (u) du. ), the values y are defined successively by the equation in accuracy. With time step hon and value y0 = y(tto k 0 a

b − a + hm k = yk−1 for N k−1 steps yields the N th left-endpoint approximation to Show that Euler’s Method with time step h y= N " b h h where m k−1 tk−1 y(b) = = Ff (u) du.+ , yk−1 + F(tk−1 , yk−1 ) . 2 2 a 21. Apply both Euler’s Method and the Euler Midpoint Method with h = 0.1 to estimate y(1.5), where y(t) satisfies y· = y with y(0) = 1. Find y(t) exactly and compute the errors in these two approximations. SOLUTION Let F(t, y) = y. With t0 = 0, y0 = 1, and h = 0.1, fifteen iterations of Euler’s method yield y(1.5) ≈ y15 = 4.177248. The Euler midpoint approximation with F(t, y) = y is h h h m k−1 = F tk−1 + , yk−1 + F(tk−1 , yk−1 ) = yk−1 + yk−1 2 2 2 2 h h yk = yk−1 + h yk−1 + yk−1 = yk−1 + hyk−1 + y 2 2 k−1 Fifteen iterations of Euler’s midpoint method yield: y(1.5) ≈ y15 = 4.471304. The exact solution to y = y, y(0) = 1 is y(t) = et ; therefore y(1.5) = 4.481689. The error from Euler’s method

is |4.177248 − 4.481689| = 0.304441, while the error from Euler’s midpoint method is |4.471304 − 4.481689| = 0.010385. " t In Exercises 23–26, use Euler’s Midpoint Method with h = 0.1 to approximate the given value of y(t). · = f (t) with initial condition y(a) = 0 is y(t) = If f (t)·is continuous on [a, b], then the solution to y f (u) du. 23. y(0.5); y = y 2 , y(0) = 0 a b−a SOLUTION = y 2 . With t0 = with 0, y0time = 0,step andhh== 0.1, weforcompute Show thatLet the F(t, Eulery)Midpoint Method N steps yields the N th midpoint approximation N " to y(b) =

b

a

f (u) du.

k

tk

mk

yk

0

0.0

0

0

1

0.1

0

0

2

0.2

0

0

3

0.3

0

0

4

0.4

0

0

5

0.5

0

0

S E C T I O N 10.3

The Logistic Equation

563

Thus, y(0.5) ≈ y5 = 0. 25. y(1); y· =· t + y, y(0) = 0 y(1); y = y 2 , y(0) = 0 SOLUTION Let F(t, y) = t + y. With t0 = 0, y0 = 0, and h = 0.1, we compute k

tk

mk

yk

0

0.0

0

0

1

0.1

0.05

0.005

2

0.2

0.16025

0.021025

3

0.3

0.282076

0.0492326

4

0.4

0.416694

0.0909021

5

0.5

0.565447

0.147447

6

0.6

0.729819

0.220429

7

0.7

0.911450

0.311574

8

0.8

1.112152

0.422789

9

0.9

1.333928

0.556182

10

1.0

1.578991

0.714081

Hence y(1) ≈ y10 = 0.714081. y(0.5);

y· = y 2 − t,

y(0) = 1

10.3 The Logistic Equation Preliminary Questions 1. Which of the following is a logistic differential equation?

y (a) y· = 2y(1 − y 2 ) (b) y· = 2y 1 − 3

x (c) y· = 2y 1 − (d) y· = 2y(1 − 3y) 4 SOLUTION The differential equations in (b) and (d) are logistic equations. The equation in (a) is not a logistic equation because of the y 2 term inside the parentheses on the right-hand side; the equation in (c) is not a logistic equation because of the presence of the independent variable on the right-hand side.

2. True or false? The logistic equation is linear. SOLUTION

False, the logistic equation is not a linear differential equation.

3. True or false? The logistic equation is separable. True, the logistic equation is a separable differential equation. 4. Let y(t) be a solution to y· = 4y(3 − y). What is lim y(t) in the following three cases:

SOLUTION

t→∞

(a) y(0) = 3

(b) y(0) = 4

SOLUTION

(a) If y(0) = 3, then y· = 0, and y(t) = 3 for all t. Thus, lim y(t) = 3. t→∞ (b) If y(0) = 4, then y· < 0, and lim y(t) = 3. t→∞ (c) If y(0) = −2, then y· < 0, and lim y(t) → −∞. t→∞

Exercises 1. Find the general solution of the logistic equation y· = 3y(1 − y/5) Then find the particular solution satisfying y(0) = 2.

(c) y(0) = −2

564

C H A P T E R 10


y· = 3y(1 − y/5) is a logistic equation with k = 3 and A = 5; therefore, the general solution is 5 . 1 − e−3t /C

y=

The initial condition y(0) = 2 allows us to determine the value of C: 2=

5 ; 1 − 1/C

1−

1 5 = ; C 2

2 so C = − . 3

The particular solution is then y=

10 5 = . 2 + 3e−3t 1 + 32 e−3t

3. Let y(t) be a solution of ·y· = 0.5y(1 − 0.5y) such that y(0) = 4. Determine lim y(t) without finding y(t) explicitly. t→∞ Find the solution of y = 2y(3 − y), y(0) = 10. SOLUTION The equation is a logistic equation with k = 0.5 and A = 2. Let F(y) = 0.5y(1 − 0.5y), and take y0 = 4. Then, F(y0 ) = 0.5(4)(1 − 0.5(4)) < 0, so, y(t) is strictly decreasing and lim y(t) = A = 2.

t→∞

5. A population of squirrels lives in a forest with a carrying capacity of 2,000. Assume logistic growth with growth a solution of y· = 5y(1 − y/A) satisfying y(0) = 10. Find lim y(t), assuming that . constantLet k =y(t) 0.6be yr−1 t→∞ (a) Find formula an initial population of (a) Aa = 15 for the squirrel population (b)P(t), A =assuming 5 (c)500 A squirrels. = 10 (b) How long will it take for the squirrel population to double? SOLUTION

(a) Since k = 0.6 and the carrying capacity is A = 2000, the population P(t) of the squirrels satisfies the differential equation P (t) = 0.6P(t)(1 − P(t)/2000), with general solution P(t) =

2000 . 1 − e−0.6t /C

The initial condition P(0) = 500 allows us to determine the value of C: 500 =

2000 ; 1 − 1/C

1−

1 = 4; C

1 so C = − . 3

The formula for the population is then P(t) =

2000 . 1 + 3e−0.6t

(b) The squirrel population will have doubled at the time t where P(t) = 1000. This gives 1000 =

2000 ; 1 + 3e−0.6t

1 + 3e−0.6t = 2;

so t =

5 ln 3 ≈ 1.83. 3

It therefore takes approximately 1.83 years for the squirrel population to double. 7. Sunset Lake is stocked with 2,000 rainbow trout and after 1 year the population has grown to 4,500. Assuming The population P(t) of mosquito larvae growing in a tree hole increases according to the logistic equation with logistic growth with a carrying capacity of 20,000, find the growth constant k (specify the units) and determine when the growth constant k = 0.3 days−1 and carrying capacity A = 500. population will increase to 10,000. (a) Find a formula for the larvae population P(t), assuming an initial population of P0 = 50 larvae. SOLUTION Since A = 20000, the trout population P(t) satisfies the logistic equation (b) After how many days will the larvae population reach 200? P (t) = k P(t)(1 − P(t)/20000), with general solution P(t) =

20000 . 1 − e−kt /C


20000 ; 1 − 1/C

1−

1 = 10; C

1 so C = − . 9

S E C T I O N 10.3


565

After one year, we know the population has grown to 4500. Let’s measure time in years. Then 4500 =

20000 1 + 9e−k

40 9 31 e−k = 81 81 k = ln ≈ 0.9605 years−1 . 31

1 + 9e−k =

The population will increase to 10000 at time t where P(t) = 10000. This gives 20000 1 + 9e−0.9605t

10000 =

1 + 9e−0.9605t = 2 1 9

e−0.9605t =

1 ln 9 ≈ 2.29 years. 0.9605

t=

9. A rumor spreads through a school with 1,000 students. At 8 AM, 80 students have heard the rumor, and by noon, half Spread of a Rumor A rumor spreads a small town. Let when y(t) be the of fraction of the will population that has the school has heard it. Using the logistic modelthrough of Exercise 8, determine 90% the students have heard the heard the rumor at time t and assume that the rate at which the rumor spreads is proportional to the product of the rumor. fraction y of the population that has heard the rumor and the fraction 1 − y that has not yet heard the rumor. SOLUTION Let y(t) be the proportion of students that have heard the rumor at a time t hours after 8 AM. In the logistic (a) Write down the differential equation satisfied by y in terms of a proportionality factor k. model of Exercise 8, we have a −1 capacity of A = 1 (100% of students) and an unknown growth factor of k. Hence, (b) Find k (in units of days ), assuming that 10% of the population knows the rumor at t = 0 and 40% knows it at 1 t = 2 days. . y(t) = (c) Using the assumptions of part (b), determine when the population will know the rumor. 1 −75% e−ktof/C The initial condition y(0) = 0.08 allows us to determine the value of C: 1 2 = ; 25 1 − 1/C

1−

1 25 = ; C 2

so C = −

2 . 23

so that y(t) =

2 . 2 + 23e−kt

The condition y(4) = .5 now allows us to determine the value of k: 1 2 ; = 2 2 + 23e−4k

2 + 23e−4k = 4;

so k =

1 23 ln ≈ 0.6106 hours−1 . 4 2

90% of the students have heard the rumor when y(t) = .9. Thus 2 9 = 10 2 + 23e−0.6106t 2 + 23e−0.6106t = t=

20 9 1 207 ln ≈ 7.6 hours. 0.6106 2

Thus, 90% of the students have heard the rumor after 7.6 hours, or at 3:36 PM. Let k = 1 and A = 1 in the logistic equation. A simpler model for the spread of a rumor assumes that the rate at which the rumor spreads is proportional Find the solutions satisfying y1 (0) = 10 and y2 (0) = −1. (with factor k) to the fraction of the population that has not yet heard the rumor. Find the time t when y1 (t) = 5. (a) Compute the solutions to this model and the model of Exercise 8 with the values k = 0.9 and y0 = 0.1. When does y2 (t) become infinite? (b) Graph the two solutions on the same axis. SOLUTION The general solution the logistic equation with k = 1 and A = 1 is (c) Which model seems more of realistic? Why?

11. (a) (b) (c)

y(t) =

1 . 1 − e−t /C

566

C H A P T E R 10


(a) Given y1 (0) = 10, we find C = 10 9 , and y1 (t) =

1 −t 1 − 10 9 e

=

10 . 10 − 9e−t

On the other hand, given y2 (0) = −1, we find C = 12 , and y2 (t) =

1 . 1 − 2e−t

y1 (t) =

10 . 10 − 9e−t

(b) From part (a), we have

Thus, y1 (t) = 5 when 5=

10 ; 10 − 9e−t

10 − 9e−t = 2;

9 so t = ln . 8

(c) From part (a), we have y2 (t) =

1 . 1 − 2e−t

Thus, y2 (t) becomes infinite when 1 − 2e−t = 0

or

t = ln 2.

13. A tissue culture grows until it has a maximum area of M cm2 . The area A(t) of the culture at time t may be modeled Reverse Logistic Equation Consider the logistic equation (with k, B > 0) by the differential equation √ −k P 1 − A P · dP = A dt =k A 1− 8 M B (a)k is Sketch the slope field of this equation. where a growth constant. /C), where C is a nonzero constant. Show that P(0) > B if C > 1 (b) The general solution is P(t) B/(1 −can ektbe 2 (a) By setting A = u , show that the = equation rewritten and 0 < P(0) < B if C < 0. (c) Show that Eq. (7) models an “extinction–explosion” That is, P(t) tends to zero if the initial populau2 ·u = 1 k 1population. tion satisfies 0 < P(0) B. M (d) Show that P = 0 is a stable equilibrium and P = B an unstable equilibrium. Then find the general solution using separation of variables. (b) Show that the general solution to Eq. (8) is A(t) = M

√ 2 Ce(k/ M)t − 1 √

Ce(k/ M)t + 1

SOLUTION

· · so that Eq. (8) becomes: (a) Let A = u 2 . This gives A = 2u u, u2 · 2u u = ku 1 − M 2 ·u = k 1 − u 2 M

Now, rewrite k u2 du = 1− dt 2 M

as

du = 12 k dt. 1 − u 2 /M

The partial fraction decomposition for the term on the left-hand side is √

1 M 1 1 = + , √ √ 2 1 − u 2 /M M +u M −u

9

S E C T I O N 10.3


567

so after integrating both sides, we obtain √ M M + u 1 ln √ = kt + C. M −u 2 2

√

Thus, √ √ √

M +u M −u

√

= Ce(k/ M)t √

√ u(Ce(k/ M)t + 1) = M(Ce(k/ M)t − 1) and u=

√

√

M

Ce(k/ M)t − 1 √

Ce(k/ M)t + 1

.

(b) Recall A = u 2 . Therefore, A(t) = M

√ 2 Ce(k/ M)t − 1 √

Ce(k/ M)t + 1

.

15. Show that if a tissue culture grows according to Eq. (8), then the growth rate reaches a maximum when A = M/3. Use the model of Exercise 13 to determine the area A(t) (t in hours) culture with initial size √ of a tissue A SOLUTION According to Equation (8),maximum the growtharea rateisofMthe=tissue is kgrowth A(1 − 2 and the M ). Therefore that the 5 cmculture constant is k = 0.06. Graph the A(0) = 0.2 cm2 , assuming solution using adgraphing √ utility. A 3 1 1 3A k A 1− = k A−1/2 − k A1/2 /M = k A−1/2 1 − =0 dA M 2 2 2 M when A = M/3. Because the growth rate is zero for A = 0 and for A = M and is positive for 0 < A < M, it follows that the maximum growth rate occurs when A = M/3. In 1751, Benjamin Franklin predicted that the U.S. population P(t) would increase with growth constant k = Further Insights and Challenges 0.028 yr−1 . According to the census, the U.S. population was 5 million in 1800 and 76 million in 1900. Assuming

17. Let y(t)growth be a solution logistic logistic with k of =the 0.028, whatequation is the predicted carrying capacity for the U.S. population? Hint: Use Eqs. (3)

and (4) to show that y dy = ky 1 − 10 dtP(t) A P0 kt = e − A Rule P0 − (a) Differentiate Eq. (10) with respect to t and useP(t) the Chain to A show that

2y d2 y 2y 1 − y 1 − = k A A dt 2 (b) Show that y(t) is concave up if 0 < y < A/2 and concave down if A/2 < y < A. (c) Show that if 0 < y(0) < A/2, then y(t) has a point of inflection at y = A/2 (Figure 5). (d) Assume that 0 < y(0) < A/2. Find the time t when y(t) reaches the inflection point. SOLUTION

(a) The derivative of Eq. (10) with respect to t is

2kyy y y 2y 2y 2y y = ky − = ky 1 − =k 1− ky 1 − = k2 y 1 − 1− . A A A A A A (b) If 0 < y < A/2, 1 − Ay and 1 − 2y A are both positive, so y > 0. Therefore, y is concave up. If A/2 < y < A, 1 − Ay > 0, but 1 − 2y A < 0, so y < 0, so y is concave down. (c) If y0 < A, y grows and lim y(t) = A. If 0 < y < A/2, y is concave up at first. Once y passes A/2, y becomes t→∞

concave down, so y has an inflection point at y = A/2. (d) The general solution to Eq. (10) is y=

A ; 1 − e−kt /C

thus, y = A/2 when A A = 2 1 − e−kt /C

568

C H A P T E R 10


1 − e−kt /C = 2 1 t = − ln(−C) k Now, C = y0 /(y0 − A), so 1 1 A − y0 y0 t = − ln = ln . k A − y0 k y0 Let

10.4 First-Order Linear Equations

y=

A 1 − e−kt /C

Preliminary Questions 1. Which of the following are first-order linear equations? · thex 2general (a) be y + y = 1 nonequilibrium solution of y = ky(1 − y/A) (b) with y + kx y>2 0. = If 1 y(t) has a vertical asymptote at t = tb , that is, if lim y(t) = ±∞, we say that the solution “blows up” at t = t . b (c) x 5 y + y t→t = ebx− (d) x 5 y + y = e y (a) Show that if 0 < y(0) < A, then y does not blow up at any time tb . equations. The equation in (b) is not linear SOLUTION The equations in (a) and (c) are first-order linear differential , which is negative hence does not iscorrespond a real (b) of Show if y(0) then yterm blows a time tbside because the that y 2 factor in > theA,second on up theatleft-hand of the equation;(and the equation in (d) not linear to because time). of the e y term on the right-hand side of the equation. (c) Show that y blows up at some positive time t if and only if y(0) < 0 (and hence does not correspond to a real 2. If α (x) is an integrating factor for y + A(x)y = bB(x), then α (x) is equal to (choose the correct answer): population). (a) B(x) (b) α (x) A(x) (d) α (x)B(x) (c) α (x) A (x) SOLUTION

The correct answer is (b): α (x) A(x).

Exercises 1. (a) (b) (c) (d)

Consider y + x −1 y = x 3 . Verify that α (x) = x is an integrating factor. Show that when multiplied by α (x), the differential equation can be written (x y) = x 4 . Conclude that x y is an antiderivative of x 4 and use this information to find the general solution. Find the particular solution satisfying y(1) = 0.

SOLUTION

(a) The equation is of the form y + A(x)y = B(x) for A(x) = x −1 and B(x) = x 3 . By Theorem 1, α (x) is defined by !

α (x) = e A(x) d x = eln x = x. (b) When multiplied by α (x), the equation becomes: x y + y = x 4. Now, x y + y = x y + (x) y = (x y) , so (x y) = x 4 . 5 (c) Since (x y) = x 4 , (x y) = x5 + C and

y=

C x4 + 5 x

(d) If y(1) = 0, we find 0=

1 +C 5

so

−

1 = C. 5

The solution, therefore, is y=

Consider

dy + 2y = e−3t . dt

1 x4 − . 5 5x

First-Order Linear Equations

S E C T I O N 10.4 2

3. Let α (x) = e x . Verify the identity (α (x)y) = α (x)(y + 2x y) and explain how it is used to find the general solution of y + 2x y = x. SOLUTION

2

Let α (x) = e x . Then (α (x)y) = (e x y) = 2xe x y + e x y = e x 2

2

2

2

2x y + y = α (x) y + 2x y .

If we now multiply both sides of the differential equation y + 2x y = x by α (x), we obtain

α (x)(y + 2x y) = x α (x) = xe x . 2

But α (x)(y + 2x y) = (α (x)y) , so by integration we find " 2 1 2 α (x)y = xe x d x = e x + C. 2 Finally, y(x) =

2 1 + Ce−x . 2

In Exercises 5–18, find the general solution of the first-order linear differential equation. Find the solution of y − y = e2x , y(0) = 1. 5. x y + y = x SOLUTION

Rewrite the equation as 1 y = 1, x

y +

which is in standard linear form with A(x) = 1x and B(x) = 1. By Theorem 1, the integrating factor is !

α (x) = e A(x) d x = eln x = x. When multiplied by the integrating factor, the rewritten differential equation becomes x y + y = x

(x y) = x.

or

Integration of both sides now yields xy =

1 2 x + C. 2

Finally, y(x) =

C 1 x+ . 2 x

7. 3x y − y = x −12 xy − y = x − x SOLUTION Rewrite the equation as y −

1 1 y = 2, 3x 3x

which is in standard form with A(x) = − 13 x −1 and B(x) = 13 x −2 . By Theorem 1, the integrating factor is !

α (x) = e A(x) d x = e−(1/3) ln x = x −1/3 . When multiplied by the integrating factor, the rewritten differential equation becomes x −1/3 y −

1 −4/3 1 = x −7/3 x 3 3

or

(x −1/3 y) =

Integration of both sides now yields 1 x −1/3 y = − x −4/3 + C. 4

1 −7/3 . x 3

569

570

C H A P T E R 10


Finally, 1 y(x) = − x −1 + C x 1/3 . 4 9. y + 3x −1 y = x + x −1 y + xy = x SOLUTION This equation is in standard form with A(x) = 3x −1 and B(x) = x + x −1 . By Theorem 1, the integrating factor is !

−1 α (x) = e 3x = e3 ln x = x 3 .

When multiplied by the integrating factor, the original differential equation becomes x 3 y + 3x 2 y = x 4 + x 2

or

(x 3 y) = x 4 + x 3 .

Integration of both sides now yields x3 y =

1 5 1 3 x + x + C. 5 3

Finally, y(x) =

1 2 1 x + + C x −3 . 5 3

y−x 11. x y = y + x −1 y = cos(x 2 ) SOLUTION Rewrite the equation as 1 y = −1, x

y −

which is in standard form with A(x) = − 1x and B(x) = −1. By Theorem 1, the integrating factor is !

α (x) = e −(1/x) d x = e− ln x = x −1 . When multiplied by the integrating factor, the rewritten differential equation becomes 1 1 1 1 1 y − 2y =− or y =− . x x x x x Integration on both sides now yields 1 y = − ln x + C. x Finally, y(x) = −x ln x + C x. 13. y + y = e x 3y x y = x 3 − SOLUTION This equation is in standard form with A(x) = 1 and B(x) = e x . By Theorem 1, the integrating factor is x !

α (x) = e 1 d x = e x . When multiplied by the integrating factor, the original differential equation becomes e x y + e x y = e2x

or

(e x y) = e2x .

Integration on both sides now yields ex y =

1 2x e + C. 2

Finally, y(x) = 15. y + (tan x)y = cos x y + (sec x)y = cos x

1 x e + Ce−x . 2



571

This equation is in standard form with A(x) = tan x and B(x) = cos x. By Theorem 1, the integrating

factor is !

α (x) = e tan x d x = eln sec x = sec x. When multiplied by the integrating factor, the original differential equation becomes sec x y + sec x tan x y = 1

(y sec x) = 1.

or

Integration on both sides now yields y sec x = x + C. Finally, y(x) = x cos x + C cos x. (ln x)y = x x 17. y −2x e y = 1 − ex y SOLUTION This equation is in standard form with A(x) = − ln x and B(x) = x x . By Theorem 1, the integrating factor is ! ex α (x) = e − ln x d x = e x−x ln x = x . x

When multiplied by the integrating factor, the original differential equation becomes x −x e x y − (ln x)x −x e x y = e x

or

(x −x e x y) = e x .

Integration on both sides now yields x −x e x y = e x + C. Finally, y(x) = x x + C x x e−x . In Exercises 19–26, solve the initial value problem. y + y = cos x 19. y + 3y = e2x ,

y(0) = −1

First, we find the general solution of the differential equation. This linear equation is in standard form with A(x) = 3 and B(x) = e2x . By Theorem 1, the integrating factor is

SOLUTION

α (x) = e3x . When multiplied by the integrating factor, the original differential equation becomes (e3x y) = e5x . Integration on both sides now yields (e3x y) =

1 5x e + C; 5

hence, y(x) =

1 2x e + Ce−3x . 5

The initial condition y(0) = −1 allows us to determine the value of C: −1 =

1 +C 5

so

6 C =− . 5

The solution to the initial value problem is therefore y(x) = 1 , y(1) 21. y +x y + y y==e xx,−2y(1) = 3= 2 x +1

1 2x 6 −3x . e − e 5 5

572

C H A P T E R 10


First, we find the general solution of the differential equation. This linear equation is in standard form with 1 and B(x) = x −2 . By Theorem 1, the integrating factor is A(x) = x+1

SOLUTION

!

α (x) = e 1/(x+1) d x = eln(x+1) = x + 1. When multiplied by the integrating factor, the original differential equation becomes ((x + 1)y) = x −1 + x −2 . Integration on both sides now yields (x + 1)y = ln x − x −1 + C; hence, y(x) =

1 1 C + ln x − . x +1 x


1 (C − 1) 2

so

C = 5.

The solution to the initial value problem is therefore y(x) =

1 1 5 + ln x − . x +1 x

= (cos x)y + 1, y( π ) = 0 23. (sin x)y y + y = sin x, y(0) = 1 4 SOLUTION First, we find the general solution of the differential equation. Rewrite the equation as

y − (cot x)y = csc x, which is in standard form with A(x) = − cot x and B(x) = csc x. By Theorem 1, the integrating factor is !

α (x) = e − cot x d x = e− ln sin x = csc x. When multiplied by the integrating factor, the rewritten differential equation becomes (csc x y) = csc2 x. Integration on both sides now yields (csc x)y = − cot x + C; hence, y(x) = − cos x + C sin x. The initial condition y(π /4) = 0 allows us to determine the value of C: √ √ 2 2 +C so C = 1. 0=− 2 2 The solution to the initial value problem is therefore y(x) = − cos x + sin x. 25. y + (tanh x)y = 1, y(0) = 3 y + (sec t)y = sec t, y(0) = 1 and y π = 1 SOLUTION First, we find the general solution4 of the differential equation. This equation is in standard form with A(x) = tanh x and B(x) = 1. By Theorem 1, the integrating factor is !

α (x) = e tanh x d x = eln cosh x = cosh x. When multiplied by the integrating factor, the original differential equation becomes (cosh x y) = cosh x. Integration on both sides now yields (cosh x y) = sinh x + C;

S E C T I O N 10.4


573

hence, y(x) = tanh x + C sech x. The initial condition y(0) = 3 allows us to determine the value of C: 3 = C. The solution to the initial value problem is therefore y(x) = tanh x + 3 sech x. 27. Find the general solution of y + ny = emx for all m, n. Note: The case m = −n must be treated separately. x 1 y + For any y m, = n, Theorem y(1)us = the 0 formula for α (x): SOLUTION 1, gives 1 + x2 (1 + x 2 )3/2 !

α (x) = e n d x = enx . When multiplied by the integrating factor, the original differential equation becomes (enx y) = e(m+n)x . If m = −n, integration on both sides yields enx y =

1 e(m+n)x + C, m+n

y(x) =

1 emx + Ce−nx . m+n

so

However, if m = −n, then m + n = 0 and the equation reduces to (enx y) = 1, so integration yields enx y = x + C

or

y(x) = (x + C)e−nx .

29. Repeat Exercise 28(a), assuming that water is pumped out at a rate of 20 gal/min. What is the limiting salt concenA 200-gal tank contains 100 gal of water with a salt concentration of 0.1 lb/gal. Water with a salt concentration tration for large t? of 0.4 lb/gal flows into the tank at a rate of 20 gal/min. The fluid is mixed instantaneously, and water is pumped out SOLUTION waterLet flows the amount tank at the same ratetank as water flows at a rate ofBecause 10 gal/min. y(t)into be the of salt in the at time t. out of the tank, the amount of water in the tank remains a constant 100 gallons. Now, let y(t) be the amount of salt in the tank (in pounds) at any time t (in (a) Set up and solve the differential equation for y(t). minutes). The net flow of salt into the tank t is (b) What is the salt concentration when the tank overflows? gal lb gal y lb 1 dy = salt rate in − salt rate out = 20 0.4 − 20 = 8 − y. dt min gal min 100 gal 5 Rewriting this linear equation in standard form, we have 1 dy + y = 8, dt 5 so A(t) = 15 and B(t) = 8. By Theorem 1, the integrating factor is !

α (t) = e (1/5) dt = et/5 . When multiplied by the integrating factor, the rewritten differential equation becomes (et/5 y) = 8et/5 . Integration on both sides now yields et/5 y = 40et/5 + C; hence, y(t) = 40 + Ce−t/5 . The initial condition y(0) = 10 allows us to determine the value of C: 10 = 40 + C

so

C = −30.

574

C H A P T E R 10


The solution to the initial value problem is therefore y(t) = 40 − 30e−t/5 . 40 = 0.4 pounds From here, we see that y(t) → 40 as t → ∞, so that the limiting concentration of salt in the tank is 100 per gallon. This is completely intuitive; as time goes on, the concentration in the tank should resemble the concentration pouring into it.

20 31. Water flowsExercise into a tank at the variablethat rate water Rin =is pumped 5 gal/min. Let gal/min and out atofthe Routis=the Repeat 28(a), assuming out at a rate 25constant gal/min.rate What limiting salt 1+t concentration for t large? V (t) be the volume of water in the tank at time t. (a) Set up a differential equation for V (t) and solve it with the initial condition V (0) = 100. (b) Find the maximum value of V . Plot V (t) and estimate the time t when the tank is empty. (c) SOLUTION

(a) The rate of change of the volume of water in the tank is given by 20 dV = Rin − Rout = − 5. dt 1+t Because the right-hand side depends only on the independent variable t, we integrate to obtain V (t) = 20 ln(1 + t) − 5t + C. The initial condition V (0) = 100 allows us to determine the value of C: 100 = 20 ln 1 − 0 + C

C = 100.

so

Therefore V (t) = 20 ln(1 + t) − 5t + 100. (b) Using the result from part (a), 20 dV = −5=0 dt 1+t when t = 3. Because ddtV > 0 for t < 3 and ddtV < 0 for t > 3, it follows that V (3) = 20 ln 4 − 15 + 100 ≈ 112.726 gal is the maximum volume. (c) V (t) is plotted in the figure below at the left. On the right, we zoom in near the location where the curve crosses the t-axis. From this graph, we estimate that the tank is empty after roughly 34.25 minutes. 120 100 80 60 40

32

34

36

38

20 10

20

30

40

In Exercises 33–35, consider a series circuit (Figure 4) consisting of a resistor of R ohms, an inductor of L henries, and A stream feeds into a lake at a rate of 1,000 m3 /day. The stream is polluted with a toxin whose concentravariable voltage source of V (t) volts (time t in seconds). The current through the circuit I (t) (in amperes) satisfies the 3 6 3 tion is 5 g/m . Assume that the lake has volume 10 m and that water flows out of the lake at the same rate differential equation: 3 of 1,000 m /day. Set up a differential equation for the concentration c(t) of toxin in the lake and solve for c(t), assuming that c(0) = 0. What is the limiting concentration for dI R 1 large t? 12 + I = V (t) dt L L R

V(t)

L

FIGURE 4 RL circuit.

S E C T I O N 10.4


575

33. Find the solution to Eq. (12) with initial condition I (0) = 0, assuming that R = 100 , L = 5 H, and V (t) is constant with V (t) = 10 V. SOLUTION

If R = 100, V (t) = 10, and L = 5, the differential equation becomes dI + 20I = 2, dt

which is a linear equation in standard form with A(t) = 20 and B(t) = 2. The integrating factor is α (t) = e20t , and when multiplied by the integrating factor, the differential equation becomes (e20t I ) = 2e20t . Integration of both sides now yields e20t I =

1 20t e + C; 10

hence, I (t) =

1 + Ce−20t . 10

The initial condition I (0) = 0 allows us to determine the value of C: 1 +C 10

0=

so

C =−

1 . 10

Finally, I (t) =

1

1 − e−20t . 10

Assume that V (t) = V is constant and I (0) = 0. Assume that R = 110 , L = 10 H, and V (t) = e−t . Solve for I (t). (a) Solve Eq. (12) with initial condition I (0) = 0. Show that lim I (t) = V /R and that I (t) reaches approximately 63% of its limiting value after L/R seconds. t→∞ (b) Use a computer algebra system to sketch the graph of the solution for 0 ≤ t ≤ 3. (c) How long does it take for I (t) to reach 90% of its limiting value if R = 500 , L = 4 H, and V = 20 V? (c) Calculate tm and I (tm ), where tm is the time at which I (t) has a maximum value.

35. (a) (b)

SOLUTION

(a) The equation R 1 dI + I = V dt L L is a linear equation in standard form with A(t) = LR and B(t) = L1 V (t). By Theorem 1, the integrating factor is !

α (t) = e (R/L) dt = e(R/L) t . When multiplied by the integrating factor, the original differential equation becomes (e(R/L) t I ) = e(R/L) t

V . L

Integration on both sides now yields (e(R/L) t I ) =

V (R/L) t + C; e R

hence, I (t) =

V + Ce−(R/L) t . R

The initial condition I (0) = 0 allows us to determine the value of C: 0=

V +C R

so

C =−

Therefore the current is given by I (t) =

V

1 − e−(R/L) t . R

V . R

576

C H A P T E R 10


(b) As t → ∞, e−(R/L) t → 0, so I (t) → VR . Moreover, when t = (L/R) seconds, we have V

L V

V 1 − e−(R/L) (L/R) = 1 − e−1 ≈ 0.632 . I = R R R R (c) Using the results from part (a) and part (b), I (t) reaches 90% of its limiting value when 9 = 1 − e−(R/L) t , 10 or when t=

L ln 10. R

With L = 4 and R = 500, this takes approximately 0.0184 seconds. Continuing with thefilled previous exercise, let Tank 2 be another tank filled with V gallons of water. Assume 37. Tank 1 in Figure 5 is with V 1 gallons of water. Water flows into the tank at a2 rate of R gal/min and out that the inky water from Tank 1 empties into Tank 2 as and the leaves 2 att the through the bottom at the same rate R. Suppose thatinIFigure gallons5,ofmixes blue instantaneously, ink are dumped into tankTank at time = 0same and inkbe in the Tank 2 at time t. in the tank at time t. rate mixed R. Let instantaneously. y2 (t) be the amount quantity of ink Let yof (t) 1

d y1 R =− y . dtR(gal/min)V1 1 (b) Solve for y1 (t) with V1 = 100, R = 10, and I = 2. (a) Explain why y1 satisfies the differential equation

Tank 1

R(gal/min)

Tank 2

R (gal/min)

FIGURE 5

(a) Explain why y2 satisfies the differential equation d y2 =R dt

y y1 − 2 V1 V2

(b) Use the solution to Exercise 36 to solve for y2 (t) if V1 = 100, V2 = 200, R = 10, I = 2, and y2 (0) = 0. Plot the solution for 0 ≤ t ≤ 120. (c) (d) Find the maximum ink concentration in Tank 2. SOLUTION

(a) The water flowing into Tank 2 has an ink concentration of y1 (t)/V1 and the water flowing out from the tank has an ink concentration of y2 (t)/V2 . Thus, y y1 d y2 − 2 . =R dt V1 V2 (b) With V1 = 100, R = 10 and I = 2, we know from the previous exercise that y1 (t) = 2e−t/10 . Substituting this expression and the given parameter values into the differential equation obtained in part (a), we have d y2 1 1 2 −t/10 1 − = 10 e y2 = e−t/10 − y . dt 100 200 5 20 2 Hence, 1 1 d y2 + y = e−t/10 . dt 20 2 5 The integrating factor for this linear equation is α (t) = et/20 so we get

1 et/20 y2 = e−t/20 , 5 and et/20 y2 = −4e−t/20 + C.

S E C T I O N 10.4


577

Thus, y2 (t) = −4e−t/10 + Ce−t/20 . The initial condition y2 (0) = 0 allows us to determine C = 4; consequently,

y2 (t) = 4 e−t/20 − e−t/10 . (c) A plot of y2 (t) is shown below. y 1.0 0.8 0.6 0.4 0.2 t 0

20 40 60 80 100 120

(d) Using the result of part (b), we find that d y2 1 −t/20 1 −t/10 − e =4 e =0 dt 10 20 when t = 20 ln 2. Thus, the maximum ink concentration in Tank 2 is y2 (20 ln 2) = 4

1 1 − 2 4

= 1.

Further Insights and Challenges 39. Use the Fundamental Theorem of Calculus and the Product Rule to verify directly that for any x 0 , the function Let α (x) be an integrating factor for y + A(x)y = B(x). The differential equation y + A(x)y = 0 is called the " x associated homogeneous equation. f (x) = α (x)−1 α (t)B(t) dt 1 x0 (a) Show that is a solution of the associated homogeneous equation. α (x) is a solution of the initial value problem C is also a solution for (b) Show that if y = f (x) is a particular solution of y + A(x)y = B(x), then f (x) + α (x) y(x 0 ) = 0, y + A(x)y = B(x), any constant C. where α (x) is an integrating factor [a solution to Eq. (3)]. SOLUTION

Remember that α (x) = A(x)α (x). Now, let " x 1 y(x) = α (t)B(t) dt. α (x) x0

Then, y(x0 ) =

" x0 1 α (t)B(t) dt = 0, α (x) x0

and " " x α (x) A(x) x α (t)B(t) dt + B(x) + α (t)B(t) dt α (x) x0 (α (x))2 x0 " x A(x) A(x) α (t)B(t) dt = B(x). = B(x) + − + α (x) α (x) x0

y + A(x)y = −

Transient Currents Suppose the circuit described by Eq. (12) is driven by a sinusoidal voltage source V (t) = V sin ω t (where V and ω are constant). (a) Show that I (t) =

2

2 2

V (R sin ω t − L ω cos ω t) + Ce−(R/L) t R2 + L 2ω 2

578

C H A P T E R 10


CHAPTER REVIEW EXERCISES 1. Which of the following differential equations are linear? Determine the order of each equation. (b) y = x 5 − 3x 4 y (a) y = y 5 − 3x 4 y √ (c) y = y − 3x y (d) sin x · y = y − 1 SOLUTION

(a) y 5 is a nonlinear term involving the dependent variable, so this is not a linear equation; the highest order derivative that appears in the equation is a first derivative, so this is a first-order equation. (b) This is linear equation; the highest order derivative that appears in the equation is a first derivative, so this is a first-order equation. √ (c) y is a nonlinear term involving the dependent variable, so this is not a linear equation; the highest order derivative that appears in the equation is a third derivative, so this is a third-order equation. (d) This is linear equation; the highest order derivative that appears in the equation is a second derivative, so this is a second-order equation. In Exercises solve Find 3–6, a value of cusing suchseparation that y = x of − variables. 2 + ecx is a solution of 2y + y = x. dy 3. = t 2 y −3 dt SOLUTION

Rewrite the equation as y 3 d y = t 2 dt.

Upon integrating both sides of this equation, we obtain: " " y 3 d y = t 2 dt t3 y4 = + C. 4 3 Thus, y=±

1/4 4 3 t +C , 3

where C is an arbitrary constant. dy 5. x x yy − y==11− x 2 dx SOLUTION

Rewrite the equation as dx dy = . 1+ y x

upon integrating both sides of this equation, we obtain " " dy dx = 1+y x ln |1 + y| = ln |x| + C. Thus, y = −1 + Ax, where A = ±eC is an arbitrary constant. In Exercises 7–10,2 solve the initial value problem using separation of variables. xy y = 2 2 1 = π 7. y = cos xx, + y(0) 4 SOLUTION First, we find the general solution of the differential equation. Because the variables are already separated, we integrate both sides to obtain " " 1 1 x sin 2x y = cos2 x d x = + cos 2x d x = + + C. 2 2 2 4


579

The initial condition y(0) = π4 allows us to determine C = π4 . Thus, the solution is: x π sin 2x + + . 2 4 4

y(x) =

9. y = x y 2 , y(1) = 2 π y = cos2 y, y(0) = SOLUTION First, we find the4general solution of the differential equation. Rewrite dy = x y2 dx Upon integrating both sides of this equation, we find "

dy = x d x. y2

as

dy = y2

−

" x dx

1 1 = x 2 + C. y 2

Thus, 1

y=−1

2 2x + C

.

The initial condition y(1) = 2 allows us to determine the value of C: 1

2=−1

2 2 ·1 +C 1 + 2C = −1

=−

2 1 + 2C

C = −1 Hence, the solution to the initial value problem is y=−1

1

2 2x − 1

2 =− 2 . x −2

11. Figure 1 shows the slope field for y· = sin y + t y. Sketch the graphs of the solutions with the initial conditions = 1, 2 = −1. y(0) = x1yy , y(0) = 0,y(3) and = y(0) y 2 1 t

0 −1 −2 −2 −1

0

1

2

FIGURE 1 SOLUTION

Which of the equations (i)–(iii) corresponds to the slope field in Figure 2? (i) y· = 1 − y 2

580

C H A P T E R 10


13. Let y(t) be the solution to the differential equation with slope field as shown in Figure 2, satisfying y(0) = 0. Sketch the graph of y(t). Then use your answer to Exercise 12 to solve for y(t). y 2 1 t

0 −1 −2 −2 −1

0

1

2

FIGURE 2

As explained in the previous exercise, the slope field in Figure 2 corresponds to the equation y· = 1 + y 2 . The graph of the solution satisfying y(0) = 0 is: SOLUTION

To solve the initial value problem y· = 1 + y 2 , y(0) = 0, we first find the general solution of the differential equation. Separating variables yields: dy = dt. 1 + y2 Upon integrating both sides of this equation, we find tan−1 y = t + C

or

y = tan(t + C).

The initial condition gives C = 0, so the solution is y = tan x. 15. Let y(t) be the solution of (x 3 ·+ 1) y·2 = y satisfying y(0) = 1. Compute approximations to y(0.1), y(0.2), and t satisfying y(2) = 1. Carry out Euler’s Method with time step h = 0.05 Let y(t) be the solution of time 4 y =step y h+= y(0.3) using Euler’s Method with 0.1. for n = 6 steps. y · SOLUTION Rewriting the equation as y = we have F(x, y) = y . Using Euler’s Method with x = 0, y = 1 and h = 0.1, we calculate

x 3 +1

x 3 +1

1 = 1.1 y(0.1) ≈ y1 = y0 + h F(x 0 , y0 ) = 1 + 0.1 · 3 0 +1 y(0.2) ≈ y2 = y1 + h F(x 1 , y1 ) = 1.209890 y(0.3) ≈ y3 = y2 + h F(x 2 , y2 ) = 1.329919 In Exercises 16–19, solve using the method of integrating factors. dy y d=y + x, 2 y(1) = 3 dx x= y + t , y(0) = 4 dt SOLUTION First, we find the general solution of the differential equation. Rewrite the equation as

17.

y −

1 y = x, x

0

0


581

which is in standard form with A(x) = − 1x and B(x) = x. The integrating factor is ! 1 1 α (x) = e − x d x = e− ln x = . x

When multiplied by the integrating factor, the rewritten differential equation becomes 1 y = 1. x Integration on both sides now yields 1 y = x + C; x hence, y(x) = x 2 + C x. The initial condition y(1) = 3 allows us to determine the value of C: 3 = 1+C

so

C = 2.

The solution to the initial value problem is then y = x 2 + 2x. 19. y + 2y = 1 + e−x , y(0) = −4 dy = y − 3t, y(−1) = 2 −x SOLUTION dt The equation is already in standard form with A(x) = 2 and B(x) = 1 + e . The integrating factor is !

α (x) = e 2 d x = e2x . When multiplied by the integrating factor, the original differential equation becomes (e2x y) = e2x + e x . Integration on both sides now yields e2x y =

1 2x e + e x + C; 2

hence, y(x) =

1 + e−x + Ce−2x . 2

The initial condition y(0) = −4 allows us to determine the value of C: −4 =

1 +1+C 2

so

C =−

11 . 2

The solution to the initial value problem is then y(x) =

11 1 + e−x − e−2x . 2 2

In Exercises 20–27, solve using the appropriate method. x)y = cos2 x, y(π ) = 2 21. y + 2(tan x y = x 2 + 1, y(1) = 10 SOLUTION First, we find the general solution of the differential equation. As this is a first order linear equation with A(x) = tan x and B(x) = cos2 x, we compute the integrating factor !

!

α (x) = e A(x) d x = e tan x d x = e− ln cos x =

1 . cos x

When multiplied by the integrating factor, the original differential equation becomes 1 y = cos x. cos x

582

C H A P T E R 10


Integration on both sides now yields 1 y = sin x + C; cos x hence, y(x) = sin x cos x + C cos x =

1 sin 2x + C cos x. 2

The initial condition y(π ) = 2 allows us to determine the value of C: 2 = 0 + C(−1)

so

C = −2.

The solution to the initial value problem is then y=

1 sin 2x − 2 cos x. 2

= t, y(1) = −3 23. (y − 1)y x y = 2y + x − 1, y( 32 ) = 9 SOLUTION First, we find the general solution of the differential equation. This is a separable equation that we rewrite as

(y − 1) d y = t dt. Upon integrating both sides of this equation, we find " " (y − 1) d y = t dt 1 y2 − y = t2 + C 2 2 y 2 − 2y + 1 = t 2 + C (y − 1)2 = t 2 + C y(t) = ± t 2 + C + 1 To satisfy the initial condition y(1) = −3 we must choose the negative square root; moreover, √ so C = 15. −3 = − 1 + C + 1 The solution to the initial value problem is then y(t) = − t 2 + 15 + 1 dw√ 1 + w 2 , w(1) t2 = 1 =k d x y + 1x y = yte , y(0) = 1. SOLUTION First, we find the general solution of the differential equation. This is a separable equation that we rewrite as

25.

k dw = d x. x 1 + w2 Upon integrating both sides of this equation, we find " " dw k = dx 2 x 1+w tan−1 w = k ln x + C w(x) = tan(k ln x + C). Because the initial condition is specified at x = 1, we are interested in the solution for x > 0; we can therefore omit the absolute value within the natural logarithm function. The initial condition w(1) = 1 allows us to determine the value of C: π 1 = tan(k ln 1 + C) so C = tan−1 1 = . 4 The solution to the initial value problem is then

π . w = tan k ln x + 4


583

y 27. y + =3ysin −x1 x y + =t +2 t 1 SOLUTION This is a first order linear equation with A(x) = x and B(x) = sin x. The integrating factor is !

α (x) = e A(x) d x = eln x = x. When multiplied by the integrating factor, the original differential equation becomes (x y) = x sin x. Integration on both sides (integration by parts is needed for the integral on the right-hand side) now yields x y = −x cos x + sin x + C; hence, y(x) = − cos x +

C sin x + . x x

dy 29. LetState A andwhether B be constants. Prove that if A > can 0, then solution of separation + Ay = approachthe the method same limit t → ∞. the differential equation be all solved using of Bvariables, of as integrating dt factors, both, or neither. SOLUTION (a) y = This y + xis2 a linear first-order equation in standard form (b)with x y integrating = y + 1 factor ! 2 α (t) = e A dt (d) = e xAty. = y

(c) y = y 2 + x 2

When multiplied by the integrating factor, the original differential equation becomes (e At y) = Be At . Integration on both sides now yields e At y =

B At e + C; A

y(t) =

B + Ce−At . A

hence,

Because A > 0,

lim y(t) = lim

t→∞

t→∞

B + Ce−At A

=

B . A

We conclude that if A > 0, all solutions approach the limit BA as t → ∞. 31. Find the solution of the logistic equation y· = 0.4y(4 − y) satisfying y(0) = 8. The trough in Figure 3 is filled with water. At time t = 0 (in seconds), water begins leaking through a circular SOLUTION can write the given equation asthe water height at time t. Find a differential equation for y(t) and solve hole at theWe bottom of radius 3 in. Let y(t) be it to determine when the trough will be half empty.

y . y· = 1.6y 1 − 4 This is a logistic equation with k = 1.6 and A = 4. Therefore, y(t) =

A 1 − e−kt /C

=

4 1 − e−1.6t /C

.


4 1 − C1

;

1−

1 1 = ; C 2

so C = 2.

Thus, y(t) =

8 4 = . 1 − e−1.6t /2 2 − e−1.6t

33. Suppose that y = ky(1 − y/8) has a solution satisfying y(0) = 12 and y(10) = 24. Find k. Let y(t) be the solution of y· = 0.3y(2 − y) with y(0) = 1. Determine lim y(t) without solving for y explicitly. t→∞

584

C H A P T E R 10


The given differential equation is a logistic equation with A = 8. Thus, y(t) =

8 . 1 − e−kt /C

The initial condition y(0) = 12 allows us to determine the value of C: 12 =

8 ; 1 − C1

1 2 = ; C 3

1−

so C = 3.

Hence, y(t) =

24 8 . = 1 − e−kt /3 3 − e−kt

Now, the condition y(10) = 24 allows us to determine the value of k: 24 =

24 3 − e−10k

3 − e−10k = 1 k=−

ln 2 ≈ −0.0693. 10

−1 When A rabbit on a deserted island increases growth grows rate k = 0.12 months 35. A lake has a population carrying capacity of 1,000 fish. Assumeexponentially that the fish with population logistically with .growth −1 the population reaches 150 rabbits (say, at time t = 0), hunters begin killing the rabbits at a rate of r rabbits per month.is constant k = 0.2 days . How many days will it take for the population to reach 900 fish if the initial population 20 fish? (a) Find a differential equation satisfied by the rabbit population P(t). (b) How large can r be without the rabbit population becoming extinct? SOLUTION

(a) The rabbit population P(t) obeys the differential equation dP = 0.12P − r, dt where the term 0.12P accounts for the exponential growth of the population and the term −r accounts for the rate of decline in the rabbit population due to hunting. (b) Rewrite the linear differential equation from part (a) as dP − 0.12P = −r, dt which is in standard form with A = −0.12 and B = −r . The integrating factor is !

!

α (t) = e A dt = e −0.12 dt = e−0.12t . When multiplied by the integrating factor, the rewritten differential equation becomes (e−0.12t P) = −r e−0.12t . Integration on both sides now yields e−0.12t P =

r −0.12t + C; e 0.12

hence, P(t) =

r + Ce0.12t . 0.12


r +C 0.12

C = 150 −

so

r . 0.12

The solution to the initial value problem is then

r r 0.12t e + . P(t) = 150 − 0.12 0.12 r < 0, then lim P(t) = −∞, and the population becomes extinct. Therefore, in order for the Now, if 150 − 0.12 t→∞

population to survive, we must have 150 −

r ≥0 0.12

or

r ≤ 18.


585

We conclude that the maximum rate at which the hunters can kill the rabbits without driving the rabbits to extinction is r = 18 rabbits per month. 37. A tank contains 100 gal of pure water. Water is pumped out at1a−rate y 2 of 20 gal/min, and polluted water with −1 x + is the in general solution y = Let y(t) . be Then the addition Show that y =ofsin(tan a toxin concentration 0.1 lb/gal is C) pumped at a rate of 15 of gal/min. theuse amount of toxinformula presentfor in the 1 + x2 tank at time t. (cos C)x + sin C sine function to show that the generalbysolution . (a) Find a differential equation satisfied y(t). may be written y = 1 + x2 (b) Solve for y(t). Plot y(t) and find the time t at which the amount of toxin is maximal. (c) SOLUTION

(a) Because water flows into the tank at a rate of 15 gallons per minute but flows out at a rate of 20 gallons per minute, there is a new outflow of 5 gallons per minute from the tank. Therefore, at any time t, there are 100 − 5t gallons of water in the tank. Now, to determine the differential equation satisfied by y(t), we employ the basic rate balance dy = toxin rate in − toxin rate out. dt The amount of toxin per minute coming into the tank is toxin rate in = concentration · water rate in = 0.1 · 15 = 1.5 To determine the toxin rate out we first compute the toxin concentration at time t: toxin concentration =

y(t) lbs of toxin = . gallons of water 100 − 5t

Since water flows out at a rate of 20 gallons per minute, toxin rate out = 20

4y y = . 100 − 5t 20 − t

Thus, the differential equation is 4y dy = 1.5 − . dt 20 − t (b) Rewrite the linear first-order differential equation found in part (a) as 4 3 dy + y= . dt 20 − t 2 The integrating factor for this equation is !

−1 α (t) = e4 (20−t) dt =

1 . (20 − t)4

When multiplied by the integrating factor, the rewritten differential equation becomes 3 1 y = . 4 (20 − t) 2(20 − t)4 Integration on both sides now yields 1 (20 − t)4

y=

1 2(20 − t)3

+ C;

hence, y(t) =

20 − t + C(20 − t)4 . 2

The initial condition y(0) = 0 allows us to determine the value of C: 0 = 10 + 204 C

so

C=−

1 . 16000

Therefore, y(t) =

20 − t (20 − t)4 − . 2 16000

lb . min

586

C H A P T E R 10


(c) A plot of y(t) is shown below. 5 4 3 2 1 5

10

15

20

To determine the time at which the amount of toxin is maximal, we note that dy 1 (20 − t)3 d 20 − t (20 − t)4 = − =− + =0 dt dt 2 16000 2 4000 when √ 3 t = 20 − 10 2 ≈ 7.4 minutes.

11 INFINITE SERIES 11.1 Sequences Preliminary Questions 1. What is a4 for the sequence an = n 2 − n? SOLUTION

Substituting n = 4 in the expression for an gives a4 = 42 − 4 = 12.

2. Which of the following sequences converge to zero? n2 (a) 2 (b) 2n n +1

(c)

−1 n 2

SOLUTION

(a) This sequence does not converge to zero: lim

n2

n→∞ n 2 + 1

= lim

x2

x→∞ x 2 + 1

= lim

1

x→∞ 1 + 1 x2

=

1 = 1. 1+0

(b) This sequence does not converge to zero: this is a geometric sequence with r = 2 > 1; hence, the sequence diverges to ∞. (c) Recall that if |an | converges to 0, then an must also converge to zero. Here, n n −1 = 1 , 2 2 which is a geometric sequence with 0 < r < 1; hence, ( 12 )n converges to zero. It therefore follows that (− 12 )n converges to zero. √ 3. Let an be the nth decimal approximation to 2. That is, a1 = 1, a2 = 1.4, a3 = 1.41, etc. What is lim an ? n→∞ √ SOLUTION lim an = 2. n→∞

4. Which sequence is defined recursively? (a) an = 2 + n −1

(b) bn =

4 + bn−1

SOLUTION

(a) an can be computed directly, since it depends on n only and not on preceding terms. Therefore an is defined explicitly and not recursively. (b) bn is computed in terms of the preceding term bn−1 , hence the sequence {bn } is defined recursively. 5. Theorem 5 says that every convergent sequence is bounded. Which of the following statements follow from Theorem 5 and which are false? If false, give a counterexample. (a) If {an } is bounded, then it converges. (b) If {an } is not bounded, then it diverges. (c) If {an } diverges, then it is not bounded. SOLUTION

(a) This statement is false. The sequence an = cos π n is bounded since −1 ≤ cos π n ≤ 1 for all n, but it does not converge: since an = cos n π = (−1)n , the terms assume the two values 1 and −1 alternately, hence they do not approach one value. (b) By Theorem 5, a converging sequence must be bounded. Therefore, if a sequence is not bounded, it certainly does not converge. (c) The statement is false. The sequence an = (−1)n is bounded, but it does not approach one limit.

588

C H A P T E R 11

INFINITE SERIES

Exercises 1. Match the sequence with the general term: a1 , a2 , a3 , a4 , . . .

General term

(a) 12 , 23 , 34 , 45 , . . .

(i) cos π n

(b) −1, 1, −1, 1, . . .

n! (ii) n 2

(c) 1, −1, 1, −1, . . .

(iii) (−1)n+1

(d) 12 , 24 , 68 , 24 16 . . .

(iv)

n n+1

SOLUTION

(a) The numerator of each term is the same as the index of the term, and the denominator is one more than the numerator; n , n = 1, 2, 3, . . . . hence an = n+1 (b) The terms of this sequence are alternating between −1 and 1 so that the positive terms are in the even places. Since cos π n = 1 for even n and cos π n = −1 for odd n, we have an = cos π n, n = 1, 2, . . . . (c) The terms an are 1 for odd n and −1 for even n. Hence, an = (−1)n+1 , n = 1, 2, . . . (d) The numerator of each term is n!, and the denominator is 2n ; hence, an = 2n!n , n = 1, 2, 3, . . . . In Exercises 3–10, calculate the first four terms of the following sequences, starting with n = 1. 1 for n = 1, 2, 3, . . . . Write out the first three terms of the following sequences. Let an = 2n − 1 2n 3. c(a) n =bn = a (b) cn = an+3 n! n+1 2 (c) dn = Setting an SOLUTION n = 1, 2, 3, 4 in the formula for cn gives (d) en = 2an − an+1 c1 =

21 2 = = 2, 1! 1

c2 =

22 4 = = 2, 2! 2

c3 =

23 8 4 = = , 3! 6 3

c4 =

24 16 2 = = . 4! 24 3

2 an+1 5. a1 = π n = 1 + an bn 3,= cos SOLUTION For n = 1, 2, 3 we have:

a2 = a1+1 = 1 + a12 = 1 + 32 = 10; a3 = a2+1 = 1 + a22 = 1 + 102 = 101; a4 = a3+1 = 1 + a32 = 1 + 1012 = 10,202. The first four terms of {an } are 3, 10, 101, 10,202. 1 1 1 7. cn = bn 1=+2 ++(−1)+n · · · + 2 3 n SOLUTION

c1 = 1; 3 1 = ; 2 2 3 1 11 1 1 ; c3 = 1 + + = + = 2 3 2 3 6 1 1 1 11 1 25 c4 = 1 + + + = + = . 2 3 4 6 4 12 c2 = 1 +

9. b1 = 2, b2 = 5, bn = bn−1 + 2bn−2 an = n + (n + 1) + (n + 2) + · · · + (2n) SOLUTION We need to find b3 and b4 . Setting n = 3 and n = 4 and using the given values for b1 and b2 we obtain: b3 = b3−1 + 2b3−2 = b2 + 2b1 = 5 + 2 · 2 = 9; b4 = b4−1 + 2b4−2 = b3 + 2b2 = 9 + 2 · 5 = 19. The first four terms of the sequence {bn } are 2, 5, 9, 19. cn = nth decimal approximation to e−1

S E C T I O N 11.1

Sequences

589

11. Find a formula for the nth term of the following sequence: 1 −1 1 2 3 4 (a) , , , ... (b) , , , . . . 1 8 27 6 7 8 SOLUTION

(a) The denominators are the third powers of the positive integers starting with n = 1. Also, the sign of the terms is alternating with the sign of the first term being positive. Thus, 1 (−1)1+1 ; a1 = 3 = 1 13

1 (−1)2+1 a2 = − 3 = ; 2 23

1 (−1)3+1 a3 = 3 = . 3 33

This rule leads to the following formula for the nth term: an =

(−1)n+1 . n3

(b) Assuming a starting index of n = 1, we see that each numerator is one more than the index and the denominator is four more than the numerator. Thus, the general term an is an =

n+1 . n+5

In Exercises 13–26, Theorem to determine the7.limit of the sequence or state that the sequence diverges. an = 41and lim bn = Determine: Suppose thatuse lim n→∞

n→∞

13. a(a) 4 (an + bn ) n = lim (b) n→∞ SOLUTION We have an = f (n) where f (x) = 4; thus, (c) lim 4bn (d) n→∞ lim an = lim f (x) = n→∞

x→∞

lim a 3 n→∞ n lim (a 2 − 2an bn ) n→∞ n

lim 4 = 4.

x→∞

9 15. an = 5 −3n2+ 1 bn = n 2n + 4 9 SOLUTION We have an = f (n) where f (x) = 5 − ; thus, x2 9 9 lim 5 − 2 = lim 5 − 2 = 5 − 0 = 5. n→∞ x→∞ n x 17. cn = −2−n 2n+1 bn = (−1) SOLUTION We have cn = f (n) where f (x) = −2−x ; thus, lim

n→∞

n 1 19. z n = cn = 4(2n ) 3 SOLUTION

−n 1 −2 = lim −2−x = lim − x = 0. x→∞ x→∞ 2

x 1 ; thus, 3 n x 1 1 = lim = 0. lim n→∞ 3 x→∞ 3

We have z n = f (n) where f (x) =

(−1)n n 2−1/n +n 21. an = z n = (0.1) 4n 2 + 1 SOLUTION

We examine the terms with n odd and n even separately. With n odd, the terms take the form −n 2 + n 4n 2 + 1

and

lim

−n 2 + n

n→∞ 4n 2 + 1

= lim

−x 2 + x

x→∞ 4x 2 + 1

1 =− . 4

On the other hand, with n even, the terms take the form n2 + n 4n 2 + 1

and

lim

n2 + n

n→∞ 4n 2 + 1

= lim

x2 + x

x→∞ 4x 2 + 1

=

1 . 4

Because these two limits are different, the general term of the given sequence does not approach one value. Hence, the sequence diverges. n an = n2 + 1

590

C H A P T E R 11

INFINITE SERIES

n 23. an = 3 n +1 We have an = f (n) where f (x) =

SOLUTION

x x3 + 1

; thus,

x √1 0 0 3/2 x x = lim = √ lim = lim = lim √ = = 0. 3 n→∞ n 3 + 1 x→∞ x 3 + 1 x→∞ x→∞ 1 1+0 x +1 1 + 13 x x 3/2

n

x

25. an = cos π n an = sin π n SOLUTION The terms in the odd places are −1 and the terms in the even places are 1. Therefore, the general term of the sequence does not approach one limit. Hence, the sequence diverges. n 27. Let an = . Find −1 2a number M such that: an = n((1 n ++ 1 n ) − 1) (a) |an − 1| ≤ 0.001 for n ≥ M. (b) |an − 1| ≤ 0.00001 for n ≥ M. Then use the limit definition to prove that lim an = 1. n→∞

SOLUTION

(a) We have n n − (n + 1) −1 = = 1 . − 1 = |an − 1| = n+1 n+1 n + 1 n + 1 1 ≤ 0.001, that is, n ≥ 999. It follows that we can take M = 999. Therefore |an − 1| ≤ 0.001 provided n+1 1 ≤ 0.00001, that is, n ≥ 99999. It follows that we can take M = (b) By part (a), |an − 1| ≤ 0.00001 provided n+1 99999. We now prove formally that lim an = 1. Using part (a), we know that n→∞

|an − 1| =

1 < , n+1

provided n > 1 − 1. Thus, Let > 0 and take M = 1 − 1. Then, for n > M, we have |an − 1| =

1 1 < = . n+1 M +1

29. Use the limit definition to prove that lim n −2 = 0. n→∞ Let bn = ( 13 )n .

see of thatM such that |bn | ≤ 10−5 for n ≥ M. aWe value (b) Use the limit definition to prove that lim bn = 0. 1 1 n→∞ |n −2 − 0| = 2 = 2 < n n

SOLUTION (a) Find

provided 1 n> √ . Thus, let > 0 and take M = √1 . Then, for n > M, we have

1 1 1 |n −2 − 0| = 2 = 2 < 2 = . n n M √ √ 31. Find lim 21/n . n→∞ Find the limit of dn = n + 3 − n. SOLUTION

Because 2x is a continuous function, lim 21/n = lim 21/x = 2limx→∞ (1/x) = 20 = 1.

n→∞

x→∞

33. Find lim n 1/n . n→∞ Show that lim b1/n is independent of b for b > 0. n→∞


Sequences

591

Let an = n 1/n . Take the natural logarithm of both sides of this expression to obtain ln an = ln n 1/n =

ln n . n

Thus, lim (ln an ) = lim n→∞

n→∞

ln n ln x 1 = lim = lim = 0. x→∞ x x→∞ x n

Because f (x) = e x is a continuous function, it follows that lim an = lim eln an = elimn→∞ (ln an ) = e0 = 1.

n→∞

n→∞

That is, lim n 1/n = 1.

n→∞

1 n 35. Find lim 1 + . (1 + n −3 )1/3 − 1 n→∞ n an = n 2 3 n 3 + 1 − n . Hint: Write an = and apply L’Hôpital’s Rule. Find the limit of n

n −3 1 SOLUTION Let an = 1 + n . Taking the natural logarithm of both sides of this expression yields

ln 1 + 1 n 1 n 1 ln an = ln 1 + = n ln 1 + . = 1 n n

n

Thus, lim (ln an ) = lim

n→∞

x→∞

ln 1 + 1x 1 x

= lim

d dx

x→∞

1 · − 1 ln 1 + 1x 1 2 x 1+ x 1 1

= lim = lim = = 1. 1 1 x→∞ x→∞ d 1 1 + 0 − 2 1+ x x

x

dx

Because f (x) = e x is a continuous function, it follows that lim an = lim eln an = elimn→∞ (ln an ) = e1 = e.

n→∞

n→∞

37. Use the Squeeze lim an , where Theorem to find n→∞ 1 n Find 1lim 1 + 2 . an = n→∞ by proving that n n4 + n8 √ SOLUTION

1 2n 4

1 ≤ an ≤ √ 2n 2

For all n > 1 we have n 4 < n 8 , so the quotient √ 41

n +n 8

That is,

is smaller than √ 41

n +n 4

and larger than √ 81

n +n 8

1 1 1 = √ = √ ; and an < 4 4 4 2n 2 n ·2 n +n 1 1 1 = √ = √ . an > 2n 4 2n 8 n8 + n8 Now, since lim √ n→∞

1 2n 4

= lim √ n→∞

1 2n 2

= 0, the Squeeze Theorem for Sequences implies that lim an = 0. n→∞

1 cos n. 39. Evaluate lim n sin Evaluate lim . n→∞ n n→∞ n 1 SOLUTION We have an = f (n) where f (x) = x sin x . Thus, lim n sin

n→∞

sin 1x 1 1 . = lim x sin = lim x→∞ x→∞ 1 n x x

sin The limit limx→0 sinx x = 1 implies that limx→∞ 1

1 x

= 1. Hence,

x

lim n sin

n→∞

1 = 1. n

.

592

C H A P T E R 11

INFINITE SERIES

Whichlim statement to the that assertion lim an = L? Explain. n )equivalent 1/n . Hint: Show Evaluate (2n + 3is n→∞ n→∞ (a) For every > 0, the interval (L − , L + ) contains at least one element of the sequence {an }. (b) For every > 0, the interval (L − , L + ) contains 3 ≤ anall≤but (2 ·at3nmost )1/n finitely many elements of the sequence {an }. 41.

SOLUTION Statement (b) is equivalent to Definition 1 of the limit, since the assertion “|an − L| < for all n > M” means that L − < an < L + for all n > M; that is, the interval (L − , L + ) contains all the elements an except (maybe) the finite number of elements a1 , a2 , . . . , a M . Statement (a) is not equivalent to the assertion lim an = L. We show this, by considering the following sequence:

n→∞

⎧ 1 ⎪ ⎪ ⎨ n an = ⎪ ⎪ ⎩1 + 1 n

for odd n for even n

Clearly for every > 0, the interval (−, ) = (L − , L + ) for L = 0 contains at least one element of {an }, but the sequence diverges (rather than converges to L = 0). Since the terms in the odd places converge to 0 and the terms in the even places converge to 1. Hence, an does not approach one limit. In Exercises 43–63, determine the limit to of the the assertion sequencethat or show the sequence diverges by using the appropriate Which statement is equivalent {an } isthat bounded? Explain. Limit Laws or theorems. (a) There exists a finite interval [m, M] containing every element of the sequence {an }. 3n 2 + n + a2finite interval [m, M] containing an element of the sequence {an }. (b) 43. an =There exists 2n 2 − 3 SOLUTION

lim

n→∞

3 3n 2 + n + 2 3x 2 + x + 2 = lim = . 2 2 x→∞ 2 2n − 3 2x − 3

√1 n 45. an = 3 + −n an = √ 2 n+4 SOLUTION By the Limit Laws for Sequences we have: 1 n 1 n 1 n = 3 + lim − . lim 3 + − = lim 3 + lim − n→∞ n→∞ n→∞ n→∞ 2 2 2 Now, n n 1 n 1 1 ≤ − ≤ . − 2 2 2 Because

n 1 = 0, n→∞ 2 lim

by the Limit Laws for Sequences, n n 1 1 = − lim = 0. lim − n→∞ n→∞ 2 2 Thus, we have 1 n = 0, − n→∞ 2 lim

and

1 n 3+ − = 3 + 0 = 3. n→∞ 2

lim

2 1− 1/3 47. bn = tan−1 4 an = 2 + 2 n n SOLUTION Because f (x) = tan−1 x is a continuous function, it follows that π 2 2 lim an = lim tan−1 1 − = tan−1 lim 1 − = tan−1 1 = . n→∞ x→∞ x→∞ x x 4

S E C T I O N 11.1

Sequences

2n2 + 1 49. cn = ln n −n bn = e3n + 4 SOLUTION

Because f (x) = ln x is a continuous function, it follows that 2x + 1 2 2x + 1 lim cn = lim ln = ln lim = ln . n→∞ x→∞ x→∞ 3x + 4 3x + 4 3

en + 3nn 51. yn = cn = 5n n + n 1/n SOLUTION We rewrite the general term of the sequence as follows:

e n 3 n en + 3n en 3n = + = + . 5n 5n 5n 5 5 n Because 0 < 5e < 1 and 0 < 35 < 1, the geometric sequences ( 5e )n and ( 35 ) converge to 0; hence, n n

e n 3 en + 3n e n 3 = lim + + lim = 0 + 0 = 0. lim = lim n→∞ n→∞ n→∞ 5 n→∞ 5 5n 5 5

en 53. an = 2 en yn 2=n n 2 SOLUTION Using the natural logarithm, we find ln an = ln

en 2n

2

2

= ln en − ln 2n = n − n 2 ln 2.

Thus, lim ln an = lim n (1 − n ln 2) = lim x (1 − x ln 2) = −∞. n→∞ x→∞

n→∞

Because f (x) = e x is a continuous function, it follows that lim an = lim eln an = elimn→∞ (ln an ) = 0.

n→∞

n→∞

−n n3 + n 2e 55. bn = an = 3 n −n 3n 2+ 4e SOLUTION We rewrite the general term of the sequence as follows: 3

an =

n

Thus,

−n

n + 2e 1+ n 3 + 2e−n n3 n3 = = 3 −n 3n + 4e 3n 3 + 4e−n 3+ 3 3

limx→∞ 1 + x e

lim an = lim = n→∞ x→∞ 3 + 4 limx→∞ 3 + x 3 ex 1 + 32 x

2 x 3 ex 4 x 3 ex

=

n

2 n 3 en . 4 3 n n e

limx→∞ 1 + 2 limx→∞ 31 x

1 1+2·0 x e = . = 1 3 + 4 · 0 3 limx→∞ 3 + 4 limx→∞ 3 x x e

3 − 4n n 57. An = 3−4 2 + 7 · 3n n bn = 2+7·4 SOLUTION Divide the numerator and denominator by 3n to obtain

3 − 4 n 3 − 4n n 3 − 4n 3 3 n n 3 an = = 23 . n = 7·3 2 2 + 7 · 3n 3n + 3n 3n + 7 We examine the limits of the numerator and the denominator: n n n 4 3 1 4 − − 3 lim = 3 · 0 − ∞ = −∞, = 3 lim lim n→∞ 3n n→∞ 3 n→∞ 3 3 whereas

n 2 2 1 + 7 = lim + lim 7 = 2 lim + lim 7 = 2 · 0 + 7 = 7. n→∞ 3n n→∞ 3n n→∞ n→∞ 3 n→∞

lim

Thus, lim an = −∞; that is, the sequence diverges. n→∞

593

594

C H A P T E R 11

INFINITE SERIES n (−4,000) 59. An = 10n Bn = n! n! (−4000)n SOLUTION Let an = and note n!

−

4000n n!

≤ an ≤

4000n . n!

For n > 4001, 4000 4000 4000 4000 4000 40004000 4000 4000 4000 4000 4000 4000 · · ··· · · · ··· · · = · · ··· · · . 1 2 4000 4001 4002 n−1 n 4000! 4001 4002 n−1 n Each one of the factors in the brackets is less than 1, so we have 0
0 for x > 0, hence f is strictly increasing on this interval. It follows that an = f (n) is also strictly increasing. We now show that M = 3 is an upper bound for an , by writing: 3n 2 3n 2 + 6 3(n 2 + 2) ≤ 2 = = 3. an = 2 n +2 n +2 n2 + 2 That is, an ≤ 3 for all n.

S E C T I O N 11.1

Sequences

595

67. Use the limit definition to prove that the limit does not change if a finite number of terms are added or removed from √ 3 Showsequence. that an = n + 1 − n is decreasing. a convergent SOLUTION Suppose that {an } is a sequence such that limn→∞ an = L. For every > 0, there is a number M such that |an − L| < for all n > M. That is, the inequality |an − L| < holds for all the terms of {an } except possibly a finite number of terms. If we add a finite number of terms, these terms may not satisfy the inequality |an − L| < , but there are still only a finite number of terms that do not satisfy this inequality. By removing terms from the sequence, the number of terms in the new sequence that do not satisfy |an − L| < are no more than in the original sequence. Hence the new sequence also converges to L.

69. Let {an } be a sequence such that lim |an | exists and is nonzero. Show that lim an exists if and only if there exists n→∞ n→∞ lim an = lim bn . Let bn = an+1 . Prove that if {a n } converges, then {bn } also converges and n→∞ n→∞ an integer M such that the sign of an does not change for n > M. Let {an } be a sequence such that lim |an | exists and is nonzero. Suppose lim an exists and let n→∞ n→∞ L = lim an . Note that L cannot be zero for then lim |an | would also be zero. Now, choose < |L|. Then there exists

SOLUTION

n→∞

n→∞

an integer M such that |an − L| < , or L − < an < L + , for all n > M. If L < 0, then −2L < an < 0, whereas if L > 0, then 0 < an < 2L; that is, an does not change for n > M. Now suppose that there exists an integer M such that an does not change for n > M. If an > 0 for n > M, then an = |an | for n > M and lim an = lim |an |.

n→∞

n∞

On the other hand, if an < 0 for n > M, then an = −|an | for n > M and lim an = lim −|an | = − lim |an |.

n→∞

n∞

n→∞

In either case, lim an exists. Thus, lim an exists if and only if there exists an integer M such that the sign of an does n→∞

not change for n > M.

n→∞

71. Show, by giving an example, that there exist divergent sequences {an } and {bn } such that {an + bn } converges. Give an example of a divergent sequence {an } such that lim |an | converges. n→∞ SOLUTION Let an = 2n and bn = −2n . Then {an } and {bn } are divergent geometric sequences. However, since n n an + bn = 2 − 2 = 0, the sequence {an + bn } is the constant sequence with all the terms equal zero, so it converges to zero. 73. Use the limit definition to prove that if {an } is a convergent sequence of integers with limit L, then there exists a Using the limit definition, prove that if {an } converges and {bn } diverges, then {an + bn } diverges. number M such that an = L for all n ≥ M. SOLUTION

Suppose {an } converges to L, and let = 12 . Then, there exists a number M such that |an − L|
a1 . If we suppose that an > an−1 for some n, it then follows that an+1 = 5 + an > 5 + an−1 = an ; hence, by mathematical induction, an+1 > an for all n.

596

C H A P T E R 11

INFINITE SERIES

Now, since {an } is increasing with upper bound, this sequence is convergent. Let lim an = L. Then, by Exercise n→∞

68, lim an+1 = L as well. It follows that n→∞

L = lim an+1 = lim n→∞

n→∞

√ 5 + an = lim (5 + an ) = 5 + lim an = 5 + L. n→∞

n→∞

That is, L2 = 5 + L L 2 − L − 5 = 0 ⇒ L 1,2 =

1±

√

1 + 20 . 2

Since an ≥ 0 for all n, the appropriate solution is: lim an =

n→∞

1+

√ 2

21

.

77. Find the limit of the sequence Let {an } be the sequence 1 1 1 + √ + · · ·√ + cn = √ 2+2 2+n n2 + 1 n n 2, 2 2, 2 2 2, . . . Hint: Show that Show that {an } is increasing and 0 ≤ an ≤ 2. Then prove that {an } converges and find the limit. n n ≤ cn ≤ 2 2 n +n n +1 SOLUTION

Since each of the n terms in the sum defining cn is not smaller than √ 12

n +n

obtain the following inequalities:

and not larger than √ 12

n +1

we

1 1 1 n =n· + ··· + = ; cn ≥ n2 + n n2 + n n2 + n n2 + n 4 56 7 n terms

1 n 1 + ··· + = . =n· cn ≤ n2 + 1 n2 + 1 n2 + 1 n2 + 1 56 7 4 1

n terms

Thus, n n ≤ cn ≤ . n2 + n n2 + 1 We now compute the limits of the two sequences: n n 1 1 lim = lim √ = lim = 1; = lim √ n 2 2 n→∞ n 2 + 1 n→∞ n→∞ n→∞ n +1 n +1 1 + 12 √ n

n n2 n n 1 1 = lim √ = lim = 1. lim = lim √ n 2 +n n→∞ n 2 + n n→∞ n→∞ n→∞ n 2 +n n √ 1 + n1 n 2 n

By the Squeeze Theorem we conclude that: lim cn = 1.

n→∞


√ n √ n! 79. LetShow bn = that an = n n! diverges. Hint: Show that n! ≥ (n/2)n/2 by observing that half of the factors of n! are n greater than or equal to n/2. n ln(n!) − n ln n k 1# ln . (a) Show that ln bn = = n n k=1 n " 1 ln x d x and conclude that bn → e−1 . (b) Show that ln bn converges to 0

SOLUTION

S E C T I O N 11.1

Sequences

597

1/n (a) Let bn = (n!)n . Then

1 1 ln (n!) − n ln n 1 n! ln (n!) − ln n = = ln (n!) − ln n n = ln n n n n n n n 1 2 3 n 2 3 n 1 k 1 1 1# · · · ··· · = ln ln . = ln + ln + ln + · · · + ln = n n n n n n n n n n n k=1 n

ln bn = ln (n!)1/n − ln n =

(b) By part (a) we have, n k 1# ln . n→∞ n n k=1

lim (ln bn ) = lim

n→∞

$ Notice that n1 nk=1 ln nk is the nth right-endpoint approximation of the integral of ln x over the interval [0, 1]. Hence, " 1 n 1# k ln = ln x d x. n→∞ n n 0 k=1 lim

We compute the improper integral using integration by parts, with u = ln x and v = 1. Then u = 1x , v = x and 1 " 1 " 1 " 1 1 ln x d x = x ln x − dx x d x = 1 · ln 1 − lim (x ln x) − x→0+ 0 0 x 0 0 1 = 0 − lim (x ln x) − x = −1 − lim (x ln x) . x→0+

x→0+

0

We compute the remaining limit using L’Hôpital’s Rule. This gives:

x→0+

1

ln x

x = lim = lim (−x) = 0. x→0+ 1 x→0+ − 12 x→0+ x

lim (x · ln x) = lim

x

Thus, lim ln bn = n→∞

" 1 0

ln x d x = −1,

and lim bn = e−1 .

n→∞

1 1 1 1 81. LetGiven cn = positive + + + b· 1· ·, + define. two sequences recursively by numbers a < n n + 1 n + 12 2n (a) Calculate c1 , c2 , c3 , c4 . an + bn bn =the interval [n, 2n] to prove that n , x −1 over n+1 = (b) Use a comparison of rectangles with theaarea underanyb= 2 " " 2n (a) Show that an ≤ bn for all n (Figure2n13). dx dx 1 1 + ≤ cn ≤ + (b) Show that {an } is increasing and {b 2n x n n n } isx decreasing. n (c) Show that Theorem to determine lim c . (c) Use the Squeeze n→∞

SOLUTION

(a)

n

bn+1 − an+1 ≤

bn − an 2

1 3 Prove that both {an } and {bn } converge c1 =and 1 +have=the ;same limit. This limit, denoted AGM(a1 , b1 ), is called the 2 2 13. arithmetic-geometric mean of a1 and b1 . See Figure √ 1 1 13 (d) Estimate AGM(1, 2) to three decimal1places. ; c2 = + + = 2 3 4 12 1 1 1 1 19 ; c3 = + + + = 3 4 5 6 20 1 1 1 1 1 743 c4 = + + + + = ; 4 5 6 7 8 840

(b) We consider the left endpoint approximation to the integral of y = 1x over the interval [n, 2n]. Since the function ! y = 1x is decreasing, the left endpoint approximation is greater than n2n dxx ; that is, " 2n 1 1 1 1 dx ≤ ·1+ ·1+ · 1 + ··· + · 1. x n n+1 n+2 2n − 1 n

598

C H A P T E R 11

INFINITE SERIES y

y=

1 x

1 2

1 3

1 1/n

x

n n+1

1 2 3

0

We express the right hand-side of the inequality in terms of cn , obtaining: " 2n 1 dx ≤ cn − . x 2n n ! We now consider the right endpoint approximation to the integral n2n dxx ; that is, " 2n 1 1 1 dx ·1+ · 1 + ··· + ·1≤ . n+1 n+2 2n x n y

y=

1 x

1 n+1 x

n+1

n

0

We express the left hand-side of the inequality in terms of cn , obtaining: cn −

" 2n 1 dx ≤ . n x n

Thus, " 2n " 2n dx dx 1 1 + ≤ cn ≤ + . x 2n x n n n (c) With " 2n 2n dx = ln x|2n = ln 2, n = ln 2n − ln n = ln x n n the result from part (b) becomes ln 2 +

1 1 ≤ cn ≤ ln 2 + . 2n n

Because lim

1

n→∞ 2n

= lim

1

n→∞ n

= 0,

it follows from the Squeeze Theorem that lim cn = ln 2.

n→∞

The nth harmonic number is the number

11.2 Summing an Infinite Series Preliminary Let an = HnQuestions − ln n.

Hn = 1 +

1 1 1 + + ··· + 2 3 n

n+1 dseries? 1. What role do partial sums play in defining the sum of an"infinite x . (a) Show that an ≥ 0 for n ≥ 1. Hint: Show that Hn ≥ x 1 (b) Show that {an } is decreasing by interpreting an − an+1 as an area.

S E C T I O N 11.2

Summing an Infinite Series

599

SOLUTION The sum of an infinite series is defined as the limit of the sequence of partial sums. If the limit of this sequence does not exist, the series is said to diverge.

2. What is the sum of the following infinite series? 1 1 1 1 1 + + + + + ··· 4 8 16 32 64 SOLUTION

This is a geometric series with c = 14 and r = 12 . The sum of the series is therefore 1 4

1

1 = 41 = . 1 2 1− 2 2 3. What happens if you apply the formula for the sum of a geometric series to the following series? Is the formula valid? 1 + 3 + 32 + 33 + 34 + · · · SOLUTION

This is a geometric series with c = 1 and r = 3. Applying the formula for the sum of a geometric series

then gives ∞ #

3n =

n=0

1 1 =− . 1−3 2

Clearly, this is not valid: a series with all positive terms cannot have a negative sum. The formula is not valid in this case because a geometric series with r = 3 diverges. 4. Arvind asserts that

∞ # 1 1 = 0 because 2 tends to zero. Is this valid reasoning? 2 n n n=1

SOLUTION Arvind’s reasoning is not valid. Though the terms in the series do tend to zero, the general term in the sequence of partial sums,

1 1 1 Sn = 1 + 2 + 2 + · · · + 2 , 2 3 n is clearly larger than 1. The sum of the series therefore cannot be zero. 5. Colleen claims that

∞ # 1 1 √ converges because lim √ = 0. Is this valid reasoning? n→∞ n n n=1

Colleen’s reasoning is not valid. Although the general term of a convergent series must tend to zero, a ∞ # 1 series whose general term tends to zero need not converge. In the case of √ , the series diverges even though its n n=1 general term tends to zero. SOLUTION

6. Find an N such that S N > 25 for the series

∞ #

2.

n=1 SOLUTION

The N th partial sum of the series is: SN =

N # n=1

2 = 2 + · · · + 2 = 2N . 4 56 7 N

7. Does there exist an N such that S N > 25 for the series

∞ #

2−n ? Explain.

n=1 SOLUTION

The series

∞ #

2−n is a convergent geometric series with the common ratio r =

n=1

1 . The sum of the series 2

is: S=

1 2

= 1. 1 − 12

Notice that the sequence of partial sums {S N } is increasing and converges to 1; therefore S N ≤ 1 for all N . Thus, there does not exist an N such that S N > 25.

600

C H A P T E R 11

INFINITE SERIES

8. Give an example of a divergent infinite series whose general term tends to zero. Consider the series

SOLUTION

∞ # 1 9 n=1 n 10

. The general term tends to zero, since lim

n→∞

1 9

= 0. However, the N th partial

n 10

sum satisfies the following inequality: SN =

1 9 1 10

+

1

1 9 2 10

+ ··· +

1 N

9 10

≥

N N

9 10

9

1

= N 1− 10 = N 10 .

1

That is, S N ≥ N 10 for all N . Since lim N 10 = ∞, the sequence of partial sums Sn diverges; hence, the series N →∞

∞ # 1

diverges.

Exercises 1. Find a formula for the general term an (not the partial sum) of the infinite series. 1 1 1 1 + + ··· (a) + + 3 9 27 81 1 5 25 125 (b) + + + + ··· 1 2 4 8 1 22 33 44 (c) − + − + ··· 1 2·1 3·2·1 4·3·2·1 1 2 1 2 + + + + ··· (d) 2 1 + 1 22 + 1 32 + 1 42 + 1 SOLUTION

(a) The denominators of the terms are powers of 3, starting with the first power. Hence, the general term is: 1 an = n . 3 (b) The numerators are powers of 5, and the denominators are the same powers of 2. The first term is a1 = 1 so, an =

n−1 5 . 2

(c) The general term of this series is, an = (−1)n+1

nn . n!

(d) Notice that the numerators of an equal 2 for odd values of n and 1 for even values of n. Thus, ⎧ 2 ⎪ ⎪ ⎨ n 2 + 1 odd n an = ⎪ ⎪ ⎩ 1 even n n2 + 1 The formula can also be rewritten as follows: n+1 1 + (−1) 2 +1 . an = n2 + 1

In Exercises compute the partial sums S2 , S4 , and S6 . Write3–6, in summation notation: 1 1 1 1 +1 + · · · (a) 11+ + 3. 1 + 2 +4 2 9+ 216+ · · · 12 13 14 1 (b) + + + + ··· SOLUTION 9 16 25 36 1 1 1 (c) 1 − + − + · · · 1 5 3 5 7 S2 = 1 + 2 = ; 4 2 125 625 3,125 15,625 (d) + + + + ··· 1 1 1 205 9 16 25 36 S4 = 1 + 2 + 2 + 2 = ; 144 2 3 4 1 1 1 1 1 5369 S6 = 1 + 2 + 2 + 2 + 2 + 2 = . 3600 2 3 4 5 6 ∞ #

(−1)k k −1

9

n=1 n 10

S E C T I O N 11.2

5.


601

1 1 1 + + + ··· 1·2 2·3 3·4

SOLUTION

1 1 1 4 2 1 + = + = = ; 1·2 2·3 2 6 6 3 2 1 1 2 1 1 4 S4 = S2 + a3 + a4 = + + = + + = ; 3 3·4 4·5 3 12 20 5 4 1 1 4 1 1 6 S6 = S4 + a5 + a6 = + + = + + = . 5 5·6 6·7 5 30 42 7 S2 =

7. Compute S5 , S10 , and S15 for the series ∞ # 1 1 1 1 1 j! S= − + − + ··· j=1 2 · 3 · 4 4 · 5 · 6 6 · 7 · 8 8 · 9 · 10

π−3 . Do your calculations support this conclusion? 4 The formula for the general term in the series is

This series S is known to converge to SOLUTION

an =

(−1)n+1 . 2n(2n + 1)(2n + 2)

Thus, S5 =

1 1 1 1 − + ··· − + = 0.035678; 2(3)(4) 4(5)(6) 8(9)(10) 10(11)(12)

S10 =

1 1 1 1 − + ··· + − = 0.035352; 2(3)(4) 4(5)(6) 18(19)(20) 20(21)(22)

S15 =

1 1 1 1 − + ··· − + = 0.035413. 2(3)(4) 4(5)(6) 28(29)(30) 30(31)(32)

Using a calculator we find

π −3 = 0.035398, 4 which is consistent with our partial sum calculations. 9. Calculate S3 , S4 , and S5 and then find the sum of the telescoping series The series ∞ # 1 1 11 1 1 SS== − + − − + ··· 0! n1!+ 1 2! n +3!2 n=1 is known SOLUTION 10−3 .

to converge to e−1 (recall that 0! = 1). Find a partial sum that approximates e−1 with an error at most

1 1 1 1 1 1 1 1 3 − + − + − = − = ; 2 3 3 4 4 5 2 5 10 1 1 1 1 1 S4 = S3 + − = − = ; 5 6 2 6 3 1 1 1 1 5 S5 = S4 + − = − = . 6 7 2 7 14 S3 =

The general term in the sequence of partial sums is 1 1 1 1 1 1 1 1 1 1 − + − + − + ··· + − = − ; SN = 2 3 3 4 4 5 N +1 N +2 2 N +2 thus,

1 1 − 2 N +2 N →∞

S = lim S N = lim N →∞

=

1 . 2

The sum of the telescoping series is therefore 12 . ∞ # 1 ∞ # 11. Write as a telescoping series and find its sum. 1 n(n using the identity , S1) Calculate S 3− 4 , and S5 and then find the sum S = n=3 2−1 4n n=1 1 1 1 1 = −

602

C H A P T E R 11

INFINITE SERIES SOLUTION

By partial fraction decomposition 1 1 1 = − , n(n − 1) n−1 n

so ∞ # 1 1 1 = − . n(n − 1) n=3 n − 1 n n=3 ∞ #

The general term in the sequence of partial sums for this series is 1 1 1 1 1 1 1 1 1 1 − + − + − + ··· + − = − ; SN = 2 3 3 4 4 5 N −1 N 2 N thus,

1 1 − N N →∞ 2

S = lim S N = lim N →∞

=

1 . 2

In Exercises 13–16, use Theorem 2 to prove that the ∞following series diverge. # of (−1)n−1 and show that the series diverges. Find a formula for the partial sums S N ∞ # n=1 13. (−1)n n 2 n=1

The general term an = (−1)n n 2 does not tend to zero. In fact, because limn→∞ n 2 = ∞, limn→∞ an does not exist. By Theorem 2, we conclude that the given series diverges. SOLUTION

15.

∞ √ # √ 2 3 0 n 1+ 1 − − + −n + · · · 2 3 4 n=11

SOLUTION

Because

√ √ √ √ n+1+ n (n + 1) − n 1 lim an = lim ( n + 1 − n) √ √ = lim √ √ = lim √ √ = 0, n→∞ n→∞ n→∞ n + 1 + n n→∞ n + 1 + n n+1+ n Theorem 2 does not help us here. Rather, we recognize that the series is a telescoping series with √ √ √ √ √ √ √ √ √ S N = ( 2 − 1) + ( 3 − 2) + ( 4 − 3) + · · · + ( N + 1 − N ) = N + 1 − 1. Because lim S N = lim

N →∞

√

N →∞

N + 1 − 1 = ∞,

the given series diverges. 17. Which of these series converge? 1 + ··· 1 + cos 1 + cos ∞ cos # 31 4 12 (a) √ −√ n n+1 n=1

(b)

∞ #

(ln n − ln(n + 1))

n=1

SOLUTION

(a) This series converges. The general term in the sequence of partial sums is 1 1 1 1 1 1 1 1 + √ −√ =1− √ . SN = 1 − √ + √ −√ + ··· + √ − √ N +1 N +1 N 3 3 2 2 4 Because

lim S N = lim

N →∞

N →∞

1− √

1

N +1

= 1,

the series converges to 1. (b) This series diverges. The general term in the sequence of partial sums is S N = (ln 1 − ln 2) + (ln 2 − ln 3) + (ln 3 − ln 4) + · · · + (ln N − ln(N + 1)) = − ln(N + 1). Because lim S N = lim − ln(N + 1) = −∞,

N →∞

we conclude the given series diverges.

N →∞

S E C T I O N 11.2


603

In Exercises 18–31, use the formula for the sum of a geometric series to find the sum or state that the series diverges. 1 1 1 + 1 + 1 + · ·1· 33 1 +34 + 352 + 3 + · · · 5 5 5 1 1 SOLUTION This is a geometric series with c = 27 and r = 3 . Thus, 1 1 ∞ n # 1 1 = 27 1 = 27 = . 2 3 18 1 − n=3 3 3

19.

21.

∞ # 3n n ∞ # 3 11n n n=3 11 n=0

SOLUTION

3 3 27 and r = 3 . Thus, This is a geometric series with c = 11 = 1331 11 27 27 ∞ # 27 3n = 13313 = 1331 = . n 8 11 968 1 − n=3 11 11

2 22 n 23 ∞ 23. 1 +# + ··· +7 · 3 + 7 72n 73 11 n=0 2 SOLUTION This is a geometric series with c = 1 and r = 7 . Thus, ∞ n # 1 1 7 2 = = 5 = . 2 7 5 1 − n=0 7 7

25.

∞ # ∞ # e3−2n e−n n=2 n=1

SOLUTION

Rewrite the series as ∞ #

e3 e−2n =

n=2

to recognize it as a geometric series with c = e3 ∞ # n=2

∞ # n=2

1 e2

2

e3

1 n e2

= 1e and r = 12 . Thus,

e3−2n =

e

1 e

e . = 2 e −1 1 − 12 e

∞ # 93n + 4n−2 ∞ # 8 +n 2n 5n n=0 5 n=0 SOLUTION Rewrite the series as

27.

n ∞ ∞ ∞ n ∞ # # # # 93n 93 4 93n + 4n−2 1 4n−2 = + + , = · n n n 5 5 5 5 16 5 n=0 n=0 n=0 n=0

0 1 4 0= which is a sum of two geometric series. The first series has c = 93 = 1 and r = 93 5 5 ; the second has c = 16 5 1 4 16 and r = 5 . Because the first series has r > 1, that series diverges. Consequently, the original series also diverges. 23 255 256 24 + + ··· +− 25 + + 5 3 − 7 7 4 742 7443 + · · · 8 2 SOLUTION This is a geometric series with c = 7 and r = 7 . Thus, n 8 ∞ # 8 2 = 7 2 = · 7 7 1− 7 n=0

29.

31.

64 8 7 49 343 + 1 +343+ 2,401 + + ··· 7+ 49 49 −7 + 8 −64 512 + ··· 8 64 512 4,096

8 7 = 8. 5 5 7

604

C H A P T E R 11


7 This is a geometric series with c = 64 49 and r = 8 . Thus,

n 64 64 ∞ # 512 64 7 = 49 7 = 49 = · . 1 49 8 49 1− 8 n=0 8 33. Which of the following series are divergent? of the following are not geometric series? ∞ Which n # 2∞ n # 7 (a) (a) 5n n=0 29n ∞ n=0 n # 5 ∞ 2 # (c) n (c) 2n n n=0 2 n=0 SOLUTION

(a) The series (b) The series (c) The series (d) The series

∞ # ∞ n (b) # 1.51 (b) n=3 n 4 ∞n=3 # ∞ (d) # (0.4)n (d)n=0 π −n n=5

∞ n # 2 2 = is a geometric series with common ratio r = . Since |r | < 1, the series converges. n 5 5 5 n=0 n=0 ∞ n # 2 ∞ #

1.5n is a geometric series with common ratio r = 1.5. Since r > 1, the series diverges.

n=3 ∞ n # 5

∞ n # 5 5 = is a geometric series with common ratio r = . Since r > 1, the series diverges. n 2 2 2 n=0 n=0 ∞ #

0.4n is a geometric series with common ratio r = 0.4. Since |r | < 1, the series converges.

n=0 ∞ # 2 of theseries following incorrect. awhy an infinite such statements that S N = is 5− . 35. Let S Explain = n be each 2 N ∞ n=1 # zero, an = 0. (a) If the general term # a10 16 n tends to # an and ann=1 ? (a) What are the values of (b) The N th partial sum of the infinite n=1 n=4 series defined by {an } is a N . ∞ (b) What is the value of a3 ? # (c) Ifa ageneral to zero, for thena . an converges. n tendsformula (c) Find n n=1 ∞ # ∞ # (d) Find the sum an . to L, then an = L. (d) If an tends n=1 n=1

SOLUTION

(a) 10 #

an = S10 = 5 −

n=1

2 102

=

249 ; 50

2 2 494 2 2 an = (a1 + · · · + a16 ) − (a1 + a2 + a3 ) = S16 − S3 = 5 − 2 − 5 − 2 = − = . 9 256 2304 3 16 n=4 16 #

(b)

2 a3 = (a1 + a2 + a3 ) − (a1 + a2 ) = S3 − S2 = 5 − 2 3

5 2 1 2 . − 5− 2 = − = 2 9 18 2

(c) Since an = Sn − Sn−1 , we have: 2 2 2 2 an = Sn − Sn−1 = 5 − 2 − 5 − − 2 = 2 2 n n (n − 1) (n − 1)

2 2 2 2 2 n − (n − 1) 2 n − n + 2n − 1 2 (2n − 1) = = = . 2 2 n 2 (n − 1)2 (n (n − 1)) (n (n − 1)) (d) The sum

∞ #

an is the limit of the sequence of partial sums {S N }. Hence:

n=1 ∞ # n=1

an = lim S N = lim N →∞

N →∞

5−

2 N2

Compute the total area of the (infinitely many) triangles in Figure 3.

= 5.

S E C T I O N 11.2

37. Use the method of Example 5 to show that

SOLUTION


605

∞ # 1 diverges. 1/3 k k=1

Each term in the N th partial sum is greater than or equal to

1 1

, hence:

N3 1 1 3 1 1 1 1 1 1 S N = 1/3 + 1/3 + 1/3 + · · · + 1/3 ≥ 1/3 + 1/3 + 1/3 + · · · + 1/3 = N · 1/3 = N 2/3 . 1 2 3 N N N N N N Since lim N 2/3 = ∞, it follows that N →∞

lim S N = ∞.

N →∞

Thus, the series

∞ #

1 diverges. 1/3 k k=1

39. A ball dropped from a height of 10 ft begins to bounce. Each time it strikes the ground, it returns to two-thirds of its ∞ # 1 infinitely many times? previousLet height. What isth thepartial total distance traveled by the ball ifS it=bounces the N sum of the harmonic series . S N be n n=1 SOLUTION The distance traveled by the ball is shown in the accompanying figure: (a) Verify the following inequality for n = 1, 2, 3. Then prove it for general n. h = 10

1 1 1 1 + n−1 + n−1 + ··· + n ≥ n−1 2 2 2 2 2 +1 2 2 +3 h+ 2h 1

3

3

(b) Prove that S diverges by showing that S N ≥ 1 + 1, 2, 4, 8 . . . , as in the following: S23 = 1 +

n 2 2 2 for 2N = 2n . Hint: Break up S h h N into n + 1 sums of length 2( 3 ) ( 3 )

1 1 1 1 1 1 1 + + + + + + 2 3 4 5 6 7 8

The total distance d traveled by the ball is given by the following infinite sum: 3 2 2 3 ∞ n # 2 2 2 2 2 2 2 d = h+2· h+2· h+2· h + · · · = h + 2h + + · · · = h + 2h . + 3 3 3 3 3 3 3 n=1 We use the formula for the sum of a geometric series to compute the sum of the resulting series:

1 d = h + 2h ·

2 3

= h + 2h(2) = 5h. 1 − 23

With h = 10 feet, it follows that the total distance traveled by the ball is 50 feet. 1 1 1 ∞+ ··· . + + # 41. Find the sum of 1 1 · 3 3 · 5 5 as a telescoping series and find its sum. Use partial fractions to rewrite · 7 n(n + 3) SOLUTION We may write this sum asn=1 ∞ # 1 1 1 1 = − . (2n − 1)(2n + 1) n=1 2 2n − 1 2n + 1 n=1 ∞ #

The general term in the sequence of partial sums is 1 1 1 1 1 1 1 1 1 1 1 1 1 1 − + − + − + ··· + − = 1− ; SN = 2 1 3 2 3 5 2 5 7 2 2N − 1 2N + 1 2 2N + 1 thus, 1 1 1 1− = , 2N + 1 2 N →∞ 2

lim S N = lim

N →∞

and ∞ #

1 1 = . (2n − 1)(2n + 1) 2 n=1 1 − . Compute S N for N = 1, 2, 3, 4. Find S by showing that Let S = n n+2 n=1 ∞ # 1

606

C H A P T E R 11

INFINITE SERIES

43. Let {bn } be a sequence and let an = bn − bn−1 . Show that

∞ #

an converges if and only if lim bn exists. n→∞

n=1 SOLUTION

Let an = bn − bn−1 . The general term in the sequence of partial sums for the series

∞ #

an is then

n=1

S N = (b1 − b0 ) + (b2 − b1 ) + (b3 − b2 ) + · · · + (b N − b N −1 ) = b N − b0 . Now, if lim b N exists, then so does lim S N and N →∞

N →∞

∞ #

an converges. On the other hand, if

n=1

lim S N exists, which implies that lim b N also exists. Thus,

N →∞

N →∞

∞ #

∞ #

an converges, then

n=1

an converges if and only if lim bn exists. n→∞

n=1

45. Find the total length of the infinite zigzag path in Figure 4 (each zag occurs at an angle of π4 ). The winner of a lottery receives m dollars at the end of each year for N years. The present value (PV) of this prize N # m(1 + r )−i , where r is the interest rate. Calculate PV if m = $50,000, r = 0.06, and in today’s dollars is PV = i=1

N = 20. What is the PV if N = ∞? π /4

π /4 1

FIGURE 4 π SOLUTION Because the angle at the lower left in Figure 4 has measure 4 and each zag in the path occurs at an angle of π4 , every triangle in the figure is an isosceles right triangle. Accordingly, the length of each new segment in the path is √1 times the length of the previous segment. Since the first segment has length 1, the total length of the path is 2

√ ∞ # √ 1 2 1 n √ = = √ = 2 + 2. 1 2 2−1 1− √ n=0 2

47. Show that if a is a positive integer, then ∞ # 1 Evaluate . Hint: that ∞ Find constantsA, B, and C such 1 1 1 1 n(n + 1)(n + 2) # n=1 = 1 + + ··· + n(n + a) a 2 a n=1 A B C 1 = + + n(n + 1)(n + 2) n (n + 1) n + 2 SOLUTION By partial fraction decomposition 1 A B = + ; n (n + a) n n+a clearing the denominators gives 1 = A(n + a) + Bn. Setting n = 0 then yields A = a1 , while setting n = −a yields B = − a1 . Thus, 1 1 1 1 = a − a = n (n + a) n n+a a

1 1 − n n+a

,

and ∞ # 1 1 1 1 = − . n(n + a) n=1 a n n+a n=1 ∞ #

For N > a, the N th partial sum is 1 1 1 1 1 1 1 1 1 1 + + + ··· + − + + + ··· + . SN = a 2 3 a a N +1 N +2 N +3 N +a Thus, 1 1 1 1 1 = lim S N = 1 + + + ··· + . n(n + a) a 2 3 a N →∞ n=1 ∞ #

Assumptions Matter ∞ ∞ # #

Show, by giving counterexamples, that the assertions of Theorem 3 are not valid if the

S E C T I O N 11.2


607

Further Insights and Challenges 49. Professor George Andrews of Pennsylvania State University observed that geometric sums can be used to calculate the derivative of f (x) = x N in a new way. By Eq. (2), 1 + r + r 2 + · · · + r N −1 = Assume that a = 0 and let x = r a. Show that

1 −rN 1−r

7

x N − aN rN − 1 = a N −1 lim x→a x − a r →1 r − 1

f (a) = lim

Then use Eq. (7) to evaluate the limit on the right. SOLUTION

According to the definition of derivative of f (x) at x = a f (a) = lim

x→a

x N − aN . x −a

Now, let x = r a. Then x → a if and only if r → 1, and

N − aN N − aN aN r N − 1 x rN − 1 a) (r = a N −1 lim = lim = lim . f (a) = lim x→a x − a r →1 r a − a r →1 a (r − 1) r →1 r − 1 By Eq. (7), 1 −rN rN − 1 = = 1 + r + r 2 + · · · + r N −1 , 1−r r −1 so

rN − 1 = lim 1 + r + r 2 + · · · + r N −1 = 1 + 1 + 12 + · · · + 1 N −1 = N . r →1 r − 1 r →1 lim

Therefore, f (a) = a N −1 · N = N a N −1 51. Cantor’s Disappearing Table (following Larry Knop of Hamilton College) Take a table of length L (Figure 6). de Fermat used of geometric series to compute the area under the graph of f (x) x N aover [0,less A].than For At stagePierre 1, remove the section length L/4 centered at the midpoint. Two sections remain, each=with length n , as 2 0 < r < 1, let F(r ) be the sum of the areas of the infinitely many right-endpoint rectangles with L/2. At stage 2, remove sections of length L/4 from each of these two sections (this stage removes endpoints L/8 of theAr table). Figure 5. Asremain, r tendseach to 1, of thelength rectangles become F(r ) tends to the area sections under theofgraph. Nowinfour sections less than L/4.narrower At stage and 3, remove the four central length L/43 , etc. 1 − r N (a) Show that that at the section has length less than L/2 and that the total amount of table (a) Show F(rN)th=stage, A N +1each remaining . 1 − r N +1 removed is " A x N d x = 1lim F(r (b) Use Eq. (7) to evaluate 1 ). 1 1 L + + + · · · + r →1 0 4 8 16 2 N +1 (b) Show that in the limit as N → ∞, precisely one-half of the table remains. This result is curious, because there are no nonzero intervals of table left (at each stage, the remaining sections have a length less than L/2 N ). So the table has “disappeared.” However, we can place any object longer than L/4 on the table and it will not fall through since it will not fit through any of the removed sections. L/16

L/4

L/16

FIGURE 6 SOLUTION

(a) After the N th stage, the total amount of table that has been removed is 1 1 1 1 L 4L 2 N −1 L 2L 1 2 N −1 1 1 = L = L + 2 + 3 + ··· + + + + · · · + + + + · · · + 4 4 8 16 4 8 16 4N 4 4 22N 2 N +1 At the first stage (N = 1), there are two remaining sections each of length L − L4 2

=

3L L < . 8 2

608

C H A P T E R 11

INFINITE SERIES

L Suppose that at the K th stage, each of the 2 K remaining sections has length less than K . The (K + 1)st stage is obtained 2 L by removing the section of length K +1 centered at the midpoint of each segment in the K th stage. Let ak and a K +1 , 4 respectively, denote the length of each segment in the K th and (K + 1)st stage. Then, L − L 1 − K1+2 a K − KL+1 L L 1 L K K +1 2 4 4 2 a K +1 = < = K < K · = K +1 . 2 2 2 2 2 2 2 L Thus, by mathematical induction, each remaining section at the N th stage has length less than N . 2 (b) From part (a), we know that after N stages, the amount of the table that has been removed is L

1 1 1 1 + + + · · · + N +1 4 8 16 2

=

N #

1

n=1

2n+1

.

As N → ∞, the amount of the table that has been removed becomes a geometric series whose sum is L

1 ∞ # 1 1 1 n = L 4 1 = L. 2 2 2 1− 2 n=1

Thus, the amount of table that remains is L − 12 L = 12 L. The Koch snowflake (described in 1904 by Swedish mathematician Helge von Koch) is an infinitely jagged “fractal” curve obtained as a limit of polygonal curves (it is continuous, but has no tangent line at any point). Begin 11.3withConvergence of Series Positive an equilateral triangle (stage 0)with and produce stageTerms 1 by replacing each edge with four edges of one-third the length, arranged as in Figure 7. Continue the process: At the nth stage, replace each edge with four edges of one-third the length. Questions Preliminary ∞ (a) Show# that the perimeter Pn of the polygon at the nth stage satisfies Pn = 43 Pn−1 . Prove that lim Pn = ∞. n→∞ 1. Let S = a . If the partial sums S N are increasing, then (choose correct conclusion) The snowflake nhas infinite length. n=1 of the original equilateral triangle. Show that (3)4n−1 new triangles are added at the nth (b) }Let A0 be the area (a) {a sequence. n is an increasing stage, each with area A0 /9n (for n ≥ 1). Show that the total area of the Koch snowflake is 85 A0 . (b) {an } is a positive sequence. SOLUTION The correct response is (b). Recall that S N = a1 + a2 + a3 + · · · + a N ; thus, S N − S N −1 = a N . If S N is increasing, then S N − S N −1 ≥ 0. It then follows that a N ≥ 0; that is, {an } is a positive sequence.

2. What are the hypotheses of the Integral Test? SOLUTION The hypotheses for the Integral Test are: A function f (x) such that an = f (n) must be positive, decreasing, and continuous for x ≥ 1.

3. Which test would you use to determine whether

∞ #

n −3.2 converges?

n=1 1 , we see that the indicated series is a p-series with p = 3.2 > 1. Therefore, the Because n −3.2 = 3.2 n series converges. SOLUTION

4. Which test would you use to determine whether SOLUTION

∞ #

1 √ converges? n+ n 2 n=1

Because 1 1 √ < n = 2 2n + n

n 1 , 2

and ∞ n # 1 n=1

2

is a convergent geometric series, the comparison test would be an appropriate choice to establish that the given series converges. 5. Ralph hopes to investigate the convergence of

∞ −n ∞ # # e 1 by comparing it with . Is Ralph on the right track? n n n=1 n=1


Convergence of Series with Positive Terms

609

No, Ralph is not on the right track. For n ≥ 1, 1 e−n < ; n n

∞ # 1 is a divergent series. The Comparison Test therefore does not allow us to draw a conclusion about the n n=1 ∞ −n # e convergence or divergence of the series . n n=1

however,

Exercises In Exercises 1–14, use the Integral Test to determine whether the infinite series is convergent.

1.

∞ # 1 n4 n=1

1 Let f (x) = 4 . This function is continuous, positive and decreasing on the interval x ≥ 1, so the Integral x Test applies. Moreover, " R " ∞ 1 1 dx 1 −4 = lim x d x = − lim −1 = . 4 3 3 3 R→∞ R→∞ x R 1 1 SOLUTION

The integral converges; hence, the series

3.

∞ # 1 also converges. n4 n=1

∞ # ∞ # n −1/31 n=1 n+3 n=1

1 1 Let f (x) = x − 3 = √ . This function is continuous, positive and decreasing on the interval x ≥ 1, so the 3 x Integral Test applies. Moreover,

SOLUTION

" ∞ 1

x −1/3 d x = lim

" R

R→∞ 1

The integral diverges; hence, the series

∞ #

x −1/3 d x =

3 lim R 2/3 − 1 = ∞. 2 R→∞

n −1/3 also diverges.

n=1 ∞ # n2 ∞ # 5. 1 3 + 9)5/2 (n √ n=25 n −4 n=5 SOLUTION

Let f (x) =

x2

5/2 . This function is positive and continuous for x ≥ 25. Moreover, because x3 + 9

f (x) =

2x(x 3 + 9)

5/2

3/2 − x 2 · 52 (x 3 + 9) · 3x 2 5

(x 3 + 9)

=

x(36 − 11x 3 ) 7/2

2(x 3 + 9)

,

we see that f (x) < 0 for x ≥ 25, so f is decreasing on the interval x ≥ 25. The Integral Test therefore applies. To evaluate the improper integral, we use the substitution u = x 3 + 9, du = 3x 2 d x. We then find " ∞

" R " R 3 +9 1 x2 x2 du d x = lim d x = lim 3 R→∞ 15634 u 5/2 R→∞ 25 (x 3 + 9)5/2 25 (x 3 + 9)5/2 1 2 2 1 = = − lim − . 9 R→∞ (R 3 + 9)3/2 156343/2 9 · 156343/2


7.

∞ # ∞ 1 # n n 2 + 12 n=1 (n + 1)3/5 n=1

∞ #

n2 5/2 also converges. 3 n=25 n + 9

610

C H A P T E R 11

INFINITE SERIES

1 . This function is positive, decreasing and continuous on the interval x ≥ 1, hence the Let f (x) = 2 x +1 Integral Test applies. Moreover, SOLUTION

" ∞ 1

" R

dx dx π π π π = lim = lim tan−1 R − = − = . 2 2 4 2 4 4 R→∞ 1 x + 1 R→∞ x +1


9.

∞ # 2 ∞ # ne−n 1 n=1 n(n + 1) n=1

SOLUTION

∞ #

1 also converges. 2+1 n n=1

Let f (x) = xe−x . This function is continuous and positive on the interval x ≥ 1. Moreover, because

2 2 2 f (x) = 1 · e−x + x · e−x · (−2x) = e−x 1 − 2x 2 , 2

we see that f (x) < 0 for x ≥ 1, so f is decreasing on this interval. To compute the improper integral we make the substitution u = x 2 , du = 2x d x. Then, we find " ∞ 1

xe−x d x = lim 2

" R

R→∞ 1


xe−x d x =

∞ #

2

" 2

2 1 1 R −u 1 e du = − lim e−R − e−1 = . 2 1 2 R→∞ 2e

ne−n also converges. 2

n=1 ∞ # ∞1 # 11. 1 2ln n n=1 n(ln n)2 n=2 SOLUTION Note that

2ln n = (eln 2 )ln n = (eln n )ln 2 = n ln 2 . Thus, ∞ #

∞ # 1 = . ln n ln 2 2 n n=1 n=1

1

1 Now, let f (x) = ln 2 . This function is positive, continuous and decreasing on the interval x ≥ 1; therefore, the Integral x Test applies. Moreover, " R " ∞ 1 dx dx = lim = lim (R 1−ln 2 − 1) = ∞, ln 2 ln 2 1 − ln 2 R→∞ R→∞ 1 x x 1 because 1 − ln 2 > 0. The integral diverges; hence, the series

13.

∞ # ln n ∞ # 1 n2 2 n=1 n −1 n=4

SOLUTION

Let f (x) =

∞ #

1 also diverges. ln n 2 n=1

ln x . Because x2

1 · x 2 − 2x ln x x (1 − 2 ln x) 1 − 2 ln x f (x) = x = = , x4 x4 x3 √ we see that f (x) < 0 for x > e ≈ 1.65. We conclude that f is decreasing on the interval x ≥ 2. Since f is also positive and continuous on this interval, the Integral Test can be applied. By Integration by Parts, we find " " ln x ln x 1 ln x d x = − + x −2 d x = − − + C; x x x x2

therefore, " R " ∞ ln x ln x ln R 1 ln 2 1 ln R 1 + ln 2 d x = lim d x = lim + − − = − lim . 2 R R 2 R→∞ 2 R→∞ 2 R→∞ R x2 x2 2

S E C T I O N 11.3


611

We compute the resulting limit using L’Hôpital’s Rule: lim

R→∞

ln R 1/R = lim = 0. R R→∞ 1

Hence, " ∞ 1 + ln 2 ln x dx = . 2 2 x 2 The integral converges; therefore, the series by adding the finite sum

∞ # ln n also converges. Since the convergence of the series is not affected n2 n=2

1 ∞ # # ln n ln n , the series also converges. 2 n n2 n=1 n=1

∞ ∞ # # 1 ∞ converges. Hint: Compare with 15. Use# the Comparison Test to show that n −3 . n 3 n + 8n n=1 n=1 2n n=1 ∞ # SOLUTION We compare the series with the p-series n −3 . For n ≥ 1, n=1

1 1 ≤ 3. n 3 + 8n n Since

∞ ∞ # # 1 1 converges (it is a p-series with p = 3 > 1), the series also converges by the Comparison 3 3 + 8n n n n=1 n=1

Test. ∞ # ∞1√ . Verify that for n ≥ 1 ∞ # # 17. Let S = 1 diverges by comparing with n −1 . Shown=1 thatn + n 2 n=2 n − 3 n=2 1 1 1 1 √ ≤ , √ ≤ √ n n+ n n+ n n

Can either inequality be used to show that S diverges? Show that SOLUTION

For n ≥ 1, n +

√

n ≥ n and n +

√

n≥

√

1 1 √ ≤ n n+ n These inequalities indicate that the series

1 1 and conclude that S diverges. √ ≥ 2n n+ n

n. Taking the reciprocal of each of these inequalities yields and

1 1 √ ≤ √ . n+ n n

∞ #

∞ ∞ ∞ # # # 1 1 1 1 √ ; however, √ is smaller than both and and n n n+ n n n=1 n=1 n=1 n=1

∞ # 1 √ both diverge so neither inequality allows us to show that S diverges. n n=1 √ √ On the other hand, for n ≥ 1, n ≥ n, so 2n ≥ n + n and

1 1 . √ ≥ 2n n+ n ∞ ∞ # # 1 1 =2 diverges, since the harmonic series diverges. The Comparison Test then lets us conclude 2n n n=1 n=1 ∞ # 1 that the larger series √ also diverges. n + n n=1

The series

In Exercises 19–31, use the Comparison Test to determine whether the infinite series# is convergent. ∞ 1 Which of the following inequalities can be used to study the convergence of √ ? Explain. ∞ 2+ n # 1 n n=2 19. n2n n=1 1 1 1 1 √ ≤ √ ≤ 2 ∞√ , n # 2 2 1 n n + n n n+≥ 1, n n SOLUTION We compare with the geometric series . For 2 n=1 1 1 ≤ n = n2n 2

n 1 . 2

612

C H A P T E R 11

INFINITE SERIES

Since

∞ n # 1

2

n=1

converges (it is a geometric series with r = 12 ), we conclude by the Comparison Test that

converges.

∞ # 1 also n n2 n=1

∞ 1/3 # ∞k # 1 2 √k k + k=1 n + 2n n=1 SOLUTION For k ≥ 1,

21.

1 k 1/3 k 1/3 ≤ 2 = 5/3 . 2 k +k k k ∞ #

5 1 is a p-series with p = > 1, so it converges. By the Comparison Test we can therefore conclude 5/3 3 k k=1 ∞ # k 1/3 also converges. that the series k2 + k k=1 The series

23.

∞ # 1√ ∞ # n 3 n=1 nn + − 13 n=4

SOLUTION

For n ≥ 1,

The series

∞ # 1 n=1

that the series

3 n2

1 n3 + 1

1 1 ≤ √ = 3/2 . 3 n n

converges, since it is a p-series with p =

3 > 1. By the Comparison Test we can therefore conclude 2

∞ #

1 also converges. 3 n=1 n + 1

∞ # sin2 k ∞ # n3 k2 5 k=1 n + 4n + 1 n=1 SOLUTION For k ≥ 1, 0 ≤ sin2 k ≤ 1, so

25.

0≤

1 sin2 k ≤ 2. 2 k k

∞ # 1 is a p-series with p = 2 > 1, so it converges. By the Comparison Test we can therefore conclude that k2 k=1 ∞ # sin2 k the series also converges. k2 k=1

The series

∞ # ∞ 4 # 1 m! + 4m n m=1 (ln n)2 n=2 SOLUTION For m ≥ 1,

27.

4 4 ≤ m = m! + 4m 4

m−1 1 . 4

∞ m−1 # 1 1 is a geometric series with r = , so it converges. By the Comparison Test we can therefore 4 4 m=1 ∞ # 4 conclude that the series also converges. m! + 4m m=1

The series

29.

∞ # 2 ∞ # 2−k 2 n k=1 3 + 3−n n=1

SOLUTION

For k ≥ 1, k 2 ≥ k and 1 ≤ k = 2 k 2 2 1

k 1 . 2

S E C T I O N 11.3

∞ k # 1

The series

2 k=1

is a geometric series with r =

conclude that the series

∞ # 1 k=1

2 2k

=

∞ #


613

1 , so it converges. By the Comparison Test we can therefore 2

2−k also converges. 2

k=1

∞ # ∞ ln n # ln n n 3 + 33 ln n n=1 n n=1 SOLUTION For n ≥ 1, 0 ≤ ln n ≤ n and

31.

ln n n 1 0≤ 3 ≤ 3 = 2. n + 3 ln n n n ∞ # 1 is a p-series with p = 2 > 1, so it converges. By the Comparison Test we can therefore conclude that n2 n=1 ∞ # ln n the series also converges. 3 n + 3 ln n n=1

The series

∞ # n∞ # 33. Does converge for any c? 1 2 Show that sin is a positive, convergent series. Hint: Use the inequality sin x ≤ x for x ≥ 0. n=1 n + c n 2 n=1 SOLUTION Because

lim

n→∞

it follows from the Divergence Test that the series

n n2 + c

= lim

n

n→∞ n

= 1 = 0,

∞ #

n diverges for all values of c. 2+c n n=1

In Exercises 34–42, use the Limit Comparison Test to prove convergence or divergence of the infinite series. 35.

∞ # ∞ 1 2 # n√ n 2 −4 n n=2 n −1 n=2

SOLUTION

We find

1 1 1 1 Let an = 2 √ . For large n, 2 √ ≈ 2 , so we apply the Limit Comparison Test with bn = 2 . n − n n − n n n 1√

an n2 n2 − n = lim = lim √ = 1. 1 n→∞ bn n→∞ n→∞ n 2 − n 2

L = lim

n

∞ # 1 1 is a p-series with p = 2 > 1, so it converges; hence, the series also converges. Because L 2 2 n n n=1 n=2 ∞ # 1 exists, by the Limit Comparison Test we can conclude that the series √ converges. 2− n n n=2

The series

∞ #

∞ # n3 ∞ # n 4 2 n=3 n −32n + 1 n=2 n − 1 SOLUTION Let

37.

n3 n3 n3 , an = = 2 = n −1 (n 2 − 1)2 n 4 − 2n 2 + 1 for n ≥ 3. For large n, n3 n2 − 1

≈ n,

so we apply the Limit Comparison Test with bn = n. We find n3

2 an n2 = lim n −1 = lim 2 = 1. L = lim n→∞ bn n→∞ n n→∞ n − 1

614

C H A P T E R 11

INFINITE SERIES

The series the series

∞ #

n diverges by the Divergence Test. Because L > 0, by the Limit Comparison Test we can conclude that

n=2 ∞ #

n3 also diverges. 4 2 n=3 n − 2n + 1

∞ n # ∞e + n # 3n + 5 2n e − n2 n=1 n(n − 1)(n − 2) n=3 SOLUTION Let

39.

en + n en + n 1 = n = n . an = 2n n 2 (e − n)(e + n) e −n e −n For large n, 1 1 ≈ n = e−n , en − n e so we apply the Limit Comparison Test with bn = e−n . We find 1

n −n an en = lim e −n = lim n = 1. n→∞ bn n→∞ e n→∞ e − n

L = lim

The series

∞ # n=1

e−n =

∞ n # 1

e n=1

is a geometric series with r = 1e < 1, so it converges. Because L exists, by the Limit

Comparison Test we can conclude that the series

∞ # en + n also converges. e2n − n 2 n=1

∞ ∞ # # ∞1 − cos 1 Hint: Compare with ∞ # # n −2 . ln n 1 n . Hint: Use L’Hˆ o pital’s Rule to compare with . n=1 n=1 2 3/2 n n n=1 n=1 1 1 SOLUTION Let an = 1 − cos , and apply the Limit Comparison Test with bn = . We find n n2

41.

− 12 sin 1x 1 − cos n1 1 − cos 1x sin 1x 1 an x = lim = lim = lim = . lim 1 1 2 n→∞ bn n→∞ x→∞ x→∞ 2 x→∞ 1 − 3 2 2

L = lim

n

x

x

x

As x → ∞, u = 1x → 0, so L=

sin 1x 1 1 sin u 1 = lim lim = . 2 x→∞ 1 2 u→0 u 2 x

∞ # 1 is a p-series with p = 2 > 1, so it converges. Because L exists, by the Limit Comparison Test we can 2 n n=1 ∞ # 1 conclude that the series also converges. 1 − cos n n=1

The series

∞ # ∞that if a ≥ 0 and lim n 2 a exists, then # an converges. Hint: Show that if M is larger than lim n 2 an , 43. Show n n −1/n n→∞ (1 − 2 ) Hint:n→∞ Compare with the harmonic series. n=1

then an n=1 ≤ M/n 2 for n sufficiently large. SOLUTION

Let lim n 2 an = L. Then there exists an integer N such that, for all n ≥ N , n→∞

|n 2 an − L|
e2 . Show that n=2 ln n n n=2 SOLUTION Using L’Hˆ opital’s Rule, 1 x −1/2 x 1/2 1 = lim x 1/2 = ∞; = lim 2 1 x→∞ ln x x→∞ x→∞ 2

lim

x

thus, there exists an integer N such that n 1/2 > 1 or ln n

n 1/2 > ln n

for all n ≥ N . Hence, 1 1 > 1/2 ln n n

and

1 1 > 2 n (ln n)

∞ # 1 also diverges. By the Comparison Test we can therefore n n=N ∞ ∞ # # 1 diverges. It follows that the series (ln n)−2 also diverges. conclude that the series 2 (ln n) n=N n=2

for all n ≥ N . The harmonic series diverges, so the series

∞ # ∞ 1 # 47. For which a does 1converge? n a ln n a converge? For which a does n=2 n(ln n) n=2 SOLUTION First consider the case a > 1. For n ≥ 3, ln n > 1 and

1 1 < a. n a ln n n ∞ ∞ # # 1 1 is a p-series with p = a > 1, so it converges; hence, also converges. By the Comparison Test a n na n=1 n=3 ∞ ∞ # # 1 1 we can therefore conclude that the series converges, which implies the series also converges. a ln n a ln n n n n=3 n=2

The series

For a ≤ 1, n a ≤ n so

1 1 ≥ n a ln n n ln n 1 . For x ≥ 2, this function is continuous, positive and decreasing, so the Integral Test applies. x ln x Using the substitution u = ln x, du = 1x d x, we find for n ≥ 2. Let f (x) =

" ∞ " R " ln R dx dx du = lim = lim = lim (ln(ln R) − ln(ln 2)) = ∞. x ln x x ln x u R→∞ R→∞ R→∞ 2 2 ln 2 The integral diverges; hence, the series ∞ #

∞ #

1 also diverges. By the Comparison Test we can therefore conclude that n ln n n=2

1 a ln n diverges. n n=2 To summarize,

the series

∞ # n=2

1 n a ln n

Use the Integral Test to show that

converges for a > 1 and diverges for a ≤ 1.

∞ # (ln n)k 2

" ∞ converges for all exponents k. You may use that

u k e−u du

616

C H A P T E R 11

INFINITE SERIES

In Exercises 49–74, determine convergence or divergence using any method covered so far. 49.

∞ #

1

n=4

n2 − 9

SOLUTION

1 1 and bn = 2 : Apply the Limit Comparison Test with an = 2 n −9 n 1

2 an n2 = 1. = lim n 1−9 = lim 2 L = lim n→∞ bn n→∞ n→∞ n − 9 2

n

∞ #

∞ #

1 1 converges, the series also converges. Because L exists, by the Limit Comparison Test 2 2 n n n=1 n=4 ∞ # 1 we can conclude that the series converges. 2−9 n n=4 √ ∞ # ∞ n 2 # 51. cos n 4n + 9 n=1 n2 n=1 √ 1 n SOLUTION Apply the Limit Comparison Test with an = and bn = √ : 4n + 9 n

Since the p-series

√

n

an n 1 = lim 4n+9 = lim = . L = lim n→∞ bn n→∞ √1 n→∞ 4n + 9 4 n

The series ∞ #

√

∞ #

1 √ is a divergent p-series. Because L > 0, by the Limit Comparison Test we can conclude that the series n n=1

n also diverges. 4n + 9 n=1 53.

∞ # ∞1 # n 2e−n n=1 3 n=1

SOLUTION

2

Because n 2 ≥ n for n ≥ 1, 3n ≥ 3n and 1

1 ≤ n = 2 3 3n

n 1 . 3

∞ n # 1

1 is a geometric series with r = , so it converges. By the Comparison Test we can therefore conclude 3 3 n=1 ∞ # 1 also converges. that the series n2 3 n=1 The series

∞ # ∞ 1 # 1 n 3/2 2ln n n=2 n + sin n n=1 SOLUTION For n ≥ 3, ln n > 1, so n 3/2 ln n > n 3/2 and

55.

1 n 3/2 ln n

1 < 3/2 . n

∞ ∞ # # 1 1 is a convergent p-series, so the series also converges. By the Comparison Test we can 3/2 3/2 n n n=1 n=3 ∞ ∞ # # 1 1 therefore conclude that the series converges. Hence, the series also converges. 3/2 ln n 3/2 ln n n n n=3 n=2

The series

57.

∞ # ∞ 1 # 21/k n 1/2 ln n n=2 k=1

SOLUTION

By L’Hôpital’s Rule 1 x −3/4 x 1/4 1 = lim x 1/4 = ∞; = lim 4 1 x→∞ ln x x→∞ x→∞ 4

lim

x


S E C T I O N 11.3

617

so there exists an integer N such that for all n ≥ N n 1/4 ≥ 1 or ln n

n 1/4 ≥ ln n.

Therefore, for n ≥ N , 1 1 ≥ 3/4 . n 1/2 ln n n ∞ #

1

∞ #

1 also diverges. By the Comparison Test we can therefore 3/4 n n=N n=1 ∞ ∞ # # 1 1 conclude that diverges. Hence, the series also diverges. 1/2 1/2 ln n n ln n n n=N n=2 The series

is a divergent p-series, so

n 3/4

∞ # ∞n # 4n n2 e n=2 5n − 2n n=1 2 SOLUTION Let f (x) = xe−x . This function is continuous and positive for x ≥ 2. Moreover, as

59.

f (x) = x(−2xe−x ) + e−x = (1 − 2x 2 )e−x , 2

2

2

we see that f (x) < 0 for x ≥ 2, so f is decreasing on this interval. The Integral Test therefore applies. Now, " ∞ " R

1 2 2 2 1 xe−x d x = lim xe−x d x = − lim e−R − e−4 = e−4 . 2 R→∞ 2 R→∞ 2 2 The integral converges; hence, the series

61.

∞ # n n n=2 e

2

also converges.

∞ n # ∞2 # n − sin n n 3 −n n=1 n3 n=1

SOLUTION

2n 2n Apply the Limit Comparison Test with an = n and bn = n : 3 −n 3 2n

n an 1 = lim 3 2−n = lim . n n→∞ bn n→∞ n→∞ 1 − nn n 3

L = lim

3

Now, lim

n

n→∞ 3n

= lim

x

x→∞ 3x

= lim

1

x→∞ 3x ln 3

= 0,

so 1 an = = 1. n→∞ bn 1−0

L = lim The series

∞ n # 2

is a convergent geometric series. Because L exists, by the Limit Comparison Test we can conclude

3 ∞ #

2n

n=1

3n − n

n=1

that the series

63.

also converges.

∞ # tan−1 n ∞ # 1 n2 n=1 n ln n −n n=2

SOLUTION

Apply the Limit Comparison Test with an = an = lim n→∞ bn n→∞

L = lim

The series series

tan−1 n 1 and bn = 2 : 2 n n

tan−1 n π n2 = lim tan−1 n = . 1 n→∞ 4 n2

∞ # 1 is a convergent p-series. Because L exists, by the Limit Comparison Test we can conclude that the 2 n n=1

∞ # tan−1 n also converges. n2 n=1

618

C H A P T E R 11

INFINITE SERIES

65.

∞ # ln n ∞ # 1 n3 n n=1 n n=1

SOLUTION

Apply the Limit Comparison Test with an =

ln n 1 and bn = 2 : n3 n ln n

an ln n 3 = 0. = lim n1 = lim n→∞ bn n→∞ n→∞ n 2

L = lim

n

The series series

∞ #

1 is a convergent p-series. Because L exists, by the Limit Comparison Test we can conclude that the n2 n=1

∞ # ln n also converges. n3 n=1

∞ # 2 + (−1)n n ∞ # 2 + (−1) n 3/2 n=1 n n=1 SOLUTION For n ≥ 1

67.

0
1, so

73.

1 1 < 1.2 . n 1.2 ln n n


S E C T I O N 11.3

619

∞ ∞ # # 1 1 is a convergent p-series, so the series also converges. By the Comparison Test we can 1.2 1.2 n n n=1 n=3 ∞ ∞ # # 1 1 converges. Finally, also converges. therefore conclude that 1.2 ln n 1.2 ln n n n n=3 n=2

The series

Approximating Infinite Sums In Exercises 75–77, let an = f (n), where f (x) is a continuous, decreasing function ∞ # ln n such that n 1.2 " ∞ n=1 f (x) d x 1

converges. 75. Show that " ∞ 1 SOLUTION

∞ #

f (x) d x ≤

an ≤ a1 +

" ∞

n=1

f (x) d x

1

4

From the proof of the Integral Test, we know that a2 + a3 + a4 + · · · + a N ≤

" N

f (x) d x ≤

1

" ∞ 1

f (x) d x;

that is, S N − a1 ≤

" ∞ 1

f (x) d x

or

S N ≤ a1 +

" ∞

f (x) d x.

1

Also from the proof of the Integral test, we know that " N 1

f (x) d x ≤ a1 + a2 + a3 + · · · + a N −1 = S N − a N ≤ S N .

Thus, " N 1

f (x) d x ≤ S N ≤ a1 +

" ∞ 1

f (x) d x.

Taking the limit as N → ∞ yields Eq. (4), as desired. ∞ # Eq. showas that 77. Let S Using = an . (4), Arguing in Exercise 75, show that n=1 M #

an +

" ∞

n=1

M+1

∞ # 1 5≤ M+1 #≤ 6 " ∞ 1.2 n f (x) d x ≤n=1 S≤ an +

M+1

n=1

f (x) d x

5

This series converges slowly. Use a computer algebra system to verify that S N < 5 for N ≤ 43,128 and S43,129 ≈ Conclude that 5.00000021. " ∞ M # 0≤S− an + f (x) d x ≤ a M+1 6 M+1

n=1

This yields a method for approximating S with an error of at most a M+1 . SOLUTION Following the proof of the Integral Test and the argument in Exercise 75, but starting with n = M + 1 rather than n = 1, we obtain

" ∞ M+1

Adding

M #

∞ #

f (x) d x ≤

an ≤ a M+1 +

" ∞

n=M+1

M+1

f (x) d x.

an to each part of this inequality yields

n=1 M # n=1

an +

" ∞ M+1

f (x) d x ≤

∞ # n=1

an = S ≤

M+1 # n=1

an +

" ∞ M+1

f (x) d x.

620

C H A P T E R 11

INFINITE SERIES M #

Subtracting

an +

n=1

" ∞ M+1

f (x) d x from each part of this last inequality then gives us 0≤S−

M #

an +

n=1

79.

" ∞ M+1

f (x) d x

≤ a M+1 .

Apply Eq. (5) with M = 40,000 to show that Use Eq. (5) with M = 43,129 to prove that ∞ # ∞ 1 1.644934066 ≤ # 1≤ 1.644934068 5.5915810 ≤n=1 n 2 1.2 ≤ 5.5915839 n n=1

Is this consistent with Euler’s result, according to which this infinite series has sum π 2 /6? 1 1 SOLUTION Using Eq. (5) with f (x) = , an = 2 and M = 40000, we find 2 x n " ∞ " ∞ ∞ # dx 1 dx ≤ ≤ S40001 + . S40000 + 2 2 x n 40001 40001 x 2 n=1 Now, S40000 = 1.6449090672; S40001 = S40000 +

1 = 1.6449090678; 40001

and " R dx dx 1 1 1 = lim = − lim − = = 0.0000249994. 40001 40001 R→∞ 40001 x 2 R→∞ R 40001 x 2

" ∞

Thus, 1.6449090672 + 0.0000249994 ≤

∞ # 1 ≤ 1.6449090678 + 0.0000249994, 2 n n=1

or 1.6449340665 ≤

Since 81.

∞ # 1 ≤ 1.6449340672. 2 n n=1

π2 ≈ 1.6449340668, our approximation is consistent with Euler’s result. 6 ∞ # ∞ Using a CAS and Eq. (6), determine the value of# n −5 to within an error less than 10−4 . n −4 to within an error less than 10−4 . Check that your Using a CAS and Eq. (6), determine the value of n=1 n=1

SOLUTION result is

−5 and a = n −5 , we have Using Eq.with (6) with = xwho n that the sum is equal to π 4 /90. consistent that off (x) Euler, proved " ∞ ∞ M+1 # # −5 −5 −5 0≤ n − n + x d x ≤ (M + 1)−5 . n=1

n=1

M+1

To guarantee an error less than 10−4 , we need (M + 1)−5 ≤ 10−4 . This yields M ≥ 104/5 − 1 ≈ 5.3, so we choose M = 6. Now, 7 #

n −5 = 1.0368498887,

n=1

and " ∞ 7

x −5 d x = lim

" R

R→∞ 7

x −5 d x = −

1 1 = 0.0001041233. lim R −4 − 7−4 = 4 R→∞ 4 · 74

Thus, ∞ # n=1

n −5 ≈

7 # n=1

n −5 +

" ∞ 7

x −5 d x = 1.0368498887 + 0.0001041233 = 1.0369540120.

S E C T I O N 11.3


621

83. Let pn denote the nth prime number ( p1 = 2, p2 = 3, etc.). It is known that there is a constant C such that The harmonic series diverges, but it does so very slowly. Show that the partial sum S N satisfies pn ≤ Cn ln n. Prove the divergence of ∞ N ≤ 1 + 1 + 1 + · · · + 1 ≤ 1 + ln N # ln 1 1 1 1 1 1 = 2+ 3+ + +N + · · · p 2 3 5 7 11 n Find an N such that S N ≥ 100. Verify that S N ≤ 10 for N ≤ 8,100.n=1 SOLUTION

Since pn ≤ Cn ln n for n ≥ 2, 1 1 1 ≥ · . pn C n ln n

1 . For x ≥ 2, this function is continuous, positive and decreasing, so the Integral Test applies. x ln x Using the substitution u = ln x, du = 1x d x, we find

Now, let f (x) =

" R " ln R " ∞ dx dx du = lim = lim = lim (ln(ln R) − ln(ln 2)) = ∞. x ln x x ln x u R→∞ R→∞ R→∞ 2 2 ln 2 The integral diverges; hence, the series the series

∞ #

1 also diverges. By the Comparison Test we can therefore conclude that n ln n n=2

∞ ∞ # # 1 1 diverges; hence, also diverges. p p n=2 n n=1 n

How far can a stack of identical books (each of unit length) extend without tipping over? The stack will not tip Further Insights and Challenges over if the (n + 1)st book is placed at the bottom of the stack with its left edge located at the center of mass of the

∞ # n converges be the center of mass ofseries the first n rbooks, measured along x-axis. first the n books (Figure Let cnagain 85. Use Integral Test 5). to prove that the geometric if 0 < r 1. 1 . Recall that if objects n=1 of mass m 1 , . . . , m n are placed along the x-axis with (a) Prove that cn+1 = cn + 2(n + 1) x SOLUTION Letoff (x) ≥center 1. Forof 0< r 1, limn→∞ r n = ∞, and the series rn n=1

n=1

diverges by the Divergence Test. ∞ # 1 ∞ # . There is a similar telescoping 87. Kummer’s Acceleration Method Suppose we wish to approximate S = − ln n 2 an , where an = (ln(ln n)) . Let S = n n=1 n=2can be computed exactly (see Example 1 in Section 11.2): series whose value n)) . (a) Show, by taking logarithms, that an = n − ln(ln(ln ∞ # 1e2 (b) Show that ln(ln(ln n)) ≥ 2 if n > C, where C = ee+ .1) = 1 n(n n=1 (c) Show that S converges. (a) Verify that

S=

∞ # 1 1 1 − + n(n + 1) n=1 n 2 n(n + 1) n=1 ∞ #

Thus for M large, S ≈1+

M #

1

n=1

n 2 (n + 1)

(b) Explain what has been gained. Why is (7) a better approximation to S than (c)

Compute 1,000 #

1 , 2 n n=1

1+

100 #

1 2 (n + 1) n n=1

Which is a better approximation to S, whose exact value is π 2 /6?

7 M # 1 ? 2 n n=1

622

C H A P T E R 11


∞ ∞ # # 1 1 both converge, and 2 n(n + 1) n n=1 n=1 # ∞ ∞ ∞ ∞ ∞ ∞ # # # # # 1 1 1 1 1 1 1 − − = S. + = + = 2 2 n(n + 1) n(n + 1) n(n + 1) n(n + 1) n n n2 n=1 n=1 n=1 n=1 n=1 n=1

(a) Because the series

Now, 1 n+1 1 n 1 − = 2 − = 2 , n(n + 1) n2 n (n + 1) n 2 (n + 1) n (n + 1) so, for M large, S ≈1+

(b) The series

M #

1 . 2 (n + 1) n n=1

∞ #

∞ # 1 1 since the degree of n in the denominator is larger. converges more rapidly than 2 n (n + 1) n2 n=1 n=1

(c) Using a computer algebra system, we find 1000 #

1 = 1.6439345667 2 n n=1

and 1 +

100 #

1 = 1.6448848903. 2 (n + 1) n n=1

The second sum is more accurate because it is closer to the exact solution

The series S =

∞ #

π2 ≈ 1.6449340668. 6

k −3 has been computed to more than 100 million digits. The first 30 digits are

11.4 Absolute andk=1 Conditional Convergence S = 1.202056903159594285399738161511


∞ the Acceleration Method of Exercise 87 with M = 100 and auxiliary series # Approximate S using 1. Suppose that S = an is conditionally convergent. Which of the following statements are correct? ∞ # n=0 (n(n + 1)(n + 2))−1 . R = ∞ # n=1 |an | may or may not converge. (a) n=0

to Exercise 46 in Section 11.2, R is a telescoping series with the sum R = 14 . (b) According S may or may not converge. ∞ # (c) |an | diverges. n=0 SOLUTION

By definition, because

∞ #

an is conditionally convergent, we know that

n=1

∞ # n=1

an converges but

∞ #

|an |

n=1

diverges. Thus: ∞ #

(a) This statement is incorrect:

|an | must diverge.

n=0

(b) This statement is incorrect: S must converge. (c) This statement is correct. 2. Which of the following statements is equivalent to Theorem 1? ∞ ∞ # # |an | diverges, then an also diverges. (a) If (b) If (c) If

n=0 ∞ # n=0 ∞ #

n=0

an diverges, then

∞ #

|an | also diverges.

n=0 ∞ #

an converges, then

n=0 SOLUTION

|an | also converges.

n=0

The correct answer is (b): If

∞ # n=0

statements (a) and (c) are not true in general.

an diverges, then

∞ # n=0

|an | also diverges. Take an = (−1)n n1 to see that

S E C T I O N 11.4

3. Lathika argues that

∞ #

Absolute and Conditional Convergence

623

√ (−1)n n is an alternating series and therefore converges. Is Lathika right?

n=1 SOLUTION

No. Although

∞ #

√ √ (−1)n n is an alternating series, the terms an = n do not form a decreasing sequence

n=1

that tends to zero. In fact, an =

√

n is an increasing sequence that tends to ∞, so

∞ #

√ (−1)n n diverges by the Divergence

n=1

Test. 4. Give an example of a series such that SOLUTION

an converges but

#

|an | diverges.

# (−1)n # 1 The series converges by the Leibniz Test, but the positive series is a divergent p-series. √ √ 3 3 n n

Exercises 1. Show that

SOLUTION

#

∞ # (−1)n converges absolutely. 2n n=0 ∞ # 1 1 is a geometric series with r = . Thus, the positive series converges, and the n 2 2 n=0

The positive series

given series converges absolutely. In Exercises determine whether series converges absolutely, conditionally, or not at all. Show3–12, that the following series the converges conditionally: 3.

∞ # (−1)n √ n n=1

SOLUTION

∞ #

1 1 1 1 1 (−1)n−1 2/3 = 2/3 − 2/3 + 2/3 − 2/3 + · · · n 1 2 3 4 n=1

Let an = √1 . Then an forms a decreasing sequence that tends to zero; hence, the series n

verges by the Leibniz Test. However, the positive series

∞ # (−1)n con√ n n=1

∞ # 1 √ is a divergent p-series, so the original series converges n n=1

conditionally. 5.

∞ # (−1)n−1n 4 ∞ # (−1) n (1.1)n n=1 n3 + 1 n=1

SOLUTION

The positive series

∞ # 1 n is a convergent geometric series; thus, the original series converges abso1.1 n=1

lutely. ∞ # (−1)n+1 ∞ # sin n n ln n n=2 n2 n=1 1 SOLUTION Let an = n ln n . Then an forms a decreasing sequence (note that n and ln n are both increasing functions ∞ # (−1)n+1 converges by the Leibniz Test. However, the positive series of n) that tends to zero; hence, the series n ln n n=2 ∞ # 1 diverges, so the original series converges conditionally. n ln n n=2

7.

∞ # sin n π n ∞ # √(−1) n 1 n=1 n=1 1 + n SOLUTION sin n π = 0 for all n, so the general term of the series is zero. Consequently, the series, as well as the positive series, converge to zero; that is, the series converges absolutely.

9.

11.

∞ cos 1 # ∞ n # n tan 1 2 n(−1) n=1 n n=4

SOLUTION

The positive series is

∞ | cos 1 | # n n=1

n2

. Because | cos n1 | n2

1 ≤ 2 n

624

C H A P T E R 11

INFINITE SERIES ∞ ∞ | cos 1 | # # 1 n imply that the series converges. 2 2 n n n=1 n=1

for all n, the Comparison Test and the convergence of the p-series Hence, the original series converges absolutely. ∞ # 1 13. Let # S∞=cos n1(−1)n+1 3 . n n=1 n n=1 S for 1 ≤ n ≤ 10. (a) Calculate n (b) Use Eq. (2) to show that 0.9 ≤ S ≤ 0.902. SOLUTION

(a) 1 S6 = S5 − 3 = 0.899782407 6 1 S7 = S6 + 3 = 0.902697859 7 1 S8 = S7 − 3 = 0.900744734 8 1 S9 = S8 + 3 = 0.902116476 9 1 S10 = S9 − 3 = 0.901116476 10

S1 = 1 1 7 S2 = 1 − 3 = = 0.875 8 2 1 S3 = S2 + 3 = 0.912037037 3 1 S4 = S3 − 3 = 0.896412037 4 1 S5 = S4 + 3 = 0.904412037 5 (b) By Eq. (2),

|S10 − S| ≤ a11 =

1 , 113

so S10 −

1 1 ≤ S ≤ S10 + 3 , 3 11 11

or 0.900365161 ≤ S ≤ 0.901867791. ∞ # (−1)n+1 ∞ decimal # to three 15. Approximate (−1)n+1 places. Use Eq. (2)n=1 to approximate to four decimal places. n4 n! n=1 ∞ # 1 (−1)n+1 SOLUTION Let S = , so that an = 4 . By Eq. (2), 4 n n n=1

|S N − S| ≤ a N +1 =

1 . (N + 1)4

To guarantee accuracy to three decimal places, we must choose N so that 1 < 5 × 10−4 (N + 1)4

or

N>

√ 4

2000 − 1 ≈ 5.7.

The smallest value that satisfies the required inequality is then N = 6. Thus, 1 1 1 1 1 S ≈ S6 = 1 − 4 + 4 − 4 + 4 − 4 = 0.946767824. 2 3 4 5 6 ∞ In Exercises 17–18,# use Eq. (2) to approximate the value of the series to within an error of at most 10−5 . n n−1 . Use a computer algebra system to calculate and plot the partial sums Sn for Let S = (−1) ∞ n2 + 1 # (−1)n+1 n=1 17. 1 ≤ n ≤ 100. Observe that the partial sums zigzag above and below the limit. n(n + 2)(n + 3) n=1

SOLUTION

Let S =

∞ #

1 (−1)n+1 , so that an = . By Eq. (2), n + 2) + 3) n + 2) (n (n (n (n + 3) n=1 |S N − S| ≤ a N +1 =

1 . (N + 1)(N + 3)(N + 4)

S E C T I O N 11.4


625

We must choose N so that 1 ≤ 10−5 (N + 1)(N + 3)(N + 4)

or (N + 1)(N + 3)(N + 4) ≥ 105 .

For N = 43, the product on the left hand side is 95,128, while for N = 44 the product is 101,520; hence, the smallest value of N which satisfies the required inequality is N = 44. Thus, S ≈ S44 =

44 #

(−1)n+1 = 0.0656746. n(n + 2)(n + 3) n=1

In Exercises determine convergence or divergence by any method. ∞ 19–26, # (−1)n+1 ln n ∞ # n! n=1 1 19. n n 3 +5 n=1 SOLUTION

For n ≥ 1 1 1 ≤ n = 3n + 5n 3

The series converges. 21.

∞ n # 1

3 n=1

n 1 . 3

is a convergent geometric series, so the Comparison Test implies that the series

∞ #

1 n + 5n also 3 n=1

∞ # (−1)n ∞ # n 2 n=1 nn 2+−1n n=2

1 This is an alternating series with an = . Because an is a decreasing sequence that converges to 2 n +1 ∞ # (−1)n converges by the Leibniz Test. zero, the series 2 n=1 n + 1 SOLUTION

23.

∞ n # 3 + (−1)n 2n ∞ # 1 5n n=1 2 n=1 n + 1

SOLUTION

The series

∞ n ∞ ∞ # # # 3 2 n (−1)n 2n = is a convergent geometric series, as is the series = . − 5n 5 5n 5 n=1 n=1 n=1 n=1 ∞ n # 3

Hence, ∞ n ∞ n ∞ # # # 3 + (−1)n 2n 3 2 n = + − 5n 5 5 n=1 n=1 n=1 also converges. ∞ # ∞ n ne−n # (−1) (−1)n+1 n=1 (2n + 1)! n=1 SOLUTION This is an alternating series with an = ne−n . Consider the function f (x) = xe−x . Using L’Hˆ opital’s Rule,

25.

x

1

lim = lim x = 0. x→∞ e x x→∞ e Moreover, f (x) =

e x − xe x 1−x = , ex e2x

so f (x) < 0 and f is decreasing for x > 1. Therefore, {an } is a decreasing sequence which converges to zero, and ∞ # (−1)n ne−n converges by the Leibniz Test. n=1

27. Show that ∞ # (−1)n n 4 2−n n=1

S=

1 1 1 1 1 1 − + − + − 2 2 3 3 4 4

converges by computing the partial sums. Does it converge absolutely?

626

C H A P T E R 11


The sequence of partial sums is S1 =

1 2

1 S2 = S1 − = 0 2 1 1 S3 = S2 + = 3 3 1 S4 = S3 − = 0 3 and, in general, ⎧ ⎨1, SN = N ⎩ 0,

for odd N for even N

Thus, lim S N = 0, and the series converges to 0. The positive series is N →∞

∞ # 1 1 1 1 1 1 1 + + + + + + ··· = 2 ; 2 2 3 3 4 4 n n=2

which diverges. Therefore, the original series converges conditionally, not absolutely. 29. Determine whether the following series converges conditionally: The Leibniz Test cannot be applied to 1 1 1 1 1 1 1 1 1 1 −1 + 1 − + ··· 1 − + 1 − 1 + 1− + ··· 3 2 − 5 + 3 2 −7 2 4+ 39 − 53 + 11 2 3 2 3 2 3 SOLUTION Although this is an alternating series, the sequence of terms {an } is not decreasing, so we cannot apply the Why not? Show that it converges by another method. Leibniz Test. However, we may express the series as ∞ # 1

n n=1

−

1 2n + 1

=

∞ #

n+1 . n(2n + 1) n=1

Using the Limit Comparison Test and comparing with the harmonic series, we find n+1 n+1 1 n(2n+1) = lim = . 1 n→∞ n→∞ 2n + 1 2 n

L = lim Because L > 0, we conclude that the series 1−

1 1 1 1 1 1 1 1 1 + − + − + − + − + ··· 3 2 5 3 7 4 9 5 11

diverges. #

#

an2 also converges. Then show by giving a counterexample # # 31. Prove variant ofifthe Leibniz Test: If {an } is a convergent. positive, decreasing sequence with lim an = 0, then that thean2following need not converge an is only conditionally n→∞ the series Prove that if and an Challenges converges absolutely, then Further Insights

a1 + a2 − 2a3 + a4 + a5 − 2a6 + · · · converges. Hint: Show that S3N is increasing and bounded by a1 + a2 , and continue as in the proof of the Leibniz Test. Following the hint, we first examine the sequence {S3N }. Now, S3N +3 = S3(N +1) = S3N + a3N +1 + a3N +2 − 2a3N +3 = S3N + a3N +1 − a3N +3 + a3N +2 − a3N +3 ≥ S3N

SOLUTION

because {an } is a decreasing sequence. Moreover, S3N = a1 + a2 −

N −1 #

2a3k − a3k+1 − a3k+2 − 2a3N

k=1

= a1 + a2 −

N −1 #

a3k − a3k+1 + a3k − a3k+2 − 2a3N ≤ a1 + a2

k=1

S E C T I O N 11.4


627

again because {an } is a decreasing sequence. Thus, {S3N } is an increasing sequence with an upper bound; hence, {S3N } converges. Next, S3N +1 = S3N + a3N +1

and

S3N +2 = S3N + a3N +1 + a3N +2 .

Given that lim an = 0, it follows that n→∞

lim S3N +1 = lim S3N +2 = lim S3N .

N →∞

N →∞

N →∞

Having just established that lim S3N exists, it follows that the sequences {S3N +1 } and {S3N +2 } converge to the same N →∞

limit. Finally, we can conclude that the sequence of partial sums {S N } converges, so the given series converges. 33. Prove the conditional convergence of Use Exercise 31 to show that the following series converges: 1 1 3 1 1 1 3 R = 11+ +1 − 2+ +1 + 1− +2 · · · 2 3−4 + 5 6 + 7 −8 S= + + ··· ln 2 ln 3 ln 4 ln 5 ln 6 ln 7 SOLUTION Using Exercise 31 as a template, we first examine the sequence {R4N }. Now, 1 1 1 3 + + − 4N + 1 4N + 2 4N + 3 4N + 4 1 1 1 1 1 1 − − − = RN + + + ≥ R4N . 4N + 1 4N + 4 4N + 2 4N + 4 4N + 3 4N + 4

R4N +4 = R4(N +1) = R4N +

Moreover, R4N = 1 +

−1 3 1 1 N# 1 1 1 3 1 1 + − − − − − ≤1+ + . 2 3 4k 4k + 1 4k + 2 4k + 3 4N 2 3 k=1

Thus, {R4N } is an increasing sequence with an upper bound; hence, {R4N } converges. Next, R4N +1 = R4N +

1 ; 4N + 1

1 1 + ; and R4N +2 = R4N + 4N + 1 4N + 2 R4N +3 = R4N +

1 1 1 + + , 4N + 1 4N + 2 4N + 3

so lim R4N +1 = lim R4N +2 = lim R4N +3 = lim R4N .

n→∞

N →∞

N →∞

N →∞

Having just established that lim R4N exists, it follows that the sequences {R4N +1 }, {R4N +2 } and {R4N +3 } converge N →∞ to the same limit. Finally, we can conclude that the sequence of partial sums {R N } converges, so the series R converges. Now, consider the positive series R+ = 1 +

1 1 3 1 1 1 3 + + + + + + + ··· 2 3 4 5 6 7 8

Because the terms in this series are greater than or equal to the corresponding terms in the divergent harmonic series, it follows from the Comparison Test that R + diverges. Thus, by definition, R converges conditionally. Matter by counterexample that the Leibniz Test does not remain true if {an } tends to 35. ShowAssumptions that the following seriesShow diverges: zero but we drop the assumption that the sequence an is decreasing. Hint: Consider 1 1 2 1 1 1 2 1 2+ 1 3− 1 4 +1 5 + 6 + 71− 8 1+ · · · 1 S 1= 1 + − + ··· + − n + ··· R= − + − + 2 4 3 8 4 16 n 2 Hint: Use the result of Exercise 33 to write S as the sum of a convergent and a divergent series. SOLUTION Let 1 1 1 1 1 1 1 1 R= − + − + − + ··· + − n+1 + · · · 2 4 3 8 4 16 n+1 2 This is an alternating series with

an =

⎧ 1 ⎪ ⎪ ⎨ k + 1 , n = 2k − 1 ⎪ ⎪ ⎩ 1 , 2k+1

n = 2k

628

C H A P T E R 11

INFINITE SERIES

Note that an → 0 as n → ∞, but the sequence {an } is not decreasing. We will now establish that R diverges. For sake of contradiction, suppose that R converges. The geometric series ∞ #

1 n+1 2 n=1 converges, so the sum of R and this geometric series must also converge; however, R+

∞ #

∞ # 1 = , n+1 n 2 n=1 n=2

1

which diverges because the harmonic series diverges. Thus, the series R must diverge. } is a rearrangement of {an } if {bn } has the same terms as {an } but occurring in a different order. Show 37. We say that {bn∞ # ∞ ∞ (ln x)a (ln n)a # # that (−1)n+1 converges for exponents a. Hint: Show is decreasing for anall converges absolutely, then Tthat = f (x)bn=also xconverges absolutely. that if {bProve n } is a rearrangement of {ann } and S = n=1

n=1

n=1

x sufficiently large. N # (This result does not hold if S is only conditionally convergent.) Hint: Prove that the partial sums |bn | are bounded. n=1

It can be shown further that S = T . SOLUTION

Suppose the series S =

∞ #

an converges absolutely and denote the corresponding positive series by

n=1

S+ =

∞ #

|an |.

n=1

Further, let TN =

N #

|bn | denote the N th partial sum of the series

n=1

∞ #

|bn |. Because {bn } is a rearrangement of {an },

n=1

we know that 0 ≤ TN ≤

∞ #

|an | = S + ;

n=1

that is, the sequence {TN } is bounded. Moreover, TN +1 =

N +1 #

|bn | = TN + |b N +1 | ≥ TN ;

n=1

that is, {TN } is increasing. It follows that {TN } converges, so the series

∞ #

|bn | converges, which means the series

n=1

∞ #

bn

n=1

converges absolutely. Assumptions Matter In 1829, Lejeune Dirichlet pointed out that the great French mathematician Augustin Louis Cauchy made a mistake in a published paper by improperly assuming the Limit Comparison Test to be valid 11.5for The Ratioseries. and Here RootareTests nonpositive Dirichlet’s two series: ∞ ∞ # # (−1)n (−1)n (−1)n √ , √ 1+ √ an+1 n an n n or lim n=1 ? 1. In the Ratio Test, is ρ equal to lim n=1 n→∞ an n→∞ an+1 the Limit Explain how they provide a counterexample to Comparison Test when the series are not assumed to be an+1 positive. SOLUTION In the Ratio Test ρ is the limit lim . n→∞ an


2. Is the Ratio Test conclusive for

SOLUTION

The general term of

∞ ∞ # # 1 1 ? ? Is it conclusive for n 2 n n=1 n=1

∞ # 1 1 is an = n ; thus, n 2 2 n=1

an+1 2n 1 1 a = 2n+1 · 1 = 2 , n

and

a 1 ρ = lim n+1 = < 1. n→∞ an 2

S E C T I O N 11.5

Consequently, the Ratio Test guarantees that the series The general term of

∞ # 1

n n=1

is an =

The Ratio and Root Tests

629

∞ # 1 converges. n 2 n=1

1 ; thus, n an+1 1 n n a = n + 1 · 1 = n + 1, n

and

a n ρ = lim n+1 = lim = 1. n→∞ an n→∞ n + 1

The Ratio Test is therefore inconclusive for the series

∞ # 1 . n n=1

3. Can the Ratio Test be used to show convergence if the series is only conditionally convergent? SOLUTION

No. The Ratio Test can only establish absolute convergence and divergence, not conditional convergence.

Exercises In Exercises 1–18, apply the Ratio Test to determine convergence or divergence, or state that the Ratio Test is inconclusive. 1.

∞ # 1 n 5 n=1

SOLUTION

With an = 51n , an+1 5n 1 1 a = 5n+1 · 1 = 5 n

Therefore, the series

3.

a 1 and ρ = lim n+1 = < 1. n→∞ an 5

∞ # 1 converges by the Ratio Test. 5n n=1

∞ # (−1)n−1n−1 ∞ # (−1) n nn n n=1 5 n=1

SOLUTION

n−1 With an = (−1) nn ,

n an+1 nn 1 1 n 1 −n 1 = · = , = 1 + a (n + 1)n+1 1 n+1 n+1 n+1 n n and


a 1 ρ = lim n+1 = 0 · = 0 < 1. n→∞ an e ∞ # (−1)n−1 converges by the Ratio Test. nn n=1

∞ # ∞ n # 3n + 2 n 2 + 13 n=1 5n + 1 n=0 SOLUTION With an =

5.

n , n 2 +1

an+1 n2 + 1 n+1 n+1 n2 + 1 , = · 2 a = (n + 1)2 + 1 · n n n + 2n + 2 n

and

a ρ = lim n+1 = 1 · 1 = 1. n→∞ a n

Therefore, for the series

∞ #

n

n=1

n2 + 1

, the Ratio Test is inconclusive.

We can show that this series diverges by using the Limit Comparison Test and comparing with the divergent harmonic series.

630

C H A P T E R 11

INFINITE SERIES

7.

∞ n # ∞2 n # 2 n 100 n=1 n n=1

SOLUTION

n With an = 2100 ,

n

100 an+1 n 100 n 2n+1 a = (n + 1)100 · 2n = 2 n + 1 n Therefore, the series

and

an+1 = 2 · 1100 = 2 > 1. ρ = lim n→∞ an

∞ # 2n diverges by the Ratio Test. n 100 n=1

∞ # 10n 3 ∞ # n n2 n=1 2 n2 n=1 3 10n , SOLUTION With an = n2

9.

2

2 n+1 an+1 2n 1 = 10 · = 10 · 2n+1 a 2 2 n 2(n+1) 10n Therefore, the series

11.

∞ # 10n n2 n=1 2

and

a ρ = lim n+1 = 10 · 0 = 0 < 1. n→∞ an

converges by the Ratio Test.

∞ n # e n ∞ # n e n n=1 n! n=1

SOLUTION

n With an = ne n ,

n an+1 nn e e en+1 n 1 −n = · = = , 1 + a (n + 1)n+1 en n+1 n+1 n+1 n n and

a 1 ρ = lim n+1 = 0 · = 0 < 1. n→∞ an e


13.

∞ n # e n converges by the Ratio Test. n n=1

∞ # ∞ n50n! # (−1) n n 4 n=0 n! n=1

SOLUTION

With an = (−1)n 4n!n , an+1 (n + 1)! 4n n+1 a = 4n+1 · n! = 4 n


and

a ρ = lim n+1 = ∞ > 1. n→∞ an

∞ #

n! (−1)n n diverges by the Ratio Test. 4 n=0

∞ # ∞1 # n! n ln n n=2 4 n n=1 1 SOLUTION With an = n ln n , an+1 1 n ln n n ln n a = (n + 1) ln(n + 1) · 1 = n + 1 ln(n + 1) , n

15.

and

an+1 ln n = 1 · lim ρ = lim . n→∞ an n→∞ ln(n + 1)

Now, lim

ln n

n→∞ ln(n + 1)

= lim

ln x

x→∞ ln(x + 1)

= lim

x→∞

1/(x + 1) x = lim = 1. x→∞ x + 1 1/x

S E C T I O N 11.5

Thus, ρ = 1, and the Ratio Test is inconclusive for the series Using the Integral Test, we can show that the series

17.

∞ 2 # ∞ n # 1 (2n + 1)! n=1 (2n)! n=1

SOLUTION

∞ #


∞ #

1 . n ln n n=2

1 diverges. n ln n n=2

2

n With an = (2n+1)! , an+1 n+1 2 1 (n + 1)2 (2n + 1)! , = a = (2n + 3)! · 2 n (2n + 3)(2n + 2) n n

and

a ρ = lim n+1 = 12 · 0 = 0 < 1. n→∞ an


∞ #

n2 converges by the Ratio Test. (2n + 1)! n=1

∞ # ∞that 2 n k 3−n converges for all exponents k. # 19. Show (n!) n=1 (2n)! n=1 SOLUTION With an = n k 3−n ,

an+1 (n + 1)k 3−(n+1) 1 1 k = = , 1+ a 3 n n k 3−n n and, for all k, an+1 1 = · 1 = 1 < 1. ρ = lim n→∞ an 3 3 Therefore, the series

∞ #

n k 3−n converges for all exponents k by the Ratio Test.

n=1 ∞ # ∞ # 2n x n2 converges if |x| < 12 . 21. Show that n converges Show that n x if |x| < 1. n=1 n=1

SOLUTION

With an = 2n x n , an+1 2n+1 |x|n+1 = 2|x| a = 2n |x|n n

Therefore, ρ < 1 and the series

∞ #

and

a ρ = lim n+1 = 2|x|. n→∞ an

2n x n converges by the Ratio Test provided |x| < 12 .

n=1 ∞ n # r n ∞ # 23. Show that rconverges if |r | < 1. n Show that converges for all r . n=1 n! n=1 n

SOLUTION

With an = rn , an+1 |r |n+1 n n a = n + 1 · |r |n = |r | n + 1 n

Therefore, by the Ratio Test, the series

∞ n # r n=1

n

an+1 = 1 · |r | = |r |. and ρ = lim n→∞ a

converges provided |r | < 1.

∞ # n! 1 n ∞ lim n # converges. Hint: Use = e. 25. Show that 1 + 2 n n→∞ n Is theren=1 any nvalue of k such that converges? k n n=1

n

631

632

C H A P T E R 11


With an = nn!n , n n an+1 1 −n n = (n + 1)! · n = = 1 + , a (n + 1)n+1 n! n+1 n n

and

a 1 ρ = lim n+1 = < 1. n→∞ an e


∞ # n! converges by the Ratio Test. n n n=1

In Exercises 26–31, assume that |an+1 /an | converges to ρ = 13 . What can you say about the convergence of the given series? 27.

∞ # ∞ # n 3 an nan n=1 n=1

SOLUTION

Let bn = n 3 an . Then b 1 n + 1 3 an+1 3 1 ρ = lim n+1 = lim a = 1 · 3 = 3 < 1. n→∞ bn n→∞ n n


∞ #

n 3 an converges by the Ratio Test.

n=1 ∞ # ∞ # 29. 3n ann 2 an n=1 n=1

SOLUTION

Let bn = 3n an . Then bn+1 3n+1 an+1 1 ρ = lim = lim = 3 · = 1. n→∞ bn n→∞ 3n an 3

Therefore, the Ratio Test is inconclusive for the series

∞ #

3n an .

n=1 ∞ # ∞ # 31. an2 n 4 an n=1 n=1

SOLUTION

Let bn = an2 . Then 2 2 b a 1 1 ρ = lim n+1 = lim n+1 = = < 1. n→∞ bn n→∞ an 3 9


∞ #

an2 converges by the Ratio Test.

n=1 ∞ # 1# ∞ 33. Is the Ratio Test conclusive for the p-series an+1 p ? a −1 converge (assume that a = 0 for all n)? n ρ = 4. Does Assume that converges to n n n=1 an n=1 SOLUTION

With an = n1p , p an+1 np n 1 = · = a (n + 1) p 1 n+1 n

Therefore, the Ratio Test is inconclusive for the p-series

and

a ρ = lim n+1 = 1 p = 1. n→∞ an

∞ # 1 . p n n=1

In Exercises 34–39, use the Root Test to determine convergence or divergence (or state that the test is inconclusive). 35.

∞ # ∞1 # 1 nn n n=1 10 n=0


With an = n1n , √ n


37.

an =

n

1 1 = nn n

and

lim

√ n

n→∞

an = 0 < 1.

∞ # 1 n converges by the Root Test. n n=0

k ∞ # k ∞ k # k 3k + 1 k=0 k + 10 k=0

SOLUTION

k k With ak = 3k+1 , √ k


39.


ak =

k

k k k = 3k + 1 3k + 1

and

lim

√ k

k→∞

ak =

1 < 1. 3

k k converges by the Root Test. 3k + 1 k=0 ∞ #

2 ∞ −n # 1 −n ∞ # 1 1+ 1n+ n=4 n n=1

SOLUTION

−n 2 With ak = 1 + n1 , √ n


an =

∞ # k=0

1+

n

1+

2 1 −n 1 −n = 1+ n n

and

lim

√ n

n→∞

an = e−1 < 1.

2 1 −n converges by the Root Test. n

In Exercises 41–52,∞determine convergence or divergence using any method covered in the text so far. # 2n 2 2 that diverges. Hint: Use 2n = (2n )n and n! ≤ n n . ∞ Prove # 2n + 4n n! n=1 41. 7n n=1 SOLUTION

Because the series ∞ n ∞ n # # 2 2 = n 7 7 n=1 n=1

and

∞ n ∞ n # # 4 4 = n 7 7 n=1 n=1

are both convergent geometric series, it follows that ∞ n ∞ n ∞ n # # # 2 + 4n 2 4 = + n 7 7 7 n=1 n=1 n=1

also converges. 43.

∞ 3 # n 3 ∞ # n 5n n=1 n! n=1

SOLUTION


45.

3

The presence of the exponential term suggests applying the Ratio Test. With an = n5n , an+1 (n + 1)3 5n an+1 1 3 1 1 3 = = · 1 = 1 < 1. · 3 = and ρ = lim 1+ a n→∞ an 5 n 5 5 5n+1 n n

∞ # 1 ∞ # 1 3 − n2 n n=2 n(ln n)3 n=2

∞ 3 # n converges by the Ratio Test. 5n n=1

633

634

C H A P T E R 11

INFINITE SERIES

This series is similar to a p-series; because

SOLUTION

1 1 1 ≈ √ = 3/2 3 3 2 n n n −n for large n, we will apply the Limit Comparison Test comparing with the p-series with p = 32 . Now, √ 1 n3 n 3 −n 2 L = lim = lim = 1. 1 3 n→∞ n→∞ n − n 2 3/2 n

The p-series with p = 32 converges and L exists; therefore, the series 47.

∞ #

1 also converges. 3 2 n=2 n − n

∞ 2 # n + 4n ∞ # 3n 44−2k+1 +9 n=1 k=1

This series is similar to a p-series; because

SOLUTION

n2 n 2 + 4n 1 ≈ √ = 2 4 3n + 9 3n 3n 4 for large n, we will apply the Limit Comparison Test comparing with the p-series with p = 2. Now, n 2 +4n 4 1 n 4 + 4n 3 = . L = lim 3n 1+9 = lim n→∞ n→∞ 3n 4 + 9 3 n2

The p-series with p = 2 converges and L exists; therefore, the series

∞ 2 # n + 4n also converges. 3n 4 + 9 n=1

∞ # 1 ∞ # sin 2 n 1 (−1) cos n n=1 n n=1 SOLUTION Here, we will apply the Limit Comparison Test, comparing with the p-series with p = 2. Now,

49.

L = lim

sin 12

n→∞

n 1 2 n

= lim

u→0

sin u = 1, u

where u = 12 . The p-series with p = 2 converges and L exists; therefore, the series n

∞ #

1 sin 2 also converges. n n=1

n ∞ # ∞ n n−1 # (−1) n + 12 √ n=1 n n=1 SOLUTION Because the general term has the form of a function of n raised to the nth power, we might be tempted to use the Root Test; however, the Root Test is inconclusive for this series. Instead, note ) *−12 12 −n 12 n/12 lim an = lim 1 + = lim = e−12 = 0. 1+ n→∞ n→∞ n→∞ n n

51.

Therefore, the series diverges by the Divergence Test. ∞

# (−2)n Further Insights and Challenges √ 53.

n=1

n Proof of the Root Test

Let S =

∞ # n=0

an be a positive series and assume that L = lim

n→∞

√ n

an exists.

n (a) Show that S converges if L < 1. Hint: #Choose R with ρ < R < 1 and show that an ≤ R for n sufficiently large. n Then compare with the geometric series R . (b) Show that S diverges if L > 1. √ SOLUTION Suppose lim n an = L exists. n→∞

S E C T I O N 11.5

(a) If L < 1, let =


635

1−L . By the definition of a limit, there is a positive integer N such that 2 √ − ≤ n an − L ≤

for n ≥ N . From this, we conclude that 0≤

√ n

an ≤ L +

for n ≥ N . Now, let R = L + . Then L +1 1+1 1−L = < = 1, 2 2 2

R=L+ and 0≤ for n ≥ N . Because 0 ≤ R < 1, the series the Comparison Test. Therefore, the series

√ n

∞ #

an ≤ R

or

0 ≤ an ≤ R n

R n is a convergent geometric series, so the series

n=N ∞ #

∞ #

an converges by

n=N

an also converges.

n=0

(b) If L > 1, let =

L −1 . By the definition of a limit, there is a positive integer N such that 2 √ − ≤ n an − L ≤

for n ≥ N . From this, we conclude that L − ≤

√ n

an

for n ≥ N . Now, let R = L − . Then R=L−

L +1 1+1 L −1 = > = 1, 2 2 2

and R≤ for n ≥ N . Because R > 1, the series Comparison Test. Therefore, the series

∞ #

√ n

an

or

R n ≤ an

R n is a divergent geometric series, so the series

n=N ∞ #

∞ #

an diverges by the

n=N

an also diverges.

n=0 ∞ n # c n!Ratio Test is inconclusive but the Root Test indicates convergence for the series , where c is a constant. 55. LetShow S = that the nn n=1 1 1 1 1 1 (a) Prove that S converges absolutely if |c| < e + and diverges > e. + · · · + 3 +if |c| + 2 4 5 2 3 2 3 2 n √ e n! (b) It is known that lim n+1/2 = 2π . Verify this numerically. n→∞ n (c) Use the Limit Comparison Test to prove that S diverges for c = e. SOLUTION n (a) With an = cn nn! ,

n an+1 |c|n+1 (n + 1)! nn 1 −n n = · n = |c| = |c| 1 + , a |c| n! n+1 n (n + 1)n+1 n and

Thus, by the Ratio Test, the series |c| > e.

a ρ = lim n+1 = |c|e−1 . n→∞ an ∞ n # c n! converges when |c|e−1 < 1, or when |c| < e. The series diverges when nn n=1

636

C H A P T E R 11

INFINITE SERIES

√ n n! (b) The table below lists the value of en+1/2 for several increasing values of n. Since 2π = 2.506628275, the numerical n evidence verifies that lim

en n!

=

n→∞ n n+1/2

2π .

n

100

1000

10000

100000

en n! n n+1/2

2.508717995

2.506837169

2.506649163

2.506630363

(c) With c = e, the series S becomes

∞ n # e n! . Using the result from part (b), nn n=1 en n! nn

L = lim √ n→∞

Because the series

√

∞ √ #

n

= lim

en n!

n→∞ n n+1/2

=

√

2π .

n diverges by the Divergence Test and L > 0, we conclude that

n=1

∞ n # e n! diverges by the Limit nn n=1

Comparison Test.

11.6 Power Series Preliminary Questions #

an x n converges for x = 5. Must it also converge for x = 4? What about x = −3? # SOLUTION The power series an x n is centered at x = 0. Because the series converges for x = 5, the radius of convergence must be at least 5 and the series converges absolutely at least for the interval |x| < 5. Both x = 4 and x = −3 are inside this interval, so the series converges for x = 4 and for x = −3. # 2. Suppose that an (x − 6)n converges for x = 10. At which of the points (a)–(d) must it also converge? 1. Suppose that

(a) x = 8

(b) x = 12

(c) x = 2

(d) x = 0

SOLUTION The given power series is centered at x = 6. Because the series converges for x = 10, the radius of convergence must be at least |10 − 6| = 4 and the series converges absolutely at least for the interval |x − 6| < 4, or 2 < x < 10. (a) x = 8 is inside the interval 2 < x < 10, so the series converges for x = 8. (b) x = 12 is not inside the interval 2 < x < 10, so the series may or may not converge for x = 12. (c) x = 2 is an endpoint of the interval 2 < x < 10, so the series may or may not converge for x = 2. (d) x = 0 is not inside the interval 2 < x < 10, so the series may or may not converge for x = 0.

3. Suppose that F(x) is a power series with radius of convergence R = 12. What is the radius of convergence of F(3x)? SOLUTION If the power series F(x) has radius of convergence R = 12, then the power series F(3x) has radius of convergence R = 12 3 = 4.

4. The power series F(x) =

∞ #

nx n has radius of convergence R = 1. What is the power series expansion of F (x)

n=1

and what is its radius of convergence? SOLUTION We obtain the power series expansion for F (x) by differentiating the power series expansion for F(x) term-by-term. Thus,

F (x) =

∞ #

n 2 x n−1 .

n=1

The radius of convergence for this series is R = 1, the same as the radius of convergence for the series expansion for F(x).

S E C T I O N 11.6

Exercises 1. Use the Ratio Test to determine the radius of convergence of

Power Series

637

∞ n # x . 2n n=0

n

SOLUTION

With an = x2n , an+1 |x|n+1 2n |x| a = 2n+1 · |x|n = 2 n

and

a |x| ρ = lim n+1 = . n→∞ an 2

|x| By the Ratio Test, the series converges when ρ = |x| 2 < 1, or |x| < 2, and diverges when ρ = 2 > 1, or |x| > 2. The radius of convergence is therefore R = 2.

3. Show that the following three power series have the same radius of convergence. Then show that (a) diverges at both ∞ # xn endpoints, endpoint at the other, and (c) converges at both endpoints. Use(b) theconverges Ratio Testattoone show that but√diverges has radius of convergence R = 2. Then determine whether it converges ∞ ∞n n ∞ n2 # xn # # x xn n=1 (a) absolutely (b) (c) or conditionally at the endpoints Rn3 =n±2. 3n n 2 3n n=1

n=1

n=1

SOLUTION

(a) With an = 31n ,

an+1 3n 1 1 a = 3n+1 · 1 = 3 n

a 1 and r = lim n+1 = . n→∞ an 3

The radius of convergence is therefore R = r −1 = 3. For the endpoint x = 3, the series becomes ∞ n ∞ # # 3 = 1, n 3 n=1 n=1

which diverges by the Divergence Test. For the endpoint x = −3, the series becomes ∞ ∞ # # (−3)n = (−1)n , n 3 n=1 n=1

which also diverges by the Divergence Test. (b) With an = n31n , an+1 1 n3n n 1 = = · a (n + 1)3n+1 1 3 n+1 n

a 1 1 and r = lim n+1 = · 1 = . n→∞ an 3 3

The radius of convergence is therefore R = r −1 = 3. For the endpoint x = 3, the series becomes ∞ ∞ # # 3n 1 = , n n3 n n=1 n=1

which is the divergent harmonic series. For the endpoint x = −3, the series becomes ∞ ∞ # # (−3)n (−1)n = , n n3 n n=1 n=1

which converges by the Leibniz Test. (c) With an = 21 n , n 3

2 an+1 n 2 3n 1 1 n = a (n + 1)2 3n+1 · 1 = 3 n + 1 n

an+1 1 2 = · 1 = 1. and r = lim n→∞ an 3 3

The radius of convergence is therefore R = r −1 = 3. For the endpoint x = 3, the series becomes ∞ ∞ # # 3n 1 = , 2 3n 2 n n n=1 n=1

which is a convergent p-series. For the endpoint x = −3, the series becomes ∞ ∞ # # (−3)n (−1)n = , n 2 3n n2 n=1 n=1

which converges by the Leibniz Test.

638

C H A P T E R 11

INFINITE SERIES ∞ # Repeat for the following 5. Show that Exercise n n x n 3diverges for all x =series: 0. ∞ ∞ # # (xn=0 − 5)n (x − 5)n (a) (b) n n 9 an = n , n9n SOLUTION With n=1 n=1

∞ # (x − 5)n n 2 9n n=1 an+1 (n + 1)n+1 an+1 1 n = = ∞. = 1 + (n + 1) and r = lim a n→∞ an nn n n

(c)

The radius of convergence is therefore R = r −1 = 0. In other words, the power series converges only for x = 0. In Exercises 7–26, find the values of x for which ∞ nthe following power series converge. # x . (a) ∞ Find the radius of convergence of # n2 n=1 7. nx n (b) Determine whether the series converges at the endpoints of the interval of convergence. n=1 SOLUTION

With an = n, an+1 n + 1 a = n n

a and r = lim n+1 = 1. n→∞ an

The radius of convergence is therefore R = r −1 = 1, and the series converges absolutely on the interval |x| < 1, or ∞ # −1 < x < 1. For the endpoint x = 1, the series becomes n, which diverges by the Divergence Test. For the endpoint x = −1, the series becomes

∞ #

n=1

(−1)n n, which also diverges by the Divergence Test. Thus, the series

n=1

nx n converges

n=1

for −1 < x < 1 and diverges elsewhere. 9.

∞ #

∞ n n # 2 x ∞ # nn(x − 3)n n=1 n=1

SOLUTION

n

With an = 2n , an+1 n 2n+1 n a = n + 1 · 2n = 2 n + 1 n

a and r = lim n+1 = 2 · 1 = 2. n→∞ an

The radius of convergence is therefore R = r −1 = 12 , and the series converges absolutely on the interval |x| < 12 , ∞ # 1 , which is the divergent harmonic series. For the or − 12 < x < 12 . For the endpoint x = 12 , the series becomes n n=1 ∞ ∞ n n # # (−1)n 2 x endpoint x = − 12 , the series becomes , which converges by the Leibniz Test. Thus, the series n n n=1 n=1 converges for − 12 ≤ x < 12 and diverges elsewhere. ∞ # xn ∞ # 11. (−5)n (x − 3)n ln n n=2 n2 n=1 1 SOLUTION With an = ln n , an+1 1 ln n ln n a = ln(n + 1) · 1 = ln(n + 1) n


The radius of convergence is therefore R = r −1 = 1, and the series converges absolutely on the interval |x| < 1, or ∞ ∞ # # 1 1 −1 < x < 1. For the endpoint x = 1, the series becomes . Because ln1n > n1 and is the divergent ln n n n=1 n=1 harmonic series, the endpoint series diverges by the Comparison Test. For the endpoint x = −1, the series becomes ∞ ∞ # # (−1)n xn , which converges by the Leibniz Test. Thus, the series converges for −1 ≤ x < 1 and diverges ln n ln n n=1 n=2 elsewhere. ∞ n # ∞x n # x (n!)2 n n=1 n2 n=1 SOLUTION With an =

13.

1 , (n!)2

2 an+1 (n!)2 1 1 = · = a ((n + 1)!)2 1 n+1 n


S E C T I O N 11.6

Power Series

639

The radius of convergence is therefore R = r −1 = ∞, and the series converges absolutely for all x. ∞ # ∞ nnn 4 (x + 4)n # (−1) x n=1 n5 n=4 SOLUTION With an = (−1)n n 4 ,

15.

an+1 (n + 1)4 1 4 = = 1 + a n n4 n

a and r = lim n+1 = 14 = 1. n→∞ an

The radius of convergence is therefore R = r −1 = 1, and the series converges absolutely on the interval |x + 4| < 1, ∞ # (−1)n n 4 , which diverges by the Divergence Test. or −5 < x < −3. For the endpoint x = −3, the series becomes ∞ #

For the endpoint x = −5, the series becomes ∞ #

n=1

n 4 , which also diverges by the Divergence Test. Thus, the series

n=1

(−1)n n 4 (x + 4)n converges for −5 < x < −3 and diverges elsewhere.

n=1 ∞ # n n n n ∞ # x n x(−1) 2 n=0 2+1 n n=0 n SOLUTION With an = 2n , an+1 n + 1 2n 1 1 = · = 1 + a 2 n 2n+1 n n

17.

a 1 1 and r = lim n+1 = · 1 = . n→∞ an 2 2

The radius of convergence is therefore R = r −1 = 2, and the series converges absolutely on the interval |x| < 2, ∞ # n, which diverges by the Divergence Test. For or −2 < x < 2. For the endpoint x = 2, the series becomes the endpoint x = −2, the series becomes

∞ #

n=1

(−1)n n, which also diverges by the Divergence Test. Thus, the series

n=1

∞ # n n n x converges for −2 < x < 2 and diverges elsewhere. 2 n=0

19.

∞ # (x − 4)n ∞ # 4nn4x n n=1 n=0

SOLUTION

With an = 14 , n

4 an+1 n4 n 1 = · = a (n + 1)4 1 n+1 n

a and r = lim n+1 = 14 = 1. n→∞ an

The radius of convergence is therefore R = r −1 = 1, and the series converges absolutely on the interval |x − 4| < 1, ∞ # 1 , which is a convergent p-series. For the endpoint or 3 < x < 5. For the endpoint x = 5, the series becomes n4 n=1 ∞ ∞ # # (−1)n (x − 4)n , which converges by the Leibniz Test. Thus, the series converges x = 3, the series becomes 4 n n4 n=1 n=1 for 3 ≤ x ≤ 5 and diverges elsewhere. ∞ # ∞n! (xn + 5)n # 2 (x + 3)n n=10 3n n=1 SOLUTION With an = n!,

21.

an+1 an+1 (n + 1)! = = ∞. = n + 1 and r = lim a n→∞ an n! n The radius of convergence is therefore R = r −1 = 0, and the series converges absolutely only for x = −5. 23.

∞ # ∞ # en (xx 2n+1 − 2)n n=12 3n + 1 n=15

640

C H A P T E R 11


With an = en , an+1 en+1 a = en = e n

a and r = lim n+1 = e. n→∞ an

The radius of convergence is therefore R = r −1 = e−1 , and the series converges absolutely on the interval |x − 2| < ∞ # 1, which diverges by the e−1 , or 2 − e−1 < x < 2 + e−1 . For the endpoint x = 2 + e−1 , the series becomes Divergence Test. For the endpoint x = 2 − e−1 , the series becomes Test. Thus, the series

∞ #

∞ #

n=1

(−1)n , which also diverges by the Divergence

n=1

en (x − 2)n converges for 2 − e−1 < x < 2 + e−1 and diverges elsewhere.

n=12 ∞ n # ∞ x n # 25. x n − 4 ln n n=1 4+2 n n=0 1 SOLUTION With an = n−4 ln n , an+1 1 − 4 lnnn 1 n − 4 ln n = , · = a n + 1 − 4 ln(n + 1) 1 n 1 + n1 − 4 ln(n+1) n

and

a r = lim n+1 = 1. n→∞ an

The radius of convergence is therefore R = r −1 = 1, and the series converges absolutely on the interval |x| < 1, ∞ ∞ # # 1 1 . Because n−41ln n > n1 and is or −1 < x < 1. For the endpoint x = 1, the series becomes n − 4 ln n n n=1 n=1 the divergent harmonic series, the endpoint series diverges by the Comparison Test. For the endpoint x = −1, the series ∞ ∞ # # (−1)n xn , which converges by the Leibniz Test. Thus, the series converges for −1 ≤ x < 1 becomes n − 4 ln n n − 4 ln n n=1 n=1 and diverges elsewhere. In Exercises ∞ 27–34, nuse Eq. (1) to expand the function in a power series with center c = 0 and determine the set of x for # (x − 2) which the expansion is valid. (n ln n)2 n=2 1 27. f (x) = 1 − 3x SOLUTION

Substituting 3x for x in Eq. (1), we obtain ∞ ∞ # # 1 (3x)n = 3n x n . = 1 − 3x n=0 n=0

This series is valid for |3x| < 1, or |x| < 13 . 1 29. f (x) = 1 f (x) 3=− x 1 + 3x SOLUTION First write 1 1 1 . = · 3−x 3 1 − x3 Substituting x3 for x in Eq. (1), we obtain ∞ n ∞ n # # x x 1 = = ; x 1− 3 3 3n n=0 n=0

Thus, ∞ n ∞ # 1# x xn 1 = . = 3−x 3 n=0 3n 3n+1 n=0

This series is valid for |x/3| < 1, or |x| < 3. f (x) =

1 4 + 3x

S E C T I O N 11.6

31. f (x) = SOLUTION

Power Series

641

1 1 + x9 Substituting −x 9 for x in Eq. (1), we obtain ∞ ∞ # # 1 9 )n = = (−x (−1)n x 9n . 1 + x9 n=0 n=0

This series is valid for | − x 9 | < 1, or |x| < 1. 1 33. f (x) = 1 f (x) 1=+ 3x 7 2 5−x SOLUTION Substituting −3x 7 for x in Eq. (1), we obtain ∞ ∞ # # 1 7 )n = = (−3x (−3)n x 7n . 1 + 3x 7 n=0 n=0 1 This series is valid for | − 3x 7 | < 1, or |x| < √ 7 . 3

35. Use the equalities 1 f (x) = 4 − 2x 3

− 13 1 1 = = 1−x −3 − (x − 4) 1 + ( x−4 3 )

to show that for |x − 4| < 3 ∞ # 1 (x − 4)n (−1)n+1 n+1 = 1−x 3 n=0 SOLUTION

Substituting − x−4 3 for x in Eq. (1), we obtain 1

1 + x−4 3

=

∞ ∞ # (x − 4)n x −4 n # = (−1)n . − 3 3n n=0 n=0

Thus, ∞ ∞ # 1 (x − 4)n (x − 4)n 1# =− (−1)n = (−1)n+1 n+1 . n 1−x 3 n=0 3 3 n=0

This series is valid for | − x−4 3 | < 1, or |x − 4| < 3. 1 37. Use the method of Exercise 35 to expand 1 in a power series with center c = 5. Determine the set of x for Use the method of Exercise 35 to expand4 − x in power series with centers c = 2 and c = −2. Determine the 1−x which the expansion is valid. set of x for which the expansions are valid. SOLUTION First write 1 1 1 = =− . 4−x −1 − (x − 5) 1 + (x − 5) Substituting −(x − 5) for x in Eq. (1), we obtain ∞ ∞ # # 1 (−(x − 5))n = (−1)n (x − 5)n . = 1 + (x − 5) n=0 n=0

Thus, ∞ ∞ # # 1 (−1)n (x − 5)n = (−1)n+1 (x − 5)n . =− 4−x n=0 n=0

This series is valid for | − (x − 5)| < 1, or |x − 5| < 1. 39. Give an example of a power series that converges for x in [2, 6). ∞ # n . Hint: Show that Evaluate n 2 n=1 (1 − x)−2 =

∞ # n=1

nx n−1

642

C H A P T E R 11

INFINITE SERIES

SOLUTION

The power series must be centered at c =

following series:

6+2 = 4, with radius of convergence R = 2. Consider the 2

∞ # (x − 4)n . n2n n=1

With an = n21n , r = lim

n2n

n→∞ (n + 1)2n+1

=

1 1 n lim = . 2 n→∞ n + 1 2

The radius of convergence is therefore R = r −1 = 2, and the series converges absolutely for |x − 4| < 2, or 2 < x < 6. ∞ ∞ # # 1 (6 − 4)n = For the endpoint x = 6, the series becomes , which is the divergent harmonic series. For the n n·2 n n=1 n=1 ∞ ∞ # # (2 − 4)n (−1)n endpoint x = 2, the series becomes = , which converges by the Leibniz Test. Therefore, the n n·2 n n=1 n=1 series converges for 2 ≤ x < 6, as desired. 41. Use Exercise 40 to prove that Prove that for −1 < x < 1 1 1 1 1 3 ∞ + − 2 4 +3 · · · − 1ln = # n n 2 3 2 = 2 (−1) 2 · 2x =31· − 2 x +4x· 2− x + · · · 1+x n=0 Use your knowledge of alternating series to find an N such that the partial sum S N approximates ln 32 to within an error ∞ n−1 n # x 2 3x.3 x4 to(−1) computex both of at most 10−3 . Confirm this using = xS N−and ln +2 − + ··· ln(1a+calculator x) = n 2 3 4 n=1 SOLUTION In the previous exercise we found that ln(1 + x) =

∞ #

(−1)n

n=0

x n+1 . n+1

Setting x = 12 yields:

n 1 ∞ ∞ # # 1 1 1 (−1)n−1 1 3 2 n−1 (−1) = − + − + ··· = ln = n 2 3 2 n=1 n n2 2 2 · 2 3 · 2 4 · 24 n=1 Note that the series for ln 32 is an alternating series with an = n21n . The error in approximating ln 32 by the partial sum S N is therefore bounded by 3 1 ln − S N < a N +1 = . 2 (N + 1)2 N +1 To obtain an error of at most 10−3 , we must find an N such that 1 < 10−3 (N + 1)2 N +1

or

(N + 1)2 N +1 > 1000.

For N = 6, (N + 1)2 N +1 = 7 · 27 = 896 < 1000, but for N = 7, (N + 1)2 N +1 = 8 · 28 = 2048 > 1000; hence, the smallest value for N is N = 7. The corresponding approximation is S7 =

1 1 1 1 1 1 1 + − + − + = 0.405803571. − 2 2 · 22 7 · 27 3 · 23 4 · 24 5 · 25 6 · 26

Now, ln 32 = 0.405465108, so

3 ln − S7 = 3.385 × 10−4 < 10−3 . 2

43. Show that for |x| < 1 Show that the following series converges absolutely for |x| < 1 and compute its sum: 1 + 2x 2 + x33 + x44 − 2x 5 6 7 8 x − x − x 5 ++x 6x −+x 7x −−x 82x+ ·+ · ·· · · F(x)= =11+ −xx − −2x x2 + 1 + x + x2 a sum of42. three geometric series with common ratio x 3 . Hint:Hint: Use Write the hintF(x) fromasExercise


Power Series

643

The terms in the series on the right-hand side are either of the form x n or −2x n for some n. Because lim

√ n

n→∞

2 = lim

√ n

n→∞

1 = 1,

it follows that lim

n→∞

n |an | = |x|.

Hence, by the Root Test, the series converges absolutely for |x| < 1. By Exercise 37 of Section 11.4, any rearrangement of the terms of an absolutely convergent series yields another absolutely convergent series with the same sum as the original series. If we let S denote the sum of the series, then

S = 1 + x3 + x6 + · · · + x + x4 + x7 + · · · − 2 x2 + x5 + x8 + · · · =

2x + 1 x 2x 2 1 + x − 2x 2 (1 − x)(2x + 1) 1 = + − = = . 1 − x3 1 − x3 1 − x3 1 − x3 (1 − x)(1 + x + x 2 ) 1 + x + x2

45. Use the power series for y = e x to show that ∞ # an x n satisfying equation y = −y with initial condition y(0) = Find a power series P(x) = 1 the1 differential 1 1 = − + − · · · n=0 e 2! 3! 4! 1. Then use Theorem 1 of Section 7.4 to conclude that P(x) = e−x . Use your knowledge of alternating series to find an N such that the partial sum S N approximates e−1 to within an error of at most 10−3 . Confirm this using a calculator to compute both S N and e−1 . SOLUTION

Recall that the series for e x is ∞ n # x x2 x3 x4 =1+x + + + + ··· . n! 2! 3! 4! n=0

Setting x = −1 yields e−1 = 1 − 1 +

1 1 1 1 1 1 − + − +··· = − + − +··· . 2! 3! 4! 2! 3! 4!

1 . The error in approximating e−1 with the partial sum S is therefore This is an alternating series with an = (n+1)! N bounded by

|S N − e−1 | ≤ a N +1 =

1 . (N + 2)!

To make the error at most 10−3 , we must choose N such that 1 ≤ 10−3 (N + 2)!

or

(N + 2)! ≥ 1000.

For N = 4, (N + 2)! = 6! = 720 < 1000, but for N = 5, (N + 2)! = 7! = 5040; hence, N = 5 is the smallest value that satisfies the error bound. The corresponding approximation is S5 =

1 1 1 1 1 − + − + = 0.368055555 2! 3! 4! 5! 6!

Now, e−1 = 0.367879441, so |S5 − e−1 | = 1.761 × 10−4 < 10−3 . 47. Find a power series # P(x) satisfying the differential equation: Let P(x) = an x n be a power series solution to y = 2x y with initial condition y(0) = 1. n=0 y − x y + y = 0 (a) Show that the odd coefficients a2k+1 are all zero. with initial condition y(0) a=2k−2 1, y (0) = 0. What is the radius of convergence of the power series? (b) Prove that a2k = and use this result to determine the coefficients a2k . ∞k # SOLUTION Let P(x) = an x n . Then n=0

P (x) =

∞ # n=1

nan x n−1

and

P (x) =

∞ # n=2

n(n − 1)an x n−2 .

10

644

C H A P T E R 11

INFINITE SERIES

Note that P(0) = a0 and P (0) = a1 ; in order to satisfy the initial conditions P(0) = 1, P (0) = 0, we must have a0 = 1 and a1 = 0. Now, P (x) − x P (x) + P(x) =

∞ #

n(n − 1)an x n−2 −

n=2

=

∞ #

∞ #

nan x n +

n=1

(n + 2)(n + 1)an+2 x n −

n=0

∞ #

an x n

n=0 ∞ #

nan x n +

n=1

∞ #

an x n

n=0

∞ # (n + 2)(n + 1)an+2 − nan + an x n . = 2a2 + a0 + n=1

In order for this series to be equal to zero, the coefficient of x n must be equal to zero for each n; thus, 2a2 + a0 = 0 and (n + 2)(n + 1)an+2 − (n − 1)an = 0, or 1 a2 = − a0 2

and an+2 =

n−1 an . (n + 2)(n + 1)

Starting from a1 = 0, we calculate a3 =

1−1 a1 = 0; (3)(2)

a5 =

2 a = 0; (5)(4) 3

a7 =

4 a5 = 0; (7)(6)

and, in general, all of the odd coefficients are zero. As for the even coefficients, we have a0 = 1, a2 = − 12 , a4 =

1 1 a2 = − ; (4)(3) 4!

a6 =

3 3 a =− ; (6)(5) 4 6!

a8 =

5 15 a6 = − (8)(7) 8!

and so on. Thus, P(x) = 1 −

1 3 15 8 1 2 x − x4 − x6 − x − ··· 2 4! 6! 8!

To determine the radius of convergence, treat this as a series in the variable x 2 , and observe that a2k+2 2k − 1 = lim r = lim = 0. k→∞ a2k k→∞ (2k + 2)(2k + 1) Thus, the radius of convergence is R = r −1 = ∞. ∞ # (−1)k (0) = 1. 2k+2initial Find a power satisfying Eq. (10)xwith condition y(0)Bessel = 0, ydifferential = is a solution of the equation of order two: 49. Prove that J2 (x) series 2k+2 k! (k + 3)! 2 k=0

x 2 y + x y + (x 2 − 4)y = 0 SOLUTION

Let J2 (x) =

∞ #

(−1)k x 2k+2 . Then 22k+2 k! (k + 2)! k=0 ∞ #

J2 (x) =

(−1)k (k + 1) 2k+1 x 22k+1 k! (k + 2)! k=0

J2 (x) =

∞ # (−1)k (k + 1)(2k + 1) 2k x 22k+1 k! (k + 2)! k=0

and x 2 J2 (x) + x J2 (x) + (x 2 − 4)J2 (x) =

∞ ∞ # (−1)k (k + 1)(2k + 1) 2k+2 # (−1)k (k + 1) 2k+2 x x + 22k+1 k! (k + 2)! 22k+1 k! (k + 2)! k=0 k=0

S E C T I O N 11.6

− = =

Power Series

645

∞ #

∞ # (−1)k (−1)k 2k+4 − x x 2k+2 22k+2 k! (k + 2)! 22k k! (k + 2)! k=0 k=0

∞ ∞ # (−1)k k(k + 2) 2k+2 # (−1)k−1 x x 2k+2 + 2k 2k 2 k!(k + 2)! 2 (k − 1)! (k + 1)! k=0 k=1 ∞ #

∞ # (−1)k (−1)k 2k+2 − x x 2k+2 = 0. 22k (k − 1)!(k + 1)! 22k (k − 1)!(k + 1)! k=1 k=1

x4 x6 x2 + − tan+−1· ·(0.5) · . to three decimal places. 51. LetUse C(x)Eq. = (4) 1 −to approximate 2! 4! 6! (a) Show that C(x) has an infinite radius of convergence. (b) Prove that C(x) and f (x) = cos x are both solutions of y = −y with initial conditions y(0) = 1, y (0) = 0. This initial value problem has a unique solution, so it follows that C(x) = cos x for all x. SOLUTION

(a) Consider the series ∞ # x 2n x2 x4 x6 (−1)n + − + ··· = . 2! 4! 6! (2n)! n=0

C(x) = 1 − 2n

x , With an = (−1)n (2n)!

an+1 |x|2 |x|2n+2 (2n)! a = (2n + 2)! · |x|2n = (2n + 2)(2n + 1) , n and

an+1 = 0. r = lim n→∞ an

The radius of convergence for C(x) is therefore R = r −1 = ∞. (b) Differentiating the series defining C(x) term-by-term, we find C (x) =

∞ #

(−1)n (2n)

n=1

∞ # x 2n−1 x 2n−1 (−1)n = (2n)! (2n − 1)! n=1

and C (x) =

∞ #

(−1)n (2n − 1)

n=1

=

∞ # n=0

(−1)n+1

∞ # x 2n−2 x 2n−2 (−1)n = (2n − 1)! n=1 (2n − 2)!

∞ # x 2n x 2n (−1)n =− = −C(x). (2n)! (2n)! n=0

Moreover, C(0) = 1 and C (0) = 0. 53. Find all values of x such that the following series converges: ∞ n2 # x Find all values of x such that F(x) =n! 1 +converges. 3x + x 2 + 27x 3 + x 4 + 243x 5 + · · · n=1 SOLUTION

Observe that F(x) can be written as the sum of two geometric series: ∞ ∞

#

# (x 2 )n + 3x(9x 2 )n F(x) = 1 + x 2 + x 4 + · · · + 3x + 27x 3 + 243x 5 + · · · = n=0

n=0

The first geometric series converges for |x 2 | < 1, or |x| < 1; the second geometric series converges for |9x 2 | < 1, or |x| < 13 . Since both geometric series must converge for F(x) to converge, we find that F(x) converges for |x| < 13 , the intersection of the intervals of convergence for the two geometric series.

∞ 3n 55. Why is it impossible to expand f (x) = |x| as a power series that converges in an interval # x around x = 0? Explain why Theorem . What is the Explain this using Theorem 3. 2 cannot be applied directly to find the radius of convergence of 5n n=1 radius of convergence of this series?

646

C H A P T E R 11


Suppose that there exists a c > 0 such that f can be represented by a power series on the interval (−c, c);

that is, |x| =

∞ #

an x n

n=0

for |x| < c. Then it follows by Theorem 3 that |x| is differentiable on (−c, c). This contradicts the well known property that f (x) = |x| is not differentiable at the point x = 0.

Further Insights and Challenges Prove that that the for bcoefficients = 0 57. Suppose of F(x) =

∞ #

an x n are periodic, that is, for some whole number M > 0, we have n=0 " ∞ # 1 − xb for1 |x| < 1 1and 1that a M+n = an . Prove that F(x) converges absolutely = dx n(n + b) b 0 1−x n=1 M−1 a + a1 x + · · · + a M−1 x F(x) = 0 Conclude that the sum on the left has the value 1 − xM

1 1 1 1 Hint: Use the hint for Exercise 42. 1 + + + ··· + b 2 3 b SOLUTION Suppose the coefficients of F(x) are periodic, with a M+n = an for some whole number M and all n. The F(x) can be written as the sum of M geometric series:

F(x) = a0 1 + x M + x 2M + · · · + a1 x + x M+1 + x 2M+1 + · · · +

= a2 x 2 + x M+2 + x 2M+2 + · · · + · · · + a M−1 x M−1 + x 2M−1 + x 3M−1 + · · · =

a1 x a x2 a a + a1 x + a2 x 2 + · · · + a M−1 x M−1 x M−1 a0 + + 2 M + · · · + M−1 M = 0 . M M 1−x 1−x 1−x 1−x 1 − xM

As each geometric series converges absolutely for |x| < 1, it follows that F(x) also converges absolutely for |x| < 1. Continuity of Power Series

11.7 Taylor Series

Let F(x) =

∞ #

an x n be a power series with radius of convergence R > 0.

n=0

(a) Prove the inequality


|x n − y n | ≤ n|x − y|(|x|n−1 + |y|n−1 ) 1. Determine f (0) and f (0) for a function f (x) with Maclaurin series Hint: x n − y n = (x − y)(x n−1 + x n−2 y + · · · + y n−1 ). 2 T (x) = 3 + 2x + 12x + 5x 3 + · · · ∞ # 2n|an |R1n converges. (b) Choose R1 with 0 < R1 < R. Use the Ratio Test to show that the infinite series M = SOLUTION The Maclaurin series for a function f has the form n=0 − F(y)| ≤ M|x − y|. (c) Use Eq. (11) to show that if |x| < R1 and |y| < R1 , then |F(x) f (0) f (0) f (0) 2 + Choosex 3R1+such · · · that |x| < R1 < R. Show that if x + at x. xHint: (0) +is continuous (d) Prove that if |x| < R, thenf F(x) 1! 2! 3! > 0 is given, then |F(x) − F(y)| ≤ for all y such that |x − y| < δ , where δ is any positive number that is less than /M and R1 − |x| (see Figure 4). f (0) Matching this general expression with the given series, we find f (0) = 3 and = 5. From this latter equation, it 3! follows that f (0) = 30. 2. Determine f (−2) and f (4) (−2) for a function with Taylor series T (x) = 3(x + 2) + (x + 2)2 − 4(x + 2)3 + 2(x + 2)4 + · · · SOLUTION

The Taylor series for a function f centered at x = −2 has the form f (−2) +

f (−2) f (4) (−2) f (−2) f (−2) (x + 2) + (x + 2)2 + (x + 2)3 + (x + 2)4 + · · · 1! 2! 3! 4!

Matching this general expression with the given series, we find f (−2) = 0 and it follows that f (4) (−2) = 48.

f (4) (−2) = 2. From this latter equation, 4!

3. What is the easiest way to find the Maclaurin series for the function f (x) = sin(x 2 )?

2 is to substitute x 2 for x in the Maclaurin series SOLUTION The easiest way to find the Maclaurin series for sin x for sin x.

S E C T I O N 11.7

Taylor Series

647

4. What is the Taylor series for f (x) centered at c = 3 if f (3) = 4 and f (x) has a Taylor expansion f (x) =

∞ # (x − 3)n n n=1

Integrating the series for f (x) term-by-term gives

SOLUTION

f (x) = C +

∞ # (x − 3)n+1 . n(n + 1) n=1

Substituting x = 3 then yields f (3) = C = 4; so f (x) = 4 +

5. (a) (b) (c)

∞ # (x − 3)n+1 . n(n + 1) n=1

Let T (x) be the Maclaurin series of f (x). Which of the following guarantees that f (2) = T (2)? T (x) converges for x = 2. The remainder Rk (2) approaches a limit as k → ∞. The remainder Rk (2) approaches zero as k → ∞. The correct response is (c): f (2) = T (2) if and only if the remainder Rk (2) approaches zero as k → ∞.

SOLUTION

Exercises 1. Write out the first four terms of the Maclaurin series of f (x) if f (0) = 3,

f (0) = 2, SOLUTION

f (0) = 4,

f (0) = 12

The first four terms of the Maclaurin series of f (x) are f (0) 2 f (0) 3 4 12 3 x + x = 2 + 3x + x 2 + x = 2 + 3x + 2x 2 + 2x 3 . f (0) + f (0)x + 2! 3! 2 6

In Exercises find thefour Maclaurin Write3–20, out the first terms ofseries. the Taylor series of f (x) centered at c = 3 if 3. f (x) = SOLUTION

1 1 − 2x

f (3) = 2,

f (3) = 1,

f (3) = 12,

f (3) = 3

1 gives Substituting 2x for x in the Maclaurin series for 1−x ∞ ∞ # # 1 (2x)n = 2n x n . = 1 − 2x n=0 n=0

This series is valid for |2x| < 1, or |x| < 12 . 5. f (x) = cos 3x x f (x) = SOLUTION Substituting 1 − x 4 3x for x in the Maclaurin series for cos x gives cos 3x =

∞ #

(−1)n

n=0

∞ # (3x)2n 9n x 2n = . (−1)n (2n)! (2n)! n=0

This series is valid for all x. 2) 7. f (x)f (x) = sin(x = sin(2x) SOLUTION

Substituting x 2 for x in the Maclaurin series for sin x gives sin x 2 =

∞ # n=0

This series is valid for all x. 9. f (x) = ln(1 4x − x 2) f (x) = e

(−1)n

∞ # (x 2 )2n+1 x 4n+2 (−1)n = . (2n + 1)! n=0 (2n + 1)!

648

C H A P T E R 11


Substituting −x 2 for x in the Maclaurin series for ln(1 + x) gives ln(1 − x 2 ) =

∞ ∞ ∞ 2n # # # (−1)n−1 (−x 2 )n (−1)2n−1 x 2n x = =− . n n n n=1 n=1 n=1

This series is valid for |x| < 1. 11. f (x) = tan−1 (x 2 ) −1/2 f (x) = (1 − x) SOLUTION Substituting x 2 for x in the Maclaurin series for tan−1 x gives tan−1 (x 2 ) =

∞ #

(−1)n

n=0

∞ # (x 2 )2n+1 x 4n+2 (−1)n = . 2n + 1 2n + 1 n=0

This series is valid for |x| ≤ 1. 13. f (x) = e x−2 2 f (x) = x 2 e x SOLUTION e x−2 = e−2 e x ; thus, e x−2 = e−2

∞ ∞ n # # x xn = . n! e2 n! n=0 n=0

This series is valid for all x. 15. f (x) = ln(1 − √ 5x) f (x) = cos x SOLUTION Substituting −5x for x in the Maclaurin series for ln(1 + x) gives ln(1 − 5x) =

∞ ∞ ∞ n n # # # (−1)n−1 (−5x)n (−1)2n−1 5n x n 5 x = =− . n n n n=1 n=1 n=1

This series is valid for |5x| < 1, or |x| < 15 , and for x = − 15 . 17. f (x) = sinh x f (x) = (x 2 + 2x)e x SOLUTION Recall that sinh x =

1 x (e − e−x ). 2

Therefore, ∞ n ∞ ∞ # # 1 # x (−x)n xn 1 − (−1)n . sinh x = − = 2 n=0 n! n=0 n! 2(n!) n=0 Now, 1 − (−1)n =

0, 2,

n even n odd

so sinh x =

∞ #

∞ # x 2k+1 x 2k+1 = . 2(2k + 1)! k=0 (2k + 1)! k=0

2

This series is valid for all x. 1 − cos(x 2 ) 19. f (x)f (x) = = cosh x x SOLUTION

Substituting x 2 for x in the Maclaurin series for cos x gives cos x 2 =

∞ # n=0

(−1)n

∞ ∞ # # (x 2 )2n x 4n x 4n (−1)n (−1)n = =1+ . (2n)! (2n)! (2n)! n=0 n=1

Thus, 1 − cos x 2 = 1 −

1+

∞ # n=1

x 4n (−1)n

(2n)!

=

∞ #

(−1)n+1

n=1

x 4n , (2n)!

S E C T I O N 11.7

Taylor Series

and ∞ ∞ # x 4n x 4n−1 1# 1 − cos(x 2 ) = = . (−1)n+1 (−1)n+1 x x n=1 (2n)! n=1 (2n)!

21. Use multiplication to find the first four terms in the Maclaurin series for f (x) = e x sin x. e x − cos x x f (x) = SOLUTION Multiply x the fifth-order Taylor Polynomials for e and sin x: x2 x3 x3 x4 x5 x5 1+x + + + + x− + 2 6 24 120 6 120 = x + x2 −

x3 x4 x4 x5 x5 x5 x3 + − + + − + + higher-order terms 6 2 6 6 120 12 24

= x + x2 +

x3 x5 − + higher-order terms. 3 30

The first four terms in the Maclaurin series for f (x) = e x sin x are therefore x + x2 +

x3 x5 − . 3 30

23. Find the first four terms of the Maclaurin series for f (x) = e x ln(1 − x). sin x x and ln(1 Find the first five of order the Maclaurin series for ffor (x)e= . SOLUTION Multiply theterms fourth Taylor Polynomials 1 − x − x): x2 x3 x4 x2 x3 x4 1+x + + + −x − − − 2 6 24 2 3 4 = −x −

x2 x3 x3 x3 x4 x4 x4 x4 − x2 − − − − − − − + higher-order terms 2 3 2 2 4 3 4 6

= −x −

3x 2 4x 3 − − x 4 + higher-order terms. 2 3

The first four terms of the Maclaurin series for f (x) = e x ln(1 − x) are therefore −x −

3x 2 4x 3 − − x 4. 2 3

. 25. Write out the first five terms of the binomial series for f (x) = (1 + x)−3/2 Write out the first five terms of the binomial series for f (x) = (1 + x)1/3 . 3 SOLUTION The first five generalized binomial coefficients for a = − 2 are 1,

3 − , 2

− 32 (− 52 ) 2!

=

15 , 8

− 32 (− 52 )(− 72 ) 3!

=−

35 , 16

− 32 (− 52 )(− 72 )(− 92 ) 4!

=

Therefore, the first five terms in the binomial series for f (x) = (1 + x)−3/2 are 1−

3 15 2 35 3 315 4 x+ x − x + x . 2 8 16 128

27. Find the first four terms of the Maclaurin for1 f (x) = e(e ) . 1 x Differentiate the Maclaurin series for . twice to find the Maclaurin series of SOLUTION With f (x) = e(e ) , we find 1−x (1 − x)3 x

x f (x) = e(e ) · e x

x x x f (x) = e(e ) · e x + e(e ) · e2x = e(e ) e2x + e x

x x f (x) = e(e ) 2e2x + e x + e(e ) e2x + e x e x

x = e(e ) e3x + 3e2x + e x and f (0) = e,

f (0) = e,

f (0) = 2e,

f (0) = 5e.

315 . 128

649

650

C H A P T E R 11

INFINITE SERIES x Therefore, the first four terms of the Maclaurin for f (x) = e(e ) are

e + ex + ex 2 +

5e 3 x . 6

π 29. Find the Taylor series for sin x at c = . 1 2 Find the first three terms of the Maclaurin series for f (x) = . Hint: First expand f (x) as a geometric 1 + sin x SOLUTION Because series.

π , sin x = cos x − 2 we obtain sin x =

∞ # (−1)n

π 2n , x− (2n)! 2 n=0

by substituting x − π2 for x in the Maclaurin series for cos x. In Exercises 31–40, find the Taylor series centered4at c. 2 What is the Maclaurin series for f (x) = x − 2x + 3? What is the Taylor series centered at c = 2? 1 31. f (x) = , c = 1 x SOLUTION Write 1 1 = , x 1 + (x − 1) 1 to obtain and then substitute −(x − 1) for x in the Maclaurin series for 1−x ∞ ∞ # # 1 (−1)n (x − 1)n . = [−(x − 1)]n = x n=0 n=0

This series is valid for |x − 1| < 1. 1√ 33. f (x)f (x) = = x,, cc = = 54 1−x SOLUTION Write 1 1 1 1 . = =− · 1−x −4 − (x − 5) 4 1 + x−5 4 1 Substituting − x−5 4 for x in the Maclaurin series for 1−x yields

∞ ∞ n # x −5 n # n (x − 5) . = = (−1) − 4 4n 1 + x−5 n=0 n=0 4 1

Thus, ∞ ∞ # (x − 5)n (x − 5)n 1 1# (−1)n = (−1)n+1 n+1 . =− n 1−x 4 n=0 4 4 n=0

This series is valid for x−5 4 < 1, or |x − 5| < 4.

35. f (x) = x 4 +43x − 1, c = 2 f (x) = x + 3x − 1, c = 0 SOLUTION To determine the Taylor series with center c = 2, we compute f (x) = 4x 3 + 3,

f (x) = 12x 2 ,

f (x) = 24x,

and f (4) (x) = 24. All derivatives of order five and higher are zero. Now, f (2) = 21,

f (2) = 35,

f (2) = 48,

f (2) = 48,

and f (4) (2) = 24. Therefore, the Taylor series is 21 + 35(x − 2) +

48 24 48 (x − 2)2 + (x − 2)3 + (x − 2)4 , 2 6 24

or 21 + 35(x − 2) + 24(x − 2)2 + 8(x − 2)3 + (x − 2)4 .

S E C T I O N 11.7

Taylor Series

651

37. f (x) = e3x , 1 c = −1 f (x) = 2 , c = 4 SOLUTION Write x e3x = e3(x+1)−3 = e−3 e3(x+1) . Now, substitute 3(x + 1) for x in the Maclaurin series for e x to obtain e3(x+1) =

∞ ∞ n # # (3(x + 1))n 3 = (x + 1)n . n! n! n=0 n=0

Thus, e3x = e−3

∞ n −3 ∞ n # # 3 3 e (x + 1)n = (x + 1)n , n! n! n=0 n=0

This series is valid for all x. 1 39. f (x) = 1, c = 3 f (x) 1=− x 2 , c = −2 1 − 4x SOLUTION By partial fraction decomposition 1

1

1 = 2 + 2 , 1−x 1+x 1 − x2 so 1

1

1 1 1 1 1 2 2 = + · . + =− · 2 x−3 −2 − (x − 3) 4 + (x − 3) 4 1+ 8 1 + x−3 1−x 2 4 1 Substituting − x−3 2 for x in the Maclaurin series for 1−x gives

∞ ∞ # (−1)n x −3 n # = = (x − 3)n , − 2 2n 1 + x−3 n=0 n=0 2 1

while substituting − x−3 4 for x in the same series gives ∞ ∞ # x −3 n # (−1)n = = (x − 3)n . − n x−3 4 4 1+ 4 n=0 n=0 1

Thus, ∞ ∞ ∞ ∞ # # 1 1# 1# (−1)n (−1)n (−1)n+1 (−1)n =− (x − 3)n + (x − 3)n = (x − 3)n + (x − 3)n n n 2 n+2 2n+3 4 n=0 2 8 n=0 4 1−x 2 2 n=0 n=0 ∞ ∞ # (−1)n+1 # (−1)n+1 (2n+1 − 1) (−1)n = + 2n+3 (x − 3)n = (x − 3)n . n+2 2n+3 2 2 2 n=0 n=0

This series is valid for |x − 3| < 2. 1 (see Example 10). 41. Find the Maclaurin 1 series for f (x) = f (x) = , c = −1 1 − 9x 2 3x − 2 SOLUTION From Example 10, we know that for |x| < 1, ∞ # 1 · 3 · 5 · · · (2n − 1) 2n 1 x , = 2 2 · 4 · 6 · · · (2n) 1−x n=0

so ∞ ∞ # # 1 1 · 3 · 5 · · · (2n − 1) 1 · 3 · 5 · · · (2n − 1) n 2n 1 = = (3x)2n = 9 x . 2 2 2 · 4 · 6 · · · (2n) 2 · 4 · 6 · · · (2n) 1 − (3x) 1 − 9x n=0 n=0

This series is valid for 9x 2 < 1, or |x| < 13 . 43. Use the first five terms of the Maclaurin series in Exercise 421to approximate sin−1 21 . Compare the result with the , that for |x| < 1, Show, by integrating the Maclaurin series for f (x) = calculator value. 1 − x2 −1

∞ # 1 · 3 · 5 · · · (2n − 1) x 2n+1

652

C H A P T E R 11


From Exercise 42 we know that for |x| < 1, sin−1 x = x +

∞ # 1 · 3 · 5 · · · (2n − 1) x 2n+1 . 2 · 4 · 6 · · · (2n) 2n + 1 n=1

The first five terms of the series are: x+ Setting x =

1 · 3 x5 1 · 3 · 5 x7 1 · 3 · 5 · 7 x9 x3 3x 5 5x 7 35x 9 1 x3 + + + =x+ + + + 2 3 2·4 5 2·4·6 7 2·4·6·8 9 6 40 112 1152

1 , we obtain the following approximation: 2

3

5

7

9 1 3 · 12 5 · 12 35 · 12 1 1 2 + + + ≈ 0.52358519539. sin−1 ≈ + 2 2 6 40 112 1152

The calculator value is sin−1 21 ≈ 0.5235988775. 45. Use the Maclaurin series for ln(1 + x) and ln(1 − x) to show that Show that for |x| < 1 1+x x3 x5 1 ln = x + x3 + x5 + · · · −1 2 tanh 1 − xx = x + 3 + 5 + ··· 3 5 What can you concluded by comparing this1 result with that of Exercise 44? . Hint: Recall that tanh−1 x = SOLUTION Using the d xMaclaurin series 1 −for x 2 ln (1 + x) and ln (1 − x), we have for |x| < 1 ln(1 + x) − ln(1 − x) = =

∞ ∞ # (−1)n−1 n # (−1)n−1 x − (−x)n n n n=1 n=1 ∞ ∞ n ∞ # # (−1)n−1 n # x 1 + (−1)n−1 n x + = x . n n n n=1 n=1 n=1

Since 1 + (−1)n−1 = 0 for even n and 1 + (−1)n−1 = 2 for odd n, ln (1 + x) − ln (1 − x) =

∞ #

2 x 2k+1 . 2k + 1 k=0

Thus, ∞ ∞ # 1 1+x 1# 2 x 2k+1 1 ln = (ln(1 + x) − ln(1 − x)) = x 2k+1 = . 2 1−x 2 2 k=0 2k + 1 2k + 1 k=0 Observe that this is the same series we found in Exercise 44; therefore, 1+x 1 ln = tanh−1 x. 2 1−x " x 2 −t 2 dt as an alternating power−6series in t. 47. UseUse the the Maclaurin expansion forxe−t to express Taylor series for cos to compute cos 1 toe within an error of at most 10 . Use the fact that cos x is an 0 alternating series with decreasing terms to estimate the error. (a) How many terms of the infinite series are needed to approximate the integral for x = 1 to within an error of at most 0.001? (b) SOLUTION

Carry out the computation and check your answer using a computer algebra system. Substituting −t 2 for t in the Maclaurin series for et yields e−t = 2

∞ ∞ # # (−t 2 )n t 2n (−1)n = ; n! n! n=0 n=0

thus, " x 0

e−t dt = 2

∞ # n=0

(−1)n

x 2n+1 . n!(2n + 1)

S E C T I O N 11.7

Taylor Series

653

(a) For x = 1, " 1 0

e−t dt = 2

∞ #

(−1)n

n=0

1 . n!(2n + 1)

1 This is an alternating series with an = n!(2n+1) ; therefore, the error incurred by using S N to approximate the value of

the definite integral is bounded by " 1 2 1 −t e dt − S N ≤ a N +1 = . 0 (N + 1)!(2N + 3) To guarantee the error is at most 0.001, we must choose N so that 1 < 0.001 (N + 1)!(2N + 3)

or

(N + 1)!(2N + 3) > 1000.

For N = 3, (N + 1)!(2N + 3) = 4! · 9 = 216 < 1000 and for N = 4, (N + 1)!(2N + 3) = 5! · 11 = 1320 > 1000; thus, the smallest acceptable value for N is N = 4. The corresponding approximation is S4 =

4 #

(−1)n 1 1 1 1 =1− + − + = 0.747486772. n!(2n + 1) 3 2! · 5 3! · 7 4! ·9 n=0

(b) Using a computer algebra system, we find " 1 0

therefore

e−t dt = 0.746824133; 2

" 1 2 −t e dt − S4 = 6.626 × 10−4 < 10−3 . 0

" x −4 In Exercises 49–52, express sin tthe dt definite integral as an infinite series and find its value to within an error of at most 10 . . Show that Let F(x) = " 1 t 0 49. cos(x 2 ) d x 0 x3 x5 x7 F(x) = x − + − + ··· 2 3 · 3!for cos 5 · 5! 7 · 7! SOLUTION Substituting x for x in the Maclaurin series x yields Evaluate F(1) to three decimal places. ∞ ∞ # # (x 2 )2n x 4n (−1)n (−1)n cos(x 2 ) = = ; (2n)! (2n)! n=0 n=0 therefore, " 1 0

cos(x 2 ) d x =

∞ # n=0

(−1)n

1 ∞ # x 4n+1 (−1)n . = (2n)!(4n + 1) (2n)!(4n + 1) n=0 0

1 ; therefore, the error incurred by using S N to approximate the value of This is an alternating series with an = (2n)!(4n+1) the definite integral is bounded by " 1 1 2 cos(x ) d x − S N ≤ a N +1 = . 0 (2N + 2)!(4N + 5)

To guarantee the error is at most 0.0001, we must choose N so that 1 < 0.0001 (2N + 2)!(4N + 5)

or

(2N + 2)!(4N + 5) > 10000.

For N = 2, (2N + 2)!(4N + 5) = 6! · 13 = 9360 < 10000 and for N = 3, (2N + 2)!(4N + 5) = 8! · 17 = 685440 > 10000; thus, the smallest acceptable value for N is N = 3. The corresponding approximation is S3 =

51.

" 2 3 " −x e 1 d−1 x 2 tan (x ) d x 0 0

3 #

(−1)n 1 1 1 =1− + − = 0.904522792. (2n)!(4n + 1) 5 · 2! 9 · 4! 13 · 6! n=0

654

C H A P T E R 11


Substituting −x 3 for x in the Maclaurin series for e x yields e−x = 3

∞ ∞ # # (−x 3 )n x 3n (−1)n = ; n! n! n=0 n=0

therefore, " 1 0

∞ #

3 e−x d x =

(−1)n

n=0

1 ∞ # x 3n+1 (−1)n . = n!(3n + 1) n!(3n + 1) n=0 0

1 ; therefore, the error incurred by using S N to approximate the value of This is an alternating series with an = n!(3n+1) the definite integral is bounded by " 1 3 1 −x e d x − S N ≤ a N +1 = . 0 (N + 1)!(3N + 4)

To guarantee the error is at most 0.0001, we must choose N so that 1 < 0.0001 (N + 1)!(3N + 4)

or

(N + 1)!(3N + 4) > 10000.

For N = 4, (N + 1)!(3N + 4) = 5! · 16 = 1920 < 10000 and for N = 5, (N + 1)!(3N + 4) = 6! · 19 = 13680 > 10000; thus, the smallest acceptable value for N is N = 5. The corresponding approximation is 5 #

(−1)n = 0.807446200. n!(3n + 1) n=0

S5 =

" 1 53–56, express the integral as an infinite series. In Exercises dx " x 10 − cos(t) x 4 + 1dt, for all x 53. t 0 SOLUTION

The Maclaurin series for cos t is cos t =

∞ #

(−1)n

n=0

∞ # t 2n t 2n (−1)n =1+ , (2n)! (2n)! n=1

so 1 − cos t = −

∞ #

(−1)n

n=1

∞ # t 2n t 2n (−1)n+1 = , (2n)! n=1 (2n)!

and ∞ ∞ # t 2n t 2n−1 1# 1 − cos t (−1)n+1 (−1)n+1 = = . t t n=1 (2n)! n=1 (2n)!

Thus, x " x ∞ ∞ # # 1 − cos(t) t 2n x 2n n+1 (−1) (−1)n+1 dt = . = t (2n)!2n (2n)!2n 0 n=1 n=1 0 " x " x 2 ) dt, for |x| < 1 ln(1t + − tsin t dt, for all x 0 t 0 SOLUTION Substituting t 2 for t in the Maclaurin series for ln(1 + t) yields

55.

ln(1 + t 2 ) =

∞ #

(−1)n−1

n=1

∞ # (t 2 )n t 2n (−1)n = . n n n=1

Thus, " x 0

" x

dt 4

,

ln(1 + t 2 ) dt =

for |x| < 1

∞ # n=1

(−1)n

x ∞ # t 2n+1 x 2n+1 (−1)n . = n(2n + 1) n(2n + 1) n=1 0

S E C T I O N 11.7

57. Which function has Maclaurin series

∞ #

Taylor Series

655

(−1)n 2n x n ?

n=0 SOLUTION

We recognize that ∞ #

∞ #

(−1)n 2n x n =

n=0

(−2x)n

n=0

1 with x replaced by −2x. Therefore, is the Maclaurin series for 1−x ∞ #

(−1)n 2n x n =

n=0

1 1 = . 1 − (−2x) 1 + 2x

In Exercises 59–62, findhas the Maclaurin first four terms Which function seriesof the Taylor series. ∞ # (−1)k (x − 3)k ? 2 3k+1 SOLUTION Substituting x for x in the Maclaurink=0 series for sin x and cos x, we find For which values of x is the expansion x 6 valid? x 10 x 14 x4 x8 x 12 2 2 2 sin(x ) cos(x ) = x − + − + ··· 1− + − + ··· 6 120 5040 2 24 720

59. f (x) = sin(x 2 ) cos(x 2 )

x6 x6 x 10 x 10 x 10 x 14 x 14 x 14 x 14 = x2 − − + + + − − − − + ··· 6 2 24 12 120 720 144 240 5040 2 2 10 4 14 = x2 − x6 + x − x + ··· . 3 15 315 61. f (x) = esin xx f (x) = e tan−1 x SOLUTION Substituting sin x for x in the Maclaurin series for e x and then using the Maclaurin series for sin x, we find sin2 x sin3 x sin4 x + + + ··· 2 6 24 2 1 1 1 x3 x3 + (x − · · · )3 + + ··· + x− + ··· =1+ x − (x − · · · )4 6 2 6 6 24

esin x = 1 + sin x +

1 2 x − 2 1 = 1 + x + x2 − 2 =1+x +

1 3 1 3 1 4 1 4 x + x − x + x + ··· 6 6 6 24 1 4 x + ··· . 8

In Exercises 63–66, find the functions with the following Maclaurin series (refer to Table 1). f (x) = sin(x 3 + 2x) x6 x9 x 12 63. 1 + x 3 + + + + ··· 2! 3! 4! SOLUTION

We recognize 1 + x3 +

∞ 3n ∞ # # x6 x (x 3 )n x9 x 12 + + + ··· = = 2! 3! 4! n! n! n=0 n=0

as the Maclaurin series for e x with x replaced by x 3 . Therefore, 1 + x3 +

3 x6 x9 x 12 + + + · · · = ex . 2! 3! 4!

55 x 5 57 x 7 53 x 3 − 43 x 3 ++4·4·x·5 − 45 x 5 + · · · 65. 1 −1 − 4x + + 42 x 2 − 3! 5! 7! SOLUTION

Note 1−

53 x 3 55 x 5 57 x 7 53 x 3 55 x 5 57 x 7 + − + · · · = 1 − 5x + 5x − + − + ··· 3! 5! 7! 3! 5! 7!

656

C H A P T E R 11

INFINITE SERIES ∞ #

= 1 − 5x +

(5x)2n+1 . (2n + 1)!

(−1)n

n=0

The series is the Maclaurin series for sin x with x replaced by 5x, so 1−

53 x 3 55 x 5 57 x 7 + − + · · · = 1 − 5x + sin(5x). 3! 5! 7!

67. When a voltage V is applied to a series circuit consisting of a resistor R and an inductor L, the current at time t is x 12 x 20 x 28 x4 − + − + ··· V 3 5 7 I (t) = 1 − e−Rt/L R Expand I (t) in a Maclaurin series. Show that I (t) ≈ V t/L if R is small. SOLUTION

t Substituting − Rt L for t in the Maclaurin series for e gives

n ∞ − Rt ∞ ∞ # # # (−1)n R n n (−1)n R n n L t =1+ t e−Rt/L = = n! n! L n! L n=0 n=0 n=1

Thus,

1 − e−Rt/L = 1 −

1+

∞ # (−1)n R n n!

n=1

L

tn

∞ # (−1)n+1 Rt n

=

n=1

n!

L

,

and I (t) =

∞ V# (−1)n+1 R n=1 n!

∞ V# Vt Rt n (−1)n+1 Rt n + = . L L R n=2 n! L

If Rt/L is small, then the terms in the series are even smaller, and we find V (t) ≈

Vt . L

√ 69. Find the Maclaurin series for f (x) = cos( x) and use it to determine f (5) (0). 20 Use substitution √ to write out the first three terms of the Maclaurin series for f (x) = e x . Explain how the SOLUTION Substituting x for x in the Maclaurin series for cos x result implies that f (k) (0) = 0 for 1 ≤ k ≤ 19. √ ∞ ∞ # # √ ( x)2n xn n cos( x) = (−1) (−1)n = . (2n)! (2n)! n=0 n=0 The coefficient of x 5 in this series is 1 f (5) (0) (−1)5 =− = , 10! 10! 5! so f (5) (0) = −

5! 1 1 =− =− . 10! 6 · 7 · 8 · 9 · 10 30240

71. Use the binomial series(8) to find f (8) (0) for f −1 (x) = 1 − x 2 . (7) Find f (0) and f (0) for f (x) = tan x. SOLUTION We obtain the Maclaurin series for f (x) = 1 − x 2 by substituting −x 2 for x in the binomial series with a = 12 . This gives 1 ∞ 1

∞ n # # 2 2 −x 2 = x 2n . 1 − x2 = (−1)n n n n=0

The coefficient of x 8 is (−1)4

1 2

4

=

1 2

n=0

1 −1 2

1 −2 2

4!

1 −3 2

=−

so f (8) (0) =

−15 · 8! = −1575. 16 · 4!

15 f (8) (0) = , 16 · 4! 8!

S E C T I O N 11.7

Taylor Series

657

3/4 converge to f (x) at x = 2? Give numerical evidence to support your 73. Does the Taylor series 3 forπ 5f (x) π=7 (1 + x) π answer. Show that π − + − + · · · converges to zero. How many terms must be computed to get within 0.01 3! 5! 7! SOLUTION of zero? The Taylor series for f (x) = (1 + x)3/4 converges to f (x) for |x| < 1; because x = 2 is not contained on this interval, the series does not converge to f (x) at x = 2. The graph below displays

SN =

N 3 # 4

n

n=0

2n

for 0 ≤ N ≤ 14. The divergent nature of the sequence of partial sums is clear. SN 15 10 5 0 −5

2

4

6

8

10 12 14

N

−10 −15 −20

75. Explain the steps required to verify that the Maclaurin series for f (x) = tan−1 x converges to f (x) at x = 0.5. Explain the steps required to verify that the Maclaurin series for f (x) = sin x converges to f (x) at x = 1. SOLUTION Recall that the Maclaurin series for tan−1 x can be obtained by term-by-term integration of the Maclaurin series for 1 2 . Now, we know that the geometric series 1+x

∞ #

(−x 2 )n

n=0 1 for |x| < 1. It then follows from Theorem 3 of Section 11.6 that the Maclaurin series for f (x) = 1+x 2 tan−1 x converges to f (x) for |x| < 1. Because x = 0.5 is inside this interval, the Maclaurin series for f (x) = tan−1 x

converges to

converges to f (x) at x = 0.5.

77. How many terms of √ the Maclaurin series of f (x) = ln(1 + x) are needed to compute ln 1.2 to within an error of at Let Make f (x) = 1 + x. most 0.0001? the computation and compare the result with the calculator value. (a) Use a graphing calculator to compare the graph of f with the graphs of the first five Taylor polynomials for f . SOLUTION Substitute x = 0.2 into the Maclaurin series for ln (1 + x) to obtain: What do they suggest about the interval of convergence of the Taylor series? ∞ the Taylor expansion ∞for f is valid for x = 1 and x = −1. (b) Investigate numerically whether or # not # (0.2)n 1 (−1)n−1 (−1)n−1 n . = ln 1.2 = n 5 n n=1 n=1 This is an alternating series with an =

1 . Using the error bound for alternating series n · 5n |ln 1.2 − S N | ≤ a N +1 =

1 , (N + 1)5 N +1

so we must choose N so that 1 < 0.0001 (N + 1)5 N +1

or

(N + 1)5 N +1 > 10000.

For N = 3, (N + 1)5 N +1 = 4 · 54 = 2500 < 10, 000, and for N = 4, (N + 1)5 N +1 = 5 · 55 = 15, 625 > 10, 000; thus, the smallest acceptable value for N is N = 4. The corresponding approximation is: S4 =

4 # (−1)n−1

5n · n n=1

=

1 1 1 1 + − = 0.182266666. − 5 52 · 2 53 · 3 54 · 4

Now, ln 1.2 = 0.182321556, so |ln 1.2 − S4 | = 5.489 × 10−5 < 0.0001. In Exercises 78–79, let f (x) =

1 (1 − x)(1 − 2x)

658

C H A P T E R 11

INFINITE SERIES

79. Find the Taylor series for f (x) at c = 2. Hint: Rewrite the identity of Exercise 78 as Find the Maclaurin series of f (x) using the identity 2 1 f (x) = 1 2 − −3f (x) − 2(x −1 − (x − 2) = − 2) − 1 − 2x 1−x SOLUTION Using the given identity, f (x) =

1 2 2 1 1 − =− . + −3 − 2(x − 2) −1 − (x − 2) 3 1 + 2 (x − 2) 1 + (x − 2) 3

1 yields Substituting − 23 (x − 2) for x in the Maclaurin series for 1−x

1 1 + 23 (x − 2)

=

∞ #

(−1)n

n=0

n 2 (x − 2)n , 3

and substituting −(x − 2) for x in the same Maclaurin series yields ∞ # 1 (−1)n (x − 2)n . = 1 + (x − 2) n=0

The first series is valid for − 23 (x − 2) < 1, or |x − 2| < 32 , and the second series is valid for |x − 2| < 1; therefore, the two series together are valid for |x − 2| < 1. Finally, for |x − 2| < 1, ) n+1 * n ∞ ∞ ∞ # # 2# 2 2 n n n n n f (x) = − (x − 2)n . (−1) (x − 2) + (−1) (x − 2) = (−1) 1 − 3 n=0 3 3 n=0 n=0 81. Use Example 11 and the approximation sin x ≈ x to show that the period T of a pendulum released at an angle θ Use the first five terms of the Maclaurin series for the elliptic function E(k) to estimate the period T of a 1-m has the following second-order approximation: pendulum released at an angle θ = π4 (see Example 11). L θ2 T ≈ 2π 1+ g 16 SOLUTION

The period T of a pendulum of length L released from an angle θ is L E(k), T =4 g

where g ≈ 9.8 m/s2 is the acceleration due to gravity, E(k) is the elliptic function of the first kind and k = sin θ2 . From Example 11, we know that E(k) =

∞ π # 1 · 3 · 5 · · · (2n − 1) 2 2n k . 2 n=0 2 · 4 · 6 · · · (2n)

Using the approximation sin x ≈ x, we have k = sin

θ θ ≈ ; 2 2

moreover, using the first two terms of the series for E(k), we find ) 2 2 * 1 π θ π θ2 = 1+ 1+ . E(k) ≈ 2 2 2 2 16 Therefore,

T =4

L E(k) ≈ 2π g

L g

θ2 1+ 16

.

Further Insights and Challenges In Exercises we we investigate thethe convergence of the binomial In this83–84, exercise show that Maclaurin expansion of theseries function f (x) = ln(1 + x) is valid for x = 1. ∞ (a) Show that for all x = −1, # a n x Ta (x) = n N N +1 x N +1 # n=0 (−1) 1 (−1)n x n + = 1+x 1+x n=0

S E C T I O N 11.7

Taylor Series

659

83. Prove that Ta (x) has radius of convergence R = 1 if a is not a whole number. What is the radius of convergence if a is a whole number? SOLUTION

Suppose that a is not a whole number. Then a (a − 1) · · · (a − n + 1) a = n n!

is never zero. Moreover, a a − n n + 1 a(a − 1) · · · (a − n + 1)(a − n) n! = , = · (n + 1)! a(a − 1) · · · (a − n + 1) n + 1 a n so, by the formula for the radius of convergence a − n = 1. r = lim n→∞ n + 1 The radius of convergence of Ta (x) therefore R = r −1 = 1. is a = 0 for all n > a. The infinite series then reduces to a polynomial of degree a, If a is a whole number, then n so it converges for all x (i.e. R = ∞). " π /2 Exercise 83,=Ta (x) converges 100, 0≤

100n = n!

100 100 100 · ··· 1 2 100

100 100 100 100100 · · < ; 101 102 n 99!n

therefore, 100n =0 n→∞ n! lim

by the Squeeze Theorem. Moreover,

3 + πn n→∞ 5n

lim

3

+ lim n→∞ 5n n→∞

= lim

π n 5

= 0 + 0 = 0.

Thus, lim an = 0 + 0 = 0.

n→∞

3 n m 21. cn = 1 + 1 bm = 1n+ m SOLUTION Write ) *3 1 n 1 n/3 = 1+ . cn = 1 + n/3 n/3

Then, because x 3 is a continuous function, *3 1 n/3 lim cn = lim 1 + = e3 . n→∞ n→∞ n/3 )

arctan(n 2 ) n(ln(n +Theorem 1) − ln n) n =Squeeze = 0. 23. Usebthe to show that lim √ n→∞ n SOLUTION

For all x, −

π π < arctan x < , 2 2

so

π /2 π /2 arctan(n 2 ) −√ < < √ , √ n n n for all n. Because

π /2 −√ n→∞ n lim

π /2 = lim √ = 0, n→∞ n

it follows by the Squeeze Theorem that arctan(n 2 ) = 0. √ n→∞ n lim

1 n 1 n 25. Given anan=example 3 − of2 a, divergent sequence {an } such that {sin an } is convergent. Give 2 3 (a) Calculate lim an . n→∞ a (b) Calculate lim n+1 . n→∞ an SOLUTION

(a) Because 1 1 3n 1 n 1 n 3 − 2 ≥ 3n − 3n = 2 3 2 3 6 and lim

3n

n→∞ 6

= ∞,


663

we conclude that lim an = ∞.

n→∞

n+1 2 1 3n+1 − 1 2n+1 n+2 n+2 3 − 2 3−0 a 3 −2 3 3 = lim = lim (b) lim n+1 = lim 2 1

n+1 = 1 − 0 = 3. n→∞ an n→∞ n − 1 2n n→∞ 3n+1 − 2n+1 n→∞ 3 2 2 3 1− 3 ∞ # n−2 and aS17 = of 2. the series 27. Calculate partial . Define the an+1 = sums an + S64with n 2 + 2n n=1 (a) Compute an for n = 2, 3, 4, 5. SOLUTION (b) Show that {an } is increasing and bounded by 3. (c) Prove that lim an exists1 and find1its value. 2 11 n→∞S = − + 0 + + =− = −0.183333; 4 3 15 24 60 1 1 2 3 4 5 287 + + + + = = 0.065079. S7 = − + 0 + 3 15 24 35 48 63 4410

8 16 32 4 29. Find the sum + 1+ 1 + 1 + · · · . 9 1− 27 +81 −243 + · · · . Find the sum 4 42 43 2 SOLUTION This is a geometric series with common ratio r = 3 . Therefore, 4

4 8 16 32 4 + + + + ··· = 9 2 = . 9 27 81 243 3 1− 3 ∞ n+1 # 2 n ∞ # 31. Find the sum 2. 3n Find the sum . n=0 e n=2 SOLUTION Note ∞ n+1 ∞ n ∞ n # # # 2 2 2 = 2 = 2 ; n n 3 3 3 n=0 n=0 n=0

therefore, ∞ n+1 # 1 2 = 2 · 3 = 6. n =2 3 1 − 23 n=0 ∞ ∞ ∞ # # # ∞ of # divergent an , bnπ such that (an + bn ) = 1. 33. Give an example series −1 2 Show that b − tan n diverges ifn=1 b = . n=1 n=1 2 n=1 n 1 SOLUTION Let an = 2 + 1, bn = −1. The corresponding series diverge by the Divergence Test; however, ∞ #

(an + bn ) =

n=1

∞ n # 1 n=1

2

=

1 2

= 1. 1 − 12

In Exercises usearea the of Integral Test to determine if the series Find35–38, the total the infinitely many circles on infinite the interval [0,converges. 1] in Figure 1. 35.

∞ #

n2

n=1

n3 + 1

SOLUTION

2 Let f (x) = 3x . This function is continuous and positive for x ≥ 1. Because

x +1

f (x) =

(x 3 + 1)(2x) − x 2 (3x 2 ) x(2 − x 3 ) = 3 , 3 2 (x + 1) (x + 1)2

we see that f (x) < 0 and f is decreasing on the interval x ≥ 2. Therefore, the Integral Test applies on the interval x ≥ 2. Now, " ∞ " R

x2 x2 1 3 + 1) − ln 9 = ∞. d x = lim d x = ln(R lim 3 R→∞ R→∞ 2 x 3 + 1 x3 + 1 2 The integral diverges; hence, the series

∞ #

n2

n=2

n3 + 1

diverges, as does the series

∞ #

n2

n=1

n3 + 1

.

664

C H A P T E R 11

INFINITE SERIES

37.

∞ # n3 ∞ # n2 n4 n=1 e (n 3 + 1)1.01 n=1

SOLUTION

Let f (x) = x 3 e−x . This function is continuous and positive for x ≥ 1. Because

4 4 4 f (x) = x 3 −4x 3 e−x + 3x 2 e−x = x 2 e−x 3 − 4x 4 , 4

we see that f (x) < 0 and f is decreasing on the interval x ≥ 1. Therefore, the Integral Test applies on the interval x ≥ 1. Now, " R " ∞

4 4 4 1 1 . x 3 e−x d x = lim x 3 e−x d x = − lim e−R − e−1 = 4 4e R→∞ R→∞ 1 1 The integral converges; hence, the series

∞ # n3 n n=1 e

4

also converges.

In Exercises ∞ 39–46, use the Comparison or Limit Comparison Test to determine whether the infinite series converges. # 1 ∞ # (2n + 1)(ln(2n + 1))2 n=1 1 39. (n + 1)2 n=1 SOLUTION

For all n ≥ 1, 0
“Butterfly 0 and set kCurve”: = 35. . Show that the trochoid a−b 5 x = at − b sin t, y= a − b cos t, 0 t≤ t ≤ T cos t − 2 cos 4t − sin x(t) = sin t e 12 T , k with G(θ , k) as in Exercise has length 2(a − b)G 29. 2 t 5 cos t − 2 cos 4t − sin y(t) = cos t e 12 SOLUTION We have x (t) = a − b cos t, y (t) = b sin t. Hence, 2 + and 2 = a 2s(t) 2 cos π . 2 t + b2 sin2 t (a) Use a computer plott)b(t) thet)speed for 0cos ≤ tt + ≤b12 x (t)2 +algebra y (t)2 system = (a −to b cos (b sin − 2ab (b) Approximate the length b(t)2for 02≤ t ≤ 10π . = a + b − 2ab cos t

The length of the trochoid for 0 ≤ t ≤ T is L=

" T a 2 + b2 − 2ab cos t dt 0

We rewrite the integrand as follows to bring it to the required form. We use the identity 1 − cos t = 2 sin2 2t to obtain L=

" T 0

(a − b)2 + 2ab − 2ab cos t dt =

" T (a − b)2 + 2ab(1 − cos t) dt 0

" T t 4ab 2 t dt (a − b)2 + 4ab sin2 dt = (a − b)2 1 + sin = 2 2 (a − b)2 0 0 " T t 1 + k 2 sin2 dt = (a − b) 2 0 " T

√

ab ). (where k = 2a−b

Substituting u = 2t , du = 12 dt, we get L = 2(a − b)

" T /2 1 + k 2 sin2 u du = 2(a − b)E(T /2, k) 0

Gm e 2 , where R by 37. TheThe acceleration to gravity on at thea surface earth g = of2the=earth 9.8 m/s path of a due satellite orbiting distanceofRthe from theis center is parametrized x = Rkm. cosUse ω t, e = 6,378 R e √ y = R sin ω t. Exercise 36(b) to show that a satellite orbiting at the earth’s surface would have period Te = 2π Re /g ≈ 84.5 min. (a) Show that the period T (the time of one revolution) is T = 2π /ω . Then estimate the distance Rm from the moon to the center of the earth. Assume that the period of the moon (sidereal (b) to Newton’s laws of motion and gravity, month) isAccording Tm ≈ 27.43 days. x x (t) = −Gm e 3 , R

y y (t) = −Gm e 3 R

where G is the universal gravitational constant and m e is the mass of the earth. Prove that has the same value for all orbits (a special case of Kepler’s Third Law).

R3 Gm e R3 = . Thus, T2 4π 2 T2

698

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S SOLUTION

By part (b) of Exercise 36, it follows that Re3 Te2

=

Gm e 4π 2 Re3 4π 2 Re 4π 2 Re ⇒ Te2 = = Gm = 2 e Gm e g 4π 2 Re

Hence,

Te = 2π

Re = 2π g

6378 · 103 ≈ 5068.8 s ≈ 84.5 min. 9.8

R3 is the same for all orbits. It follows that this quotient is the same for the T2 satellite orbiting at the earth’s surface and for the moon orbiting around the earth. Thus, 3 Tm 2/3 Rm Re3 = 2 ⇒ R m = Re . Te Tm2 Te In part (b) of Exercise 36 we showed that

Setting Tm = 27.43 · 1440 = 39499.2 minutes, Te = 84.5 minutes, and Re = 6378 km we get 39499.2 2/3 Rm = 6378 ≈ 384154 km. 84.5

12.3 Polar Coordinates Preliminary Questions 1. If P and Q have the same radial coordinate, then (choose the correct answer): (a) P and Q lie on the same circle with the center at the origin. (b) P and Q lie on the same ray based at the origin. SOLUTION Two points with the same radial coordinate are equidistant from the origin, therefore they lie on the same circle centered at the origin. The angular coordinate defines a ray based at the origin. Therefore, if the two points have the same angular coordinate, they lie on the same ray based at the origin.

2. Give two polar coordinate representations for the point (x, y) = (0, 1), one with negative r and one with positive r . The point (0, 1) is on the y-axis, distant one unit from the origin, hence the polar representation with positive r is (r, θ ) = 1, π2 . The point (r, θ ) = −1, π2 is the reflection of (r, θ ) = 1, π2 through the origin, hence we must add π to return to the original point. We obtain the following polar representation of (0, 1) with negative r :

3π π . (r, θ ) = −1, + π = −1, 2 2 SOLUTION

3. Does a point (r, θ ) have more than one representation in rectangular coordinates? The rectangular coordinates are determined uniquely by the relations x = r cos θ , y = r sin θ . Therefore a point (r, θ ) has exactly one representation in rectangular coordinates. SOLUTION

4. Describe the curves with polar equations (a) r = 2 (b) r 2 = 2

(c) r cos θ = 2

SOLUTION

(a) Converting to rectangular coordinates we get

x 2 + y 2 = 2 or

x 2 + y 2 = 22 .

This is the equation of the circle of radius 2 centered at the origin. √ (b) We convert to rectangular coordinates, obtaining x 2 + y 2 = 2. This is the equation of the circle of radius 2, centered at the origin. (c) We convert to rectangular coordinates. Since x = r cos θ we obtain the following equation: x = 2. This is the equation of the vertical line through the point (2, 0). 5. If f (−θ ) = f (θ ), then the curve r = f (θ ) is symmetric with respect to the (choose the correct answer): (a) x-axis (b) y-axis (c) origin SOLUTION The equality f (−θ ) = f (θ ) for all θ implies that whenever a point (r, θ ) is on the curve, also the point (r, −θ ) is on the curve. Since the point (r, −θ ) is the reflection of (r, θ ) with respect to the x-axis, we conclude that the curve is symmetric with respect to the x-axis.

S E C T I O N 12.3

Polar Coordinates

699

Exercises 1. Find polar coordinates for each of the seven points plotted in Figure 17. y (x, y) = (2 3, 2)

3 2 1 1 2 3

x

FIGURE 17 SOLUTION

We mark the points as shown in the figure. y A

F(2 3, 2) E

x

B C

D G(2 3, −2)

Using the data given in the figure for the x and y coordinates and the quadrants in which the point are located, we obtain:

√

r = (−3)2 + 32 = 18 ⇒ (r, θ ) = 3√2, 3π (A): 4 θ = π − π4 = 34π y A 3 2 3π 4

x

(B):

r =3 ⇒ (r, θ ) = (3, π ) θ=π y

π B

x

3

√

√ 5 ≈ 2.2 22 + 12 = ⇒ (r, θ ) ≈ (C): 5, 3.6 −1 1 −1 −1 θ = tan −2 = tan 2 = π + 0.46 ≈ 3.6 r=

y

3.6 x C

(D):

2.2

√

√ r = 12 + 12 = 2 ≈ 1.4 ⇒ (r, θ ) ≈ 2, 54π π 5 π θ=π+ 4 = 4 y

5π 4

1.4 D

x

700

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S

(E):

r=

θ

12 + 12=

= tan−1

1 1

√

2 ≈ 1.4

= π4

√

⇒ (r, θ ) ≈

2, π4

y

E π

1.4

√ √ 2 2 3 + 22 = 16 = 4 (F):

2 = tan−1 √1 = π6 θ = tan−1 √ r=

2 3

4

x

⇒ (r, θ ) = 4, π6

3

y F(2 3, 2) 4

π 6

x

(G): G is the reflection of F about the x axis, hence the two points have equal radial coordinates, and the angular coordinate of G is obtained from the angular coordinate of F: θ = 2π − π6 = 116π . Hence, the polar coordinates of

G are 4, 116π . 3. Convert from rectangular to polar coordinates: Plot the points with polar coordinates (a) (1, 0) π (b) (4, 34π ) (a) (2, 6 ) (c) (−2, 2)

√ (b) (3, 3) π (c) (3, √ −2) (d) (−1, 3)

(d) (0, π6 )

SOLUTION

(a) The point (1, 0) is on the positive x axis distanced one unit from the origin. Hence, r = 1 and θ = 0. Thus, (r, θ ) = (1, 0).

√

√

√ 2 √ 3 π −1 2+ = (b) The point 3, 3 is in the first quadrant so θ = tan . Also, r = 3 3 = 12. Hence, 3 6

√ π 12, 6 . (r, θ ) = (c) The point (−2, 2) is in the second quadrant. Hence, 3π π 2 θ = tan−1 = tan−1 (−1) = π − = . −2 4 4

√ √ Also, r = (−2)2 + 22 = 8. Hence, (r, θ ) = 8, 34π .

√ (d) The point −1, 3 is in the second quadrant, hence,

θ Also, r =

(−1)2 +

= tan−1

√

√ 3 π 2π = tan−1 − 3 = π − = . −1 3 3

√ 2 √

3 = 4 = 2. Hence, (r, θ ) = 2, 23π .

5. Convert from polar to rectangular coordinates: Use to polar coordinates (make sure your choice of θ gives the correct π ) a calculator to convert from rectangular (b) (6, 34π ) (c) (5, − π2 ) (a) (3, 6 quadrant): SOLUTION (a) (2, 3) (b) (4, −7) (a) Since r =−8) 3 and θ = π6 , we have: (c) (−3, (d) (−5, 2) √ π 3 x = r cos θ = 3 cos = 3 · ≈ 2.6 6 2 ⇒ (x, y) ≈ (2.6, 1.5) . 1 π y = r sin θ = 3 sin = 3 · = 1.5 6 2

S E C T I O N 12.3

Polar Coordinates

701

(b) For 6, 34π we have r = 6 and θ = 34π . Hence, 3π ≈ −4.24 4 3π ≈ 4.24 y = r sin θ = 6 sin 4

x = r cos θ = 6 cos

⇒

(x, y) ≈ (−4.24, 4.24) .

(c) Since r = 5 and θ = − π2 we have

π =5·0=0 x = r cos θ = 5 cos − 2

π y = r sin θ = 5 sin − = 5 · (−1) = −5 2

⇒

(x, y) = (0, −5)

7. Which of the following are possible polar coordinates for the point P with rectangular coordinates (0, −2)? from polar to rectangular coordinates:

Convert π 7π (a) (a) 2, (0, 0) (b) 2, (b) (−4, π3 ) (c) (0, π6 ) 2 2 3π 7π (c) −2, − (d) −2, 2 2

7π π (f) 2, − (e) −2, − 2 2 SOLUTION The point P has distance 2 from the origin and the angle between O P and the positive x-axis in the positive

direction is 32π . Hence, (r, θ ) = 2, 32π is one choice for the polar coordinates for P. y 3π 2

x

0 P

The polar coordinates (2, θ ) are possible for P if θ − 32π is a multiple of 2π . The polar coordinate (−2, θ ) are possible for P if θ − 32π is an odd multiple of π . These considerations lead to the following conclusions: (a) 2, π2 π2 − 32π = −π ⇒ 2, π2 does not represent P.

(b) 2, 72π 72π − 32π = 2π ⇒ 2, 72π represents P.

(c) −2, − 32π − 32π − 32π = −3π ⇒ −2, − 32π represents P.

(d) −2, 72π 72π − 32π = 2π ⇒ −2, 72π does not represent P. (e) −2, − π2 − π2 − 32π = −2π ⇒ −2, − π2 does not represent P.

(f) 2, − 72π − 72π − 32π = −5π ⇒ 2, − 72π does not represent P. 1 9. FindDescribe the equation in polarsector coordinates of the through theinorigin θ . slope 2 . each shaded in Figure 18 line by inequalities r andwith

A line of slope m = 12 makes an angle θ0 = tan−1 21 ≈ 0.46 with the positive x-axis. The equation of the line is θ ≈ 0.46, while r is arbitrary. SOLUTION

11. Which of the two equations, r = 2 sec θ and r = 2 csc θ , defines a horizontal line? What is the slope of the line θ = 3π ? SOLUTION The equation r = 2 csc θ is 5 the polar equation of a horizontal line, as it can be written as r = 2/ sin θ , so r sin θ = 2, which becomes y = 2. On the other hand, the equation r = 2 sec θ is the polar equation of a vertical line, as it can be written as r = 2/ cos θ , so r cos θ = 2, which becomes x = 2. In Exercises 12–17, convert to an equation in rectangular coordinates. 13. r = sin θ r =7 SOLUTION

Multiplying by r and substituting y = r sin θ and r 2 = x 2 + y 2 gives r 2 = r sin θ x 2 + y2 = y

702

C H A P T E R 12


We move the y and then complete the square to obtain x 2 + y2 − y = 0 2 1 2 1 = x2 + y − 2 2

Thus, r = sin θ is the equation of a circle of radius 12 and center 0, 12 . 15. r = 2 csc θ r = 2 sin θ SOLUTION We multiply the equation by sin θ and substitute y = r sin θ . We get r sin θ = 2 y=2 Thus, r = 2 csc θ is the equation of the line y = 2. 1 17. r = 1 r 2=− cos θ cos θ − sin θ SOLUTION We multiply the equation by 2 − cos θ . Then we substitute x = r cos θ and r = x 2 + y 2 , to obtain r (2 − cos θ ) = 1 2r − r cos θ = 1

2 x 2 + y2 − x = 1 Moving the x, then squaring and simplifying, we obtain

2 x 2 + y2 = x + 1

4 x 2 + y 2 = x 2 + 2x + 1 3x 2 − 2x + 4y 2 = 1 We complete the square: 2 3 x 2 − x + 4y 2 = 1 3 1 2 4 3 x− + 4y 2 = 3 3

2 x − 13 y2 + 1 =1 4 9

3

This is the equation of the ellipse shown in the figure:

y

π 2

π

0.5

1 1.5 x

0π

3π 2

In Exercises 18–21, convert to an equation in polar coordinates. 19. x = 5 x 2 + y2 = 5 SOLUTION Substituting x = r cos θ gives the polar equation r cos θ = 5 or r = 5 sec θ . 21. x y = 1 y = x2

Polar Coordinates


703

We substitute x = r cos θ , y = r sin θ to obtain (r cos θ ) (r sin θ ) = 1 r 2 cos θ sin θ = 1

Using the identity cos θ sin θ = 12 sin 2θ yields r2 ·

sin 2θ = 1 ⇒ r 2 = 2 csc 2θ . 2

23. Find the values of θ in the plot of r = 4 cos θ corresponding to points A, B, C, D in Figure 19. Then indicate the Match the equation with its description: portion of the graph traced out as θ varies in the following intervals: π ≤θ ≤π (i) Vertical (a) 0(a)≤ rθ = ≤ 2π2 (b)line (c) π ≤ θ ≤ 32π 2 (ii) Horizontal line (b) θ = 2 (c) r = 2 sec θ (iii) Circle y (d) r = 2 csc θ (iv) Line through origin B 2

C 2 −2

A x 4

D

FIGURE 19 Plot of r = 4 cos θ .

The point A is on the x-axis hence θ = 0. The point B is in the first quadrant with x = y = 2 hence

2 = tan−1 (1) = π . The point C is at the origin. Thus, 2 4

SOLUTION

θ

= tan−1

r = 0 ⇒ 4 cos θ = 0 ⇒ θ =

π 3π , . 2 2

The point D is in the fourth quadrant with x = 2, y = −2, hence 7π π −2 θ = tan−1 = tan−1 (−1) = 2π − = . 2 4 4 0 ≤ θ ≤ π2 represents the first quadrant, hence the points (r, θ ) where r = 4 cos θ and 0 ≤ θ ≤ π2 are the points on the circle which are in the first quadrant, as shown below: y

x

If we insist that r ≥ 0, then since π2 ≤ θ ≤ π represents the second quadrant and π ≤ θ ≤ 32π represents the third quadrant, and since the circle r = 4 cos θ has no points in the left x y -plane, then there are no points for (b) and (c). However, if we allow r < 0 then (b) represents the semi-circle y

y

x

and (c) like (a) represent

x

25. Match each equation in rectangular coordinates with its equation in polar coordinates. Suppose that (x, y) has polar coordinates (r, θ ). Find the polar coordinates of the following: 2 + y2 = 2 2 (1 + 2 sin2 θ ) = 4 (a) x(a) (i) r(−x, (x, −y) (b) −y) (c) (−x, y) (d) (y, x) 2 2 (b) x + (y − 1) = 1 (ii) r (cos θ + sin θ ) = 4 (iii) r = 2 sin θ (c) x 2 − y 2 = 4 (d) x + y = 4 (iv) r = 2 SOLUTION

(a) Since x 2 + y 2 = r 2 , we have r 2 = 4 or r = 2.

704

C H A P T E R 12


(b) Using Example 7, the equation of the circle x 2 + (y − 1)2 = 1 has polar equation r = 2 sin θ . (c) Setting x = r cos θ , y = r sin θ in x 2 − y 2 = 4 gives

x 2 − y 2 = r 2 cos2 θ − r 2 sin2 θ = r 2 cos2 θ − sin2 θ = 4. We now use the identity cos2 θ − sin2 θ = 1 − 2 sin2 θ to obtain the following equation:

r 2 1 − 2 sin2 θ = 4. (d) Setting x = r cos θ and y = r sin θ in x + y = 4 we get: x+y=4 r cos θ + r sin θ = 4 so r (cos θ + sin θ ) = 4 27. ( 12 ,

√ center in rectangular coordinates is Show that r = sin θ + cos θ is the equation of the circle of radius 1/ 2 whose are the polar equations of the lines parallel to the line r cos(θ − π3 ) = 1? 1 ). What Then find the values of θ between 0 and π such that ( θ , r ( θ )) yields the points A, B, C, and D in Figure 20. 2 y A

D

( 12 , 12 ) B

C

x

FIGURE 20 Plot of r = sin θ + cos θ . SOLUTION

We show that the rectangular equation of r = sin θ + cos θ is

1 2 1 1 2 + y− = . x− 2 2 2

We multiply the polar equation by r and substitute r 2 = x 2 + y 2 , r sin θ = y, r cos θ = x. This gives r = sin θ + cos θ r 2 = r sin θ + r cos θ x 2 + y2 = y + x Transferring sides and completing the square yields x 2 − x + y2 − y = 0 1 2 1 1 1 1 2 + y− = + = x− 2 2 4 4 2 The point A corresponds to θ = π2 . Hence, r = sin π2 + cos π2 = 1 + 0 = 1. That is, for the point A we have (θ , r ) = π is, sin θ + cos θ = 0. Solving for 0 ≤ θ ≤ π we get sin θ = − cos θ or 2 , 1 . The point B corresponds to r = 0, that

tan θ = −1, hence θ = 34π . That is, (θ , r ) = 34π , 0 . The point C corresponds to θ = 0. Hence, r = sin 0 + cos 0 = 1. That is, (θ , r ) = (0, 1). The point D is on the line y = x, hence θ = π4 . The corresponding value of r is √ π π 2 √ = 2. r = sin + cos = 2 4 4 2

√ Thus, (θ , r ) = π4 , 2 .

29. Sketch the graph of r = 3 cos θ − 1 (see Example 8). Sketch the curve r = 12 θ (the spiral of Archimedes) for θ between 0 and 2π by plotting the points for θ = 0, π4 , π2 , . . . , 2π .

S E C T I O N 12.3

Polar Coordinates

705

SOLUTION We first choose some values of θ between 0 and π and mark the corresponding points on the graph. Then we use symmetry (due to cos (2π − θ ) = cos θ ) to plot the other half of the graph by reflecting the first half through the x-axis. Since r = 3 cos θ − 1 is periodic, the entire curve is obtained as θ varies from 0 to 2π . We start with the values θ = 0, π6 , π3 , π2 , 23π , 56π , π , and compute the corresponding values of r :

r = 3 cos 0 − 1 = 3 − 1 = 2 ⇒ A = (2, 0) √

π π 3 3 r = 3 cos − 1 = − 1 ≈ 1.6 ⇒ B = 1.6, 6 2 6

π π 3 r = 3 cos − 1 = − 1 = 0.5 ⇒ C = 0.5, 3 2 3

π π r = 3 cos − 1 = 3 · 0 − 1 = −1 ⇒ D = −1, 2 2 2π 2π r = 3 cos − 1 = −2.5 ⇒ E = −2.5, 3 3 5π 5π − 1 = −3.6 ⇒ F = −3.6, r = 3 cos 6 6 r = 3 cos π − 1 = −4 ⇒ G = (−4, π ) The graph begins at the point (r, θ ) = (2, 0) and moves toward the other points in this order, as θ varies from 0 to π . Since r is negative for π2 ≤ θ ≤ π , the curve continues into the fourth quadrant, rather than into the second quadrant. We obtain the following graph: π 2

2π 3

π 3 π 6

5π 6

C

B G

A

π D

0

F

E

Now we have half the curve and we use symmetry to plot the rest. Reflecting the first half through the x axis we obtain the whole curve: y

B

1 C

A 2

G x 4

D E

F

31. Figure 21 displays the graphs of r = sin 2θ in rectangular coordinates and in polar coordinates, where it is a “rose Sketch the graph of r = cos θ − 1. with four petals.” Identify (a) the points in (B) corresponding labeled A–I tothe points in (A), and (b) the parts of the curve in (B) corresponding to the angle intervals 0, π2 , π2 , π , π , 32π , and 32π , 2π . r

y B

F

A

C

E

G

I 2

3 2

2

D

x

H

(I) Graph of r = sin 2 in rectangular coordinates

(II) Graph of r = sin 2 in polar coordinates

FIGURE 21 Rose with four petals. SOLUTION

706

C H A P T E R 12


(a) The graph (I) gives the following polar coordinates of the labeled points:

θ = 0,

A:

r =0

π 2π , r = sin =1 4 4 π C: θ = , r = 0 2 3π 2 · 3π , r = sin = −1 D: θ = 4 4 E: θ = π , r = 0 B: θ =

5π , r =1 4 3π , r =0 G: θ = 2 7π , r = −1 H: θ = 4 I : θ = 2π , r = 0.

θ=

F:

Since the maximal value of |r | is 1, the points with r = 1 or r = −1 are the furthest points from the origin. The corresponding quadrant is determined by the value of θ and the sign of r . If r0 < 0, the point (r0 , θ0 ) is on the ray θ = −θ0 . These considerations lead to the following identification of the points in the x y plane. Notice that A, C, G, E, and I are the same point. y

3π 4 H

π

B 4

7π 4

=

π 2

A,C,E,G,I

r = −1 π

π 4

=

r=1 x 2π

r=1 =

r = −1

5π 4

=

5π F 4

3π 4

D 7π

3π 2

4

(b) We use the graph (I) to find the sign of r = sin 2θ : 0 ≤ θ ≤ π2 ⇒ r ≥ 0 ⇒ (r, θ ) is in the first quadrant. π ≤ θ ≤ π ⇒ r ≤ 0 ⇒ (r, θ ) is in the fourth quadrant. π ≤ θ ≤ 3π ⇒ r ≥ 0 ⇒ (r, θ ) is in the third quadrant. 2 2 3π ≤ θ ≤ 2π ⇒ r ≤ 0 ⇒ (r, θ ) is in the second quadrant. That is, 2 3π ≤ 2

y ≤ 2π

0≤

≤

π 2

x

π≤

≤

π ≤ 2

3π 2

≤π

x3 Sketch thethe curve r =r sin θ . First in the table -values below and plot the corresponding Plot cissoid = 23sin θ tan fill θ and show thatof itsrequation in rectangular is y 2 =points .of the coordinates , πx . Then curve. Notice that the three petals of the curve correspond to the angle intervals 0, π3 , π3 , 23π , and 2π3 − plot r = sin 3 θ in rectangular coordinates and label the points on this graph corresponding to (r, θ ) in the table. SOLUTION Using a CAS we obtain the following curve of the cissoid:

33.

θ r

0

π 12

π 6

π 4

π 3

5π 12

···

11π 12

π

S E C T I O N 12.3

y

Polar Coordinates

707

π 2

π

1

2

3 x

0π

3π 2

We substitute sin θ = ry and tan θ = xy in r = 2 sin θ tan θ to obtain r =2

y y · . r x

Multiplying by r x, setting r 2 = x 2 + y 2 and simplifying, yields r 2 x = 2y 2 (x 2 + y 2 )x = 2y 2 x 3 + y 2 x = 2y 2 y 2 (2 − x) = x 3 so y2 =

x3 2−x

35. Show that r = a cos θ + b sin θ is the equation of a circle passing through the origin. Express the radius and center Prove that r = 2a cos θ is the equation of the circle in Figure 22 using only the fact that a triangle inscribed in a (in rectangular coordinates) in terms of a and b. circle with one side a diameter is a right triangle. SOLUTION We multiply the equation by r and then make the substitution x = r cos θ , y = r sin θ , and r 2 = x 2 + y 2 . This gives r 2 = ar cos θ + br sin θ x 2 + y 2 = ax + by Transferring sides and completing the square yields x 2 − ax + y 2 − by = 0 2 2

a 2 a b b b a 2 2 2 = + + y −2· y+ x −2· x + 2 2 2 2 2 2

x−

a 2 b 2 a 2 + b2 + y− = 2 2 4

√

2 2 This is the equation of the circle with radius a 2+b centered at the point a2 , b2 . By plugging in x = 0 and y = 0 it is clear that the circle passes through the origin. 2 θ − sin2 θ to find a polar equation of the hyperbola x 2 − y 2 = 1. θ = costo 37. UseUse the the identity cos 2exercise previous write the equation of the circle of radius 5 and center (3, 4) in the form r = a cos θ + b sin θ . We substitute x = r cos θ , y = r sin θ in x 2 − y 2 = 1 to obtain SOLUTION

r 2 cos2 θ − r 2 sin2 θ = 1 r 2 (cos2 θ − sin2 θ ) = 1 Using the identity cos 2θ = cos2 θ − sin2 θ we obtain the following equation of the hyperbola: r 2 cos 2θ = 1

or r 2 = sec 2θ .

39. Show that cos 3θ = cos3 θ − 3 cos θ sin2 θ and use this identity 2 to find an equation in rectangular coordinates for Find equation the curve r =ancos 3θ . in rectangular coordinates for the curve r = cos 2θ .

708

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S SOLUTION We use the identities cos(α + β ) = cos α cos β − sin α sin β , cos 2α = cos2 α − sin2 α , and sin 2α = 2 sin α cos α to write

cos 3θ = cos(2θ + θ ) = cos 2θ cos θ − sin 2θ sin θ = (cos2 θ − sin2 θ ) cos θ − 2 sin θ cos θ sin θ = cos3 θ − sin2 θ cos θ − 2 sin2 θ cos θ = cos3 θ − 3 sin2 θ cos θ Using this identity we may rewrite the equation r = cos 3θ as follows: r = cos3 θ − 3 sin2 θ cos θ

(1)

Since x = r cos θ and y = r sin θ , we have cos θ = rx and sin θ = ry . Substituting into (1) gives: r=

x 3 r x3

y 2 x

−3

r

r

3y 2 x

r= 3 − 3 r r

We now multiply by r 3 and make the substitution r 2 = x 2 + y 2 to obtain the following equation for the curve: r 4 = x 3 − 3y 2 x 2 (x 2 + y 2 ) = x 3 − 3y 2 x

In Exercises 41–45, find an equation coordinates of the with given description. Use the addition formula for in thepolar cosine to show that theline lineL L with polar equation r cos(θ − α ) = d has the d.πShow that L has slope m = − cot α and y-intercept equation in rectangular coordinates (cos α )x + (sin α )y = 41. The d point on L closest to the origin has polar coordinates 2, 9 . . sin α SOLUTION In Example 5, it is shown that the polar equation of the line where (r, α ) is the point on the line closest to the origin is r = d sec (θ − α ). Setting (d, α ) = 2, π9 we obtain the following equation of the line:

π . r = 2 sec θ − 9 √ 43. L isThe tangent = the 2 origin 10 at the with rectangular coordinates pointtoonthe L circle closestr to haspoint rectangular coordinates (−2, 2). (−2, −6). SOLUTION y

x

(−2, −6)

Since L is tangent to the circle at the point (−2, −6), this is the point on L closest to the center of the circle which is at the origin. Therefore, we may use the polar coordinates (d, α ) of this point in the equation of the line: r = d sec (θ − α )

(1)

We thus must convert the coordinates (−2, −6) to polar coordinates. This point is in the third quadrant so π < α < 32π . We get

√ √ d = (−2)2 + (−6)2 = 40 = 2 10 −6 α = tan−1 = tan−1 3 ≈ π + 1.25 ≈ 4.39 −2 Substituting in (1) yields the following equation of the line: √ r = 2 10 sec (θ − 4.39) . L has slope 3 and is tangent to the unit circle in the fourth quadrant.

S E C T I O N 12.3

Polar Coordinates

709

45. y = 4x − 9. SOLUTION

Substituting y = r sin θ and x = r cos θ in y = 4x − 9, gives r sin θ = 4r cos θ − 9 4r cos θ − r sin θ = 9 r (4 cos θ − sin θ ) = 9

so r=

9 4 cos θ − sin θ

47. Distance Formula Use the Law of Cosines (Figure 23) to show that the distance d between two points with polar Show that the polar equation of the line y = ax + b can be written in the form coordinates (r, θ ) and (r0 , θ0 ) is b r= d2 = r 2 + r02 − 2rr0 cos(θ − θ0 ) 2 sin θ − a cos θ y

(r, q ) d r q

r0 q0

(r0, q 0)

x

FIGURE 23 SOLUTION

Note that the angle between the line segments r and r0 has measurement θ − θ0 . Thus, by the Law of

Cosines, d 2 = r 2 + r02 − 2rr0 cos (θ − θ0 ) 49. Show that the cardiod r = 1 + sin θ has equation Use the distance formula (2) to show that the circle of radius 9 whose center has polar coordinates 5, π4 has equation: x 2 + y 2 = (x 2 + y 2 − x)2

π θ and multiply by r . This gives r = r 2 − r sin θ . SOLUTION We write the equation of the cardioid in 10r the form 56 cos 1θ = − r sin= r2 − 4 We now make the substitution r = x 2 + y 2 and r sin θ = y to obtain

x 2 + y 2 = x 2 + y 2 − y. Finally, we square both sides to obtain

2 x 2 + y2 = x 2 + y2 − y . 51. The Derivative in Polar Coordinates A polar curve r = f (θ ) has parametric equations (since x = r cos θ and For a > 0, a lemniscate curve is the set of points P such that the product of the distances from P to (a, 0) and y = r sin θ ): 2 (−a, 0) is a . Show that the equation of the lemniscate is: x = f (θ ) cos θ , y = f (θ ) sin θ (x 2 + y 2 )2 = 2a 2 (x 2 − y 2 ) Apply Theorem 1 of Section 12.1 to prove the formula Then find the equation in polar coordinates. To obtain the simplest form of the equation, use the identity cos 2θ = cos2 θ − sin2 θ . Plot the lemniscate forday = 2 iff (you have a computer θ ) cos θ+ f (θ ) sin θalgebra system. 3 = dx − f (θ ) sin θ + f (θ ) cos θ where f (θ ) = d f /d θ . SOLUTION

By the formula for the derivative we have y (θ ) dy = dx x (θ )

(1)

We differentiate the functions x = f (θ ) cos θ and y = f (θ ) sin θ using the Product Rule for differentiation. This gives y (θ ) = f (θ ) sin θ + f (θ ) cos θ x (θ ) = f (θ ) cos θ − f (θ ) sin θ

710

C H A P T E R 12


Substituting in (1) gives f (θ ) cos θ + f (θ ) sin θ f (θ ) sin θ + f (θ ) cos θ dy = . = dx f (θ ) cos θ − f (θ ) sin θ − f (θ ) sin θ + f (θ ) cos θ 53. Find the equation in rectangular coordinates of the tangent line to r π= 4 cos 3θ at θ = π6 . Use Eq. (3) to find the slope of the tangent line to r = θ at θ = 2 and θ = π . SOLUTION We have f (θ ) = 4 cos 3θ . By Eq. (3), m=

4 cos 3θ cos θ − 12 sin 3θ sin θ . −4 cos 3θ sin θ − 12 sin 3θ cos θ

Setting θ = π6 yields

4 cos π2 cos π6 − 12 sin π2 sin π6 −12 sin π6 π 1 π π = π π m= = tan = √ . π 6 −12 cos 6 −4 cos 2 sin 6 − 12 sin 2 cos 6 3

We identify the point of tangency. For θ = π6 we have r = 4 cos 36π = 4 cos π2 = 0. The point of tangency is the origin. The tangent line is the line through the origin with slope √1 . This is the line y = √x . 3

3

Show that for the circle r = sin θ + cos θ , Further Insights and Challenges 55. (a) (b) (c) (d)

dy cos 2θ + sin 2θ Let c be a fixed constant. Explain the relationship between the graphs of: = dx cos 2θ − sin 2θ y = f (x + c) and y = f (x) (rectangular) c)slopes and r of = the f (θtangent ) (polar)lines at points A, B, C in Figure 20 and find the polar coordinates of the points at rCalculate = f (θ +the which the tangent line is horizontal. y = f (x) + c and y = f (x) (rectangular) r = f (θ ) + c and r = f (θ ) (polar)

SOLUTION

(a) For c > 0, y = f (x + c) shifts the graph of y = f (x) by c units to the left. If c < 0, the result is a shift to the right. It is a horizontal translation. y f(x + c)

f(x) c x

(b) As in part (a), the graph of r = f (θ + c) is a shift of the graph of r = f (θ ) by c units in θ . Thus, the graph in polar coordinates is rotated by angle c as shown in the following figure: p 2

f(q ) p

0 c f(q + c)

3p 2

(c) y = f (x) + c shifts the graph vertically upward by c units if c > 0, and downward by (−c) units if c < 0. It is a vertical translation. (d) The graph of r = f (θ ) + c is a shift of the graph of r = f (θ ) by c units in r . In the corresponding graph, in polar coordinates, each point with f (θ ) > 0 moves on the ray connecting it to the origin c units away from the origin if c > 0 and (−c) units toward the origin if c < 0, and vice-versa for f (θ ) < 0.

Area and Arc Length in Polar Coordinates

S E C T I O N 12.4

711

p 2

y c

c

p

1

c

0

c x

c

3p 2

c>0 Use a graphing utility to convince yourself that graphs of the polar equations r = f 1 (θ ) = 2 cos θ − 1 and 57. of period π , that why. is, f (x) f (x + ). Explain periodicity is r = f 2 (θ )Let = 2f (x) cos θbe+a1periodic have thefunction same graph. Then 2explain Hint:=Show that2πthe points ( how f 1 (θ this + π ), θ + π ) and in the graph of: ( f 2 (reflected θ ), θ ) coincide. (a) y = f (x) in rectangular coordinates SOLUTION The graphs of r = 2 cos θ − 1 and r = 2 cos θ + 1 in the x y -plane coincide as shown in the graph obtained (b) r = f (θ ) in polar coordinates using a CAS. y

π 2

y

2 π −2

2

1

2

3 x

0π

x

−2 3π 2

Recall that (r, θ ) and (−r, θ + π ) represent the same point. Replacing θ by θ + π and r by (−r ) in r = 2 cos θ − 1 we obtain −r = 2 cos (θ + π ) − 1 −r = −2 cos θ − 1 r = 2 cos θ + 1 Thus, the two equations define the same graph. (One could also convert both equations to rectangular coordinates and note that they come out identical.) Plot the limaçon curves r = a + cos θ for several values of a. (a) Describe how the shape changes as a increases. What happens for a < 0? 12.4(b) Area and Arc Length in Polar Coordinates A closed curve is called convex if, for any two points P and Q in the interior of the curve, the segment P Q is also contained in the interior. Figure 24 shows that the limaçon is convex if a is large and is not convex if a is small. Preliminary Experiment Questions with a computer algebra system to estimate the smallest value a for which it is convex. 1. True or False: under51the withifpolar equation r= f (θ ) is equal to the θintegral (θ ).θ = −a/2. = 0, πofor fcos (c) Use Eq. (3)The of area Exercise to curve show that the tangent line is vertical, then either SOLUTION is false. Consider circle lines r = to 1. the Its left areaofisthe π , yet theifintegral of21and from 0 to 2πvertical is 2π . (d) ShowThe thatstatement the limaçon has three verticalthe tangent y-axis 1 2?coordinates are best suited to finding the area (choose one): 2. Polar (a) Under a curve between x = a and x = b. (b) Bounded by a curve and two rays through the origin. SOLUTION Polar coordinates are best suited to finding the area bounded by a curve and two rays through the origin. The formula for the area in polar coordinates gives the area of this region.

3. True or False: The formula for area in polar coordinates is valid only if f (θ ) ≥ 0. SOLUTION

The statement is false. The formula for the area " 1 β f (θ )2 d θ 2 α

always gives the actual (positive) area, even if f (θ ) takes on negative values.

712

C H A P T E R 12


4. The horizontal line y = 1 has polar equation r = csc θ . Which area is represented by the integral (Figure 13)? (a) ABCD

(b) ABC

" 1 π /2 2 csc θ d θ 2 π /6

(c) ACD

y D 1

C y=1

A

B

x

2

FIGURE 13 SOLUTION This integral represents an area taken from θ = π /6 to θ = π /2, which can only be the triangle ACD, as seen in part (c).

Exercises

1. Sketch the area bounded by the circle r = 5 and the rays θ = π2 and θ = π , and compute its area as an integral in polar coordinates.

SOLUTION The region bounded by the circle r = 5 and the rays θ = π 2 and θ = π is the shaded region in the figure. The area of the region is given by the following integral: " " 1 π 2 25

π 25π 1 π 2 = r dθ = 5 dθ = π− 2 π /2 2 π /2 2 2 4 π

= 2 y

=π

x

3. Calculate the area of the circle r = 4 sin θ as an integral in polar coordinates (see Figure 4). Be careful to choose the Sketchofthe region bounded by the line r = sec θ and the rays θ = 0 and θ = π3 . Compute its area in two ways: correct limits integration. as an integral in polar coordinates and using geometry. SOLUTION The equation r = 4 sin θ defines a circle of radius 2 tangent to the x-axis at the origin as shown in the figure: π

= 2 y 2π 3

=π

5π 6

π 3

π 6

x

=π

The circle is traced as θ varies from 0 to π . We use the area in polar coordinates and the identity sin2 θ =

1 (1 − cos 2θ ) 2

to obtain the following area: " " " π " π 1 π sin 2θ π 1 π 2 r dθ = sin2 θ d θ = 4 A= (4 sin θ )2 d θ = 8 (1 − cos 2θ ) d θ = 4 θ − 2 0 2 0 2 0 0 0 sin 2π =4 π− − 0 = 4π . 2 Compute the area of the shaded region in Figure 14 as an integral in polar coordinates.


S E C T I O N 12.4

713

5. Find the total area enclosed by the cardioid r = 1 − cos θ (Figure 15). y

−2

x

−1

FIGURE 15 The cardioid r = 1 − cos θ . SOLUTION

We graph r = 1 − cos θ in r and θ (cartesian, not polar, this time): r 2

1

π 2

π

3π 2

2π

We see that as θ varies from 0 to π , the radius r increases from 0 to 2, so we get the upper half of the cardioid (the lower half is obtained as θ varies from π to 2π and consequently r decreases from 2 to 0). Since the cardioid is symmetric with respect to the x-axis we may compute the upper area and double the result. Using cos2 θ =

cos 2θ + 1 2

we get " π " π

" 1 π 2 1 − 2 cos θ + cos2 θ d θ r dθ = (1 − cos θ )2 d θ = 2 0 0 0 " π " π cos 2θ + 1 1 3 − 2 cos θ + cos 2θ d θ = 1 − 2 cos θ + dθ = 2 2 2 0 0 π 3 1 3π = θ − 2 sin θ + sin 2θ = 2 4 2 0

A =2·

The total area enclosed by the cardioid is A = 32π . 7. Find the area of one leaf of the “four-petaled rose” r = sin 2θ (Figure 16). Find the area of the shaded region in Figure 15. SOLUTION We consider the graph of r = sin 2θ in cartesian and in polar coordinates: y

r

r = 1,

1

π 4

A

A π 4

=

π 2

3π 4

x

π

−1

We see that as θ varies from 0 to π4 the radius r is increasing from 0 to 1, and when θ varies from π4 to π2 , r is decreasing back to zero. Hence, the leaf in the first quadrant is traced as θ varies from 0 to π2 . The area of the leaf (the four leaves have equal areas) is thus A=

" " 1 π /2 2 1 π /2 2 r dθ = sin 2θ d θ . 2 0 2 0

Using the identity sin2 2θ =

1 − cos 4θ 2

714

C H A P T E R 12


we get A=

" 1 π /2 1 cos 4θ 1 θ 1 sin 4θ π /2 sin 2π π π = − dθ = − − − 0 = 2 0 2 2 2 2 8 2 4 8 8 0

The area of one leaf is A = π8 ≈ 0.39. 2θ (Figure your limits of 9. FindProve the area one of the lemniscate equation r 2 = cos that enclosed the total by area ofloop the four-petaled rose rwith = sin 2θ is equal to one-half the 17). areaChoose of the circumscribed integration carefully. circle (Figure 16). y

−1

x

1

FIGURE 17 The lemniscate r 2 = cos 2θ .

SOLUTION

We sketch the graph of r 2 = cos 2θ in the r 2 , θ plane; for − π4 ≤ θ ≤ π4 : r2 1

−π

π 1 4

4

We see that as θ varies from − π4 to 0, r 2 increases from 0 to 1, hence r also increases from 0 to 1. Then, as θ varies from 0 to π4 , r 2 , so r decreases from 1 to 0. This gives the right-hand loop of the lemniscate. y

r = 0, r =

π 4

π 4

= r=1 r = 0, r =

x

π 4

We compute the area enclosed by the right-hand loop, by the following integral: " "

π 1 1 π /4 2 1 π /4 1 sin 2θ π /4 1 π r dθ = cos 2θ d θ = = sin − sin − = 2 −π /4 2 −π /4 2 2 −π /4 4 2 2 2 Since the lemniscate is symmetric with respect to the y-axis, the total area A enclosed by the lemniscate is twice the area enclosed by the right-hand loop. Thus, A =2·

1 = 1. 2

11. Find the area enclosed by the cardioid r = a(1 + cos θ ), where a > 0. Sketch the spiral r = θ for 0 ≤ θ ≤ 2π and find the area bounded by the curve and the first quadrant. SOLUTION The graph of r = a (1 + cos θ ) in the r θ -plane for 0 ≤ θ ≤ 2π and the cardioid in the x y-plane are shown in the following figures: y r

= 2a

π 2

,r=a

= π, r = 0

a

π 2

π

3π 2

r = a (1 + cos θ )

= 3π ,r=a 2

= 0, r = 2a

x

2π

The cardioid r = a (1 + cos θ ), a > 0


S E C T I O N 12.4

715

As θ varies from 0 to π the radius r decreases from 2a to 0, and this gives the upper part of the cardioid. The lower part is traced as θ varies from π to 2π and consequently r increases from 0 back to 2a. We compute the area enclosed by the upper part of the cardioid and the x-axis, using the following integral (we use the identity cos2 θ = 12 + 12 cos 2θ ): " " " 1 π 2 1 π 2 a2 π

r dθ = a (1 + cos θ )2 d θ = 1 + 2 cos θ + cos2 θ d θ 2 0 2 0 2 0 " " a2 π 1 1 a2 π 3 1 = + 2 cos θ + cos 2θ d θ 1 + 2 cos θ + + cos 2θ d θ = 2 0 2 2 2 0 2 2 π 2 2 a 3θ 1 a 1 3π a 2 3π = + 2 sin θ + sin 2θ = + 2 sin π + sin 2π − 0 = 2 2 4 2 2 4 4 0 Using symmetry, the total area A enclosed by the cardioid is A=2·

3π a 2 3π a 2 = 4 2

13. Find the area of region A in Figure 18. Find the area of the intersection of the circles r = sin θ and r = cos θ . y

r = 4 cos

r=1

A

−1

1

2

4

x

FIGURE 18 SOLUTION We first find the values of θ at the points of intersection of the two circles, by solving the following equation for − π2 ≤ x ≤ π2 : 1 1 4 cos θ = 1 ⇒ cos θ = ⇒ θ1 = cos−1 4 4 y

= 1.32 r = 4 cos

r=1 x

= −1.32

We now compute the area using the formula for the area between two curves: A=

" " 1 θ1

1 θ1

16 cos2 θ − 1 d θ (4 cos θ )2 − 12 d θ = 2 −θ1 2 −θ1

Using the identity cos2 θ = cos 22θ +1 we get θ1 " " 1 θ1 1 1 θ1 16 (cos 2θ + 1) − 1 dθ = A= (8 cos 2θ + 7) d θ = (4 sin 2θ + 7θ ) 2 −θ1 2 2 −θ1 2 −θ1

= 4 sin 2θ1 + 7θ1 = 8 sin θ1 cos θ1 + 7θ1 = 8 1 − cos2 θ1 cos θ1 + 7θ1 Using the fact that cos θ1 = 14 we get

√ A=

15 + 7cos−1 2

1 ≈ 11.163 4

15. Find the area of the inner loop of the limaçon with polar equation r = 2 cos θ − 1 (Figure 20). Find the area of the shaded region in Figure 19, enclosed by the circle r = 12 andπ a petal ofπ the curve r = cos 3θ . SOLUTION We consider the graph of r = 2 cos θ − 1 in cartesian and in polar, for − ≤x≤ : Hint: Compute the area of both the petal and the region inside the petal and outside2the circle.2

716

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S y π 3

r 1

x π − 2

π − 3

π 3

π 2

−1

−

π 3

r = 2 cos θ − 1 As θ varies from − π3 to 0, r increases from 0 to 1. As θ varies from 0 to π3 , r decreases from 1 back to 0. Hence, the inner loop of the limaçon is traced as θ varies from − π3 to π3 . The area of the shaded region is thus " " " 1 π /3 1 π /3

1 π /3 2 4 cos2 θ − 4 cos θ + 1 d θ r dθ = (2 cos θ − 1)2 d θ = 2 −π /3 2 −π /3 2 −π /3 " π /3 " 1 1 π /3 = (2 (cos 2θ + 1) − 4 cos θ + 1) d θ = (2 cos 2θ − 4 cos θ + 3) d θ 2 −π /3 2 −π /3 π /3

π 1 1 2π π 2π −π = (sin 2θ − 4 sin θ + 3θ ) = sin − 4 sin + π − sin − − 4 sin − 2 2 3 3 3 3 −π /3 √ √ √ 3 3 3 4 3 = − +π =π − ≈ 0.54 2 2 2

A=

17. Find the area of the part of the circle r = sin θ + cos θ in the fourth quadrant (see Exercise 27 in Section 12.3). Find the area of the region between the inner and outer loop of the limaçon r = 2 cos θ − 1. SOLUTION The value of θ corresponding to the point B is the solution of r = sin θ + cos θ = 0 for −π ≤ θ ≤ π . y

B

A

C

x

That is, sin θ + cos θ = 0 ⇒ sin θ = − cos θ ⇒ tan θ = −1 ⇒ θ = −

π 4

At the point C, we have θ = 0. The part of the circle in the fourth quadrant is traced if θ varies between − π4 and 0. This leads to the following area: A=

" " " 1 0 1 0 2 1 0 r 2 dθ = sin θ + 2 sin θ cos θ + cos2 θ d θ (sin θ + cos θ )2 d θ = 2 −π /4 2 −π /4 2 −π /4

Using the identities sin2 θ + cos2 θ = 1 and 2 sin θ cos θ = sin 2θ we get: " 1 cos 2θ 0 1 0 θ− A= (1 + sin 2θ ) d θ = 2 −π /4 2 2 −π /4 −π cos 2 1 π π 1 π 1 1 1 = − − − = − = − ≈ 0.14. 0− 2 2 4 2 2 4 2 8 4 19. Find the area between the two curves in Figure 22(A). Compute the area of the shaded region in Figure 21. SOLUTION We compute the area A between the two curves as the difference between the area A1 of the region enclosed in the outer curve r = 2 + cos 2θ and the area A2 of the region enclosed in the inner curve r = sin 2θ . That is, A = A1 − A2 .


S E C T I O N 12.4

717

y r = 2 + 2cos r = sin

A A2

x

In Exercise 8 we showed that A2 = π2 , hence, A = A1 −

π 2

(1)

We compute the area A1 . y

A1 x

Using symmetry, the area is four times the area enclosed in the first quadrant. That is, A1 = 4 ·

" " π /2 " π /2

1 π /2 2 4 + 4 cos 2θ + cos2 2θ d θ r dθ = 2 (2 + cos 2θ )2 d θ = 2 2 0 0 0

Using the identity cos2 2θ = 12 cos 4θ + 12 we get " π /2 " π /2 1 1 9 1 A1 = 2 dθ = 2 + cos 4θ + 4 cos 2θ d θ 4 + 4 cos 2θ + cos 4θ + 2 2 2 2 0 0 π /2 9θ 9π sin 4θ sin 2π 9π =2 =2 + + 2 sin 2θ + + 2 sin π − 0 = 2 8 4 8 2 0

(2)

Combining (1) and (2) we obtain A=

π 9π − = 4π . 2 2

21. Find the area inside both curves in Figure 23. Find the area between the two curves in Figure 22(B). SOLUTION The area we need to find is the area of the shaded region in the figure. y

r = 2 + sin 2

D A x C B

r = 2 + cos 2

We first find the values of θ at the points of intersection A, B, C, and D of the two curves, by solving the following equation for −π ≤ θ ≤ π : 2 + cos 2θ = 2 + sin 2θ cos 2θ = sin 2θ tan 2θ = 1 ⇒ 2θ =

π π πk + πk ⇒ θ = + 4 8 2

The solutions for −π ≤ θ ≤ π are A:

θ=

π . 8

718

C H A P T E R 12


3π . 8 7π θ=− . 8 5π θ= . 8

B: θ = − C: D:

Using symmetry, we compute the shaded area in the figure below and multiply it by 4: π 2 5π 8

π 8

A1 π

0π

r = 2 + cos 2 −

π 2

" 5π /8

" 5π /8 1 4 + 4 cos 2θ + cos2 2θ d θ · (2 + cos 2θ )2 d θ = 2 2 π /8 π /8 " 5π /8 " 5π /8 1 + cos 4θ =2 dθ = 4 + 4 cos 2θ + (9 + 8 cos 2θ + cos 4θ ) d θ 2 π /8 π /8 √ sin 4θ 5π /8 5π π π π 5π 1 5π 9π = 9θ + 4 sin 2θ + = 9 − + 4 sin − sin + sin − sin = −4 2 4 π /8 8 8 4 4 4 2 2 2

A = 4 · A1 = 4 ·

23. Figure 24 suggests that the circle r = sin θ lies inside the spiral r = θ . Which inequality from Chapter 2 assures us Find the area of the region that lies inside one but not both of the curves in Figure 23. that this is the case? Find the area between the curves r = θ and r = sin θ in the first quadrant. y r= 1 r = sin x

FIGURE 24 SOLUTION The inequality |sin θ | ≤ |θ | assures us that for all θ , the corresponding point on the spiral has radial coordinate with absolute value greater than that of the corresponding point on the circle. Hence, the circle lies inside the spiral. y

x

The area between the circle and the spiral in the first quadrant is traced as θ varies from 0 to π2 . y r=

r = sin x

Using the formula for the area between two curves and the identity sin2 θ = 12 − 12 cos 2θ we obtain " " 1 π /2 2 1 π /2 2 1 1 A= θ − sin2 θ d θ = θ − + cos 2θ d θ 2 0 2 0 2 2


S E C T I O N 12.4

1 = 2

719

π /2 θ3 θ π π3 π 1 sin π 1 π3 − + sin 2θ − + −0 = − ≈ 0.25 = 3 2 4 2 24 4 4 48 8 0

25. Find the length of the spiral r = θ for 0 ≤ θ ≤ A. Calculate the total length of the circle r = 4 sin θ as an integral in polar coordinates. SOLUTION We use the formula for the arc length. In this case f (θ ) = θ , f (θ ) = 1. Using integration formulas we get: A " A " A 1 θ 2 2 2 2 2 S= θ + 1 dθ = θ + 1 dθ = θ + 1 + ln |θ + θ + 1| 2 2 0 0 0 A 2 1 = A + 1 + ln | A + A2 + 1| 2 2 y

x

The spiral r = θ 27. Sketch the segment r = sec θ for 0 ≤ θ ≤ A. Then compute its length in two ways: as an integral in polar Findand the using lengthtrigonometry. of r = θ 2 for 0 ≤ θ ≤ π . coordinates SOLUTION

The line r = sec θ has the rectangular equation x = 1. The segment AB for 0 ≤ θ ≤ A is shown in the

figure. y C sec A

A

D 1

x

Using trigonometry, the length of the segment AB is L = AB = 0B tan A = 1 · tan A = tan A Alternatively, we use the integral in polar coordinates with f (θ ) = sec(θ ) and f (θ ) = tan θ sec θ . This gives L=

A " A " A " A 1 + tan2 θ sec θ d θ = sec2 θ d θ = tan θ = tan A. (sec θ )2 + (tan θ sec θ )2 d θ = 0

0

0

0

The two answers agree, as expected.

θ Then in two ways: asθan=integral coordinates the circle r= sin θ +rcos θ . Hint: its Uselength the identity 1 + cos 2 cos2 in polar 29. FindSketch the length of the cardioid = θ1. + cos compute to evaluate the and arc 2 using trigonometry. length integral. SOLUTION

In the equation of the cardioid, f (θ ) = 1 + cos θ . Using the formula for arc length in polar coordinates

we have: L=

" β α

f (θ )2 + f (θ )2 d θ

(1)

We compute the integrand:

2 2 2 2 f (θ ) + f (θ ) = (1 + cos θ ) + (− sin θ ) = 1 + 2 cos θ + cos2 θ + sin2 θ = 2 (1 + cos θ ) We identify the interval of θ . Since −1 ≤ cos θ ≤ 1, every 0 ≤ θ ≤ 2π corresponds to nonnegative value of r . Hence, θ varies from 0 to 2π . By (1) we obtain L=

" 2π 2(1 + cos θ ) d θ 0

720

C H A P T E R 12


Since 1 + cos θ = 2 cos2 (θ /2), and by the symmetry of the graph, " π " π θ π L =2 2(1 + cos θ ) d θ = 2 2 cos(θ /2) d θ = 8 sin = 8 2 0 0 0 31. FindFind the the length of the equiangular spiral r = eθ rfor ≤θ2π . located in the first quadrant. length of the cardioid with equation = 01 ≤ + θcos θ SOLUTION Since f (θ ) = e , by the formula for the arc length we have: " 2π " 2π " 2π 2 2 f (θ )2 + f ( θ ) d θ + 2e2θ d θ eθ + eθ d θ = L= 0

0

0

2π √

√ " 2π θ √ √

= 2 e d θ = 2eθ = 2 e2π − e0 = 2 e2π − 1 ≈ 755.9 0

0

In Exercises 33–36, express the length of the2 curve as an integral but do not evaluate it. Find the length of the curve r = cos θ . 33. r = ea θ , 0 ≤ θ ≤ π SOLUTION We use the formula for the arc length in polar coordinates. For this curve f (θ ) = ea θ , f (θ ) = aea θ , hence, " π " π " π L= f (θ )2 + f (θ )2 d θ = e2a θ + a 2 e2a θ d θ = 1 + a 2 ea θ d θ 0

0

0

−1 35. r =r (2 ≤π θ ≤ 2π =− sincos 2θθ, ) 0 ,≤ θ0 ≤ SOLUTION We have f (θ ) = (2 − cos θ )−1 , f (θ ) = −(2 − cos θ )−2 sin θ , hence,

2 −2 −4 2 2 f (θ ) + f (θ ) = (2 − cos θ ) + (2 − cos θ ) sin θ = (2 − cos θ )−4 (2 − cos θ )2 + sin2 θ

= (2 − cos θ )−2

√ 4 − 4 cos θ + cos2 θ + sin2 θ = (2 − cos θ )−2 5 − 4 cos θ

Using the integral for the arc length we get L=

" 2π √ 5 − 4 cos θ (2 − cos θ )−2 d θ . 0

The inner loop of r = 2 cos θ − 1 (see Exercise 15)


37. Suppose

that the polar coordinates of a moving particle at time t are (r (t), θ (t)). Prove that the particle’s speed is equal to (dr/dt)2 + r 2 (d θ /dt)2 . SOLUTION

The speed of the particle in rectangular coordinates is:

ds = x (t)2 + y (t)2 dt

(1)

We need to express the speed in polar coordinates. The x and y coordinates of the moving particles as functions of t are x(t) = r (t) cos θ (t),

y(t) = r (t) sin θ (t)

We differentiate x(t) and y(t), using the Product Rule for differentiation. We obtain (omitting the independent variable t) x = r cos θ − r (sin θ ) θ y = r sin θ − r (cos θ ) θ Hence, 2 2 2 2 x + y = r cos θ − r θ sin θ + r sin θ + r θ cos θ = r cos2 θ − 2r r θ cos θ sin θ + r 2 θ sin2 θ + r sin2 θ + 2r r θ sin2 θ cos θ + r 2 θ cos2 θ

2 2 2 2 = r cos2 θ + sin2 θ + r 2 θ sin2 θ + cos2 θ = r + r 2 θ 2

2

2

2

(2)

Substituting (2) into (1) we get ds = dt

r 2 + r 2θ 2 =

2 dr 2 dθ + r2 dt dt

Compute the speed at time t = 1 of a particle whose polar coordinates at time t are r = t, θ = t (use Exercise 37). What would the speed be if the particle’s rectangular coordinates are x = t, y = t? Why is the speed increasing

S E C T I O N 12.5

Conic Sections

721

12.5 Conic Sections Preliminary Questions 1. Which of the following equations defines an ellipse? Which does not define a conic section? (b) −4x + 9y 2 = 0 (a) 4x 2 − 9y 2 = 12 (c) 4y 2 + 9x 2 = 12

(d) 4x 3 + 9y 3 = 12

SOLUTION

(a) This is the equation of the hyperbola

2 √x − 3

2 y

= 1, which is a conic section.

√2 3

(b) The equation −4x + 9y 2 = 0 can be rewritten as x = 94 y 2 , which defines a parabola. This is a conic section. 2

2 + x2 = 1, hence it is the equation of an (c) The equation 4y 2 + 9x 2 = 12 can be rewritten in the form √y √ 3

3

ellipse, which is a conic section. (d) This is not the equation of a conic section, since it is not an equation of degree two in x and y. 2. For which conic sections do the vertices lie between the foci? SOLUTION

If the vertices lie between the foci, the conic section is a hyperbola. y y Vertex Focus Vertex Vertex Focus Vertex

Focus F1

Focus F2

Vertex

x

F1

F2

x

Vertex

ellipse: foci between vertices 3. What are the foci of

x 2 a

+

y 2 b

hyperbola: vertices between foci

= 1 if a < b?

2 2 If a < b the foci of the ellipse ax + by = 1 are at the points (0, c) and (0, −c) on the y-axis, where c = b2 − a 2 . SOLUTION

y b

F1 = (0, c)

a

x

F2 = (0, −c)

x 2 a

2 + by = 1; a < b

4. For a hyperbola in standard position, the set of points equidistant from the foci is the y-axis. Use the definition P F1 − P F2 = ±K to explain why the hyperbola does not intersect the y-axis. SOLUTION The points on the hyperbola are the point such that the difference of the distances to the two foci is ±k, for some constant k > 0. For the points on the y-axis, this difference is zero, hence they are not on the hyperbola. Therefore, the hyperbola does not intersect the y-axis.

5. What is the geometric interpretation of the quantity

b in the equation of a hyperbola in standard position? a

The vertices, i.e., the points where the focal axis intersects the hyperbola, are at the points (a, 0) and (−a, 0). The values ± ab are the slopes of the two asymptotes of the hyperbola. SOLUTION

722

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S b a

y=− x

y

y=

b x a

b

(−a, 0)

(a, 0)

x

−b

Hyperbola in standard position

Exercises In Exercises 1–8, find the vertices and foci of the conic section. 1.

x 2 9

+

y 2 4

=1

√ This is an ellipse in standard position with a = 9 and b = 4. Hence, c = 92 − 42 = 65 ≈ 8.06. The foci are at F1 = (−8.06, 0) and F2 = (8.06, 0), and the vertices are (9, 0) , (−9, 0), (0, 4) , (0, −4). SOLUTION

x2 y2 +x 2 = 1y 2 =1 9 4 + 4 9 x 2 y 2 SOLUTION Writing the equation in the from 3 + 2 = 1 we get an ellipse with a = 3 and b = 2. Hence √ c = 32 − 22 = 5 ≈ 2.24. The foci are at F1 = (−2.24, 0) and F2 = (2.24, 0) and the vertices are (3, 0), (−3, 0), (0, 2), (0, −2).

x 2 y 2

x− 2 y = 2 1 5. 4 −9 =1 9 4 SOLUTION For this hyperbola a = 4, b = 9 and c = 42 + 92 ≈ 9.85. The foci are at F1 = (9.85, 0) and F2 = (−9.85, 0) and the vertices are A = (4, 0) and A = (−4, 0). x −2 3 2 2 y + 1 2 y− x =1 7. =14 7 − 9 4 x 2 y 2 SOLUTION We first consider the hyperbola 7 − 4 = 1. For this hyperbola, a = 7, b = 4 and c = 72 + 42 ≈ 8.06. Hence, the foci are at (8.06, 0) and (−8.06, 0) and the vertices are at (7, 0) and (−7, 0). Since the given hyperbola 2 2 is obtained by translating the center of the hyperbola x7 − 4y = 1 to the point (3, −1), the foci are at F1 = (8.06 + 3, 0 − 1) = (11.06, −1) and F2 = (−8.06 + 3, 0 − 1) = (−5.06, −1) and the vertices are A = (7 + 3, 0 − 1) = (10, −1) and A = (−7 + 3, 0 − 1) = (−4, −1). 3.

consider ellipse 9–12, In Exercises the y+1 2 x −3 2 + =1 4 7

y−9 2 x − 12 2 + =1 5 7

Find the equation of the translated ellipse. 9. Translated so that its center is at the origin SOLUTION

Recall that the equation (x − h)2 (y − k)2 + =1 2 a b2

describes an ellipse with center (h, k). Thus, for our ellipse to be located at the origin, it must have equation y2 x2 + =1 52 72 11. Translated three units down Translated four units to the right

Conic Sections


723

Recall that the equation (y − k)2 (x − h)2 + =1 2 a b2

describes an ellipse with center (h, k). Thus, for our ellipse to be moved 3 units down, it must have equation (y − 6)2 (x − 12)2 + =1 2 5 72 In Exercises 13–16,sofind the equation the ellipse with the given properties. Translated its center is (−4,of−9) 13. Vertices at (±9, 0) and (0, ±16) x 2 y 2 SOLUTION The equation is a + b = 1 with a = 9 and b = 16. That is,

x 2

+

9

y 2 =1 16

15. Foci (0, ±6) and two vertices at (±4, 0) Foci (±6, 0) and two vertices at (±10, 2 y 2 x0) SOLUTION The equation of the ellipse is a + b = 1. The foci are (0, ±c) on the y-axis with c = 6, and two vertices are at (±a, 0) with a = 4. We use the relation c = b2 − a 2 to find b: b = a 2 + c2 = 42 + 62 ≈ 7.2 Therefore the equation is

x 2 4

+

y 2 = 1. 7.2

In Exercises 17–22, find the equation of the hyperbola with the given properties. Foci (0, ±3) and eccentricity 34 17. Vertices (±3, 0) and foci at (±5, 0) x 2 y 2 SOLUTION The equation is a − b = 1. The vertices are (±a, 0) with a = 3 and the foci (±c, 0) with c = 5. We use the relation c = a 2 + b2 to find b: √ b = c2 − a 2 = 52 − 32 = 16 = 4 Therefore, the equation of the hyperbola is

x 2 3

−

y 2 4

= 1.

19. Vertices (±4, 0) and asymptotes y = ±3x Vertices (0, ±5) and foci ±8) x (0, 2 2 SOLUTION The equation is a − by = 1. The vertices are at (±a, 0) with a = 4 and the asymptotes are y = ± ab x with ab = 3. Hence b = 3a = 3 · 4 = 12, and the equation of the hyperbola is

x 2 4

−

y 2 =1 12

21. Vertices (0, −5), (0, 4) and foci (0, −8), (0, 7) Vertices (±3, 0) and asymptotes y = ± 12 x

−5+4 = −0.5. Thus, the equation has the form y+0.5 2 − x 2 = 1. SOLUTION The center of the parabola is at a b 2 Since b = 4.5 and c = 7.5, we quickly find that a = 6, giving us the equation

y + 0.5 2 x 2 − = 1. 4.5 6

In Exercises 23–30, find the equation √ of the parabola with the given properties. Foci (0, ±3) and eccentricity 2 23. Vertex (0, 0), focus (2, 0) 2

SOLUTION 2

y Since the focus is on the x-axis rather than on the y-axis, the equation is x = 4c . Since c = 2 we get

x = y8 . Vertex (0, 0), focus (0, 12 )

724

C H A P T E R 12


25. Vertex (0, 0), directrix y = −5 SOLUTION

1 x 2 . The directrix is y = −c with c = 5, hence y = 1 x 2 . The equation is y = 4c 20

27. Focus (0, 4), directrix y = −4 Vertex (0, 0), directrix y = − 18 SOLUTION The focus is (0, c) with c = 4 and the directrix is y = −c with c = 4, hence the equation of the parabola is y=

1 2 x2 x = . 4c 16

29. Focus (2, 0), directrix x = −2 Focus (0, −4), directrix y = 4 SOLUTION The focus is on the x-axis rather than on the y-axis and the directrix is a vertical line rather than horizontal as in the parabola in standard position. Therefore, the equation of the parabola is obtained by interchanging x and y in 2

1 x 2 . Also, by the given information c = 2. Hence, x = 1 y 2 = 1 y 2 or x = y . y = 4c 4c 4·2 8

In Exercises finddirectrix the vertices, Focus31–40, (−2, 0), x = 2foci, axes, center (if an ellipse or a hyperbola) and asymptotes (if a hyperbola) of the conic section. 31. x 2 + 4y 2 = 16 SOLUTION

We first divide the equation by 16 to convert it to the equation in standard form:

x 2 y 2 4y 2 x2 y2 x2 + =1 + =1⇒ + =1⇒ 16 16 16 4 4 2 √ For this ellipse, a = 4 and b = 2 hence c = 42 − 22 = 12 ≈ 3.5. Since a > b we have: • The vertices are at (±4, 0), (0, ±2). • The foci are F1 = (−3.5, 0) and F2 = (3.5, 0). • The focal axis is the x-axis and the conjugate axis is the y-axis. • The ellipse is centered at the origin.

33. 4x 2+ y 2 = 16 y+5 2 x −3 2 − = 1 by 16 to rewrite it in the standard form: SOLUTION We divide the equation 16 49

x 2 y 2 y2 x2 y2 4x 2 + =1 + =1⇒ + =1⇒ 16 16 4 16 2 4 This is the equation of an ellipse with a = 2, b = 4. Since a < b the focal axis is the y-axis. Also, c = √ 12 ≈ 3.5. We get:

42 − 22 =

• The vertices are at (±2, 0), (0, ±4). • The foci are (0, ±3.5). • The focal axis is the y-axis and the conjugate axis is the x-axis. • The center is at the origin.

35. 4x 2 −23y 2 + 8x + 30y = 215 3x − 27y 2 = 12 SOLUTION Since there is no cross term, we complete the square of the terms involving x and y separately:

4x 2 − 3y 2 + 8x + 30y = 4 x 2 + 2x − 3 y 2 − 10y = 4(x + 1)2 − 4 − 3(y − 5)2 + 75 = 215 Hence, 4(x + 1)2 − 3(y − 5)2 = 144 3(y − 5)2 4(x + 1)2 − =1 144 144 2 2 y−5 x +1 − √ =1 6 48 2 2

This is the equation of the hyperbola obtained by translating the hyperbola x6 − √y = 1 one unit to the left and 48 √ √ √ five units upwards. Since a = 6, b = 48, we have c = 36 + 48 = 84 ∼ 9.2. We obtain the following table:

S E C T I O N 12.5

Standard position

Translated hyperbola

vertices

(6, 0), (−6, 0)

(5, 5), (−7, 5)

foci

(±9.2, 0)

(8.2, 5), (−10.2, 5)

focal axis

The x-axis

y=5

conjugate axis

The y-axis

x = −1

center

The origin

(−1, 5)

asymptotes

y = ±1.15x

y = −1.15x + 3.85 y = 1.15x + 6.15

Conic Sections

725

37. y = 4(x −24)2 y = 4x

1 SOLUTION By Exercise 36, the parabola y = 4x 2 has the vertex at the origin, the focus at 0, 16 and its axis is the y-axis.

Ourparabola is a translation of the standard parabola four units to the right. Hence its vertex is at (4, 0), the focus 1 and its axis is the vertical line x = 4. is at 4, 16 8x − 10y = 20 39. 4x 2 +225y 2 − 8y + 6x 2 − 36x − 64y + 134 = 0 SOLUTION Since there are no cross terms this conic section is obtained by translating a conic section in standard position. To identify the conic section we complete the square of the terms involving x and y separately:

2 4x 2 + 25y 2 − 8x − 10y = 4 x 2 − 2x + 25 y 2 − y 5 1 2 −1 = 4(x − 1)2 − 4 + 25 y − 5 1 2 2 = 4(x − 1) + 25 y − − 5 = 20 5 Hence, 1 2 = 25 l4(x − 1)2 + 25 y − 5 1 2 4 =1 (x − 1)2 + y − 25 5 2 x −1 1 2 + y − =1 5 5 2

2 This is the equation of the ellipse obtained by translating the ellipse in standard position

right and 15 unit upward. Since a = 52 , b = 1 we have c =

x

5 2

+ y 2 = 1 one unit to the

5 2 − 1 ≈ 2.3, so we obtain the following table: 2

Vertices

Standard position

± 52 , 0 , (0, ±1)

Foci

(−2.3, 0) , (2.3, 0)

Focal axis

The x-axis

Conjugate axis

The y-axis

Center

The origin

Translated ellipse

1 ± 52 , 15 , 1, 15 ± 1

−1.3, 15 , 3.3, 15 y = 15 x =1

1, 15

In Exercises 41–44, use the Discriminant Test to determine the type of the conic section defined by the equation. You may y 2 − 2x 2 + 8x − 9y − 12 = 0 assume that the equation is nondegenerate. Plot the curve if you have a computer algebra system. 41. 4x 2 + 5x y + 7y 2 = 24 SOLUTION

Here, D = 25 − 4 · 4 · 7 = −87, so the conic section is an ellipse.

43. 2x 2 −28x y − 3y 2 −24 = 0 2x − 8x y + 3y − 4 = 0

726

C H A P T E R 12


Here, D = 64 − 4 · 2 · (−3) = 88, giving us a hyperbola. b e2 for standard ellipse of eccentricity e. 45. Show x 2 that − 2xay = + y 21+−24x − 8a = 0 SOLUTION

SOLUTION

By the definition of eccentricity: e=

c a

(1)

a 2 − b2 . Substituting into (1) and simplifying yields 2 b a 2 − b2 a 2 − b2 = 1− = e= a a a2

For the ellipse in standard position, c =

We square the two sides and solve for ab : e2 = 1 −

2 2 b b b ⇒ = 1 − e2 ⇒ = 1 − e2 a a a

47. Explain why the dots in Figure 22 lie on a parabola. Where are the focus and directrix located? Show that the eccentricity of a hyperbola in standard position is e = 1 + m 2 , where ±m are the slopes of the y asymptotes. y = 3c y = 2c y=c x y = −c

FIGURE 22 SOLUTION All the circles are centered at (0, c) and the kth circle has radius kc. Hence the indicated point Pk on the kth circle has a distance kc from the point F = (0, c). The point Pk also has distance kc from the line y = −c. That is, the indicated point on each circle is equidistant from the point F = (0, c) and the line y = −c, hence it lies on the parabola with focus at F = (0, c) and directrix y = −c. y P3 3c P 2 2c

(0, c) P1

3c x

2c y = −c

49. Kepler’s First Law states that the orbits of the planets around the with the sun at one focus. The orbit

xsun 2 are ellipses y 2 of PlutoShow has anthat eccentricity of approximately = to 0.25 the perihelion−(closest = distance to the(x sun) orbit is the equation of the tangent eline theand hyperbola 1 at a point is 0 , yof 0 ) Pluto’s approximately 2.7 billion miles. Find the aphelion (farthest distance afrom thebsun). By is = an 1 ellipse in standard position, as shown in the figure. SOLUTION We define an x y-coordinate system so thatAx the−orbit x y where A = 02 and B = 02 . a b

y

A'(−a, 0)

Sun F1(c, 0)

A(a, 0)

The aphelion is the length of A F1 , that is a + c. By the given data, we have c ⇒ c = 0.25a a a − c = 2.7 ⇒ c = a − 2.7

0.25 = e =

x

S E C T I O N 12.5

Conic Sections

727

Equating the two expressions for c we get 0.25a = a − 2.7 0.75a = 2.7 ⇒ a =

2.7 = 3.6, c = 3.6 − 2.7 = 0.9 0.75

The aphelion is thus A F0 = a + c = 3.6 + 0.9 = 4.5 billion miles. In Exercises 51–54, find the polar equation of the conic with given eccentricity and directrix. Kepler’s Third Law states that the ratio T /a 3/2 is equal to a constant C for all planetary orbits around the sun, the3period (time for a complete orbit) and a is the semimajor axis. x= 51. ewhere = 12 ,T is (a) Compute C in units of 1days and kilometers, given that the semimajor axis of the earth’s orbit is 150 × 106 km. SOLUTION Substituting e = 2 and d = 3 in the polar equation of a conic section we obtain (b) Compute the period of Saturn’s orbit, given that its semimajor axis is approximately 1.43 × 109 km. 1 · 3the perihelion and aphelion of Saturn. (c) Saturn’s orbit has eccentricity 3 3 ede = 0.056. Find 2 = = ⇒r = r= 1 + e cos θ 2 + cos θ 2 + cos θ 1 + 1 cos θ 2

53. e = 1, e=

x =4

1 , x = −3 SOLUTION 2 We substitute e = 1 and d = 4 in the polar equation of a conic section to obtain

r=

1·4 4 4 ed = = ⇒r = 1 + e cos θ 1 + 1 · cos θ 1 + cos θ 1 + cos θ

In Exercises 355–58, identify the type of conic, the eccentricity, and the equation of the directrix. e = 2 , x = −4 8 55. r = 1 + 4 cos θ Matching with the polar equation r = 1+eedcos θ we get ed = 8 and e = 4 yielding d = 2. Since e > 1, the conic section is a hyperbola, having eccentricity e = 4 and directrix x = 2 (referring to the focus-directrix definition (11)). SOLUTION

8 57. r = 8 r 4=+ 3 cos θ 4 + cos θ ed SOLUTION We first rewrite the equation in the form r = 1+e cos θ , obtaining r=

2 1 + 34 cos θ

Hence, ed = 2 and e = 34 yielding d = 83 . Since e < 1, the conic section is an ellipse, having eccentricity e = 34 and directrix x = 83 . 59. Show 12that r = f 1 (θ ) and r = f 2 (θ ) define the same curves in polar coordinates if f 1 (θ ) = − f 2 (θ + π ), = to show that the following define the same conic section: and use rthis 4 + 3 cos θ −de de , r= 1 − e cos θ 1 + e cos θ The curve r = f 2 (θ ) can be parametrized using θ as a parameter: r=

SOLUTION

x = f 2 (θ ) cos θ

(1)

y = f 2 (θ ) sin θ Using the identities cos (θ + π ) = − cos θ and sin (θ + π ) = − sin θ , we obtain the following parametrization for r = f 1 (θ ): x = f 1 (θ ) cos θ = − f 2 (θ + π ) cos θ = f 2 (θ + π ) (− cos θ ) = f 2 (θ + π ) cos (θ + π ) y = f 1 (θ ) sin θ = − f 2 (θ + π ) sin θ = f 2 (θ + π ) (− sin θ ) = f 2 (θ + π ) sin (θ + π ) Using t = θ + π as the parameter we get x = f 2 (t) cos t

(2)

728

C H A P T E R 12


y = f 2 (t) sin t The parametrizations (1) and (2) define the same curve. We now consider the polar equations: r=

de = f 1 (θ ) , 1 − e cos θ

r=

−de = f 2 (θ ) 1 + e cos θ

(3)

The following equality holds: − f 2 (θ + π ) = −

de −de = = f 1 (θ ) 1 + e cos (θ + π ) 1 − e cos θ

We conclude that the polar equations in (3) define the same conic section. 4 rectangular coordinates. 61. Find the aequation of the ellipse = Find polar equation for therhyperbola withinfocus at the origin, directrix x = −2, and eccentricity e = 1.2. 2 + cos θ SOLUTION

We use the polar equation of the ellipse with focus at the origin and directrix at x = d: r=

ed 1 + e cos θ

Dividing the numerator and denominator of the given equation by 2 yields r=

2 . 1 + 12 cos θ

We identify the values e = 12 and ed = 2. Hence, d = 2e = 4. The focus-directrix of the ellipse is P O = eD. That is, x 2 + y 2 = 12 |4 − x|. y

Directrix P(x, y)

D x

d=4

0

We square the two sides, simplify, and complete the square to obtain

4 x 2 + y 2 = 16 − 8x + x 2 3x 2 + 8x + 4y 2 8 3 x 2 + x + 4y 2 3 4 2 3 x+ + 4y 2 3

2 x + 43 y2 + 16 64

9

3

= 16 = 16 =

64 3

=1⇒

2 x + 43 8 3

⎞2 y + ⎝ 4 ⎠ = 1. ⎛

√

3

ed de ed have x-coordinates 63. Let e > 1. Show that the verticesde of the hyperbola r = and . 1 + e cos θ e + 1 e 1 center C, Let C be the ellipse r = , where e < 1. Express the x-coordinates of the vertices A, A−, the 1 + e cos θ SOLUTION Since focus the focus at the of origin and the hyperbola and the second F2 inis terms d and e (Figure 23). is to the right (see figure), the two vertices have positive x coordinates. The corresponding values of θ at the vertices are θ = 0 and θ = π . Hence, since e > 1 we obtain de = de x A = |r (0)| = 1 + e cos 0 1 + e de de = = de x A = |r (π )| = 1 + e cos π 1 − e e − 1

S E C T I O N 12.5

Conic Sections

729

Further Insights and Challenges 65. Verify Theorem 4 in2.the case 0 < e < 1. Hint: Repeat the proof of Theorem 4, but set c = d/(e−2 − 1). Verify Theorem SOLUTION We follow closely the proof of Theorem 4 in the book, which covered the case e > 1. This time, for 0 < e < 1, we prove that P F = e P D defines an ellipse. We choose our coordinate axes so that the focus F lies on the x-axis with coordinates F = (c, 0) and so that the directrix is vertical, lying to the right of F at a distance d from F. As suggested by the hint, we set c = −2d , but since we are working towards an ellipse, we will also need to let e −1 b = a 2 − c2 as opposed to the c2 − a 2 from the original proof of Theorem 4. Here’s the complete list of definitions: d , c = −2 e −1

c , e

a=

b=

a 2 − c2

The directrix is the line x = c + d = c + c(e−2 − 1) = ce−2 =

a e

Now, the equation PF = e · PD for the points P = (x, y), F = (c, 0), and D = (a/e, y) becomes

(x − c)2 + y 2 = e · (x − (a/e))2 Returning to the proof of Theorem 4, we see that this is the same equation that appears in the middle of the proof of the Theorem. As seen there, this equation can be transformed into y2 x2 =1 − 2 2 2 a a (e − 1) and this is equivalent to x2 y2 =1 + a2 a 2 (1 − e2 ) Since a 2 (1 − e2 ) = a 2 − a 2 e2 = a 2 − c2 = b2 , then we obtain the equation of the ellipse y2 x2 + =1 a2 b2 Reflective Property the1,Ellipse Exercises we prove that the focal radiifocus at a point on an ellipse make Verify that if of e> then Eq. In (12) defines a67–69, hyperbola of eccentricity e, with

x its

y at2 the origin and directrix 2 at x = d. equal angles with the tangent line. Let P = (x0 , y0 ) be a point on the ellipse + = 1 (a > b) with foci a b F1 = (−c, 0) and F2 = (c, 0), and eccentricity e (Figure 24). R1 = ( 1,

1)

y P = (x 0, y 0) R2 = ( 2,

2)

1

2

F1 = (−c, 0)

F2 = (c, 0)

x

x 2

y 2

FIGURE 24 The ellipse

a

+

b

= 1.

67. Show that P F1 = a + x0 e and P F2 = a − x 0 e. Hints: (a) Show that P F1 2 − P F2 2 = 4x0 c. (b) Divide the previous relation by P F1 + P F2 = 2a, and conclude that P F1 − P F2 = 2x 0 e. SOLUTION

Using the distance formula we have 2

P F1 = (x 0 + c)2 + y 2 ;

2

P F2 = (x 0 − c)2 + y 2

Hence, 2

2

P F1 − P F2 = (x 0 + c)2 + y 2 − (x 0 − c)2 − y 2

730

C H A P T E R 12


= (x 0 + c)2 − (x 0 − c)2 = x 0 2 + 2x 0 c + c2 − x 0 2 + 2x 0 c − c2 = 4x0 c That is, P F1 + P F2 = 4x0 c. We use the identity u 2 − v 2 = (u − v) (u + v) to write this as P F1 − P F2 P F1 + P F2 = 4x0 c

(1)

2 2 Since P lies on the ellipse ax + by = 1 we have P F1 + P F2 = 2a

(2)

Substituting in (1) gives

P F1 − P F2 · 2a = 4x0 c

We divide by a and use the eccentricity e = ac to obtain P F1 − P F2 = 2x 0 e 69. Define R1 and R2 as in the figure, so that F1 R1 and F2 R2 are perpendicular to the y x tangent line. Show that tangent line at P is Ax + By = 1, where A = 02 and B = 02 . α1 the + cequation α2 −ofc the A a b (a) Show that = = . β1 β2 B (b) Use (a) and the distance formula to show that F1 R1 β = 1 F2 R2 β2 (c) Solve for β1 and β2 :

β1 = (d) Show that

B(1 + Ac) , A2 + B 2

β2 =

B(1 − Ac) A2 + B 2

F1 R1 P F1 = . Conclude that θ1 = θ2 . F2 R2 P F2

SOLUTION

(a) Since R1 = α1 , β1 and R2 = α2 , β2 lie on the tangent line at P, that is on the line Ax + By = 1, we have A α1 + B β1 = 1

and

A α2 + B β2 = 1

β

1 and it is perpendicular to the tangent line having slope − BA . Similarly, the slope of The slope of the line R1 F1 is α1 +c β2 and it is also perpendicular to the tangent line. Hence, the line R2 F2 is α2 −c

A α1 + c = β1 B

and

A α2 − c = . β2 B

(b) Using the distance formula, we have 2

R1 F1 = (α1 + c)2 + β12 Thus,

2

R1 F1 = β12

α1 + c 2 +1 β1

(1)

By part (a), α1β+c = BA . Substituting in (1) gives 1 2

R1 F1 = β12

A2 + 1 = β12 1 + B −2 A2 2 B

Likewise, 2

R2 F2 = (α2 − c)2 + β2 2 = β2 2

α2 − c 2 +1 β2

(2)

S E C T I O N 12.5

Conic Sections

731

but since α2β−c = BA , we get that 2 2

R2 F2 = β22

A2 +1 . B2

(3)

Dividing, we find that 2

β12 = 2 β22 R2 F2 R1 F1

β = 1, β2 R2 F2 R1 F1

so

as desired. (c), (d) In part (a) we show that ⎧ ⎪ ⎨ A α 1 + B β1 = 1 β1 B ⎪ = ⎩ α1 + c A . Substituting in (2) we obtain Solving for β1 gives β1 = B(1+Ac) A2 +B 2 B 2 (1 + Ac)2 (1 + Ac)2 A2 B 2 (1 + Ac)2 ( A2 + B 2 ) 2 R1 F1 = = 1 + = 2 2 B2 A2 + B 2 ( A2 + B 2 ) ( A2 + B 2 ) B 2 Similarly solving the equations in part (a) for β2 and using equation (3) yields ⎧ ⎪ ⎨ A α2 + B β2 = 1 B (1 − Ac) ⇒ β2 = B β2 ⎪ A2 + B 2 = ⎩ α2 − c A Using the distance formula for R2 F2 we have

2

R2 F2 = (α2 − c)2 + β22 = β22

(4)

A2 − c 2 +1 β2

Substituting Aβ2 −c = BA from part (a) and β2 in (3) we get 2 B 2 (1 − Ac)2 2 R2 F2 = 2 2 2 A +B

B 2 (1 − Ac)2 A2 + B 2 (1 − Ac)2 +1 = = 2 2 2 B A + B2 A2 + B 2 B 2

A2

Using the expression for R1 F1 and R2 F2 obtained above, we get R1 F1 R2 F2

√1+Ac =

A2 +B 2

√1−Ac

=

A2 +B 2

1 + Ac 1 − Ac

Substituting c = ea and A = x02 we obtain a

R1 F1 R2 F2

=

1 + x0 ea 2

1 + xa0 e a + x0 e a = x0 e = a − x e 1 − x0 ea 1 − 0 a a2

0 Now by Exercise 67, we have P F1 = a + x0 e and P F2 = a − x0 e, where P = (x 0 , y0 ). By Exercise 69, R1 F1 = a+ex a−ex0 . R2 F2 Using these results we derive the following equality:

P F1 P F2

=

R1 F1 R2 F2

⇒

R1 F1 P F1

=

R2 F2 P F2

By R1 F1 = sin θ1 and R2 F2 = sin θ2 we get sin θ1 = sin θ2 , which implies that θ1 = θ2 since the two angles are acute. PF PF 1

2

x2 Show R in Figure is independent of the ypoint P. focus (0, c), and the vertex at the origin, as 71. Show thatthat y =the length is theQequation of a 25 parabola with directrix = −c, 4c stated in Theorem 3. SOLUTION

The points P = (x, y) on the parabola are equidistant from F = (0, c) and the line y = −c.

732

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S y P(x, y) F(0, c) x y = −c

That is, by the distance formula, we have

PF = PD x 2 + (y − c)2 = |y + c|

Squaring and simplifying yields x 2 + (y − c)2 = (y + c)2 x 2 + y 2 − 2yc + c2 = y 2 + 2yc + c2 x 2 − 2yc = 2yc x 2 = 4yc ⇒ y =

x2 4c

Thus, we showed that the points that are equidistant from the focus F = (0, c) and the directrix y = −c satisfy the 2 equation y = x4c . 73. If we rewrite the general equation of degree two (13) in terms of variables x and y that are related to x and y Consider two ellipses in standard position: by equations (14) and (15), we obtain a new equation of degree two in x and y of the same form but with different 2 2 coefficients: x y E1 : + =1 A x 2 + B x y + C ya21 + D x +b1E y + F = 0 2 2 x y (a) Show that B = B cos 2θ + (C − A) sin 2θE. : + =1 2 a b 2 2 (b) Show that if B = 0, then we obtain B = 0 for We say that E 1 is similar to E 2 under scaling if there a factor r > 0 such that for all (x, y) on E 1 , the point 1 exists −1 A − C θ similar = cotunder (r x, r y) lies on E 2 . Show that E 1 and E 2 are scaling if and only if they have the same eccentricity. 2 B Show that any two circles are similar under scaling. This proves that it is always possible to eliminate the cross term Bx y by rotating the axes through a suitable angle. SOLUTION

(a) If we plug in x = x cos θ − y sin θ and y = x sin θ + y cos θ into the equation Ax 2 + Bx y + C y 2 + Dx + E y + F = 0, we will get a very ugly mess. Fortunately, we only care about the x y term, so we really only need to look at the Ax 2 + Bx y + C y 2 part of the formula. In fact, we only need to pull out those terms which have an x y in them. From the Ax 2 term, after replacing x with x = x cos θ − y sin θ , we will get an x y term of −2 Ax y cos θ sin θ . Likewise, from the Bx y term we will get Bx y cos2 θ − sin2 θ , and from the C y 2 term we get 2C x y cos θ sin θ . Adding these together, we see that the (new) B , the coefficient of the (new) x y term, will be −2 A cos θ sin θ + B cos2 θ − sin2 θ + 2C x y cos θ sin θ which simplifies to B cos 2θ + (C − A) sin 2θ , as desired. (b) Setting B = 0, we get 0 = B cos 2θ + (C − A) sin 2θ , so B cos 2θ = ( A − C) sin 2θ , so cot 2θ = A−C B , giving us 1 cot−1 A−C . 2θ = cot−1 A−C , and thus θ = B B 2

CHAPTER REVIEW EXERCISES 1. Which of the following curves pass through the point (1, 4)? (b) c(t) = (t 2 , t − 3) (a) c(t) = (t 2 , t + 3) (c) c(t) = (t 2 , 3 − t) SOLUTION

curves.

(d) c(t) = (t − 3, t 2 )

To check whether it passes through the point (1, 4), we solve the equations c(t) = (1, 4) for the given


733

(a) Comparing the second coordinate of the curve and the point yields: t +3=4 t =1 We substitute t = 1 in the first coordinate, to obtain t 2 = 12 = 1 Hence the curve passes through (1, 4). (b) Comparing the second coordinate of the curve and the point yields: t −3=4 t =7 We substitute t = 7 in the first coordinate to obtain t 2 = 72 = 49 = 1 Hence the curve does not pass through (1, 4). (c) Comparing the second coordinate of the curve and the point yields 3−t =4 t = −1 We substitute t = −1 in the first coordinate, to obtain t 2 = (−1)2 = 1 Hence the curve passes through (1, 4). (d) Comparing the first coordinate of the curve and the point yields t −3=1 t =4 We substitute t = 4 in the second coordinate, to obtain: t 2 = 42 = 16 = 4 Hence the curve does not pass through (1, 4). 3. Find parametric equations for the circle of radius 2 with center (1, 1). Use the equations to find the points of interFind parametric equations for the line through P = (2, 5) perpendicular to the line y = 4x − 3. section of the circle with the x- and the y-axes. SOLUTION

Using the standard technique for parametric equations of curves, we obtain c(t) = (1 + 2 cos t, 1 + 2 sin t)

We compare the x coordinate of c(t) to 0: 1 + 2 cos t = 0 1 2 2π t =± 3

cos t = −

Substituting in the y coordinate yields √ √ 2π 3 1 + 2 sin ± =1±2 =1± 3 3 2 √ Hence, the intersection points with the y-axis are (0, 1 ± 3). We compare the y coordinate of c(t) to 0: 1 + 2 sin t = 0 1 2 π t =− 6

sin t = −

or

7 π 6

734

C H A P T E R 12


Substituting in the x coordinates yields √

π √ 3 =1+2 =1+ 3 1 + 2 cos − 6 2 √

π √ 7 3 1 + 2 cos π = 1 − 2 cos =1−2 =1− 3 6 6 2 √ Hence, the intersection points with the x-axis are (1 ± 3, 0). 5. Find a parametrization c(θ ) of the unit circle such that c(0) = (−1, 0). Find a parametrization c(t) of the line y = 5 − 2x such that c(0) = (2, 1). SOLUTION The unit circle has the parametrization c(t) = (cos t, sin t) This parametrization does not satisfy c(0) = (−1, 0). We replace the parameter t by a parameter θ so that t = θ + α , to obtain another parametrization for the circle: c∗ (θ ) = (cos(θ + α ), sin(θ + α ))

(1)

We need that c∗ (0) = (1, 0), that is, c∗ (0) = (cos α , sin α ) = (−1, 0) Hence cos α = −1 sin α = 0

⇒

α=π

Substituting in (1) we obtain the following parametrization: c∗ (θ ) = (cos(θ + π ), sin(θ + π )) 7. Find a path c(t) that traces the line y = 2x + 1 from (1, 3) to (3, 7) for 0 ≤ t ≤ 1. Find a path c(t) that traces the parabolic arc y = x 2 from (0, 0) to (3, 9) for 0 ≤ t ≤ 1. SOLUTION Solution 1: By one of the examples in section 12.1, the line through P = (1, 3) with slope 2 has the parametrization c(t) = (1 + t, 3 + 2t) But this parametrization does not satisfy c(1) = (3, 7). We replace the parameter t by a parameter s so that t = α s + β . We get c∗ (s) = 1 + α s + β , 3 + 2(α s + β ) = (α s + β + 1, 2α s + 2β + 3) We need that c∗ (0) = (1, 3) and c∗ (1) = (3, 7). Hence, c∗ (0) = (1 + β , 3 + 2β ) = (1, 3) c∗ (1) = (α + β + 1, 2α + 2β + 3) = (3, 7) We obtain the equations 1+β =1 3 + 2β = 3

α+β +1=3

⇒

β = 0, α = 2

2α + 2β + 3 = 7 Substituting in (1) gives c∗ (s) = (2s + 1, 4s + 3) Solution 2: The segment from (1, 3) to (3, 7) has the following vector parametrization: (1 − t) 1, 3 + t 3, 7 = 1 − t + 3t, 3(1 − t) + 7t = 1 + 2t, 3 + 4t The parametrization is thus c(t) = (1 + 2t, 3 + 4t) Sketch the graph c(t) = (1 + cos t, sin 2t) for 0 ≤ t ≤ 2π and draw arrows specifying the direction of motion.


735

In Exercises 9–12, express the parametric curve in the form y = f (x). 9. c(t) = (4t − 3, 10 − t) SOLUTION

We use the given equation to express t in terms of x. x = 4t − 3 4t = x + 3 t=

x +3 4

Substituting in the equation of y yields y = 10 − t = 10 −

x 37 x +3 =− + 4 4 4

That is, y=−

37 x + 4 4

2 3 1 2 − 4) + 11. c(t)c(t) = =3 (t −3 +, t1, t t t SOLUTION

We use the given equation to express t in terms of x: x =3−

2 t

2 =3−x t 2 t= 3−x Substituting in the equation of y yields y=

3 1 3−x 8 2 + + = 3 3−x 2/(3 − x) 2 (3 − x)

13. Findx all points by the path c(t) = (t 2 , sin t). Plot c(t) with a graphing utility. = tan t, visited y = sectwice t SOLUTION For every point, if the curve passes through it twice, then its x coordinate and y coordinate are obtained twice by the functions x(t), y(t). We first calculate the x coordinate of these points. Since x(t) = t 2 , every x coordinate is obtained twice—once for t and once for −t. We now check for which of the above x coordinates, the y coordinates are equal as well. We substitute the above condition in the formula for the y coordinates. We have sin t = sin(−t). Since sin t = − sin(−t) for all t ∈ R, we can add it to the former equation. We obtain 2 sin t = 0 or t = π k where k ∈ Z. We substitute the t values in the parametric equation to find the desired points. We obtain (π 2 k 2 , 0)

where k ∈ Z

The path c(t) is shown in the following figure. y

1 x 1 −1

In Exercises 14–17, calculate

dy at the point indicated. dx

15. c(θ ) = (tan23θ , cos 2θ ), θ = π4 c(t) = (t + t, t − 1), t = 3

2

3

736

C H A P T E R 12

PAR A M E T R I C E Q U AT I O N S , P O L A R C O O R D I N AT E S , A N D C O N I C S E C T I O N S SOLUTION

The parametric equations are x = tan2 θ , y = cos θ . We use the theorem on the slope of the tangent line to

find dd xy : dy

dy − sin θ cos3 θ = dd θx = =− 2 dx 2 2 tan θ sec θ dθ

We now substitute θ = π4 to obtain cos3 π4 d y 1 = − =− √ d x θ =π /4 2 4 2 17. c(t) = (ln t, t3t 2 − t), P = (0, 2) c(t) = (e − 1, sin t), t = 20 SOLUTION The parametric equations are x = ln t, y = 3t 2 − t. We use the theorem for the slope of the tangent line to find dd xy : dy

6t − 1 dy = ddtx = 1 = 6t 2 − t dx

(1)

t

dt

We now must identify the value of t corresponding to the point P = (0, 2) on the curve. We solve the following equations: ln t = 0 3t 2 − t = 2 Substituting t = 1 in (1) we obtain

⇒

t =1

d y = 6 · 12 − 1 = 5 dx P

19. Find the points on (t + sin t, t − 2 sin t) where the tangent is vertical or horizontal. 1 Find the point on the cycloid c(t) = (t − sin t, 1 − cos t) wheredthe y tangent line has slope 2 . SOLUTION We use the theorem for the slope of the tangent line to find d x : dy

1 − 2 cos t dy = ddtx = dx 1 + cos t dt

We find the values of t for which the denominator is zero. We ignore the numerator, since when 1 + cos t = 0, 1 − 2 cos t = 3 = 0. 1 + cos t = 0 cos t = −1 t = π + 2π k

where k ∈ Z

We now find the values of t for which the numerator is 0: 1 − 2 cos t = 0 1 = 2 cos t 1 = cos t 2 π t = ± + 2π k 3

where k ∈ Z

Note that the denominator is not zero at these points. Thus, we have vertical tangents at t = π + 2π k and horizontal tangents at t = ±π /3 + 2π k. 21. Find the speed at t = π4 of a particle whose position at time t seconds is c(t) = (sin 4t, cos 3t). Find the equation of the Bézier curve with control points SOLUTION We use the parametric definition to find the speed. We obtain P3 (1, −1) P0 = (−1, −1), P 1 = (−1, 1), P2 = (1, 1),

ds 2 2 2 2 = ((sin 4t) ) + ((cos 3t) ) = (4 cos 4t) + (−3 sin 3t) = 16 cos2 4t + 9 sin2 3t dt At time t = π4 the speed is

3π ds 1 √ 2 2 = 16 cos π + 9 sin = 16 + 9 · = 20.5 ≈ 4.53 dt t=π /4 4 2


737

23. Find the length of (3et − 3, 4et + 7) for 0 ≤ t ≤ 1. Find the speed (as a function of t) of a particle whose position at time t seconds is c(t) = (sin t + t, cos t + t). SOLUTION Weparticle’s use the formula arc length, to obtain What is the maximalfor speed? s=

" 1 " 1 ((3et − 3) )2 + ((4et + 7) )2 dt = (3et )2 + (4et )2 dt 0

0

1 " 1 " 1 " 1 9e2t + 16e2t dt = 25e2t dt = 5et dt = 5et = 5(e − 1) = 0

0

0

0

In Exercises 24–25, let c(t) = (e−t cos t, e−t sin t). 25. Find the first positive value of t0 such that the tangent line to c(t0 ) is vertical and calculate the speed at t = t0 . Show that c(t) for 0 ≤ t < ∞ has finite length and calculate its value. dy dy SOLUTION The curve has a vertical tangent where lim d x = ∞. We first find d x using the theorem for the slope of t→t0

a tangent line: dy

(e−t sin t) −e−t sin t + e−t cos t dy = = ddtx = −t dx (e cos t) −e−t cos t − e−t sin t dt

=−

sin t − cos t cos t − sin t = cos t + sin t sin t + cos t

We now search for t0 such that lim dd xy = ∞. In our case, this happens when the denominator is 0, but the numerator t→t0

is not, thus: sin t0 + cos t0 = 0 cos t0 = − sin t0 cos −t0 = sin −t0 π −t0 = − π 4 3 t0 = π 4 We now use the formula for the speed, to find the speed at t0 .

ds = ((e−t sin t) )2 + ((e−t cos t) )2 dt

= (−e−t cos t − e−t sin t)2 + (−e−t sin t + e−t cos t)2

= e−2t (cos t + sin t)2 + e−2t (cos t − sin t)2

√ = e−t cos2 t + 2 sin t cos t + sin2 t + cos2 t − 2 sin t cos t + sin2 t = e−t 2 Next we substitute t = 34 π , to obtain √ √ e−t0 2 = e−3π /4 2 27. Convert the points (x, y) = (1, −3), (3, −1) from rectangular to polar coordinates. Plot c(t) = (sin 2t, 2 cos t) for 0 ≤ t ≤ π . Express the length of the curve as a definite integral and approximate SOLUTION convertalgebra the given points from cartesian coordinates to polar coordinates. For the first point we have it using a We computer system. r=

x 2 + y2 =

θ = arctan

√ 12 + (−3)2 = 10

y = arctan −3 = 5.034 x

For the second point we have r= =

√

x 2 + y2 =

10θ = arctan

32 + (−1)2

y −1 = arctan = −0.321, 5.961 x 3

Convert the points (r, θ ) = 1, π6 , 3, 54π from polar to rectangular coordinates.

738

C H A P T E R 12


29. Write (x + y)2 = x y + 6 as an equation in polar coordinates. SOLUTION

We use the formula for converting from cartesian coordinates to polar coordinates to substitute r and θ for

x and y: (x + y)2 = x y + 6 x 2 + 2x y + y 2 = x y + 6 x 2 + y 2 = −x y + 6 r 2 = −(r cos θ )(r sin θ ) + 6 r 2 = −r 2 cos θ sin θ + 6 r 2 (1 + sin θ cos θ ) = 6 r2 =

6 1 + sin θ cos θ

r2 =

6 1 + sin22θ

r2 =

12 2 + sin 2θ

4 is the polar equation of a line. 31. Show that r = 2 cos θ θ an equation in rectangular coordinates. Write r = 7 cos θ − sin as cos θ − sin θ SOLUTION We use the formula for converting from polar coordinates to cartesian coordinates to substitute x and y for r and θ : r=

4 7 cos θ − sin θ

1=

4 7r cos θ − r sin θ

1=

4 7x − y

7x − y = 4 y = 7x − 4 We obtained a linear function. Since the original equation in polar coordinates represents the same curve, it represents a straight line as well. 33. Calculate the area of the circle r = 3 sin θ bounded by the rays θ = π3 and θ = 23π . Convert the equation SOLUTION We use the formula for area in polar coordinates to obtain 9(x 2 + y 2 ) = (x 2 + y 2 − 2y)2 " 2π /3 " 2π /3 " 2π /3 9 sin 2θ 2π /3 1 2 dθ = 9 2 θ dθ = 9 (3 sin θ ) sin (1 − cos 2 θ ) d θ = θ − A = to polar coordinates and plot with a graphing utility. 2 π /3 2 π /3 4 π /3 4 2 π /3 √ √ √ 1 4π 2π 9 π 1 9 π 9 π 3 3 3 − sin − sin = − − − = + = 4 3 2 3 3 4 3 2 2 2 4 3 2 35. The equation r = sin(n θ ), where n ≥ 2 is even, is a “rose” of 2n petals (Figure 1). Compute the total area of the the area of one petal. thethat graph of r not = sin 4θ and flower andPlot show it does depend on calculate n. y

y

x

n = 2 (4 petals)

y

x

n = 4 (8 petals)

FIGURE 1

x

n = 6 (12 petals)


739

SOLUTION We calculate the total area of the flower, that is, the area between the rays θ = 0 and θ = 2π , using the formula for area in polar coordinates: " " 1 2π 1 sin 4n θ 2π 1 2π 2 sin 2n θ d θ = (1 − cos 4n θ ) d θ = θ− A= 2 4 4 4n

0

=

0

0

π π 1 − (sin 8n π − sin 0) = 2 16n 2

π for every n ∈ Z, the area is independent of n. 2 37. Find the shaded area in Figure 3. Calculate the total area enclosed by the curve r 2 = cos θ esin θ (Figure 2). Since the area is

y 1

−2

r = 1 + cos 2q x

−1

1

2

−1

FIGURE 3 SOLUTION

We first find the points of intersection between the unit circle and the function. 1 = 1 + cos 2θ cos 2θ = 0 π 2θ = + π n 2 π π θ= + n 4 2

We now find the area of the shaded figure in the first quadrant. This has two parts. The first, from 0 to π /4, is just an octant of the unit circle, and thus has area π /8. The second, from π /4 to π /2, is found as follows: " " " 1 π /2 1 π /2 1 π /2 3 1 A= (1 + cos 2θ )2 d θ = 1 + 2 cos 2θ + cos2 2θ d θ = + 2 cos 2θ + cos 4θ d θ 2 π /4 2 π /4 2 π /4 2 2 π /2 1 3θ 1 1 3π = = + sin 2θ + sin 4θ −1 2 2 8 2 8 π /4 The total area in the first quadrant is thus 516π − 12 ; multiply by 2 to get the total area of 58π − 1. 39. Figure showsof thethe graph r =polar e0.5θequation sin θ forr0=≤θθin≤Figure 2π . Use Calculate the5length curveofwith 4. a CAS to approximate the difference in length between the outer and inner loops. y 10 5 x

−6

3

FIGURE 5 SOLUTION We note that the inner loop is the curve for θ ∈ [0, π ], and the outer loop is the curve for θ ∈ [π , 2π ]. We express the length of these loops using the formula for the arc length. The length of the inner loop is 0.5θ 2 " π " π e sin θ s1 = (e0.5θ sin θ )2 + ((e0.5θ sin θ ) )2 d θ = eθ sin2 θ + + e0.5θ cos θ d θ 2 0 0

and the length of the outer loop is s2 =

" 2π π

eθ sin2 θ +

0.5θ 2 e sin θ + e0.5θ cos θ d θ 2

We now use the CAS to calculate the arc length of each of the loops. We obtain that the length of the inner loop is 7.5087 and the length of the outer loop is 36.121, hence the outer one is 4.81 times longer than the inner one.

740

C H A P T E R 12


In Exercises 40–43, identify the conic section. Find the vertices and foci. 41. x 2 − 22 = 4 y 2

x2y + =the 1 equation by 4 to obtain SOLUTION 3 We divide 2

y 2 − √ =1 2 2 √ 2 √ This is a hyperbola in standard position, its foci are ± 22 + 2 , 0 = (± 6, 0), and its vertices are (±2, 0).

x 2

43. (y − 3)2 =1 2x22 − 1 (2x + 2 y) = 4 − (x − y)2 SOLUTION We simplify the equation: (y − 3)2 = 2x 2 − 1 2x 2 − (y − 3)2 = 1 ⎞2

⎛

⎝ x ⎠ − (y − 3)2 = 1 1 √

2

This is a hyperbola shifted 3 units on the y-axis. Therefore, its foci are

vertices are ± √1 , 3 .

2 √1 + 1, 3 = ± 32 , 3 and its ± 2

2

3 √ 45. FindFind the the equation of aofstandard hyperbola with two vertices at (±8, 0) and asymptotes = 0). ± x. equation a standard ellipse with vertices at (±8, 0) and foci (± y 3, 4 Since the asymptotes of the hyperbola are y = ± 34 x, and the equation of the asymptotes for a general hyperbola in standard position is y = ± ab x, we conclude that ab = 34 . We are given that the vertices are (±8, 0), thus a = 8. We substitute and solve for b: SOLUTION

3 b = a 4 b 3 = 8 4 b=6 Next we use a and b to construct the equation of the hyperbola:

x 2 y 2 − = 1. 8 6 1. 47. Find the the equation of aofstandard ellipse with foci (±8,(8, 0) 0) andand eccentricity 8 Find equation a standard parabola withatfocus directrix x = −8. 2 2 SOLUTION If the foci are on the x-axis, then a > b, and c = a − b . We are given that e = 18 , and c = 8. Substituting and solving for a and b yields

c a c = a 2 − b2

e=

1 8 = 8 a 64 = a 8 = 642 − b2 64 = 642 − b2 b2 = 64 · 63 √ b = 8 63 We use a and b to construct the equation of the ellipse:

x 2 y 2 + = 1. √ 64 8 63


741

49. Show that the “conic section” with equation2x 2 − 4x +2y 2 + 5 = 0 has no points. Find the asymptotes of the hyperbola 3x + 6x − y − 10y = 1. SOLUTION We complete the squares in the given equation: x 2 − 4x + 4y 2 + 5 = 0 x 2 − 4x + 4 − 4 + 4y 2 + 5 = 0 (x − 2)2 + 4y 2 = −1 Since (x − 2)2 ≥ 0 and y 2 ≥ 0, there is no point satisfying the equation, hence it cannot represent a conic section. 51. The orbit of Jupiter is an ellipse with the sun at a focus. Find the eccentricity of the orbit if the perihelion (closest x dy 6 2 aphelion the sun) equals 816 × 10e.6 km. distanceShow to thethat sun)the equals 740 × 10 1) theholds = (ekm−and on a (farthest standard distance ellipse ortohyperbola of eccentricity relation dx y SOLUTION For the sake of simplicity, we treat all numbers in units of 106 km. By Kepler’s First Law we conclude that the sun is at one of the foci of the ellipse. Therefore, the closest and farthest points to the sun are vertices. Moreover, they are the vertices on the x-axis, hence we conclude that the distance between the two vertices is 2a = 740 + 816 = 1556 Since the distance between each focus and the vertex that is closest to it is the same distance, and since a = 778, we conclude that the distance between the foci is c = a − 740 = 38 We substitute this in the formula for the eccentricity to obtain: e=

c = 0.0488. a

Refer to Figure 24 in Section 12.5. Prove that the product of the perpendicular distances F1 R1 and F2 R2 from the foci to a tangent line of an ellipse is equal to the square b2 of the semiminor axes.

Single Variable Calculus Student Solutions Manual

Complete Solutions Manual for Single Variable Calculus Early Transcedentals

Calculus: Single Variable

Calculus: Single Variable