Developments in Petroleum Science, 1
GEOCHEMISTRY OF OILFIELD WATERS
This Page Intentionally Left Blank
Developmen...
139 downloads
1279 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
Developments in Petroleum Science, 1
GEOCHEMISTRY OF OILFIELD WATERS
This Page Intentionally Left Blank
Developments in Petroleum Science, 1
GEOCHEMISTRY OF OILFIELD WATERS A. GENE COLLINS Bartlesville Energy Research Center Bureau o f Mines United States Department of the Interior Bartlesville, Oklahoma, U.S.A.
ELSEVIER SCIENTIFIC PUBLISHING COMPANY Amsterdam
-
Oxford
-
New York 1975
ELSEVIER SCIENTIFIC PUBLISHING COMPANY 335 Jan van Galenstraat P.O. Box 211,Amsterdam, The Netherlands AMERICAN ELSEVIER PUBLISHING COMPANY, INC. 52 Vanderbilt Avenue New York, New York 10017
Library of Congress Card Number: 73-89149 ISBN 0-444-41183-6 With 132 illustrations and 87 tables Copyright 0 1975 by Elsevier Scientific Publishing Company, Amsterdam All rights reserved. No part of this publication may be reproduced, stored in a retrieva system, or transmitted, in any form or by any means, electronic, mechanical, photo copying, or otherwise without the prior written permission of the publisher, Elseviei Scientific Publishing Company, Jan van Galenstraat 335,Amsterdam Printed in The Netherlands
PREFACE
The purpose of this book is to provide information relevant to the analytical chemistry and geochemistry of oilfield waters. The book attempts to recognize the importance of subsurface oilfield waters as they are related t o origin, migration, accumulation, and maturation of oil and gas and thus their relationship t o exploration for and production of oil and gas. One chapter presents a simplistic introduction t o the origin of oilfield waters. Because oilfield waters can constitute an environmental pollution hazard, the book describes and comments on methods of their disposal or of recovering valuable constituents from them. The numerous references indicate that the book relies heavily upon the work of others. The reader will vastly expand his knowledge of the subject by consulting these references. The writer appreciates the understanding and thoughtfulness of his Wife, Barbara, and children, Sandy and Mike, during the preparation of part of this book at our home. He acknowledges With appreciation the criticisms, opinions, and suggestions of various portions of the book by O.C. Baptist, W.H. Caraway, P.H. Dickey, G.L. Gates, R.V. Huff, P.H. Jones, and C.C. Linville. M.E. Crocker and Ms. C.A. Pearson, did an invaluable service of proof-reading and index preparation. He extends appreciation t o Ms. D.J. Forbes, Ms. M.G. Goff, and Ms. J. Haimson for typing the manuscript; t o D.W. Anderson, Ms. E.S. Baldwin, J.A. Chidester, G.E. Fletcher, R.M. Horn, and W.A. McClung for preparing the figures; and to authors, book publishers, companies, and technical journals who granted permission t o use various illustrations. Permission t o publish this manuscript was granted by the Director of the United States Bureau of Mines. Bureau of Mines officials who generously helped obtain this permission were: J.S. Ball, R.T. Johansen, and J.W.Watkins. Finally inasmuch as it is the writer’s belief that this book is not perfect, he takes this opportunity to solicit constructive criticism from its readers. A. GENE COLLINS Bartlesville Energy Research Center U.S.Bureau of Mines Bartlesville, Oklahoma
This Page Intentionally Left Blank
CONTENTS
Preface
. . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 1. Introduction References . . . . .
V
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 2 . Sampling subsurface oilfield waters . . . Drill-stem test . . . . . . . . . . . . . . Sample containing dissolved gases . . . . . . . Sampling at the flow line . . . . . . . . . . . Sampling at the wellhead . . . . . . . . . . . Sample for determining unstable properties or species . Sample for stable-isotope analysis . . . . . . . Sample containers . . . . . . . . . . . . . Tabulation of sample description . . . . . . . References . . . . . . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 3. Analysis of oilfield waters for some physical properties and inorganic chemical constituents . . . . . . . . . . . . . . . . Quality control . . . . . . . . . . . . . . . . . . . . . . . Preliminary sample treatment . . . . . . . . . . . . . . . . . . Reporting the analytical results . . . . . . . . . . . . . . . . . Synthetic brine . . . . . . . . . . . . . . . . . . . . . . . Determination of pH . . . . . . . . . . . . . . . . . . . . . Determination of Eh . . . . . . . . . . . . . . . . . . . . . Suspended solids . . . . . . . . . . . . . . . . . . . . . . . Resistivity . . . . . . . . . . . . . . . . . . . . . . . . . Specific gravity . . . . . . . . . . . . . . . . . . . . . . . TITRIMETRIC METHODS . . . . . . . . . . . . . . . . . . Acidity. alkalinity. and borate boron . . . . . . . . . . . . . . . Calcium and magnesium . . . . . . . . . . . . . . . . . . . Ammonium nitrogen . . . . . . . . . . . . . . . . . . . . Chloride . . . . . . . . . . . . . . . . . . . . . . . . . Bromide and iodide . . . . . . . . . . . . . . . . . . . . . Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . Carbon dioxide . . . . . . . . . . . . . . . . . . . . . . Sulfide . . . . . . . . . . . . . . . . . . . . . . . . . Sulfur compounds . . . . . . . . . . . . . . . . . . . . . FLAME SPECTROPHOTOMETRIC .METHODS . . . . . . . . . . . Lithium . . . . . . . . . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . . . . . Potassium . . . . . . . . . . . . . . . . . . . . . . . . Rubidium and cesium . . . . . . . . . . . . . . . . . . . . Manganese . . . . . . . . . . . . . . . . . . . . . . . . Strontium . . . . . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . . . . .
1 5
. .
. . .
. . .
. . . . . . . . . . . . . . . . . .
7 8
12 13 14 14 15 16 17 17 19 19 22 25 27 27 29 31 32 35 37 37 40 43 44 45 47 50 51 52 53 54 57 58 59 61 62 63
VIII
CONTENTS
ATOMIC ABSORPTION METHODS . . . . . . . . . . Interferences . . . . . . . . . . . . . . . . . . . Burners and solvents . . . . . . . . . . . . . . . . . Lithium . . . . . . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . Potassium . . . . . . . . . . . . . . . . . . . . Magnesium (1) . . . . . . . . . . . . . . . . . . . Calcium (1) . . . . . . . . . . . . . . . . . . . . . Magnesium (2) . . . . . . . . . . . . . . . . . . . Calcium (2) . . . . . . . . . . . . . . . . . . . . Strontium . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . Manganese . . . . . . . . . . . . . . . . . . . . Iron . . . . . . . . . . . . . . . . . . . . . . Copper . . . . . . . . . . . . . . . . . . . . . Zinc . . . . . . . . . . . . . . . . . . . . . . Lead(1) . . . . . . . . . . . . . . . . . . . . . Lead(2) . . . . . . . . . . . . . . . . . . . . . EMISSION SPECTROSCOPY . . . . . . . . . . . . . Barium, boron, iron, manganese, and strontium . . . . . . . Beryllium . . . . . . . . . . . . . . . . . . . . Aluminum . . . . . . . . . . . . . . . . . . . . MASS SPECTROMETRIC METHODS FOR STABLE ISOTOPES . Deuterium . . . . . . . . . . . . . . . . . . . . Oxygen-18 . . . . . . . . . . . . . . . . . . . . COLORIMETRIC METHODS . . . . . . . . . . . . Interferences . . . . . . . . . . . . . . . . . . . Iron . . . . . . . . . . . . . . . . . . . . . . Copper . . . . . . . . . . . . . . . . . . . . . Nickel . . . . . . . . . . . . . . . . . . . . . Lead . . . . . . . . . . . . . . . . . . . . . . Zinc . . . . . . . . . . . . . . . . . . . . . . Cadmium . . . . . . . . . . . . . . . . . . . . Phosphate . . . . . . . . . . . . . . . . . . . . Silica . . . . . . . . . . . . . . . . . . . . . . Nitrate nitrogen . . . . . . . . . . . . . . . . . . Arsenic . . . . . . . . . . . . . . . . . . . . . Fluoride . . . . . . . . . . . . . . . . . . . . . Iodide . . . . . . . . . . . . . . . . . . . . . Selenium . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . GRAVIMETRIC METHODS . . . . . . . . . . . . . Sulfate . . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . OTHER METHODS . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . Dissolved solids . . . . . . . . . . . . . . . . . . Spent acid . . . . . . . . . . . . . . . . . . . . Acetic acid solutions . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . Chapter 4 . Interpretation of chemical analysis of oilfield waters Calculating probable compounds . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
. . . .
. . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . .
. . . .
. . . .
.
65 66 66 68 68 70 71 72 74 75 76 77 78 79 80 80 81 : 82 . 83 . 83 89 90 . 91 91 91 . 92 93 94 96 98 99 101 103 105 107 107 108 109 110 111 114 . 114 114 115 116 116 117 118 120 121
. . . . . . . 125 . . . . . . . 125
CONTENTS
IX
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
127 128 132
Chapter 5 . Significance of some inorganic constituents and physical properties of oilfield waters . . . . . . . . . . . . . . . . . . . . . Lithium . . . . . . . . . . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . Potassium . . . . . . . . . . . . . . . . . . . . . . . . . Rubidium . . . . . . . . . . . . . . . . . . . . . . . . . Cesium . . . . . . . . . . . . . . . . . . . . . . . . . . Beryllium . . . . . . . . . . . . . . . . . . . . . . . . . Magnesium . . . . . . . . . . . . . . . . . . . . . . . . . Calcium . . . . . . . . . . . . . . . . . . . . . . . . . . Strontium . . . . . . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . . . . . . Manganese . . . . . . . . . . . . . . . . . . . . . . . . . Iron . . . . . . . . . . . . . . . . . . . . . . . . . . . Copper . . . . . . . . . . . . . . . . . . . . . . . . . . Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . Mercury . . . . . . . . . . . . . . . . . . . . . . . . . . Lead . . . . . . . . . . . . . . . . . . . . . . . . . . . Cadmium . . . . . . . . . . . . . . . . . . . . . . . . . Boron . . . . . . . . . . . . . . . . . . . . . . . . . . Aluminum . . . . . . . . . . . . . . . . . . . . . . . . . Silica . . . . . . . . . . . . . . . . . . . . . . . . . . . Ammonium nitrogen . . . . . . . . . . . . . . . . . . . . . Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . Arsenic . . . . . . . . . . . . . . . . . . . . . . . . . . Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . Sulfur . . . . . . . . . . . . . . . . . . . . . . . . . . Selenium . . . . . . . . . . . . . . . . . . . . . . . . . Fluorine . . . . . . . . . . . . . . . . . . . . . . . . . . Chlorine . . . . . . . . . . . . . . . . . . . . . . . . . . Bromine . . . . . . . . . . . . . . . . . . . . . . . . . . Iodine . . . . . . . . . . . . . . . . . . . . . . . . . . Significance of some physical properties . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .
133 133 136 138 140 141 141 142 143 145 147 149 149 150 151 151 152 152 153 155 156 157 158 158 158 159 160 161 161 162 164 166 174
Chapter 6. Organic constituents in saline waters . Nitrogen-free organic compounds . . . . . Hydrocarbons containing nitrogen . . . . . Fatty acids . . . . . . . . . . . . . . Naphthenic and humic acids . . . . . . . Determination of oil in water . . . . . . . Organic acids in oilfield brines . . . . . . References . . . . . . . . . . . . . .
188
Determining a sought compound Graphic plots . . . . . . . References . . . . . . . .
. . . . . . . . Chapter 7. Origin of oilfield waters . . . . . . Definitions of some water terms . . . . . . . Sedimentary rocks . . . . . . . . . . . . Composition of oilfield waters . . . . . . .
. . . . . . . . . . . .
Research studies related t o the originof oilfield brines
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. 177 . 178 . 182 183 . 185 . 186 . 188
. 193 . 194 195 . 213 . 219
X
CONTENTS
Conclusions References .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 8 . Classification of oilfield waters Palmer’s classification . . . . . . . Sulin’s classification . . . . . . . . Modification of Sulin’s system by Bojarski Chebotarev’s classification . . . . . . Schoeller’s system . . . . . . . . Oilfield brine analyses . . . . . . . Application of the classification systems . Discussion . . . . . . . . . . . Conclusions . . . . . . . . . . References . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
245 246
. . . . . . . . . . . . 253 . . . . . . . . . . . . 254
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . 257 . . 260 . . 262 . . 267 . . 272 . . 274 . . 289 . . 291
. . . . . . . . . . . . . .
291
Chapter 9. Some effects of water upon the generation. migration. accumulation. and alteration of petroleum . . . . . . . . . . . . . . . . . Compaction . . . . . . . . . . . . . . . . . . . . . . . . Generation and migration . . . . . . . . . . . . . . . . . . . . Accumulation . . . . . . . . . . . . . . . . . . . . . . . . Alteration . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .
293 294 295 298 299 304 304
Chapter 10. Geochemical methods of exploration for petroleum and natural gas . . 307 Introduction . . . . . . . . . . . . . . . . . . . . . . . . 307 Hydrogeochemical research and methods . . . . . . . . . . . . . . . 313 Reviews . . . . . . . . . . . . . . . . . . . . . . . . . . 322 Case history of the Delaware sand (Bell Canyon formation). Texas. by Visher (1961) 322 Formation water maps of others areas . . . . . . . . . . . . . . . . 330 Concluding remarks . . . . . . . . . . . . . . . . . . . . . . 335 References . . . . . . . . . . . . . . . . . . . . . . . . . 337 Chapter 11. Geopressured reservoirs . . Geopressure . . . . . . . . . . . Origin of abnormal pressures . . . . . Abnormal pressures in the Gulf Coast area Detection of abnormal pressures . . . . References . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . . . . . . . . . . .
. . . . Chapter 12. Compatibility of oilfield waters . . . . . . . Wellbore and formation damage . . . . . . . . . . . Solubility of calciumcompounds invarioussaltsolutions . .
. . . . . . . . . . . . . . . . . . . . . . . . 367 . . . . . 368 . . . . . 370 Solubilities of the sulfates of barium and strontium in saline solutions . . . . . 372 Experimental determination of some solubilities of the sulfates of barium and strontium . . . . . . . . . . . . . . . . . . . . . . . . . 372 Resultsand discussion of the experimental investigation . . . . . . . . . . 373 Brine stabilization . . . . . . . . . . . . . . . . . . . . . . 380 Mixing of subsurface waters . . . . . . . . . . . . . . . . . . . 382 References . . . . . . . . . . . . . . . . . . . . . . . . . 386
.
Chapter 13 Valuable minerals in oilfield waters . . Recovery of iodine and bromine from oilfield brines
. . . . . . .
. . . . . . .
343 343 344 346 362 364
. . . . . . . . . . . 389 . . . . . . . . . . . 390
CONTENTS
XI
Minerals recovered from saline waters . . . . Fresh-water production . . . . . . . . . . Preliminary economic evaluation . . . . . Disposal brines . . . . . . . . . . . . Worth and value estimates . . . . . . . . . Conclusions . . . . . . . . . . . . . References . . . . . . . . . . . . . .
. . . . . . .
Chapter 1 4.Subsurface disposal . . . . . . . History of brine disposal operations . . . . . . Subsurface injection . . . . . . . . . . . . Present-day technology in subsurface disposal . . Economics and oilfield brine disposal . . . . . Injection well versus disposal well . . . . . . Acceptable geologic areas . . . . . . . . . . Suitable disposal zones . . . . . . . . . . . Evaluation of the disposal zone . . . . . . . Considerations during drilling and well completion . Fluid travel . . . . . . . . . . . . . . . Hazards of underground waste disposal . . . . State regulations and tax incentives . . . . . . Costs of disposal systems . . . . . . . . . . Conclusions . . . . . . . . . . . . . . References . . . . . . . . . . . . . . .
.
. . . . . . .
. . . . . . . 392 . . . . . 401 . . . . . . 402 . . . . . 411 . . . . . 411 . . . . . 417 . . . . . 417
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
. . . . . . . 419 . . . . . . . 419 . . . . . . 420 . . . . . . . 421 . . . . . . . 422 . . . . . . . 424 . . . . . . 425 . . . . . . 426 . . . . . . . 427 . . . . . . . 432 . . . . . . 433 . . . . . . . 434 . . . . . . . 434 . . . . . . 437 . . . . . . 438 . . . . . . 438 . . . . . . . 441 . . . . . . . 441 . . . . . . . 443 . . . . . . . 447 . . . . . . . 449 . . . . . . . 449 . . . . . . . 453 . . . . . . . 457 . . . . . . . 459
Chapter 1 5 Solubilities of some silicate minerals in saline waters Composition and structure of minerals studied . . . . . . Silicate solubilities a t 25°C and 1 atm . . . . . . . . . Experimental equipment . . . . . . . . . . . . . Experimental method . . . . . . . . . . . . . . Fundamental equations . . . . . . . . . . . . . . Experimental dataandempirical equations . . . . . . . Summary and conclusions . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . .
.
Chapter 16 Environmental impact of oil- and gas-well drilling. production and associated waste disposal practices . . . . . . . . . . . . Drilling . . . . . . . . . . . . . . . . . . . . . . . . . . Production . . . . . . . . . . . . . . . . . . . . . . . . . Disposal . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . Reference Index Subject Index .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. 461 461 467 471 474
*
477 485
This Page Intentionally Left Blank
Chapter 1.
INTRODUCTION
Petroleum, known to exist long before an oil well was drilled, first found limited use as a medicine, lubricant, and waterproofing agent. The American Indians knew of several oil and gas springs and gave this information t o the early American settlers. Early settlements were commonly located close to salt licks which supplied salt to the population. Often these salt springs were contaminated with petroleum, and many of the early efforts to acquire more salt by digging wells were rewarded by finding unwanted increased amounts of oil and gas associated with the saline waters. In the Appalachians: many saline water springs occurred along the crests of anticlines (Rogers and Rogers, 1843). In 1855 it was found that distillation of petroleum produced a light oil similar to coal oil, which was better than whale oil as an illuminant (Howell, 1934, p.2). This knowledge spurred the search for saline waters which contained oil. Colonel Edward Drake, utilizing the methods of the salt producers, drilled a well on Oil Creek, near Titusville, Pennsylvania, in 1859. He struck oil at a depth of 21 m, and this first oil well produced about 35 barrels of oil per day (Dickey, 1959). The early oil producers did not realize the significance of the oil and saline waters occurring together. In fact, it was not until 1938 that the existence of interstitial water in oil reservoirs was generally recognized (Schilthuis, 1938). Torrey (1966) was convinced as early as 1928 that dispersed interstitial water existed in oil reservoirs, but his belief was rejected by his colleagues because most of the producing oil wells did not produce any water upon completion. Occurrences of mixtures of oil and gas with water were recognized by Griswold and Munn (1907), but they believed that there was a definite separation of the oil and water, and that oil, gas, and water mixtures did not occur in the sand before a well tapped the reservoir. It was not until 1928 that the first commercial laboratory for the analysis of rock cores was established (Torrey, 1966); the first core tested was from the Bradford Third Sand (from the Bradford field, McKean County, Pennsylvania). The percent saturation and percent porosity of this core were plotted versus depth t o construct a graphic representation of the oil and water saturation. The soluble mineral salts that were extracted from the core led Torrey to suspect that water was indigenous t o the oil productive sand. Shortly thereafter a test well was drilled near Custer City, Pennsylvania, which encountered higher than average oil saturation in the lower part of the Bradford Sand. This high oil saturation resulted from the action of an un-
2
INTRODUCTION
suspected flood, the existence of which was not known when the location for the test well had been selected. The upper part of the sand was not cored. Toward the end of the cutting of the first core with a Baker cable tool core barrel, oil began t o come into the hole so fast that it was not necessary t o add water for the cutting of the second section of the sand. The lower 1 m of the Bradford Sand therefore was cut with oil in a hole free from water. Two samples from this section were preserved in sealed containers for saturation tests, and both of them, when analyzed, had a water content of approximately 20% of pore volume. This well made about 10 barrels of oil per day and no water after being shot with nitroglycerine. Thus, the evidence developed by the core analysis and the productivity test after completion provided a satisfactory indication of the existence of immobile water, indigenous t o the Bradford Sand oil reservoir, which was held in its pore system and which could not be produced by conventional pumping methods (Torrey, 1966). Fettke (1938) was the first t o report the presence of water in an oilproducing sand. However, he thought that it might have been introduced by the drilling process. It was recognized by Munn (1920)that moving underground water might be the primary cause of migration and accumulation of oil and gas. However, this theory had little experimental data t o back it until Mills (1920)conducted several laboratory experiments on the effect of moving water and gas on water-oil-as-sand and water-oil-sand systems. Mills concluded that “the up-dip migration of oil and gas under the propulsive force of their buoyancy in water, as well as the migration of oil, either up or down dip, caused by hydraulic currents, are among the primary factors influencing both the accumulation and the recovery of oil and gas.” This theory was seriously questioned and completely rejected by many of his contemporaries. Rich (1923)postulated that “hydraulic currents, rather than buoyancy, are effective in causing accumulation of oil or its retention.” He did not believe that the hydraulic accumulation and flushing of oil required a rapid movement of water, but rather that the oil was an integral constituent of the rock fluids and that it could be carried along with them whether the movement was very slow or relatively rapid. The effect of water displacing oil during production was not recognized in the early days of the petroleum industry in Pennsylvania. Laws were passed, however, to prevent operators from injecting water into the oil reservoir sands through unplugged wells. In spite of these laws, some operators at Bradford surreptitiously opened the well casing opposite shallow groundwater sands in order to start a waterflood in the oil sands. Effects of artificial waterfloods were noted in the Bradford field, McKean County, Pennsylvania, in 1907, and became evident about 5 years later in the nearby oilfields of New York (Torrey, 1950). Volumetric calculations of the oil-reservoir volume which were made for engineering studies of these waterflood opera-
a
INTRODUCTION
3
tions proved that interstitial water was generally present in the oil sands. Publications by Garrison (1935) and Schilthuis (1938) give detailed information concerning the distribution of water and oil in porous rocks, and of the origin and occurrence of “connate” water with information concerning the relationship of water saturation to formation permeability. The word “connate” was first used by Lane and Gordon (1908) to mean interstitial water that was deposited with the sediments. The processes of rock compaction and mineral diagenesis result in the expulsion of large amounts of water from sediments and movement out of the deposit through the more permeable rocks. It is therefore highly unlikely that the water now in any pore is the same as that which was there when the particles that surround it were deposited. White (1957) redefined connate water as “fossil” water; it has been out of contact with the atmosphere for an appreciable part of a geologic time period. Connate water is thus distinguished from meteoric water which has entered the rocks in geologically recent times, and from juvenile water which has come from deep in the earth’s crust and has never been in contact with the atmosphere. Meanwhile petroleum engineers and geologists had learned that waters associated with petroleum could be identified with regard to the reservoir in which they occurred by a knowledge of their chemical characteristics. Commonly the waters from different strata differ considerably in their dissolved chemical constituents, making the identification of a water from a particular strata easy. However, in some areas the concentrations of dissolved constituents in waters from different strata do not differ significantly, and the identification of such waters is difficult or impossible. The amount of water produced with the oil often increases as the amount of oil produced decreases. If this is edge water, nothing can be done about it. If it is bottom water, the well can be plugged back. However, it often is intrusive water from a shallow sand gaining access t o the well from a leaky casing or faulty completion and this can be repaired. Enormous quantities of water are produced with the oil in some fields, and it is necessary to separate the oil from the water. Most of the oil can be removed by settling. Often, however, an oil-in-water emulsion forms which is very difficult t o break. In such cases, the oil is heated and various surfaceactive chemicals are added to induce separation. In the early days, the water was dumped on the ground where it seeped below the land surface. Until about 1930, the oilfield waters were disposed into local drainage, frequently killing fish and even surface vegetation. After 1930, it became common practice t o evaporate the water in earthen pits or to inject it into the producing sand or another deep aquifer. The primary concern in such disposal practice is to remove all oil and basic sediment from the waters before pumping them into injection wells, to prevent clogging of the pore spaces in the formation receiving the waste water. Chemical compatibility of waste water and host aquifer water must also be assured. Waters produced with petroleum are growing in importance. In years past,
4
INTRODUCTION
these waters were considered waste and had t o be disposed of in some manner. Injection of these waters into the petroleum reservoir rock serves three purposes: (1) it produces additional petroleum (secondary recovery); (2) it utilizes a potential pollutant; and (3) in some areas it controls land subsidence. The volume of water produced with petroleum in the United States is very large. In 1970, daily production was about 3.78 trillion liters of water with about 1.51 trillion liters of oil. In older fields, the production is frequently 95%water and 5%petroleum. . Waterflooding in petroleum production is expanding rapidly, and in 1970 one-third to one-half of the production in the United States came from fields into which water was injected. The volume of injected water has grown each year. In many fields the volume of petroleum produced by secondary recovery by waterflooding is equal to the volume recovered by primary met hods. To inject these waters into reservoir rocks, suspended solids and oil must be removed from the waters to prevent plugging of the porous formations. Water injection systems require separators, filters, and, in some areas, deoxygenating equipment utilizing chemical and physical control methods to minimize corrosion and plugging in the injection system. In waterflooding most petroleum reservoirs, the volume of produced water is not sufficient t o efficiently recover the additional petroleum. There. fore, supplemental water must be added t o the petroleum reservoir. The use of waters from other sources requires that the blending of the produced water with supplemental water must yield a chemically stable mixture so that plugging solids will not be formed. For example, a produced water containing considerable calcium should not be mixed with a water containing considerable carbonate because calcium carbonate may precipitate and prevent injection of the flood water. The design and successful operation of a secondary recovery waterflood requires a thorough knowledge of the composition of the waters used. Chemical analyses of waters produced with oil are useful in oil production problems, such as identifying the source of intrusive water, planning waterflood and salt-water disposal projects, and treating t o prevent corrosion problems in primary and secondary recovery. Electrical well-log interpreta tion requires a knowledge of the dissolved solids concentration and composi tion of the interstitial water. Such information also is useful in correlationof stratigraphic units and of the aquifers within these units, and in studiesof the movement of subsurface waters. It is impossible to understandthe processes that accumulate petroleum or other minerals without insight in to the nature of these waters.
REFERENCES
5
References Dickey, P.A., 1959. The first oil well. J. Pet. Technol., 11:14-26. Fettke, C.R., 1938. Bradford oil field, Pennsylvania, and New York. Pa. Geol. Surv., Fourth Ser., Bull., M21:l-454. Garrison, A.D., 1935. Selective wetting of reservoir rock and its relation to oil production. In: Drilling and Production Practice. American Petroleum Institute, New York, N.Y., pp.130-140. Griswold, W.T. and Munn, M.J., 1907. Geology of oil and gas fields in Steubenville, Burgettstown and Claysville Quadrangles, Ohio, West Virginia and Pennsylvania. U.S. Geol. Sum. Bull., No.318, 196 pp. Howell, J.V., 1934. Historical development of the structural theory of accumulation of oil and gas. In: W.E. Wrather and F.H. Lahee (Editors), Problems of Petroleum Geology. American Association of Petroleum Geologists, Tulsa, Okla., pp.1-23. Lane, A.C. and Gordon, W.C., 1908. Mine waters and their field assay. Bull. Geol. SOC. A m . , 19:501-512. Mills, R. van A., 1920. Experimental studies of subsurface relationships in oil and gas fields. Econ. Geol., 15:398-421. Munn, M.J., 1920. The anticlinal and hydraulic theories of oil and gas accumulation. Econ. Geol., 4:509-529. Rich, J.L., 1923. Further notes on the hydraulic theory of oil migration and accumulation. Bull. Am. Assoc. Pet. Geol., 7:213-225. Rogers, W.B. and Rogers H.D., 1843. On the connection of thermal springs in Virginia with anticlinal axes and faults. A m . Geol. Rep., 1313.323-347. Schilthuis, R.J., 1938. Connate water in oil and gas sands. In: Petroleum Development and Technology, AIME, pp.199-214. Torrey, P.D., 1950. A review of secondary recovery of oil in the United States. In: Secondary Recovery of Oil in the United States. American Petroleum Institute, New York, N.Y., pp.3-29. Torrey, P.D., 1966. The discovery of interstitial water. Prod. Monthly, 30:8-12. White, D.E., 1957. Magmatic, connate, and metamorphic water. Bull. Geol. SOC.A m . , 68:1659-1682.
This Page Intentionally Left Blank
Chapter 2.
SAMPLING SUBSURFACE OILFIELD WATERS
Subsurface waters associated with petroleum are subjected to forces that promote mixing and homogeneity, but the assumption cannot be made that they are so well mixed that no attention to sampling technique is required. Localized conditions within an aquifer are commonly such that a given subsurface body of water may not be of uniform composition. The composition of subsurface water commonly changes with depth, and also laterally in the same aquifer. Changes may be brought about by the intrusion of other waters, and by discharge from and recharge to the aquifer. It is thus difficult to obtain a representative sample of a given subsurface body of water because any one sample is a very small part of the total mass, which may vary widely in composition. To develop a comprehensive picture of the composition characteristics of the total mass, it is generally necessary t o obtain and analyze many samples. Also, the samples may change with time as gases come out of solution and supersaturated solutions approach saturation. The sampling sites should be selected, if possible, t o fit into a comprehensive network t o cover an oil-productive geologic basin. Considerations in selecting sampling sites are as follows: (1)Which sites will better fit into an overall plan to evaluate the chemistry of the waters on a broad basis? (2) Which sites will yield the better information for correlation with data obtained from other sites? (3) Which sites are more representative of the total chemistry of brines from a given area? The value of the sample is directly proportional to the facts known about its source; therefore, sites should be selected for which the greater source knowledge is available. For surveillance purposes, samples can be collected from the same site at sufficiently frequent intervals that no important change in quality will occur between sampling times. Change in composition may result from changes in rate of water movement, pumpage rates, or infiltration of other water. Changes that can occur in petroleum-associated water are illustrated in Table 2.1. Well 1 shows the sort of change that commonly occurs. The water from well 2 did not change between 1947 and 1957, within the accuracy of the analytical determination. Water from well 3 changed drastically, suggesting the intrusion of water from a different source.
SAMPLING SUBSURFACE OILFIELD WATERS
8 TABLE 2.1
Changes in the composition of petroleum-associated waters (mg/l) Constituent
Well 2
Well 1
Well 3
1947
1957
1947
1957
1956
1959
Sodium and potassium Magnesium Calcium Bicarbonate Sulfate Chloride
29,062 1,100 5,900 34 14 58,500
25,000 1,200 5,500 12 50 51,800
46,038 2,011 14,200 24 3 102,100
45,924 2,200 14,400 12 52 102,800
1,491 30 60 600 200 2,000
856 2 10 1,800 0 300
Total dissolved solids
94,610
83,562
164,376
165,388
4,381
2,968
~-
-
_.
There is a tendency for some petroleum-associated waters t o become more dilute as the oil reservoir is produced. Such dilution may result from the movement of dilute water from adjacent compacting clay beds into the petroleum reservoir as pressure declines with the continued removal of oil and brine (Wallace, 1969). The composition of petroleum-associated water varies with the position within the geologic structure from which it is obtained. For example, if the water table is tilted, the more dilute water probably will be on the structurally high side. In some cases the salinity will increase upstructure t o a maximum at the point of oil-water contact. Usually this is caused by infiltrating meteoric waters. Few of the samples collected by drill-stem test are truly representative formation-water samples. During drilling, the pressure in the well bore is intentionally maintained higher than that in the formations. Filtrate from the drilling mud seeps into the permeable strata, and this filtrate is the first liquid to enter the test tool. The most truly representative formation-water sample usually is obtained after the oil well has produced for a period of time and all extraneous fluids adjacent t o the wellbore have been flushed out. Samples taken immediately after the well is completed may be contaminated with drilling muds, with drilling fluids, and/or with well completion fluids, such as filtrate from cement, tracing fluids, and acids, which contain many different chemicals.
Drill-stem test The drill-stem test, if properly made, can provide a reliable formation water sample. Mud filtrate will be the first fluid to enter the drill-stem test tool, and it will be found at the top of the fluid column immediately below
DRILL-STEM TEST
9
tester
Multiple closed i pressure s a m p l e
Flow stream pressure recorde V e r t i c a l and rot
Locked down
Blanked o f f pressure r e c o r d
RUNNING IN HOLE
Fig. 2.1. Multiple closed-in-pressure subsurface sampler. (Courtesy of the Halliburton Company.)
10
SAMPLING SUBSURFACE OILFIELD WATERS Droin YOIV4
' Rubber doughnut
Mud
Oroln
Floottnq p l s t o n h
SAMPLE UNIT,' for l o w p e r m o b i l i t i e s
Sample
v o l v e ( l o c k s open)
Dump chamber
"
L
MECHANICAL U N I T F I M - A
- J
SAMPLE U N I T
RECORDED TESTER POSITION
m E ? SP
Tester positioning depth
SURFACE CONTROL I N 0 ICATIDNS
S A M P L I N G PRESSURE Pod set
Tool open
I n i t i a l shut-in prss1ure
action
Sampling pressure
3
F i n a l shul-in pressure Hydroltotic head
RECORDED LOG
Fig. 2.2. Formation interval tester. (Courtesy of Schlumberger Well Services Company.)
DRILL-STEM TEST
11
the oil. At some point down the column a representative formation-water sample can be found. The point is variable and will be influenced by rock characteristics, mud pressure, type of mud, and duration of the test. It is best to sample the water after each stand of pipe is removed. Normally, the total dissolved solids content will increase downwards and become constant when pure formation water is obtained, if the concentration continues t o increase all the way t o the bottom, no representative sample can be obtained. A test that flows water will give even higher assurance of an uncontaminated sample. If only one drill-stem test water sample is taken for analysis, it should be taken just above the tool, as this is the last water to enter the tool and is least likely t o show contamination. Fig. 2.1 and 2.2 illustrate two drill-stem test tools with their various components. Fig. 2.1 illfistrates a Halliburton Company tool; Fig. 2.2 illustrates a Schlumberger Well Services Company tool. Other companies supply equally adequate tools, and reference t o specific brands throughout this test is made for identification only and does not imply endorsement by the US. Bureau of Mines. The drill-stem test can provide pressure head and head decline and buildup data useful in permeability calculation (Bredehoeft, 1965) and other information for the determination of additional reservoir conditions, such as the gas/oil ratio and reservoir depletion (McAlister et al., 1965). A stratigraphic interval of interest is isolated in the drilled hole by use of packers attached t o the drill string. Opening the tester valve in the test string allows the formation fluid to enter the drill pipe. Pressures are recorded by gages in the bottom of the test tool. To insure that a representative sample is obtained, the pH, resistivity, and chloride content of samples taken at intervals down the drill pipe can be determined. Usually a transition zone will be found below which apparently uncontaminated formation water will be located. The pH, resistivity, and chloride content will vary above the transition zone, and they will become constant below it. The sample taken for analysis in the laboratory can yield positive evidence of contamination, if present. The two most indicative tests are pH and the color of a filtered sample. If the filtered sample remains tan or brown and the color cannot be removed even with pressure filtration, it probably is contaminated with drilling-mud filtrate. A sample can be placed on a white-spot plate for color evaluation. For positive identification of the presence of mud filtrate, a sample of the drilling mud used in drilling the well can be obtained and allowed t o react with distilled water, the reacted water .is analyzed to determine the mud-contributed ions, and the suspected contaminated sample is analyzed t o determine if it contains these ions. Analyses of water obtained from a drill-stem test of Smackover Limestone water in Rains County, Texas, show how errors can be caused by improper sampling of drill-stem test water. Analyses of top, middle, and bottom samples taken from a 15-m fluid recovery are shown in Table 2.11. These data show an increase in salinity with depth in the drill pipe, indicating that the first water was contaminated by mud filtrate (Noad, 1962). The middle
SAMPLING SUBSURFACE OILFIELD WATERS
12 TABLE 2.11
Drill-stem test recovery of Smackover Limestone water Constituent
Concentration (mg/l) top
Sodium Calcium Magnesium Bicarbonate Sulfate Chloride Total dissolved solids
middle -
bottom
29,600 8,100 600 500 2,000 59,900
43,500 13,100 900 500 1,300 91,800
71,800 22,400 1,400 400 500 154,000
101,000
151,000
251,000
sample is approximately half mud filtrate and half formation water. The bottom sample is the most representative of Smackover water. No single procedure is universally applicable for obtaining a sample of oilfield water. For example, information may be desired concerning the dissolved gas or hydrocarbons in the water, or the reduced species present such as ferrous or manganous compounds. Sampling procedures applicable to the desired information must be used. Sample containing dissolved gases Knowledge of certain dissolved hydrocarbon gases is used in exploration. Methane is quite soluble in water, but samples must be collected in a sampler that keeps the subsurface pressure on the sample until it is opened in the laboratory. The testing tool is kept open until the head of water in the drill pipe is equalized with the formation pressure or until water flows at the surface. The pressure equalization may require 4 or more hours. However, a surface recording subsurface pressure gage can be lowered into the drill pipe to determine when the pressure has equalized. After equalization of pressure, formation-water samples can be obtained by lowering a subsurface sampler into the drill pipe (Buckley et al.,1958). Zarrella et al. (1967) determined the content of dissolved benzene. For this it is not necessary t o use a subsurface sampler; the samples are caught in buckets on opening the pipe string, and immediately transferred from the buckets t o new narrow-necked glass or metal containers. A preferred method of obtaining a sample for subsequent gas analysis is t o catch the aqueous sample in a metal container of about one-quart capacity. This sample is immediately transferred to another metal sample container. The second container should be filled completely t o the top, then the sides of the can are lightly squeezed t o allow for fluid expansion, and the lid is sealed tightly. A foil-lined (not plastic) lid should be used. If possible, the
13
SAMPLING AT THE FLOW LINE
sample should be analyzed immediately. If this is not possible, cool or freeze the sample. Sampling at the flow line Another method of obtaining a sample for analyses for dissolved gases is to place a sampling device in a flow line. Fig. 2.3 illustrates such a device.
-I
container
Valve P i p e line
f
Valve
Rubber tube
-
Fig. 2.3. Flow-line sampler.
The device is connected to the flow line, and water is allowed t o flow into and through the container, which is held above the flow line, until 10 or more volumes of water have flowed through. The lower valve on the sample container is closed and the container removed. If any bubbles are present in the sample, the sample is discarded and a new one is obtained.
SAMPLING SUBSURFACE OILFIELD WATERS
14
Sampling at the wellhead It is common practice in the oil industry to obtain a sample of formation water from a sampling valve at the wellhead. A plastic or rubber tube can be used to transfer the sample from the sample valve into the container. The source and sample container should be flushed t o remove any foreign material before a sample is taken. After flushing the system, the end of the tube is placed in the bottom of the container, and several volumes of fluid are displaced before the tube is slowly removed from the container and the container is sealed. Fig. 2.4 illustrates a method of obtaining a sample at the wellhead. An extension of this method is to place the sample container in a larger container, insert the tube to the bottom of the sample container, allow the brine to overflow both containers, withdraw the tube, and cap the sample under the fluid. At pumping wellheads the brine will surge out in heads and will be mixed with oil. In such situations a larger container equipped with a valve at the bottom can be used as a surge tank or an oil-water separator, or both. To use this device, place the sample tube in the bottom of the large container, open the wellhead valve, rinse the large container with the well fluid, allow the large container t o fill, and withdraw a sample through the valve at the bottom of the large container. This method will serve to obtain samples that are relatively oil-free. We1 l h e a d
O i l and water
Fig. 2.4. Schematic of method of obtaining a sample at the wellhead.
Sample for determining unstable properties or species The pH, Eh, and various species of elements are unstable and will change with changes in pressure and temperature, and when the formation water is exposed to the atmosphere. The pH of the sample will change because of the oxidation of reduced species, because of release of dissolved gases, and because of hydrolysis reactions such as: H+
c03-*
+ HCO,-+
H+
H,CO,
SAMPLE FOR STABLE-ISOTOPE ANALYSIS
15
Because the pH of the formation water sample will change, the pH should be determined using a flowing sample. A pH/Eh flow sampling chamber (Collins, 1964) is shown in Fig. 2.5. The Eh determination is difficult and for corroboration it should be checked using a knowledge of the dissolved
Fe+*/Fe+3ratio of the water. Ther mocompensator
Fig. 2.5. Flow chamber for use in determining pH and Eh at the wellhead.
Ferrous iron will oxidize to ferric and should be determined immediately after collecting a fresh sample. Some of the other dissolved constituents that should be determined immediately after securing a fresh sample are oxygen, hydrogen sulfide, thiosulfate, and manganous manganese.
Sample for stable-isotope analysis A sample that is t o be analyzed for stable isotopes should be collected with care. If possible, such a sample should be taken at reservoir temperatures and pressures to minimize any isotope fractionation. However, because this usually is impossible, caution should be exercised t o insure that a representative sample is collected at the prevailing wellhead temperature and pressure. The sample should be collected at the wellhead. If this proves impossible,
16
SAMPLING SUBSURFACE OILFIELD WATERS
it may be feasible t o collect the sample from a nonheated separator or heater; samples are not to be taken of water that has been heated or treated with any chemicals. Glass sample bottles (about 100 ml usually is sufficient) should be used, and the sample should overflow the bottle. The bottle should be closed with a cap equipped with a plastic insert, and the top should be sealed with wax to minimize exchange reactions with air. Sample containers Various factors influence the type of sample container that is selected. Containers that are used include polyethylene, other plastics, hard rubber, metal cans, and borosilicate glass. Glass will absorb various ions such as iron and manganese, and may contribute boron or silica to the aqueous sample. Plastic and hard rubber containers are not suitable if the sample is to be analyzed to determine its organic content. A metal container is used by some laboratories if the sample is to be analyzed for dissolved hydrocarbons such as benzene. TABLE 2.111 Description needed for each petroleum-associated water sample __
--
Sample number Field Farm or lease Well No. in the of Section Township Range __ County State Operator Operator’s address (main office) -___ Date Sample obtained by -_____ Address Representing Sample obtained from (lead line, separatory flow tank, etc.) Elevation of well ___ Completion date of well Name of productive zone from which sample is produced Sand -Shale Lime Other Names of formations Name of productive formation well passes through Bottom of formation Depths: Top of formation Top of producing zone Bottom of producing zone Total depth drilled Present depth Bottom hole pressure and date of pressure Bottom hole temperature b e any chemicals If yes, Date of last workover added to treat well? -what? Well production Initial Present Casing service record: Oil, barrelslday Water, barrels/day _______ Gas, cubic feetlday Method of production (primary or secondary)
__
~
~
~
~
Remarks: (such as casing leaks, communication, or other pays in same well, lease, or field)
TABULATION OF SAMPLE DESCRIPTION
17
The type of container selected is dependent upon the planned use of the analytical'data. Probably the more satisfactory container, if the sample is to be stored for some time before analysis, is the polyethylene bottle. All polyethylenes are not satisfactory because some contain relatively high amounts of metals contributed by catalysts in their manufacture. The approximate metal content of the plastic can be determined using a qualitative emission spectrographic technique. If the sample is transported during freezing temperatures, the plastic container is less likely to break than glass. The practice of obtaining two samples and acidifying one sample so that the heavy metals will stay in solution works better if the plastic container is used. Some of the heavy metals are adsorbed by glass even if the sample is acidified. Tabulation of sample description The sample is of little value if detailed information concerning it is not available. Information such as that in Table 2.111 should be obtained for each sample of petroleum-associated water, and for certain types of studies, additional information may be needed. References Bredehoeft, J.D., 1965. The drill-stem test: the petroleum industry's deep-well pumping test. Ground Water, 3:15-23. Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958.Distribution of dissolved hydrocarbons in subsurface waters. In: L.C. Weeks (Editor), Habitat of Oil. American Association Petroleum Geologists, Tulsa, Okla., pp.850-882. Collins, A.G., 1964. Eh and pH of oilfield waters. Prod. Monthly, 28:ll-12. McAlister, J.A., Nutter, B.P. and Lebourg, M., 1965. A new system of tools for better control and interpretation of drill-stem tests. J. Pet. Technol., 17 :207-214. Noad, D.F., 1962. Water analysis data, interpretation and applications. J. Can. Pet. TechnoL , 1 :82-89. Wallace, W.E., 1969. Water production from abnormally pressured gas reservoirs in South Louisiana, J. Pet. Technol., 21 :969-982. Zarrella, W.M., Mousseau, R.J., Coggeshall, N.E., Norris, M.S. and Schrayer, G.T., 1967. Analysis and significance of hydrocarbons in subsurface brines. Geochim. Cosmochim. Acta, 31 :1155-1166.
This Page Intentionally Left Blank
Chapter 3. ANALYSIS OF OILFIELD WATERS FOR SOME PHYSICAL PROPERTIES AND INORGANIC CHEMICAL CONSTITUENTS
Water analyses are used by the petroleum industry in studies related to subsurface formation identification, pollution problems, water compatibilities, corrosion, water-quality control, waterflooding, and exploration. Efforts to standardize methods applicable to analyzing oilfield waters have been made by the American Petroleum Institute (1968),and currently similar efforts are being made by the American Society for Testing and Materials. The methods discussed in this chapter include wet chemical procedures for calcium, magnesium, barium, carbon dioxide, sulfide, sulfur compounds, selenium, oxygen, spent acid, fluoride, chloride, bromide, and iodide. Instrumental methods are described for pH, Eh, specific gravity, resistivity, suspended solids, acidity, alkalinity, oxygen isotopes, ammonium nitrogen, phosphate, boron, arsenic, copper, nickel, lead, manganese, zinc, cadmium, and silica. Also described are emission and atomic absorption methods for lithium, sodium, potassium, rubidium, cesium, magnesium, calcium, barium, manganese, zinc, copper, iron, and lead; and emission spectroscopic methods for aluminum, beryllium, boron, iron, manganese, and strontium. ._ The methods used to analyze oilfield waters should be capable of producing precise and accurate results. Methods applicable to analyzing fresh waters may or may not be directly applicable to a petroleum-associated water, but in general such a method will need modification or complete redevelopment because the petroleum-associated water contains a more complex and concentrated array of dissolved salts than the fresh water. Quality control Data provided by the analytical laboratory are used in decision-making, and the data must describe precisely and accurately the characteristics or concentrations of the constituents in the sample. Usually an approximate or incorrect result is less valuable than no result because it leads t o faulty interpretations. The analyst needs t o be aware of his responsibility to provide results that reliably describe the sample. Further, he should know that the procedures that he uses, his professional competence, and his reported values may be used or challenged. To meet any challenge his results must be adequately
20
ANALYSIS OF OILFIELD WATERS
documented. The value of research investigations which use oilfield brine analyses depends upon the validity of the laboratory results. A program to insure the reliability of analytical data is mandatory because of the importance of the laboratory results and the actions that they produce. An established routine control program applied to analytical tests is necessary to assure the precision and accuracy of the final results. The use of spiked samples can measure quality, while the use of analytical-grade reagents is a control measure. Quality control varies with the type of as*sis. For example, the frequent standardization of the titrant used in a titration is an element of quality control, while instrument calibration in an instrumental method is also a quality control function. The specific methodology employed should be carefully documented regardless of the method used; thus the data user or reviewer can apply the associated precision and accuracy when interpreting the laboratory data.
Choosing an analytical method Widespread use of an analytical method indicates that it probably is reliable and will produce valid results. Use of a little-known procedure forces the data user t o accept the judgment of the analyst. The following criteria are useful in selecting analytical methods: (1)The desired constituent should be measured with sufficient precision and accuracy in the presence of the interferences normally found in oilfield waters. (2) The method must utilize the skills and equipment available in the oilfield water laboratory. (3) The method should be sufficiently tested and used by several laboratories to establish its validity.
Precision Precision is the reproducibility among replicate observations, and in quality control it is determined on actual water samples containing interfering constituents. Several methods to determine precision are available and the following is representative: (1) Study four separate concentration levels, including a low concentration near the sensitivity level of the method, two intermediate concentrations, and a concentration near the upper limit of application of the method. (2) Make seven replicate determinations at each of the concentrations tested. (3) To allow for changes in conditions, the precision study should use at least 2 hours of normal laboratory operation. (4) To permit the maximum interferences in sequential operation, the samples should be run in the following order: high, low, intermediate, intermediate. Repeat this series seven times to obtain the desired replication.
21
QTJALITY CONTROL TABLE 3.1 Precision data on oilfield brine samples for boron
___ ~~
Concentrations of boron found (mg/l)
Sample
Taylor -
.
Eagle Ford
-
Average Standard deviation
Paluxy
_ _ _ _-
Douglas -
10.1 10.1 10.2 10.3 10.1 10.2 10.2
15.2 15.3 15.1 15.2 15.3 15.2 15.1
20.1 20.1 20.3 20.2 20.3 20.3 20.1
30.3 30.2 30.1 30.1 30.3 30.2 30.1
10.2 0.1
15.2 0.1
20.2 0.1
30.2 0.2
(5) The precision statement includes a range of standard deviations over the t A e d range of concentrations. Thus, four standard deviations are obtained over a range of four concentrations, but the statement contains only the extremes of standard deviations and concentrations studied. An example of data generated from such an approach is shown in Table 3.1. Using the data of Table 3.1 the precision statement would read: “In a single ldboratory, using oilfield water samples containing concentrations of 10.2 and 30.2 mg B/1, the standard deviation was kO.1.” Accuracy The degree of difference between observed and actual values is accuracy. The accuracy of a method can be determined as follows: (1) Add known amounts of the constituent t o be determined to actual samples at concentration levels where the precision of the method is adequate. The added amount should double the concentration of the lowconcentration sample and bring the concentration of an intermediate sample to about 75% of the upper limit of application of the method. (2) Make seven replicate determinations at each concentration. (3) Report the accuracy as the percent recovery found in the spiked sample, where the percent at each concentration is the mean of the seven replicate tests. Table 3.11 illustrates the application of this approach, where two of the samples used in the precision study, Table 3.1, were used. An appropriate accuracy statement is: “In a single laboratory, using oilfield water samples containing concentrations of 20.2 and 35.3 mg B/1, recoveries were 100.0% and 100.3%,respectively.”
ANALYSIS OF OILFIELD WATERS
22
TABLE 3.11 Accuracy data on oilfield brine samples for boron Sample
Concentrations of boron found (mg/l) Taylor (added 10 mg/l boron)
Paluxy (added 15 mg/l boron)
20.2 20.2 20.1 20.1 20.3 20.3 20.4
35.5 35.4 35.2 35.2 35.3 35.2 35.1
Average
20.2
35.3
Percent recovery
20-2 ] x 100 = 100.0 [ 10.2 + 10
[ 20.2 + 15
35.3
] x 100 = 100.3
The precision and accuracy data are valuable in determining that the analyst and the method are capable of generating valid data. Once this is proven, the data can be used to evaluate systematic performance. This can be done by using spiked samples about 10% of the time to determine that the accuracy is favorable, and evaluating replicate samples to determine that the precision is favorable. Preliminary sample treatment The following determinations should be made in the field immediately after sampling: (1) temperature (in "C),(2) pH, (3) dissolved oxygen, (4) resistivity, ( 5 ) acidity, (6) alkalinity, (7) sulfide, and (8) carbon dioxide. If possible, the oilfield water sample should be filtered immediately after sampling in the field. A preferred method-is to use pressure filtration through a 0.45-micrometer (pm) membrane filter. A liter of filtrate usually is sufficient and the following determinations can be made on aliquots: (1) iodide, (2) bromide, (3) chloride, (4) selenium, ( 5 ) sulfate, (6) nitrogen, (7) phosphate, (8) silica, (9) boron, (10) potassium, (11) sodium, and (12) lithium. If a field-filtered sample cannot be provided, a laboratory-filtered sample may be substituted with slightly less confidence in the reported data.
PRELIMINARY SAMPLE TREATMENT
23
Standard solutions Examples of standardization procedures are given for some of the methods. The concentrations of standard solutions are indicated as the weight of a given element equivalent to, or contained in, 1 ml of solution. The strength of acids and bases are given in terms of molarities or normalities.
Accuracy of measurements In the instructions for making the analysis and preparing the solutions, significant figures are utilized to define the accuracy of weights and measures. Required accuracy for measurement of volume in the analysis and preparation of reagents is also shown. Standard solutions are prepared in and measured from volumetric glassware.
Reagent chemicals and solutions All of the chemicals used in the analytical procedures should conform to the specifications of the Committee on Analytical Reagents of the American Chemical Society. Chemicals not listed by this organization can be tested according to procedures given by Rosin (1955). Primary standard chemicals can be obtained from the National Bureau of Standards or from companies marketing chemicals of the same purity. Water used to dilute samples or t o prepare chemical solutions should first be demineralized by passage through mixed cation-anion exchange resins or by distillation. Its specific conductance a t 25°C must not exceed 1.5 pmho/ cm, and it should be stored in polyethylene bottles. Carbon-dioxide-free water may be prepared by boiling and cooling demineralized water immediately before use. Its pH should be between 6.2 and 7.2. Ammonia-free water should be prepared by passing distilled water through a mixed-bed ion-exchange resin.
Sampling A field-filtered acidified sample also should be taken. It is pressured filtered using a 0.45-pm membrane filter and then the filtrate is immediately acidified to a pH of 3.0 or less with reagent-grade HCl. The acidified sample is used for the following determinations: (1) aluminum, (2) arsenic, (3) barium, (4) cadmium, (5) calcium, (6) copper, (7) iron, (8) lead, (9) magnesium, (10) manganese (11)nickel, (12) strontium, and (13) zinc.
ANALYSIS OF OILFIELD WATERS
24 TABLE 3.111 Units in which water analyses may be reported milligrams per liter = mg/l
1 grain per U.S. gallon = 17.12 mg/l
part per million = ppm
1 grain per Imperial gallon = 14.3 mg/l
milligrams per liter ppm = specific gravity of the water
1 ppm = 0.012 milligram atom per liter
T o convert compounds expressed as parts per million t o ions expressed,as parts per million (where compound is A, Bm): ppm ion A = ppm compound A,
,(atomic weight A) Bm molecular weight A, B,
ppm ion B = ppm compound A,
,(atomic weight B) Bm molecular weight A, , B
To convert parts per million t o equivalents per million (epm): Example: sample contains 28.3 ppm Ca” , what is the concentration of calcium in epm? Solution; atomic weight Ca = 40.08; valence = 2; equivalent weight =
40.08 2 28.3 ppm Ca+’ =
= 20.04; then:
28 3 20.04
= 1.41 epm Ca+’
Titrime tric analysis (ml x N of standard solution) x
milliequivalent weight of determined ion ml of sample used
106
= mg/l of determined ion
Gravimetric analysis (grams of preninitnta\ .,.F.’U”’,
weight of determined element ,. atomic molecular - weig ,ht of precipitate Y
..._.
ml of sample used %O
106
=. me/l . . _.- determined ~
element
= parts per thousand or g/kg = mass in grams of silver required to precipitate the halogens in
Chlorinity (CZ)
328.5233 g of sea water = total amount of solid material, in grams contained in 1 kg of
Salinity (S)
sea water when all of the bromide and iodide have been replaced by the equivalent amount of chloride, when all of the carbonate is converted t o oxide and when all the organic matter is completely oxidized %o
S = 1.805 x
%o
Cl + 0.03
REPORTING THE ANALYTICAL RESULTS
25
Reporting the analytical results A study conducted by the American Petroleum Institute (1968) indicated that some laboratories reported the results of oilfield water analysis as parts per million (ppm) or as milligrams per liter (mg/l) without regard to the specific gravity of the sample. For example, a sample with a specific gravity of 1.200 containing 12,000 mg/l of calcium does not contain 12,000 ppm of calcium but contains 12,000/1.200 = 10,000 pprn of calcium. Such an error obviously is more serious in reporting the analytical results for a brine than in reporting the results for a fresh water. The unit ppm means parts per million by weight, while the unit mg/l means milligrams per liter or weight per unit volume; therefore, they are not interchangeable until the volume is changed to a unit weight. Table 3.111 indicates the relation between various units of measurement. Because the American Petroleum Institute now recommends that oilfieldwater analysis be reported in units of mg/l, other associations will no doubt recommend the same uniform practice. Such standardization implements studies concerned with the chemistry and geochemistry of waters.
Sign i f ican t figures The term significant figure (Ballinger et al., 1972) is used rather loosely to describe some judgment of the number of reportable digits in a result. Often the judgment is not soundly based and meaningful digits are lost or meaningless digits are accepted. Proper use of significant figures gives an indication of the reliability of the analytical method used. The following definitions and rules are suggested for retention of significant figures. A number is an expression of quantity. A figure or digit is any of the characters 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, which, alone or in combination, serves to express a number. A significant figure is a digit that denotes the amount of the quantity in the place in which it stands. Reported values should contain only significant figures. A value is made up of significant figures when it contains all digits known to be true and one last digit in doubt. For example, if a value is reported as 18.8 mg/l, the “18” must be a firm value while the “0.8” is somewhat uncertain and may be “0.7” or “0.9”. The number zero m a y or may not be a significant figure: (a) Final zeros after a decimal point are always significant figures. For example, 9.8 g t o the nearest milligram is reported as 9.800 g. (b) Zeros before a decimal point with other preceding digits are significant. With no other preceding digit, a zero before the decimal point is not significant. (c) If there are no digits preceding a decimal point, the zeros after the
26
ANALYSIS OF OILFIELD WATERS
decimal point but preceding other digits are not significant. These zeros only indicate the pbsition of the decimal point. (d) Final zeros in a whole number may or may not be significant. A good measure of the significance of one or more zeros before or after another digit is to determine whether the zeros can be dropped by expressing the number in exponential form. If they can, the zeros are not significant. For example, no zeros can be dropped when expressing a weight of 100.08 g in exponential form; therefore, the zeros are significant. However, a weight of 0.0008 g can be expressed in exponential form as 8 x g, and the zeros are not significant. Significant figures reflect the limits of the particular method of analysis. It must be decided beforehand whether this number of significant digits is sufficient for interpretation purposes. If not, there is little that can be done within the limits of normal laboratory operations t o improve these values. If more significant figures are needed, a further improvement in method or selection of another method will be required t o produce an increase in significant figures. Once the number of significant figures is established for a type of analysis, data resulting from such analyses are reduced according t o set rules for rounding off.
R o unding-of f numbers Rounding off of numbers is a necessary operation in all analytical areas. It is automatically applied by the limits of measurement of every instrument and all glassware. However, it is often applied in chemical calculations incorrectly by blind rule or prematurely, and in these instances can seriously affect the final results. Rounding off should normally be applied only as follows.
Round ing-of f rules (a) If the figure following those to be retained is less than 5 , the figure is dropped, and the retained figures are kept unchanged. As an example, 11.443 is rounded off to 11.44. (b) If the figure following those to be retained is greater than 5, the figure is dropped, and the last retained figure is raised by 1.As an example, 11.446 is rounded off to 11.45. (c) When the figure following those t o be retained is 5 , and there are no figures other than zeros beyond the 5, the figure is dropped, and the last place figure retained is increased by 1 if it is an odd number, or it is kept unchanged if an even number. As an example, 11.435 is rounded off t o 11.44,while 11.425 is rounded off to 11.42. Rounding-off single arithmetic operations (a) Addition: when adding a series of numbers, the sum should be rounded off to the same number of decimal places as the addend with the
SYNTHETIC BRINE
27
smallest number of places. However, the operation is completed with all decimal places intact, and rounding off is done afterward. As an example, 11.1+ 11.12 + 11.13 = 33.35, and the sum is rounded off t o 33.4. (b) Subtraction : when subtracting one number from another, rounding off should be completed before the subtraction operation, to avoid invalidation of the whole operation. (c) Multiplication: when two numbers of unequal digits are t o be multiplied, all digits are carried through the operation; then the product is rounded off to the number of significant digits of the less accurate number. (d) Division: when two numbers of unequal digits are t o be divided, the division is carried out on the two numbers using all digits. Then the quotient is rounded off t o the number of digits of the less accurate of the divisor or dividend. (e) Powers and roots: when a number contains n significant digits, its root can be relied on-for n digits, but its power can rarely be relied on for n digits. Synthetic brine Synthetic brine solutions are used in many of the analytical procedures for analyzing oilfield waters (American Petroleum Institute, 1968). Such solutions are a necessity in the development of analytical methods to study the effects of possible interfering ions. Often these synthetic solutions are used as an integral part of the analytical technique (Collins, 1967). Preparation of a fairly stable synthetic brine involves saturating distilled water with carbon dioxide by bubbling carbon dioxide through it, followed by adding the bicarbonate and sulfate compounds to one portion of the C 0 2-saturated water, adding the alkali chlorides to one portion, and adding the alkaline earth chlorides to one portion. The alkali chloride solution is mixed with the bicarbonate-sulfate solution, and t o this mixture the alkaline earth chloride solution is added. Carbon dioxide is bubbled through the synthetic brine to mix it, and the synthetic brine container is sealed immediately after removing the carbon dioxide source. Determination of pH The pH of the water can be determined with a pH meter which utilizes the principle of measuring the electrical potential between an indicator electrode and a reference electrode (Potter, 1956, p.56). pH meters measure the electrical potential between two suitable electrodes immersed in the solution to be tested. The reference electrode assumes a constant potential, and the indicating electrode assumes a potential dependent on the pH of the solution. Electrode potential is the difference in poteptial between the electrode and the solution in which it is immersed. The calomel electrode, which is a widely used reference electrode in water analysis, consists of a mercurycalomel rod immersed in a saturated solution of potassium chloride; this
ANALYSIS OF OILFIELD WATERS
28
electrode has a potential of +0.246 V. Electrical connection with the sample is provided through porous fibers sealed into the immersion end. A hydrogen-ion-selective glass electrode is normally used as an indicating electrode. The glass electrode has several features that recommend it for pH measurements. Among the most important are that it is not affected by oxidizing or reducing substances in the sample and that it can be used to measure the pH of turbid samples and/or colloidal suspensions. The basic design is a silver-silver chloride or mercury-mercurous chloride electrode immersed in a solution of known pH and the whole completely sealed in glass. The mechanism by which the glass membrane responds to hydrogen-ion activity involves absorption of hydrogen ions on both sides of the membrane proportionally to the activity of the hydrogen ions in solution. The cell for measuring the pH of a solution is of the following type: Ago :AgC1
1I
solution of known pH;
glass membrane;
glass electrode
solution of unknown pH
I 1 Hgo :HgC1
The voltage of the glass electrode is a logarithmic function of the difference in hydrogen-ion activity of the solutions on either side of the glass membrane. To measure this voltage an electron-tube voltmeter is used because the resistance of the glass membrane is so great. The pH should be determined at the time of sampling. A device similar to that shown in Fig. 2.5, can be used, or the electrodes can be placed in a container and then a stream of the sample allowed to flow from the oilwater separator (Fig. 2.4.) into the container while the pH is measured. If accurate results are desired, at least two pH buffer solutions should be used to calibrate the pH meter and electrodes before determining the pH. Because TABLE 3.IV pH buffer solutions (pH values of NBS standards from 0-30°C) Temperature 0.5M potassium ("C) tetroxalate
0 10 15 20 25 30
1.67 1.67 1.67 1.68 1.68 1.69
Potassium acid tartrate (sat. at 25OC)
-
0.05M potassium acid phthalate
0.025M potassium dihydrogen phosphate + 0.025M sodium dihydrogen phosphate
0.01M sodium tetraborate
4.01 4.00
6.98 6.92 6.90 6.88 6.86 6.85
9.46 9.33 9.27 9.22 9.18 9.14
-
4.00
3.56 3.55
4.00 4.01 4.01
-
DETERMINATION O F Eh
29
TABLE 3.V Performance characteristics of typical pH meters Normal scale
Expanded scale
~~~
Range Smallest scale division Accuracy Reproducibility Temperature compensation Input impedance -
1 pH (* 100 mV)
0-14 pH (+ 1,400 mV) 0.1 pH (10 mV) f 0.05 pH (5 5 mV) f 0.02 pH (+ 2 mV)
0.005 pH (0.5 mV) f 0.002 pH (+ 2%of reading) f 0.002 pH (+ 0.2 mV)
O - I O O ~ C(manual or automatic) 1014
> 1013
>
the pH probably will fall between 5 and 7, the standard pH buffer solutions used could be for pH 5 and pH 7. Standard buffer solutions, covering a range of pH, may be purchased from almost any chemical supply house and are satisfactory for routine use. Table 3.IV gives a list of NBS buffers (easily made in the laboratory) and the resulting pH at several different temperatures. An idea of the effect of temperature on pH may be obtained by observing temperature versus pH of various buffers shown in Table 3.IV. Theoretically, the potential response of the electrode system changes 0.20 mV per pH unit per degree Celsius. Since all pH meters measure potential but read out in pH, a variable compensation is used. A rough rule of thumb is that temperature compensation is about 0.05 pH unit per 5' increase in temperature. Performance data of a conventional and an expanded scale pH meter are shown in Table 3.V. Determination of Eh The Eh, called the oxidation-reduction potential or the redox potential, is a measure of the relative intensity of oxidizing or reducing conditions in a chemical system. It is expressed in volts, and at equilibrium it is related to the proportions of oxidized and reduced species present. Standard equations of chemical thermodynamics express the relationships (Collins, 1964). Eo is the standard potential of a redox system when unit activities of participating substances are present under standard conditions. Eo is related to standard free energy change in a reaction by the equation:
A P
= -nfEo
where n is the number of unit negative charges (electrons) shown in the redox reaction and f is the Faraday constant in units that give a potential in volts (94,484absolute coulombs). Standard free energy values are given in texts such as that of Latimer (1952).
ANALYSIS OF OILFIELD WATERS
30
When the system is not under standard conditions, the redox potential is expressed by the Nernst equation: R T log oxidized species) Eh=Eo+ nf (reduced species) where R is the gas constant (1.987 calories per degree mole) and T is the temperature in degrees Kelvin. Geochemical literature and biochemical literature such as that of Pourbaix (1949) use increasing positive potential values to represent increasing oxidizing systems, and decreasing potential values to represent reducing systems. The sign of Eh used in this manner is opposite to standard American practice in electrochemistry. Reagents. An Eh standard' which can be used is a solution of M/300 K3Fe(CN)6 and M/300 KqFe(CN), in M/10 KC1 (Zobell, 1946). The Eh of this mixture is 0.430 V at 25'C. Equipment. A pH meter equipped with a thermometer, a glass electrode, a calomel electrode, a platinum electrode and a thermocompensating electrode. Eh flowchamber, a design similar to Fig. 2.5 can be used. Procedure. Buff the platinum electrode lightly with a fine abrasive cloth and wipe it carefully with a dry soft tissue. Install the glass electrode, the calomel electrode, the platinum electrode, the thermocompensator, and the thermometer in the flowchamber. Standardize the instrument using the Eh standard. Connect a line t o the wellhead or waterline and install an oil-water separator if oil and water both are present. Connect the flowchamber to the waterline, allowing the water to flow into the bottom and out the top. Make certain that all air bubbles are excluded at the top. Take at least three readings of the Eh (in mV), and the temperature at 10-minute intervals. These readings should agree; if they do not, continue making readings until three successive readings do agree. Make certain the water is continually flowing, that there are no air bubbles in the flowchamber, and that the solution is being stirred. It may be necessary t o remove and rebuff the platinum electrode. Calculation. Because a thermocompensator is used in determining the pH, a temperature correction need not be made. However, if a thermocompensator is not used, a temperature correction should be made. The Eh value is obtained by algebraically adding the measured voltage E and the voltage of the constant voltage reference electrode, which in this case is the saturated calomel electrode. The potential of the saturated calomel electrode at 25OC is 0.242 V. Therefore, if the millivolt reading of the sample is +300:
31
SUSPENDED SOLIDS
Eh = E + voltage of reference electrode Eh= +300+242=+542 However, if the millivolt reading is -300, then:
E = +0.242 - 7.6x
( t - 24)
Note: the calculations for Eh are correct only if the temperature of the brine is 25OC at the time of measurement. If the temperature is not 25"C, a correction should be made. For example, the potential of the saturated calomel electrode is 0.246 V at 20°C and 0.238 V at 30°C. The following formula can be used to obtain the correct potential:
E
= +0.242 - 7.6 x
( t - 24)
where t is in degrees Celsius. Suspended solids Various inorganic and organic materials are found in petroleum-associated water. Knowledge of the composition of such material is useful in determining the source of the material and what treatments can be used t o remove it or prevent it from recurring. Such material may be particles of oxides of the metals from well casings, pumps, or precipitates caused by oxidation of the formerly reduced species, such as iron or manganese. Other suspended solids may be silt, sand, and clay. An estimation of the amount of material in suspension can be accomplished by using a turbidimeter (Rainwater and Thatcher, 1960).This is done by comparing the intensity of light passing through the solution with the Tyndall effect produced by lateral illumination of the solution with the same source of light.
Turbidimeter Instruments for the measurement of turbidity employ principles of design related to transmission or reflectance of light. The lack of a primary standard for turbidity, however, has resulted in a complete absence of uniformity among the available instruments. Further, the Jackson candle turbidimeter, which does not depend upon the use of a primary standard, is a primitive instrument, subject to many interferences, and the measurements generally are not reproducible. Recently developed turbidimeters often use for calibration a suspension of formazin permanently embedded in a cylinder of Lucite. These instruments produce reproducible readings up t o 40 Jackson candle units (JCU), and samples containing turbidities in excess of 40 JCU should be diluted to
32
ANALYSIS O F OILFIELD WATERS
values below this level and the results multiplied by the correct dilution factor. To obtain maximum accuracy and precision the following precautions should be observed: (a) Protect the Lucite standard from scratches, nicks, and fingerprints. (b) While calibrating the instrument, use a constant orientation of the Lucite standard. (c) Use a homogeneous sample in the sample cuvette; do not take readings until finely dispersed bubbles have disappeared. (d) Dilute samples containing. excess tubidity to some value below 40 JCU; take reading, and multiply results by correct dilution factor. Suspended solids analysis
To determine the composition of the suspended solids they can be removed by filtration using a 0.45-pm membrane or less porous filter. The filtered solids can then be subjected to chemical analysis. To determine the exact composition of the solids may require the filtration of a large sample in order to procure enough solid material. The heavy-metal content can be determined by subjecting a portion of the sample t o an emission spectrometric analysis; X-ray diffraction can be used to determine which, if any, clays are present; extraction with organic solvents followed by infrared mass spectrometric, chromatographic, and gas chromatographic analysis will give an indication of organic compounds present; thermogravimetric analysis will provide clues; wet chemical analysis can be used t o determine many of the anions; and X-ray fluorescence can be used to determine some of the anions. Resistivity The resistivity of petroleum-associated waters is used in electric log interpretations (Wyllie, 1963), and for such use the values must be adjusted t o the formation temperature. This can be done by referring t o curves such as those shown in Fig.3.1, which gives resistivity values for sodium chloride solutions. The resistivity of a formation water will not be exactly the same as that of a pure sodium chloride solution of equal dissolved solid (DS) content, but for practical purposes the assumption that the resistivities are approximately equal is satisfactory. It is possible to calculate the resistivity from water-mineral analysis by using methods such as those developed by Dunlap and Hawthorne (1951) or Jones (1944). The calculated values are less accurate and usually lower than the directly measured resistivities. The direct-measurement method is essentially the electrical resistance of a cube of oilfield water. In well-logging practice, the edge of the cube considered is 1 m in length. Therefore, resistivity of an oilfield water is expressed in ohm-meters ( a m ) . Temperature has a profound effect on resistivity; therefore, all resistivities
RESISTIVITY
33
R E S I S T I V I T Y . .hm-n.ttrl
Fig. 3.1. Plots of resistivity of aqueous solutions containing various concentrations of sodium chloride.
should be determined at a known constant temperature. The sample should be freshly filtered and free of oil. Nonionized silica and other materials in suspension in an oilfield water can affect the resistivity determination, but in general such interferences can be ignored. Cell polarization can be troublesome with highly mineralized waters and will vary directly with the current that flows between the electrodes and inversely with the frequency of the current. High input voltage t o the bridge or low cell resistance (highly mineralized waters) increases the likelihood of polarization. Cell resistance can be increased by increasing the cell constant. Reagents. The necessary reagents are standard potassium chloride solutions of l.OOON, O.lOOON, and 0.01OON (use only certified reagent-grade KC1 that has been oven-dried t o constant weight at 110OC);chromic-sulfuric acid cleaning solution; platinizing solution (dissolve 3 g of chloroplatinic acid and 0.02 g of lead acetate in 100 ml of water); and a 10%aqueous sulfuric acid solution. Equipment. The necessary resistivity measurement equipment includes a Wheatstone bridge; resistivity cells, either dip or pipet type, with platinum
34
ANALYSIS OF OILFIELD WATERS
electrodes; water bath, complete with stirrer, thermostat, and thermometer, with 0.loC graduations; source of alternating current, 25- t o 60-cycle a.c. galvanometer, and an appropriate isolating transformer. Selection of the cell constant is limited by the accuracy and sensitivity of the bridge when measuring very high and very low resistivities. Also, current frequency should not be excessively high since a.c. resistance is a complex function of frequency; e.g., at frequencies necessary to avoid polarization, the differences between a.c. resistance and d.c. resistance may be appreciable unless the cell has been carefully designed t o minimize this difference. In essence, the ideal single apparatus for measurement of resistivity throughout a wide range necessarily incorporates compromises between low input voitage, high cell constant, high current frequency, and accuracy and sensitivity of the bridge.
Cell preparation To prepare the cell, clean it with chromic-sulfuric acid solution and rinse thoroughly with water. Immerse the cell or fill it, depending upon whether a dip or pipet cell is used, in the platinizing solution. Connect the electrodes of the cell to three dry cells (1-1/2 V each) in parallel through a limiting resistance of approximately 1,000 52. Reverse the direction of the current once a minute for 6 minutes or until the shiny platinum surface is covered with a dense black coating. Repeat the electrolytic process using 10% sulfuric acid solution to remove chlorine. Remove the electrodes, rinse with distilled water, and store in distilled water. Note: new cells should be cleaned and platinized before use. They should be cleaned and replatinized whenever the readings become erratic or when the platinum black flakes off.
Cell resistance To determine the cell resistance using the standard potassium chloride solutions, adjust the temperature of each potassium chloride solution to exactly 25OC and obtain a reading in ohms for each solution with the cell. Calculate the cell constant using the following formula:
C = R K C l x specific conductance of standard KC1 solution where R K C l = reading obtained in ohms for standard KC1 solution. Note: the specific conductivities of the standard KC1 solutions a t 25°C are as follows (Hodgman et al., 1962, p. 2690):
1.OON KCl = 0.11173 mho/cm 0.1ON KC1 = 0.012886 mho/cm 0.01N KCl = 0.0014114 mho/cm
SPECIFIC GRAVITY
35
Method of determination Procedure. To determine the resistivity of the petroleum-associated water, filter the sample to remove oil and transfer the sample t o the cell or cell container and place it in a water bath. Allow sample sufficient time to adjust to bat!i temperature, and measure resistance of sample and record the temperature to nearest 0.1Oc. Calculation. The resistivity calculation is dependent upon the type of cell and bridge used, but in general the following formula will apply:
R,
D2 V
=-
4LXT
where R, = resistivity of water, a m ; V = difference in potential between potential-measuring electrodes, V; I = current flowing through the cell, A; D = inside diameter of potential-measuring electrodes, m; and L = distance between potential-measuring electrodes, m. Because D and L are constant for any one cell and I is held constant for most waters, these values can be combined into a single constant, K , and the following simplified equation used :
R,
= KV
Calculated resistivity The resistivity of petroleum-associated waters often is calculated using the laboratory analysis (Dunlap and Hawthorne, 1951). The concentrations of the ionic constituents are used in the calculation method. Dunlap and Hawthorne (1951) caution users of their calculation method that the sulfate factor 0.50 may give unreliable results if the water contains appreciable concentrations. of sulfate. If the sulfate concentration exceeds 2,500 mg/l, a factor of 0.40 will give a better calculated resistivity value. Specific gravity Specific gravity is the ratio of the weight of a given volume of material to the weight of an equal volume of some other material used as a standard (Mellon, 1956, p.306), and pure water is the usual standard for liquids and solids. Depending upon the accuracy desired, the specific gravity of a petroleum-associated water can be determined with a pycnometer, specific gravity balance, or hydrometer. Because any oil in or on the sample will interfere with the specific gravity determination, the sample should be filtered.
TABLE 3.W Approximate relation of specific gravity (Sp. gr.) to mg/l of dissolved solids (DS)
-
Sp. gr.
DS
Sp.gr.
DS
Sp. gr.
DS
Sp.gr.
DS
Sp. gr.
DS
Sp.gr.
DS
1.ooo
0 1,400 2,800 4,200 5,600 7,000 8,300 9,700 11,100 12,400 13,700 15,200 16,600 17,800 19,100 20,500 21,900 23,200 24,500 25,900 27,300 28,500 29,800 31,000 32,400 33,900 35,100 36,400 37,700 39,100 40,400 41,700 43,000 44,300 45,600 46,900 48,300 49,500
1.038 1.039 1.040 1.041 1.042 1.043 1.044 1.045 1.046 1.047 1.048 1.049 1.050 1.051 1.052 1.053 1.054 1.055 1.056 1.057 1.058 1.059 1.060 1.061 1.062 1.063 1.064 1.065 1.066 1.067 1.068 1.069 1.070 1.07 1 1.072 1.073 1.074 1.075
'50,800 52,000 53,300 54,600 55,900 57,100 58,300 59,600 60,900 62,100 63,400 64,600 65,900 67,100 68,400 69,600 70,900 72,000 73,300 73,600 75,800 77,100 78,200 79,400 80,600 81.800 83,100 84,300 85,600 86,700 87,800 89,100 90,300 91,500 92,700 93,900 95,100 96,200
1.076 1.077 1.078 1.079 1.080 1.081 1.082 1.083 1.084 1.085 1.086 1.087 1.088 1.089 1.090 1.091 1.092 1.093 1.094 1.095 1.096 1.097 1.098 1.099 1.100 1.101 1.102 1.103 1.104 1.105 1.I06 1.107 1.108 1.109 1.110 1.111 1.112 1.113
97,400 98,700 99,800 101,000 102,200 103,400 104,600 105,800 106,900 108,100 109,300 110,400 111,600 112,800 114,000 115,100 116,200 117,400 118,600 119,600 120,800 122,000 123,100 124,400 125,500 126,700 127,800 128.800 130,000 131,100 132,300 133,400 134,500 135,600 136,800 137,900 139,100 140.1 00
1.114 1.115 1.116 1.117 1.118 1.119 1.120 1.121 1.122 1.123 1.124 1.125 1.126 1.127 1.128 1.129 1.130 1.131 1.132 1.133 1.134 1.135 1.136 1.137 1.138 1.139 1.140 1.141 1.142 1.183 1.144 1.145 1.146 1.147 1.148 1.149 1.150 1.151
141,200 142,300 143,400 144,500 145,600 146,700 147,900 148,900 150.00 0 151,100 152,100 153,200 154,400 155,500 156,600 157,700 158,800 159,900 161,000 162,000 163,100 164,100 165,200 166,200 167,300 168,400 169,400 170,400 171,500 172,500 173,600 174,700 175.7 00 176,800 177,900 178,900 180,000 181,100
1.152 1.153 1.154 1.155 1.156 1.157 1.158 1.159 1.160 1.161 1.162 1.163 1.164 1.165 1.166 1.167 1.168 1.169 1.170 1.171 1.172 1.173 1.174 1.175 1.176 1.177 1.178 1.179 1.180 1.181 1.182 1.183 1.184 1.185 1.186 1.187 1.188 1.189
182,100 183,200 184,200 185,300 186,300 187,400 188,400 189,500 190,500 191,600 192,600 193,600 194,700 195,700 196,700 197,800 198.800 199,800 200,900 201,900 202,900 203,900 204,900 206,000 207,000 208,000 209,000 210,000 211,000 212,000 213,000 214,000 215,000 216,000 217,000 218,000 219.000 22o;ooo
1.190 1.191 1.192 1.193 1.194 1.195 1.196 1.197 1.198 1.199 1.200 1.201 1.202 1.203 1.204 1.205 1.206 1.207 1.208 1.209 1.210 1.211 1.212 1.213 1.214 1.215 1.216 1.217 1.218 1.219 1.220 1.221 1.222 1.223 1.224 1.225
221,000 222,000 223,?00 224,000 225,000 226,000 227,000 228,000 229,000
1.001 1.002 1.003 1.004 1.005 1.006 1.007 1.008 1.009 1.010 1.011 1.012 1.013 1.014 1.015 1.016 1.017 1.018 1.019 1.020 1.021 1.022 1.023 1.024 1.025 1.026 1.027 1.028 1.029 1.030 1.031 1.032 1.033 1.034 1.035 1.036 1.037
230,000 230,800 231,800 232,800 233,700 234,700 235,700 236,700 237,600 238,600 239,500 240,500 241,500 242,400 243,400 244,300 245,300 246,200 247,700 248,100 249,100 250,000 250,900 251,900 252,800 253,800 254,700
8
3 m
TITRIMETRIC METHODS
37
Knowledge of the specific gravity of the sample is necessary to convert the analytical data determined for the sample from milligrams per liter to parts per million. In addition, the specific gravity will give an indication of the amount of dissolved solids present in the sample, as indicated in Table 3.VI. TITRIMETRIC METHODS
Acidity, alkalinity, and borate boron If the pH of the water is less than 4.5, the water possesses what is called “mineral-acid acidity”. The acidity of a petroleum-associated water may indicate a contaminant because of acid treatment of the well or it could indicate the presence of various dissolved gases and salts. Most petroleumassociated waters contain little or no acidity. If a water contains acidity, it does not contain alkalinity. The acidity of a water is determined by adding a standard base such as 0.02N sodium hydroxide to the water until the pH of the water is 4.5 (Collins et al., 1961) as monitored with a pH meter. To obtain a value close to natural conditions, the acidity should be determined at the sampling point. The alkalinity of a water is determined by adding a standard acid such as 0.05N hydrochloric acid t o the water and recording the volume used to neutralize it to pH 8.1 and pH 4.5. The amounts of hydroxide, carbonate, and/or bicarbonate can then be calculated using the relationships shown in Table 3.VII. Because the alkalinity will change when the sample is exposed to the atmosphere, the alkalinity should be determined as rapidly as possible after sampling.
TABLE 3.VII Relationships for determining alkalinity after neutralization with a standard acid Volume of standard acid used
P=O P = 1/2T P = 1/2T P > 1/2T P=T
100 ml, take 50 ml 100 ml, take 20 ml 100 ml, take 1 0 ml 500 ml, take 10 ml 500 ml, take 10 ml 500 ml, take 10 ml 1,000 ml, take 10 ml -
Reagents. The necessary reagents include silver nitrate, standard solution, 0.05N;potassium or sodium chromate, neutral 5% aqueous solution; and nitric acid, 0.1N (nitrous free); and sodium bicarbonate. Equipment. The necessary equipment includes a hotplate, a 10-ml microburet, flasks, and pipets.
Procedure. After removal of interferences and selection of correct aliquot size, dilute the sample to 20 ml or more, adjust the pH to 8.3 with sodium bicarbonate or 0.1N nitric acid, add 1 ml of a 5% aqueous potassium chromate solution, and titrate with an 0.05N silver nitrate solution until the red endpoint just persists. Calculation: ml AgN03 x N x 35,500 = mg,l clml sample The precision and accuracy of the method are about 1%and 2%, respectively, of the amount present. Bromide and iodide Bromide and iodide are present in almost all petroleum-associated waters. In the following procedure, iodide is selectively oxidized t'o iodate with bromine water; excess bromine is reacted with sodium formate. The iodate reacts with added iodide t o produce iodine which is titrated with thiosulfate. Hypochlorite is added to another sample to oxidize both bromide and iodide to bromate and iodate, respectively. Excess hypochlorite is reacted with sodium formate, and the bromate and iodate are reacted with iodide t o liberate iodine for titration with thiosulfate.
46
ANALYSIS OF OILFIELD WATERS
Iron, manganese, and organic matter can interfere but are removed in the procedure. Fluoride is added to mask interference from any remaining traces of iron.
Reagents. The necessary reagents include a 2% ammonium molybdate solution; glacial acetic acid; calcium hydroxide; calcium carbonate; 0.05N hydrochloric acid; 6N hydrochloric acid; potassium iodide; sodium fluoride; starch indicator solution; 0 . O l N sodium thiosulfate (standardize prior t o use); 3.8M sodium formate (prepare fresh daily); saturated bromine water; and methyl red indicator solution. Equipment. The necessary equipment includes a mechanical shaker, 200-ml bottles, a hot-water bath, flasks, pipets, and microburets. Procedure. To remove iron, manganese, and organic matter from the sample, add exactly 100 ml of sample to a stoppered bottle. Add 1 g of calcium hydroxide, and place the mixture in a shaker for 1 hour. Allow the mixture to stand overnight and filter through a dry folded filter, discarding the first 20 ml that comes through. Brines with specific gravities of less than 1.009 may be filtered without standing overnight. Prepare a blank in the same manner. Transfer an aliquot of the filtrate containing 1-2 mg of iodide t o a 250-ml Erlenmeyer flask. Add sufficient water t o make the total volume 75 ml, and three drops of methyl red indicator. Add 0.05N hydrochloric acid until the mixture is just slightly acid, add 10 ml of sodium acetate solution, 1 ml of glacial acetic acid, and 4 ml of bromine water, and allow to stand for 5 minutes. Next add 2 ml of sodium formate solution, blow out any bromine vapor from the neck of the flask, and wash down the sides with water. When the solution is completely colorless, add 0.2 g of sodium fluoride and 0.5 g of potassium iodide. Mix until dissolved and add 15 ml of 6N hydrochloric acid. Titrate with 0.01N sodium thiosulfate using starch indicator. Disregard any return of blue color after the endpoint. Record this titration volume for the iodide calculation. Transfer another aliquot of the filtrate containing 1-2 mg of bromide t o a 250-ml Erlenmeyer flask and add sufficient water to make the total volume 75 ml. Add 10 ml sodium hypochlorite solution and approximately 0.4 g of calcium carbonate (or enough so that approximately 0.1 g will remain after the next step). Adjust the pH of the solution with 3N hydrochloric acid to between 5.5 and 6.0 and heat in a water bath t o 90°C for 10 minutes. (A small amount of undissolved calcium carbonate should remain at this point.) Remove the flask and cautiously add 10 ml of sodium formate solution, return the flask to the water bath, and keep the contents hot for 5 minutes more and observe the timing very closely. Rinse down the inside of the flask with a few milliliters of distilled water and allow the solution to cool t o room temperature. (Do not use a cold water bath.) To the ambient solution
TITRlMETRIC METHODS
47
add three drops of ammonium molybdate solution, 0.5 g sodium fluoride (if iron is present), and 0.5 g potassium iodide, mix until dissolved, and acidify with 15 ml of 6N hydrochloric acid. Titrate with 0 . O W sodium thiosulfate using starch indicator. Disregard any return of blue color after the endpoint. Record this titration for the bromide calculation. Calculations. Iodide: ml of Na, S2O 3 for sample - ml of Na, S2O3 for blank corrected ml of Na, S2O 3 :
=
(ml x N) N a 2 S 2 0 3x 21,150- mg/l Iml sample Bromide: ml of Na2S2O 3 for sample - ml of Na2 S2O3 for blank = corrected ml of Na2S203: (ml x N) Na, S 2 0 3 x 13.320 - mg/l I- x 0.63 = mg/l Brml sample The precision and accuracy of the method are about 3%and 676, respectively, of the amounts of bromide and iodide present. Oxygen The solubility of a gas varies directly with pressure and inversely with temperature and usually is reduced by the presence of dissolved minerals. Most petroleum-associated waters contain little or no dissolved oxygen in situ at depth. Knowledge of the dissolved oxygen content of waters that are to be reinjected for waterflooding or disposal is needed to determine treatment required t o prevent corrosion. Instrumental and wet chemical methods (American Petroleum Institute, 1968) are available for the determination of dissolved oxygen. Instrumental methods usually are modifications of the rotating platinum electrode method (Marsh, 1951), but with them the residual current (when no oxygen is present) is difficult to determine. The modified Winkler method probably is the most accurate wet chemical method available (Watkins, 1954). In the Winkler method for quantitatively determining dissolved oxygen in water, a glass-stoppered bottle is completely filled with the water to be tested. Manganous sulfate (MnS04) and potassium hydroxide (KOH) are added, forming a precipitate of manganous hydroxide (Mn(OH), ) in accordance with the following reaction: MnS04 + 2KOH + Mn (OH), + K2S04 The manganous hydroxide combines with the oxygen dissolved in the water to form a higher oxide of uncertain composition, assumed t o be manganese hydroxide (MnO(OH), ), as follows: 2Mn (OH), + 0,
+
MnO (OH),
ANALYSIS OF OILFIELD WATERS
48
On acidification in the presence of an iodide, the higher oxide of manganese liberates a quantity of iodine stoichiometrically equivalent t o the quantity of dissolved oxygen present in the sample in the following manner: MnO (OH), + 2H2 SO4 +. Mn(S04 )Z + 3H20 Mn(S04), + 2KI +. MnS04 + K2S04 + I2 The quantity of iodine liberated is determined by titrating an aliquot portion of the sample with a standard solution of sodium thiosulfate ( N a 2 S 2 0 3 )using starch solution as an indicator, as shown by the equation: 2Na2S, O3 + I,
+. Na,
S40, + 2NaI
The iodine modification of the Winkler method depends upon the conversion of any hydrogen sulfide t o hydrogen iodide and free sulfur by reducing the iodine added t o the brine. This reaction proceeds as follows: H,S+I2 + 2 H I + S Tests have shown that interfering substances other than hydrogen sulfide that might be present in oilfield brines also are counteracted by the iodine added.
Reagents. It is important to use sterile glassware or polyethylene bottles in preparing and storing reagents for this test to prevent contamination and t o make longer storage of reagents possible without appreciable changes in their normality. Iodine solutions, 0.5N and 0.W. Hydrogen sulfide water: saturate distilled water (which has been boiled and cooled recently t o drive off dissolved oxygen) with hydrogen sulfide gas. Starch solution. Manganous sulfate solution: dissolve 480 g of manganous sulfate (MnS04 *4H20) or 400 g of manganous sulfate (MnS04 *2H20) in distilled water, filter, and dilute to 1 liter. Alkaline iodide solution: dissolve 700 g of potassium hydroxide (KOH) or 500 g of sodium hydroxide (NaOH) and 150 g of potassium iodide (KI), or 135 g of sodium iodide (NaI).in distilled water, and dilute the solution t o 1 liter. If a white carbonate precipitate is formed, separate the precipitate by settling and then siphoning off the supernatant liquid. The solution should give no color with starch indicator when diluted and acidified, which indicates the absence of nitrates, iodates, and ferric salts. Sulfuric acid, concentrated. Sodium thiosulfate solution, 0.1N. Standard sodium thiosulfate solution, 0.025N.
TITRIMETRIC METHODS
49
Equipment. The necessary equipment includes glass-stoppered bottles, pipets, flasks, and microburets. Sa mp 1ing Care must be taken t o obtain uncontaminated samples of water for determining dissolved gases. Glass-stoppered bottles should be used for sample containers. To determine dissolved oxygen in water, 300-ml bottles with pointed, ground-glass stoppers and overflow lips of the type used for biochemical oxygen-demand tests are particularly suitable. These bottles are so designed that samples may be obtained without contamination by atmospheric oxygen and so the necessary chemical reagents may be introduced during the analysis without excessive overflow from the lip of the bottle. Before a sample is taken, rinse the bottle three times with the water to be sampled and fill through a rubber tube extending to the bottom of the bottle. A quantity of water equal t o at least three times the capacity of the bottle should be allowed to overflow the bottle, and the rubber tube should be withdrawn slowly so that the space in the bottle occupied by the tube is filled simultaneously with water. The glass stopper, when placed in the mouth of the bottle, will displace all excess water. If any bubbles are seen, the sample is immediately analyzed. If the temperature of the water taken for analysis of dissolved gases is above 2OoC, a cooling coil should be used to cool the sample before the water enters the bottle. It is important that the samples contain no included atmospheric oxygen or carbon dioxide, as errors may be introduced into many of the analyses if extraneous oxygen or carbon dioxide is present in the water.
Procedure. All reagents in the following steps 1 through 8 should be added slowly and carefully under the surface of the water near the bottom of the bottle, using pipets, permitting the displaced water t o overflow the top of the bottle. The quantities of reagents added should be recorded for use in the final calculation. After each reagent is added, the stopper should be carefully replaced and the bottle inverted gently several times so as not to introduce air into the bottle while adding and mixing reagents. Collect the sample as described previously. Add excess 0.5N iodine solution to give the sample a yellow color and let stand 5 minutes. Add saturated hydrogen sulfide water until the sample is a very light straw-yellow, and 1ml of starch solution as an indicator. Add dilute hydrogen sulfide water until the blue color just disappears and then add, drop by drop, 0.1N iodine solution until a faint blue color persists. Add 1 ml of manganous sulfate solution, 1 ml of alkaline iodide solution, and 1ml of concentrated sulfuric acid, letting it run down the neck of the bottle. Transfer 200 ml of the solution by pipet from the sample bottle t o a 500-ml Erlenmeyer flask. Titrate the 200-ml sample in the Erlenmeyer flask with 0.025N sodium thiosulfate solution. The starch indicator should be
ANALYSIS OF OILFIELD WATERS
50
added when the yellow color of free iodine has been almost eliminated by the sodium thiosulfate titration, and the titration should be continued until one drop changes the solution from a light blue to colorless. (Subsequent blue recoloration should be disregarded.) If no hydrogen sulfide or other interfering substances are present, the first six steps of the determination may be eliminated, using only the part of the procedure starting with the addition of the alkaline iodide solution. Calculation. The dissolved oxygen content of the water is determined by the following equations:
v = 200 x-(Y-1) u=-200 w V
where U = dissolved oxygen content, ppm; V = volume of sample titrated, ml; W = volume of 0.025N sodium thiosulfate required, ml; X = volume of sample bottle, ml; Y = total volume of all reagents added, ml; and 1 = the 1 ml of acid added, which does not change the effective oxygen-tested volume of the sample because it is added after all the oxygen has been absorbed. The factor used to take into account the volume of reagents added may involve a slight error, because it is based on the assumption that the reagents contain no dissolved oxygen.
Carbon dioxide Petroleum-associated waters containing carbon dioxide and bicarbonate or carbonate will contain a weak acid H2C03 or its salt, which buffers the solution. This combination controls the pH of waters in the range of about pH 4.5-8.0. Such buffering is caused by the presence of slightly dissociated acids or bases, and when H+ or OH- ions are added they first convert the undissociated acid or base to its salt or vice versa. Loss of carbon dioxide will disturb the carbon dioxide-bicarbonatecarbonate buffer systems. For example, the pH probably will change and precipitation of calcium carbonate or other compounds may occur. An increase in carbon dioxide will shift t h e . carbon dioxide-carbonatebicarbonate equilibria, allowing more material such as calcite t o go into solution. Bacterial reduction of sulfate can cause the amount of dissolved carbon dioxide and hydrogen sulfide in petroleum-associated waters t o be quite high. Several hundred milligrams per liter of C 0 2 can be present in such waters. Knowledge of the amount of carbon dioxide in solution is useful in carbonate equilibria studies (Garrels and Christ, 1965) and in water compatibility studies (Watkins, 1954).
TITR IMETR IC METHODS
51
Reagents. The necessary reagents are 0.05N sodium carbonate solution and phenolphthalein indicator solution. Procedure. Collect the water sample in the same manner used in taking the sample t o be analyzed for dissolved oxygen. Pipet 100 ml of the water into a flask and add five drops of phenolphthalein indicator. If the sample turns red, no free carbon dioxide is present; if it remains colorless, titrate the sample with the standard sodium carbonate solution to a red endpoint. Calcula tion : ml Na2CO, x N x 22,000 = mg/l C02 ml sample Sulfide
As mentioned above, the bacterial reduction of sulfate causes some petroleum-associated waters t o contain appreciable concentrations of hydrogen sulfide. Knowledge of the amount of dissolved sulfide present is necessary for corrosion and water compatibility studies (Watkins, 1954). The following method depends upon the reduction of iodine by the hydrogen sulfide in the brine, as shown by the following equation: H2S+I2 + 2 H I + S Because of the unstable nature of the hydrogen sulfide in solution in waters and brines, the sulfide is not titrated directly. To prevent the loss of hydrogen sulfide t o the air, an excess of iodine solution is added, and the sample is back-titrated with standard sodium thiosulfate solution, in accordance with the following equation: 2Na2 S2 0, + I2
+. Na2S4 O6 + 2NaI
Experiments conducted by the US. Bureau of Mines indicate that residual reducing agents that cannot be removed by aeration or boiling are present in some oilfield brines. Brine from the Arbuckle (siliceous) Limestone formation originally containing 96 mg/l hydrogen sulfide showed such residual reducing agents t o equal 9 mg/l of hydrogen sulfide after air has been bubbled through the brine for 28 hours. This dropped to 4 mg/l after standing another 24 hours. Further tests in which the hydrogen sulfide was driven off by boiling indicated the presence of 5 mg/l residual reducing agents. When the brine was neutralized with hydrochloric acid (using methyl orange indicator) before boiling, residual reducing agents equal t o 2 mg/l hydrogen sulfide remained.
ANALYSIS OF OILFIELD WATERS
52
Reagents. The necessary reagents are potassium iodide, standard sodium thiosulfate, 0.1N and 0.01N; standard iodine solutions, 0.1N and 0.01N; and starch indicator solution. Procedure. Collect the sample in a glass-stoppered bottle (approximately 200-mi capacity) in the manner previously described for dissolved oxygen. Analysis should be made as soon as possible after sampling. Pipet 5 ml of 0.1N or 0.01N standard iodine solution, depending upon the hydrogen sulfide concentration expected, into each of two Erlenmeyer flasks. It may be necessary to use a larger quantity of 0.01N solution if the hydrogen-sulfide content of the sample is high. Add approximately 1 g of potassium iodide crystals t o each flask. (This step usually may be omitted in determinations on brine samples because of the high mineral content of the water.) Add 50 ml of distilled water t o the flask to be used for a blank determination, and pipet 50 ml of the water sample into the other flask. Titrate both the distilled water blank and the water sample with standard sodium thiosulfate solution of the same normality as the iodine solution used, adding 1 ml of starch indicator near the end of the titration. Record the milliliters of thiosulfate used in each titration. Calculation. Subtract the milliliters of thiosulfate solution used for the sample from the milliliters used for the blank and use the difference in the following formula: (ml x N ) I2 - (ml x N) N a 2 S 2 0 3x 17,000 = mg/l H 2 S ml sample
Sulfur compounds The redox potential of petroleum-associated waters indicates that sulfur compounds other than sulfate and sulfide may exist in solution. When the water is brought t o the land surface, the change in pressure and temperature will affect the redox potential and, if the sample is allowed t o come into contact with the atmosphere, the equilibria of the sample will start t o change immediately. Better methods are needed t o determine the composition of a water in situ. The following method can be used to gain a semiqualitative estimation of the amomts of thiosulfate, sulfite, and sulfide in a water.
Reagents. Zinc carbonate suspension: add zinc acetate to a solution of sodium carbonate, filter and wash the precipitate with several volumes of cold water. Prepare the zinc carbonate suspension by vigorously shaking the precipitate with water. The other reagents are glycerol iodine, 0.01N; sodium thiosulfate, 0.01N; starch indicator solution; glacial acetic acid; and formaldehyde.
FLAME SPECTROPHOTOMETRIC METHODS
53
Determination o f thiosulfate, sulfate, and sulfide Procedure. Collect a water sample as described in the dissolved oxygen procedure. Pipet 100 ml of the sample into a 300-ml flask, and add 20 ml of glycerol, 100 ml of an aqueous suspension of zinc carbonate, and 70 ml of distilled water. Shake the mixture vigorously for 1 minute, filter, and discard the precipitate. Pipet 50 ml of the filtrate into a flask and add 5 ml of formaldehyde, and 3 ml of acetic acid, add starch indicator and titrate to the blue endpoint with 0.01N iodine. Record the amount of iodine used to calculate thiosulfate (A). Pipet another 50-ml aliquot of the filtrate into another flask; add 0.01N iodine until the solution remains yellow. Add starch indicator and titrate to a colorless endpoint with 0.Ol.N sodium thiosulfate. Record the amount of iodine used for thiosulfate plus sulfite (B). Pipet 25 ml of water that was not treated with the zinc carbonate into a flask and add an excess of 0.Ol.N iodine, 3 ml of acetic acid, add starch indicator and titrate t o the colorless endpoint with 0.01N sodium thiosulfate, sulfite, and sulfide (C). Calculations. Milliliters iodine used in A = X ml
X ml
x N x 112,000 = mg/l S2 03-2 ml sample
Milliliters iodine used in A - milliliters iodine used in B = Y ml
Y ml x N x 40,000 = mg/l SO,-2 ml sample Milliliters iodine used in C - milliliters iodine used in B = 2 ml 2 ml x N x 16,000
ml sample
= mg/l
S-’
FLAME SPECTROPHOTOMETRIC METHODS
When a metal salt in solution is sprayed into a flame, the solvent evaporates and the salt decomposes and vaporizes, producing atoms. Some of these atoms can be raised t o an excited state by the thermal energy of the flame, although a major portion of the atoms present in the flame remain at the grourid state. The return of the excited atoms to the ground state results in the emission of radiant energy characteristic of the element atomized. The quantitative measurement of this radiation is the basis of emission flame spectrophotometry, and the essential difference between this form of analysis and classical arc-emission spectrography is the temperature of the source used to excite the atoms. Because the g a s a i r and gas-oxygen flames
54
ANALYSIS OF OILFIELD WATERS
are much cooler than the spark and arc sources used in spectrography, analysis by emission flame spectrophotometry is usually limited t o the more easily excitable elements - lithium, sodium, and potassium. Instrumentation requirements include: (1) A method of introducing the sample into the flame for vaporization. (2) A method of detecting and recording the radiation intensity emitted. (3) A method of selecting the correct wavelength, ordinarily a variable monochromator. A more complete discussion of the theory and instrumentation can be found in books by Burriel-Marti and Ramirez-Munoz (1957) and Dean (1960), as well as in publications of commercial instrument manufacturers. Lithium Lithium usually is calculated as a part of the sodium content in reporting the results of oilfield water analyses rather than being determined and reported separately. One of the more accurate methods t o determine lithium in petroleum-associated waters is the flame spectrophotometric method (Collins, 1962). Reagents. The reagents are lithium, standard solutions, 0.1 mg/ml and 0.01 mg/ml; and n-propanol. Equipment. The necessary equipment includes a flame spectrophotometer, 10-ml'microburets, and volumetric flasks. Preliminary calibration curves. Preliminary calibration curves are useful in determining approximately how much lithium is in the sample and in determining the optimum amount of standard lithium solution t o use in the analysis. Because n-propanol is easier t o work with, it usually is used; however, if additional sensitivity is needed, the acetone-n-amyl alcohol mixture can be used (Collins, 1965). To prepare the preliminary calibration curves, transfer t o 50-ml volumetric flasks aliquots of diluted standard lithium solution containing the following amounts of lithium: 0.01 mg, 0.05 mg, 0.1 mg, 0.15 mg, and 0.2 mg. Add 20 ml of n-propanol to each flask and dilute to volume with distilled water. Aspirate, burn, and record the emission intensity of each of these five standards at 670.8 mp and their background at about 665 mp. Record several peaks for each standard at various sensitivity levels and slit widths. Plot the results on linear graph paper by plotting milligrams of lithium versus intensity. Prepare a curve for each sensitivity level and slit width used, as illustrated in Fig. 3.2. The sensitivity of the instrument will determine the optimum concentrations of lithium and this will require some experimentation. The analyst may find it convenient t o scan all the emission lines of
FLAME SPECTROPHOTOMETRIC METHODS
26 -
-
2 22-
-
-x 20180 0
L
5
4
1
'I
I
I
55
I
0.01 ~ mm r l i t 1,620 volta to I T 1 F W 6836 tOppri 02 5 p s i C2H2 12.5 mm burner height
-
-
-
10-
L
-
mg L i / m l 5 0 % n - P R O P A N O L
Fig. 3.2. Preliminary calibration curves for use in selecting optimum standard additions: Instrument: 0.01-mm slit, 1,620 V to ITT FW 6836, 1 0 psi 02,5 psi C2HZ,and 12.5-mm burner height.
interest; e.g., lithium, sodium, potassium, rubidium, cesium, and perhaps others. This will give information concerning what elements are present.
Procedure. To determine the amount of lithium in the petroleum-associated water, transfer an aliquot of about 10 ml of brine to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to volume with distilled water. (The size of the aliquot will vary with the sample. The specific gravity can be used to help decide the aliquot size. For a brine with a specific gravity of 1.1, an aliquot of 10 ml or less probably will be sufficient.) Aspirate the sample into the flame and read or record the emission intensity of the background at 665 mp and lithium line at 670.8 mp. With these readings and the preliminary calibration curves, calculate approximately how much lithium is in the sample. Determine an aliquot size that will contain about 0.05 mg of lithium. Transfer equal aliquots to three 50-ml volumetric flasks. Add no lithium standard to the first flask, 0.05 mg to the second flask, and 0.1 mg to the third flask. Add 20 ml of n-propanol t o each flask and dilute to volume with distilled water. Aspirate and record the background at 665 mp and the emission intensity of each sample at 670.8 mp. Optimum accuracy is attained by this method when the two standard additions are respectively equal to and twice the amount of lithium in the sample. Care should be taken that too much lithium is not present in the final samples, because self-absorption will cause errors.
ANALYSIS OF OILFIELD WATERS
56 5
4 v)
13
r
2 3 W a c a .x2 I
u 1
0
L I
COI
2s
I
3
I
I
I
4 5 6 tENTRATION OF STANDARD ADDITIONS
-
I 7
Fig. 3.3. Standard-addition calculation graph. In this ideal case the unknown would contain 2 x the dilution factor ( 2 could be 2 mg or 2 pg or whatever unit the analyst used).
Calculation. A graph can be used in the calculation, as illustrated in Fig. 3.3. Plot the concentrations in milligrams of the standard-addition samples on the horizontal axis of linear graph paper and the emission intensities on the vertical axis. Plot the emission intensity of the sample to which no standard lithium soiution was added at 0 concentration. The plot should produce a straight line as shown in Fig. 3.3. Multiply the chart reading at 0 concentration by 2, place this value on the y-axis, and draw a line parallel to the x-axis until it intersects the line plotted. From this point, draw a line parallel to the y-axis until it intersects the x-axis. The vrlue obtained in milligrams can be converted to milligrams per liter by the following formula: mg Li x 1,000 = mg/l Li+ ml sample The formula, shown in Table 3.X1, can be used to calculate the amount of lithium in the sample, using the flame spectrophotometric readings in lieu of the graph method. Optimum accuracy is attained with this method using either type of calculation when the two standard additions respectively are equal to and twice the amount of lithium that is present in the sample. The addition of alcohols t o the aqueous phase before aspiration into the flame increases the sensitivity of the flame method, allowing the use of more dilute solutions and consequently less dissolved solids, which reduces burner plugging. The average precision and accuracy of the lithium method are about 2% and 4%, respectively, of the amount present.
57
FLAME SPECTROPHOTOMETRIC METHODS TABLE 3.XI C Formula for standard-addition calculation C, = (rx - r b ) r r, where the following are true*: . .-
Solution
Concentration
Reading
Unknown
c,
r,
Mixture
c,
.
= c, +
c
r
*C is a standard addition.
Sodium The flame spectrophotometer offers an excellent instrumental technique for determining sodium in a petroleum-associated water. The flames containing alkali metals give strong resonance lines of these metals plus some additional continyous radiation. The strongest line for sodium results from a transition between the lowest excited level and the ground state. The yellow doublet of sodium at 589.0-589.6 mp results from such a transition. Reagents. The necessary reagents are sodium standard solutions, 1 mglml and 0.01 mg/ml; and n-propanol. Preliminary Calibration curves. Preliminary calibration curves similar to those shown in Fig. 3.2 should be used to determine the approximate amount of sodium in the sample. These curves are prepared in the same manner as the lithium curves, except that standard sodium solutions are used; the emission intensity of the sodium at 589 mp is determined, minus a background at about 582 mp. Procedure. To analyze the petroleum-associated water, transfer an aliquot of water t o a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to volume with distilled water. (The size of the aliquot will vary with the sample. The specific gravity can be used to help decide the aliquot size. For a water with a specific gravity of 1.1,an aliquot of 1ml or less probably will be sufficient.) Aspirate the sample into the flame and record the emission intensity of the background at 582 mp and sodium line at 589 mp. With these readings, calculate approximately how much sodium is in the sample by using the preliminary calibration curves. Determine the aliquot size that will contain about 0.05 mg of sodium. Transfer equal aliquots t o three 50-ml volumetric flasks. Add no sodium
ANALYSIS OF OILFIELD WATERS
58
standard to the first flask, 0.05 mg to the second flask, and 0.1 mg t o the third flask. Add 20 ml of n-propanol to each flask and dilute to volume with water. Aspirate and record the emission intensity of each sample at 589 mp and its background at 582 mp. Calculation. Use the graph or formula illustrated in the lithium method. The value obtained in milligrams can be converted to milligrams per liter by the following formula: mg Na x 1,000 = mg/l Na' ml sample The precision and accuracy of the method are approximately 3%and 6%, respectively, of the amount of sodium present. Some elements, when present in the solution being analyzed, will cause a change in the emission intensity of the sodium. The use of a standard addition technique largely compensates for these interferences. Potassium Potassium usually is included with sodium without any differentiation in reporting the results of brine analyses, although potassium is known to be present in many oilfield brines. Potassium compounds often are dissolved before sodium compounds; however, they do not remain dissolved as readily because they are readily adsorbed and enriched in clays. In sea water and oilfield brines, only a small part of the originally dissolved potassium remains in solution. The fact that many oilfield brines are low in potassium with respect to sodium, whereas surface waters and young volcanic waters are enriched in potassium with respect to sodium, is an important criterion in identifying the sources of brines. The flame spectrophotometer provides a sensitive method for the determination of potassium. The strongest lines for potassium detection in a flame are the doublet at 766.5 and 769.9 mp. Reagents. The necessary reagents are potassium standard solution, 0.1 mg/ml; and n-propanol. Preliminary calibration curves. Preliminary calibration curves are useful in determining the approximate amount of potassium in the sample, so that the optimum sample size for standard addition can be selected for the analysis. These curves can be prepared in the same manner used in the preparation of the lithium preliminary calibration curves (Fig.3.2) except that standard potassium solutions are used. The emission intensity of the potassium line at 766.5 mp minus the background at about 750 mp can be used in preparing the curves.
FLAME SPECTROPHOTOMETRIC METHODS
59
Procedure. To determine the amount of potassium in the sample, transfer an aliquot of sample to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to volume with distilled water. The specific gravity can be used to help decide the aliquot size. For a brine with a specific gravity of 1.1,an aliquot of 5 ml or less probably will be sufficient. Aspirate the sample into the flame and record the emission intensity of the background at 750 mp and potassium line at 766.5 mp. With this reading, use the preliminary calibration curves and calculate approximately how much potassium is in the sample. Determine an aliquot size that will contain about 0.05 mg of potassium. Transfer equal aliquots t o three 50-ml volumetric flasks. Add no potassium standard to the first flask, 0.05 mg to the second flask, and 0.1 mg t o the third flask. Add 20 ml of n-propanol to each flask and dilute t o volume with distilled water. Aspirate and record the emission intensity of each sample at 766.5 mp and the background at 750 mp. Optimum accuracy is attained by this method when the two standard additions are respectively equal to and twice the amount of potassium in the sample. Care should be taken that too much potassium is not present in the final samples, because self-absorption will cause errors. Calculation. The graph or formula illustrated in the lithium method can be used. The value obtained in milligrams can be converted to milligrams per liter by the following formula: mg K x 1,000 = mg/l K+ ml sample The precision and accuracy of the method are approximately 2% and 4% of the amount present. Several elements can interfere in the flame analysis of potassium. Elements which ionize easily will lower the degree of ionization of potassium, and elements which are difficult to ionize or have high ionization energies will give the opposite effect. By using the Saha equation (Herrmann and Alkemade, 1963), it is possible t o estimate such interferences. Generally, the use of a standard addition compensates for interferences. Rubidium and cesium The flame spectrophotometer provides one of the most sensitive methods available for determining rubidium and cesium. Cesium has a pair of emission lines at 852.1 and 894.4 mp. Both lines are of about equal intensity, but water produces a molecular band system at 900 mp which can interfere at 894.4 mp. Rubidium also has two strong lines in the red region at 780.0 and 794.8 mp. It is necessary t o use a photomultiplier with an S-1 response t o detect cesium and rubidium at the levels found in many waters. Examples of such tubes are ITT type 6836/FW118, RCA types 1P22 and 7102, and DuMont
60
ANALYSIS OF OILFIELD WATERS
type 6911. Such tubes also are useful for lithium and potassium determinations. Several elements can interfere in the determination of cesium and rubidium. However, because a solvent extraction or standard-addition technique is used most interferences are either removed or compensated (Collins, 1965).
Reagents. The necessary reagents are cesium standard solution, 0.01 mg/ml; rubidium standard solution, 0.01 mg/ml; buffer solution, pH 6.6 (adjust the pH of a 1M sodium citrate solution to 6.6 with 0.5M nitric acid); sodium tetraphenylboron, 0.05M (dissolve 0.855 g of sodium tetraphenylboron in distilled water and dilute t o 50 ml - prepare a fresh solution daily); nitroethane; hydrochloric acid, 0.1N; sodium hydroxide, 0.W;synthetic brine solution. Procedure. To determine the amount of rubidium and cesium in the petroleum-associated water, transfer an aliquot of brine containing 0.005 to 0.05 mg of cesium and rubidium to a 100-ml beaker and add 25 ml of the citrate buffer solution. Transfer the solution to a 125-ml Teflon-stoppered separatory funnel and adjust to 100-ml volume. Add 2 ml of 0.05M sodium tetraphenylboron aqueous solution and 1 0 ml of nitroethane, and shake the mixture vigorously for 2 minutes. Allow the phases t o separate for 30 minutes, after which time withdraw the aqueous phase. Centrifuge the nitroethane phase. Determine the cesium and rubidium emission intensities by burning the nitroethane phase in the flame spectrophotometer and automatically scanning the 780.0 mp, 794.8 mp, and 894.4 mp lines. Calibration curves. Prepare calibration curves by using appropriate portions of the standard cesium and rubidium solutions. Add 5 ml of synthetic brine solution t o each standard sample before buffering and extraction. Plot the resultant emission intensities versus milligrams of cesium or rubidium or linear graph paper. Calculation. Determine the milligrams of cesium or rubidium in the sample by referring t o the calibration curves. The milligrams can be converted to mg/l by the following formula: mgx 1,000 = mg/l Cs+ or Rb' ml sample Fig.3.4 illustrates the relative emission intensities obtained with cesium and rubidium in nitrobenzene, nitroethane, 1-nitropropane, and 2-nitropropane. 15 ml of each of these solvents.are used t o extract 0.1 mg each of cesium and rubidium tetraphenylboron from aqueous solutions. The organic phases then are aspirated directly into the flame, and the peaks scanned automatically. Good resolution is obtained with a 0.01 mm slit width. Amy1 alcohol gives poorer results than nitrobenzene.
FLAME SPECTROPHOTOMETRIC METHODS NITROBENZENE
NITROETHANE
I- NITROPROPANE
63
2 - NITROPROPANE
1
Fig. 3.4. Relative intensities obtained by burning organic solvents containing tetraphenylboron salts of cesium and rubidium.
Standard-addition technique to determine rubidium Some waters contain sufficient rubidium to enable use of the standardaddition technique. To analyze such waters, preliminary calibration curves similar to those used to determine lithium (Fig.3.2) are recommended, to aid in selecting the optimum amount of standard rubidium solution to use. Manganese The amounts of sodium, potassium , calcium, and strontium in most petroleum-associated waters are too high t o permit determination of manganese with the flame spectrophotometer without preliminary separations. These interferences can be obviated by extracting the manganese into a chloroform 8-hydroxyquinoline solution. The chloroform is removed by evaporation, and the manganese hydroxyquinoline is dissolved in n-propanol. This solution is burned in the flame spectrophotometer, and the emission intensity of its resonance triplet at 403.2 mp is recorded (Collins, 1962). Reagents. The necessary reagents are standard manganese solution (dissolve 0.583 g of manganese dioxide in 10 ml of hydrochloric acid and dilute to 1 liter with distilled water, transfer a 100-ml aliquot of this solution t o another
62
ANALYSIS OF OILFIELD WATERS
1-liter flask, add 10 ml of hydrochloric acid, and dilute to volume with distilled water; ( 1 ml of this solution contains 10 pg of manganese); chloroform solution of 8-hydroxyquinoline (dissolve 1.O g of 8-hydroxyquinoline in 100 ml of chloroform); hydrogen peroxide (3% solution); ammonium hydroxide ( 3 N ) ; sodium potassium tartrate (10% solution); ammonium fluoride (5%solution); n-propanol; and chloroform.
Procedure. Transfer an aliquot of brine containing up to 150 pg of manganese to a 100-ml beaker; add 1ml of hydrogen peroxide, 5 ml of ammonium fluoride, and 10 ml of sodium potassium tartrate; and adjust the pH of the mixture to 9.0 with ammonium hydroxide. Transfer the solution to a 125-ml Teflon-stoppered separatory funnel, add 10 ml of 8-hydroxyquinoline chloroform solution, and bring the mixture t o equilibrium by shaking it vigorously for 1minute. Draw the chloroform phase off into a 100-ml beaker and strip the aqueous phase by an additional extraction with chloroform. Evaporate the combined chloroform extracts to dryness over a hotplate, taking care t o prevent the residue from charring. Dissolve the residue in n-propanol and make to 50 ml volume with n-propanol. Aspirate the n-propanol solution directly into the flame and determine the net emission by subtracting the background emission at 400 mp. Calculate the amount of manganese in the sample from a calibration curve prepared by adding known amounts of manganese t o a synthetic brine solution. The calibration curve should be linear for up t o 150 pg of manganese when the emission intensity is plotted versus micrograms of manganese on linear graph paper. Calculation : pg Mn (from curve) = mg/l Mn +* ml sample The intensity of the emission of manganese in a flame spectrophotometer is enhanced by a factor of 16 by using n-propyl alcohol rather than water as the solvent. With this increased intensity, the sensitivity of the method is about 1 mg/l, although additional sensitivity is attainable by concentrating the brine by evaporation. The precision of the method is about 3%,and the accuracy is about 6% of the amount present.
Strontium Several flame photometric methods are available for determining strontium in oilfield brines; a standard curie may be unreliable if there are instrument changes, such as a slightly plugged burner, change of resistance in the amplifying circuit, or other variables. Chemical precipitation of strontium as the sulfate does not satisfactorily separate strontium from barium
FLAME SPECTROPHOTOMETRIC METHODS
63
and calcium without several preliminary separations. Precipitations as the carbonate or oxalate have the same disadvantages, and precipitation as the nitrate and subsequent solvent extraction of calcium with butylcellosolve still leaves barium in the precipitate. The use of a standard addition flame photometric method gives reproducible results without the necessity of several separations. Reagents. The necessary reagents are standard strontium solution, 1mg/ml; and n-propanol. Preliminary calibration curves. To determine approximately how much strontium is present in the samples, it is advantageous t o prepare preliminary calibration curves. A procedure similar to that used in the lithium method can be used, except that the strontium emission should be determined at 680 mp with a background reading at 690 mp. The data are plotted in a manner similar to Fig. 3.3. Procedure. To determine the amount of strontium, transfer an aliquot of brine to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute t o volume with distilled water. Aspirate the sample into the flame and read; record the emission intensity of the background at 690 mp and the strontium line at 680 mp. With these readings and the preliminary calibration curves, calculate approximately how much strontium is in the sample. Determine an aliquot size that will contain about 1.0 mg of strontium. Transfer equal aliquots t o three 50-ml volumetric flasks. Add no strontium standard t o the first flask, 1.0 mg t o the second flask, and 2.0 mg t o the third flask. Add 20 ml of n-propanol to each flask and dilute t o volume with distilled water. Aspirate and record the background at 690 mp and the emission intensity of each sample at 680 mp. Calculation. A graph can be used in the calculation as illustrated in Fig.3.3. The value obtained in milligrams can be converted t o milligrams per liter by the following formula: mg Sr x 1,000 = mg/l Sr+* ml sample The formula, shown in Table 3.X1, can be used to calculate the amount of strontium in the sample using the flame spectrophotometric readings in lieu of the graph method. Barium A flame spectrophotometric method was developed which utilizes the chromate precipation followed by dissolution in nitric acid, mixing with an alcohol, and burning in the flame (Collins, 1962). The flame method is
64
ANALYSIS OF OILFJELD WATERS
subject to few interferences except from calcium, but by using the chromate precipitation, calcium is eliminated and barium is concentrated. Reagents. The necessary reagents are barium standard solution, 1 mg/ml; ammonium chromate solution (dissolve 10 g of ammonium chromate in distilled water and dilute to 100 ml); 10%ammonium acetate aqueous solution; nitric acid ( 4 N ) ;n-propanol; acetic acid; and synthetic brine solution (use carbon dioxide-saturated distilled water and dissolve the following amounts of constituents in 1 liter of water: sodium bicarbonate, 0.4 g; sodium chloride, 61 g; potassium. chloride, 5 g; calcium chloride, 19 g; magnesium chloride, 1 2 g; and strontium chloride, 5 g). Procedure. Transfer an aliquot of the sample containing 0.5-15 mg of barium t o a 100-ml beaker, add 1 ml of the ammonium acetate solution, 10 ml of the ammonium chromate solution, and adjust the pH t o 4.6 using acetic acid. Cover the beaker with a watchglass; heat the solution t o near boiling (90°C), remove from the hotplate, and allow to stand for 1 hour. Filter the solution through a 0.45-pm membrane filter using vacuum. Take care to transfer all of the precipitate from the beaker to the filter funnel. Use ammonium chromate solution rather than distilled water to aid in this transfer. Wash the precipitate with 50 ml of ammonium chromate or until strontium and calcium are absent. Wash the precipitate with 50 ml of hot water to remove excess chromate. Add 5 ml of 4N nitric acid t o the filter and swirl the solution on the filter gently to dissolve the precipitate. A clean test tube should be placed below the filter to catch the dissolved precipitate. When all of the precipitate is dissolved, turn on the vacuum and catch the solution in the test tube. Repeat this procedure using an additional 5 ml of 4N nitric acid. Transfer the solution from the test tube t o a 50-ml volumetric flask. Carefully wash the test tube with two 5-ml portions of water. Add 25 ml of n-propanol, dilute to 50 ml volume with water, and mix the solution thoroughly. Burn the sample in the flame spectrophotometer and record the emission intensity at 873 mp and the background at 900 mp. Prepare calibration curves by adding up t o 25 mg of barium to 10 ml portions of the synthetic brine followed by analysis according t o the foregoing procedure, and use in the calculation. Calculation: mg Ba x 1,000 = mg/l Ba+’ ml sample
ATOMIC ABSORPTION METHODS
65
ATOMIC ABSORPTION METHODS
Atomic absorption is complementary to flame spectroscopy. The spectra emitted are analyzed by absorption of resonance lines by free atoms of a constituent in the vapor phase. The unexcited or ground-state atoms produced in the flame can absorb radiant energy when supplied by a suitable external radiation source at a frequency coinciding with that of the emission frequencies of the element atomized. The measurement of this radiation absorbed forms the basis of absorption flame spectrophotometry - or atomic absorption spectrophotometry, as it is usually called. At temperatures up t o 2,7OO0C, ground-state atoms usually account for more than 90%of the atoms in the vapor phase. Hollow cathode discharge tubes generally are used as a light source. The sensitivity of detection does not depend upon the spectral response of the light receiver, since the absorption coefficient is a measure of the relative intensity of the light which passes through an absorption cell versus that which does not. Additional theory can be found in a book by Robinson (1966). Atomic absorption is useful in water and brine analysis, and there are several publications on the subject. Publications oriented to oilfield and sea water analysis are Fabricand et al. (1966), and Angino and Billings (1967). Table 3.XII illustrates the sensitivities that can be obtained using atomic absorption t o determine some metals in aqueous solutions. The sensitivities listed are obtainable if no interferences are present. Interference usually
TABLE 3.XII Approximate sensitivities for some metals to atomic absorption Metal
Wavelength --
Aluminum Barium Beryllium Cadmium Calcium Chromium Copper Iron Lead Magnesium Manganese Mercury Nickel Silver Sodium Zinc
Sensitivity (mg/l)
Fuel and oxidant
(A)
3093 5536 2348 2288 4226 3579 3247 2483 2833 2852 2794 2536 2320 3281 5890 21 38
1.o 0.2 0.1 0.04 0.08 0.15 0.2 0.3 0.5 0.02 0.15 0.01 0.15 0.1 0.03 0.04
nitrous oxide-acetylene nitrous oxide-acetylene nitrous oxide-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene
. -
66
ANALYSIS OF OILFIELD WATERS
results from lack of absorption of atoms bound in molecular combination in the flame and can occur when the flame is not hot enough to dissociate the molecule. It also occurs when a dissociated atom immediately oxidizes to a compound that cannot dissociate further at the temperature of the flame. Interferences
Ionization When a significant number of the atoms of the element being determined are ionized in the flame, an error in the analysis can result. This ionization is because of excessive flame temperature, which, however, can be changed to control this interference. Another type of interference can be caused by the presence in the sample of other, more easily ionizable elements than the one sought. The resulting increase can be controlled by the addition of a sufficient amount of the interfering element t o both sample and standards t o produce a “plateau” in the absorbance above which no further increase occurs. Che m ica 1 A chemical interference is caused by the formation, in the flame, of salts of the element sought which are difficult to decompose, thus reducing the amount of the element available for absorption. The formation of such compounds may often be precluded by the addition of another element, such as lanthanum, which forms a less-soluble salt with the interfering anion than does the element desired. The interfering anion is thus removed from the flame, and the interference is eliminated. Phosphate combines with calcium and magnesium and produces an interference; however, the addition of lanthanum largely overcomes this interference. Addition of an excess of a cation having a similar or lower ionization potential usually reduces interference problems.
Matrix Matrix interference is caused by unequal amounts of dissolved solids in the standards and samples. This can cause error because of differences in aspiration rates through the atomizer. Often this can be controlled by matching the specific gravities of the standards and samples or by adding salts t o the standards. Burners and solvents Various types of burners are used with atomic absorption spectrophotometers. For example, a Boling burner usually is used for aqueous solutions,
ATOMIC ABSORPTION METHODS
67
while a premix burner is used for organic solutions. A nitrous oxide burner head with. a 2-inch slot is used for determining aluminium, barium, and beryllium because overheating is often encountered wit,h a 3-inch slot burner. The use of concentration steps, such as solvent extraction of a chelated compound, enables sensitivities lower than those shown in Table 3.XII to be achieved. For example, aluminium and beryllium can be complexed with 8-quinolinol and extracted with chloroform; cadmium and lead can be complexed with ammonium pyrrolidine dithiocarbamate and extracted with methyl isobutyl ketone. When burning the organic solvents, it usually is necessary t o reduce the fuel air ratio because the burning organic solvent contributes to the fuel supply producing an undesirable luminescent flame and may also lift the flame off the burner. An optimum fuel/air ratio can be found by noting the characteristics of the flame before burning the organic solvent and then reducing the fuel flow, while burning the organic solvent until the flame characteristics are similar t o those noted before the organic solvent was burned. Ramirez-Munoz (1968) provides additional information. Burner height is very important and adjustment often is necessary when changing from one element t o another. Some instruments have a Vernier adjustment for reproducing burner-height settings and some do not. Fig. 3.5 illustrates a device which can be used for reproducing exact burner height (Ballinger et al., 1972).
0-m
from hollow cathode lamp
Fig. 3.5. Device for reproducing burner height for emission and atomic absorption spectrometers.
68
ANALYSIS OF OILFIELD WATERS
Lithium Lithium is determined at the 6707.8 A wavelength with an air-acetylene flame.
Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of potassium. Reagents. The necessary reagents are: (1)Potassium solution: see reagents preparation under “Sodium”. (2) Standard lithium solution: obtain commercially or dissolve 5.324 g of lithium carbonate, Liz CO, , in a minimum volume of one part Hz0 to one part of HC1 (1+ 1).Dilute to 1liter with water. 1ml of this solution contains 1,000 pg of lithium. Preliminary calibration. Prepare standard lithium solutions containing 1-5 pg/l of lithium using the standard lithium solution and 50-ml volumetric flasks. Add to each of these and to a blank, 0.5 ml of the potassium stock solution. Aspirate these standards and the blank as recommended in the calcium method and determine the absorbance at a wavelength of 6707.8 A. Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.05 mg of lithium. Add 0.5 ml of the potassium stock solution, dilute t o volume with water, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.05 mg of lithium. Transfer equal aliquots containing about 0.05 mg of lithium t o three 50-ml volumetric flasks. Add no lithium standard t o the first flask, 0.05 mg of the lithium standard t o the second flask, and 0.10 mg t o the third. Add 0.5 ml of the potassium stock solution t o each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig.3.3, or Table 3.XI: mgLix 1,000 = mg/l Li+ ml sample
Precision. In a single laboratory using oilfield water samples containing concentrations of 90 and 190 mg Li+/l, the standard deviations were k 3 and +5, respectively. The recoveries were 100.6% and 92.996, respectively. Sodium Two wavelengths are used: the 5890-5896
A doublet for the 1-mg/l
69
ATOMIC ABSORPTION METHODS
a
aliquots and the 3302-3303 doublet for the 100-mg/l aliquots. Because of the wide .range of sodium concentrations found in brines, the higher wavelength can be used for the lower gravity brines and the lower wavelength for the higher gravity brines, thus avoiding making two dilutions with some of the heavier brines. It is usually necessary t o make a preliminary determination so that the correct aliquot can be used with the standard additions.
Interferences. Ionization interference is usually overcome by adding potassium. Reagents. The necessary reagents are: (1) Potassium solution: dissolve 190.70 g of potassium chloride, KC1, in water and dilute to 1 liter. 1 ml of this solution contains 100 mg of potassium. (2) Standard sodium solution: obtain commercially or dissolve 25.420 g of sodium chloride in 1 liter of water. 1 ml of this solution contains 10 mg of sodium. Dilute 1 0 ml of this solution t o a liter. 1 ml of this solution contains 100 pg of sodium. Preliminary calibration. Prepare standard solutions containing 1.O-5.0 and 100-500 pg/ml of sodium using the standard sodium solutions and 50-ml volumetric flasks. Add to each of these, and to a blank, 0.5 ml of the potassium stock solution. Aspirate these standards and blank as recommended in for the the calcium method and determine the absorbance at 5890-5896 for the 100-500 pg/ml Na 1.0-5.0 pg/ml Na solutions and at 3302-3303 solutions.
a
a
Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing either about 0.05 mg or about 5 mg of sodium. Add 0.5 ml of the potassium stock solution, dilute t o volume, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings. Determine the aliquot size that will contain either about 0.05 mg or 5 mg of sodium, depending on the wavelength to be used. Transfer equal aliquots t o three 50-ml volumetric flasks. For the 0.05-mg aliquots, add no sodium standard t o the first flask, 0.05 mg of sodium standard to the second flask, and 0.10 mg to the third. For the 5-mg aliquots, add no sodium standard to the first flask, 5 mg t o the second, and 10 mg t o the third. Add 0.5 ml of the potassium stock solution t o each flask and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mg Na x 1,000 = mg/l Na+ ml sample
70
ANALYSIS OF OILFIELD WATERS
Precision. In a single laboratory using oilfield water samples containing concentrations of 22,700 and 43,200 mg Na+/l, the standard deviations were +485 and ?1,890, respectively. The recoveries were 100.8% and 100.9%, respectively. Potassium Potassium is determined at the 7664.9 A wavelength with an air-acetylene flame.
Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of sodium. Reagents. The necessary reagents are: (1) Sodium solution: dissolve 254.20 g of sodium chloride in 1 liter of water. 1 ml of this solution contains 100 mg of sodium. (2) Standard potassium solution: obtain commercially or dissolve 1.907 g of potassium chloride, KCl, in 1liter of water. 1 ml of this solution contains 1,000 pg of potassium. Preliminary calibration. Prepare standard solutions containing 1-5 pg/l of potassium using the standard potassium solution and 50-ml volumetric flasks. Add 0.5 ml of the sodium stock solution t o each of these and to a blank. Aspirate these standards and the blank as recommended in the calcium method and determine the absorbance at a wavelength of 7664.9 A. Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.05 mg of potassium. Add 0.5 ml of the sodium stock solution, dilute t o volume with water, and aspirate. Calculate the approximate potassium concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.05 mg of potassium. Transfer equal aliquots containing about 0.05 mg of potassium t o three 50-ml volumetric flasks. Add no potassium standard to the first flask, 0.05 mg of the potassium standard t o the second flask, and 0.10 mg to the third. Add 0.5 ml of the sodium stock solution to each flask and dilute t o volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mg K x 1,000 = mg/l K+ ml sample
ATOMIC ABSORPTION METHODS
71
Precision. I n a single laboratory using oilfield water samples containing concentrations of 456 and 5,680 mg K+/1, the standard deviations were *25 and +325, respectively. The recoveries were 93.7% and 97.8%, respectively. Magnesium (1) Magnesium is determined at the 2852.1 acetylene flame.
A wavelength with an air-
Interferences. The silicon and aluminum suppression of the magnesium absorption is generally removed by the addition of lanthanum or by the use of a nitrous oxide-acetylene flame.
Reagents. The reagents are: (1)Lanthanum solution (same as used in the calcium procedure). (2) Standard magnesium solution: obtain commercially or dissolve 1.OOO g of magnesium ribbon in a minimum of (1 + 1) HC1, and dilute to 1 liter with 1%(v/v) HC1. 1ml of this solution contains 1,000 pg of magnesium per ml and should be made up daily t o use for the standard additions.
Preliminary calibration. Prepare standard solutions containing 0.1-0.5 pg/l of magnesium using the standard magnesium solution and 50-ml volumetric flasks. Add t o each of these and t o a blank 5 ml of the stock lanthanum solution. Aspirate as suggested in the calcium method and determine the absorbance at 2852.1 A.
Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine often can be used as a guide in estimating the size of an aliquot containing about 0.005 mg of magnesium. Add 5 ml of the lanthanum stock solution, dilute t o volume with water, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings, and determine the aliquot size that will contain about 0.005 mg magnesium. Fig. 3.6 illustrates a plot of the concentration of magnesium found in some oilfield brines compared to their specific gravity. This figure cannot necessarily be applied to all oilfield brines, however, because some will contain more and some less. The concentrations of magnesium in brines from the same formation at about the same depth often are similar. Transfer equal aliquots containing about 0.005 mg magnesium to three 50-ml volumetric flasks. Add no magnesium standard t o the first flask, 0.005 mg t o the second flask, and 0.010 mg t o the third. Add 5 ml of the lanthanum stock solution t o each of the three flasks and dilute t o volume. Aspirate and record the absorbance readings for each sample.
72
ANALYSIS OF OILFIELD WATERS
-L
4.000
1.00
1.05
1.10
1.15
1.20
I. 5
SPECIFIC G R A V I T Y
Fig. 3.6. Relationship of the concentration of magnesium to specific gravity for some oilfield brines.
Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mgMgx 1,000 = mg/l Mg+2 rnl sample Precision. In a single laboratory using oilfield water samples containing concentrations of 1,470 and 2,000 mg Mg+2/1, the standard deviations were k36 and +128, respectively. The recoveries were 97.3%and 103.2% respectively. Calcium (1) Calcium is determined at the 4226.7 A wavelength with an air-acetylene flame.
ATOMIC ABSORPTION METHODS
73
Interferences. The chemical suppressions caused by silicon, aluminium, and phosphate are controlled by adding lanthanum. The lanthanum also controls a slight ionization interference. A pH above 7 causes low calcium values, so dilute HC1 is added t o standards and samples. For samples containing large amounts of silica, it often is preferable t o use the nitrous oxide-acetylene flame. The analysis appears to be free from chemical suppressions, but a large amount of alkali salt should be added' t o control ionization interferences. Reagents. The reagents are: (1) Lanthanum solution: wet 58.65 g of L a 2 0 3 with water, add 250 ml concentrated HC1 very slowly until the material is dissolved and dilute t o 1 liter. This provides a 5% lanthanum solution in 25% (v/v) HC1. (2) Standard calcium solution: obtain commercially or prepare by adding 50 ml of water t o 0.2497 g of primary standard calcium carbonate, CaC03. Add dropwise a minimum volume of HC1 to dissolve all of the CaCO, and dilute to 1 liter. 1ml of solution contains 100 pg of calcium. Preliminary calibration. Use the standard calcium solution (1ml-100 pg Ca) and transfer the following amounts t o six 50-ml volumetric flasks. To the first flask add 0.5 ml, t o the second 1.0 ml, t o the third 1.5 ml, to the fourth 2.0 ml, and to the fifth 2.5 ml; and the sixth flask should have 0.0 ml. To each flask add 5 ml of the lanthanum solution and sufficient distilled water to adjust the volume t o 50.0 ml. The first flask now contains 1.0 pg/ml Ca, the fifth contains 5.0 pg/ml Ca, and the sixth is a blank. Aspirate these five standards and the blank into an air-acetylene flame and determine the absorbance at 4226.7 A. If the atomic absorption instrument has curvature correction controls, make the necessary adjustments to obtain a linear relationship between absorbance and the actual concentration of the standards. If the instrument does not have these controls, plot the results on linear graph paper as illustrated in Fig. 3.2 by substituting absorbance for intensity. Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.05 mg of calcium. Add 5 ml of the lanthanum stock solution, dilute t o volume with water, aspirate the sample into an airacetylene flame, and determine the absorbance of 4226.7 A. Calculate the approximate sample concentration from the preliminary calibration readings, and determine the aliquot size that will contain 0.05 mg of calcium. Transfer equal aliquots containing 0.05 mg Ca+2 t o three 50-ml volumetric flasks. Add no calcium standard t o the first flask, 0.5 mg to the second flask, and 0.10 mg t o the third.
74
ANALYSIS OF OILFIELD WATERS
Add 5 ml of the lanthanum stock solution to each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample.
Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig.3.3, or Table 3.XI: mg Ca x 1,000 = mg/l Ca+’ ml sample Precision. In a single laboratory using oilfield water samples containing concentrations of 17,400 and 32,500 mg Cat’ /1, the standard deviations were k430 and +1,090, respectively. The recoveries were 103.5% and 100.3’36, respectively. Magnesium (2) The following method for the determination of magnesium in an oilfield water was supplied through courtesy of the Halliburton Company (1970), and can be used t o determine all concentrations of the magnesium ion in a brine.
Reagents. The necessary reagents are magnesium standard solution, 1 mg/ml; lanthanum solution, 1g/ml; and hydrochloric acid. Magnesium standard working so 1ut ions Pipet 1.0 ml of the magnesium standard stock solution into a 1liter flask, add 11.0 ml t o a second 1-liter flask, and add 21.0 ml to a third 1-liter flask. To each flask add 50 ml of concentrated hydrochloric acid, 10 ml of the lanthanum stock solution, and dilute each to an overall 1,000 ml volume with water. This yields standards of 1.0, 11.0, and 21.0 mg/l of magnesium in the first, second, and third flasks, respectively.
Procedure. Filter the sample with the micropore filter apparatus t o remove solids and traces of hydrocarbons from the water. Transfer, by means of “Lambda” pipet or volumetric transfer pipet, an aliquot of sample t o contain not more than 1.0 mg magnesium into a 100-ml volumetric flask. Add 5.0 ml hydrochloric acid, 1.0 ml lanthanum stock solution, and sufficient water to dilute to exactly the 100-ml mark. Mix thoroughly. Aspirate the 5-mg/l standard through the burner, positioning the burner angle as necessary until the recorder indicates a stable reading of about 25% absorption using a wavelength setting of 2852 a. Record the reading and aspirate distilled water through the burner until the recorder returns t o the original baseline. Next, aspirate the sample through the burner until a maximum stable reading is obtained on the recorder. Record the reading and if the sample
ATOMIC ABSORPTION METHODS
75
reading on the recorder is greater than the 5-mg/l standard, aspirate the 9-mg/l standard through the burner until a maximum stable reading is obtained. Record the reading and if the sample reading on the recorder is less than the 5-mg/l standard, aspirate the 1-mg/l standard through the burner until a maximum, stable reading is obtained, and record the reading.
Calculations:
where %A = percent absorption of high standard; %I2 = percent absorption of low standard; %A, = percent absorption of sample; mg/ll = mg Mgt2 /1 of high standard; mg/12 = mg Mg+2/1 of low standard; mg/l, = mg Mg+2/1 of sample; and DF= dilution factor of sample (100/ml sample). Derivation of above equation: %A 1-%A 2 - %A s-%A 2 mg/l1-mg/l2 mg/l,-mg/l2
or :
when mg/l, = 11,mg/12 = 1 ; A mg/l(l-2) = 1 0 when mg/ll = 21, mg/12 = 11;A mg/l( - 2 = 1 0 mg/ll -mg/12 = 10, when standards of 21 mg/l and 11 mg/l or 11 mg/l and 1mg/l are used.
Calcium (2) The same apparatus used in determining magnesium by atomic absorption can be used t o determine calcium.
Reagents. The necessary reagents are calcium standard solution, 1 mg/ml; lanthanum solution, 1g/ml; and hydrochloric acid. Calcium standard working solutions Pipet 1.0 ml of the calcium standard stock solution into a 1-liter flask, add 11.0 ml t o a second 1-liter flask, and 21.0 ml to a third 1-liter flask. To each flask add 50 ml of concentrated hydrochloric acid, and 10 ml of the lanthanum stock solution, and dilute each to an overall 1,000 ml volume
76
ANALYSIS OF OILFIELD WATERS
with water. This yields standards of 1.0, 11.0, and 21.0 mg/l of calcium in the first, second, and third flasks, respectively.
Procedure. Filter the sample through the micropore filter apparatus to remove solids and traces of hydrocarbons from the water. Transfer, by means of micropipet or volumetric transfer pipet, an aliquot of sample containing not more than 2.0 mg calcium into a 100-ml volumetric flask. Add 5.0 ml hydrochloric acid, 1.0 ml lanthanum stock solution, and sufficient water t o dilute to exactly the 100-ml mark and mix thoroughly. Aspirate the ll mg/l standard through the burner, positioning the burner angle as necessary until the recorder reaches a maximum stable reading of about 22% absorption using a wavelength setting of 4227 A. Record the reading and aspirate distilled water through the burner until the recorder returns t o the original baseline. Remove and aspirate the sample through the burner until a maximum stable reading is obtained on the recorder. Record the reading and aspirate distilled water through the burner until the recorder returns t o the original baseline. If the sample reading on the recorder is greater than the 11 mg/l standard, aspirate the 21 mg/l standard through the burner until a maximum stable reading is obtained. Record the reading and if the sample reading on the recorder is less than the 11 mg/l standard, aspirate the 1 mg/l standard through the burner until a maximum, stable reading is obtained. Record the reading.
Calculations: (%A,--%A 2 ) 10 + mg/12 x DF = mg/l Ca+2 %A1 -76 2 where %A = percent absorption of high standard; %A2 = percent absorption of low standard; %A, = percent absorption of sample; mg/ll = mg Ca+2/lof high standard; mg/12 = mg Ca+2/l of low standard; mg/l, = mg Ca+?/l of sample; and DF = dilution factor of sample (100/ml sample).
Strontium Strontium is determined at the 4607 A wavelength with an air-acetylene flame.
Interferences. The chemical suppression caused by silicon, aluminum, and phosphate is controlled by adding lanthanum. The lanthanum also controls ionization interference. The nitrous oxide-acetylene flame can be used t o control chemical interferences, but a large excess of alkali salt should be added t o control ionization.
ATOMIC ABSORPTION METHODS
77
Reagents. The necessary reagents are: (1) Lanthanum solution (same as used in the calcium standard-addition procedure). (2) Standard strontium solution: obtain commercially or dissolve 2.415 g of strontium nitrate, Sr(N03)2,in 1 liter of 1% (v/v) HNO,. 1 ml of the solution contains 1,000 pg of strontium. Preliminary calibration. Prepare standard strontium solutions containing 1-10 pg/ml of strontium using the standard strontium stock solution and 50 ml of volumetric flasks. Add to each of these and t o a blank, 5 ml of the lanthanum stock solution. Aspirate these standards and the blank as suggested in the calcium method and determine the absorbance of strontium at 4607 A. Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.1 mg of strontium (see Fig. 3.6). Add 5 ml of the lanthanum stock solution, dilute t o volume, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings, and determine the aliquot size that will contain about 0.1 mg strontium. Transfer equal aliquots containing about 0.1 mg of strontium t o three volumetric flasks. Add no strontium standard to the first flask, 0.1 mg to the second, and 0.2 t o the third. Add 5 ml of the lanthanum stock solution to each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mg Sr x 1,000 = mg/l Sr+2 ml sample
Precision. In a single laboratory using oilfield water samples containing concentrations of 840 and 2,250 mg Sr+2/1,the standard deviations were +48 and +110, respectively. The recoveries were 106.8%and 103.1%, respectively.
Barium Barium is determined at the 5336 nitrous-oxide flame.
A wavelength with an acetylene
Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of sodium.
78
ANALYSIS OF OILFIELD WATERS
Reagents. The necessary reagents are: (1) Sodium solution: see reagents preparation under “Potassium”. (2) Standard barium solution: obtain commercially or dissolve 1.5161 g of BaClz in 1 liter of water. 1 ml of this solution contains 1,000 pg of barium. Preliminary calibmtion. Prepare standard barium solutions containing 2-1 0 pg/ml of barium using the standard barium solution and 50-ml volumetric flasks. Add t o each of these and to a blank, 0.5 ml of the sodium stock solution. Aspirate these standards and the blank as recommended in the calcium method and determine the absorbance at a wavelength of 5336 8. Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.1 mg of barium. Add 0.5 ml of the sodium stock solution, dilute t o volume with water, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.1 mg of barium. Transfer equal aliquots containing about 0.1 mg of barium to three 50-ml volumetric flasks. Add no barium standard t o the first flask, 0.1 mg of the barium standard t o the second flask, and 0.2 mg to the third. Add 0.5 ml of the sodium stock solution t o each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3 or Table 3.XI: mg Ba x 1,000 = mg/l Ba+’ ml sample
Precision. In a single laboratory using oilfield water samples containing concentrations of 7 and 8 mg Ba+’/l, the standard deviations were k0.5 and kO.9, respectively. The recoveries were 108.2% and 97.3% respectively. Manganese Manganese is determined at the 2794.8 8 wavelength with an air-acetylene flame.
Reagents. The necessary reagent is a standard manganese solution: obtain commercially or dissolve 1.000 g of manganese in a minimum volume of (1 +1)nitric acid. Dilute t o 1 liter with 1%(v/v) HC1.l ml of solution contains 1mg of manganese. Preliminary calibration. Prepare standard manganese solutions containing 1-5 pg/ml using the standard manganese solution and 50-ml volumetric flasks. Aspirate these standards arid a blank as recommended in the calcium method, and determine the absorbance at a wavelength of 2794.8 8.
ATOMIC ABSORPTION METHODS
79
Procedure. Transfer an aliquot containing about 0.05 mg of manganese to a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 0.05 mg of manganese. Transfer equal aliquots containing about 0.05 mg of manganese to three 50-ml volumetric flasks. Add no manganese standard t o the first flask, 0.05 mg of the manganese standard to the second flask, and 0.10 mg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample. ’
Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mg Mn x 1,000 = mg/l Mn+’ ml sample
Precision. In a single laboratory using oilfield water samples containing concentrations of 20 and 97 mg Mn+2/1,the standard deviations were k 1 and +3, respectively. The recoveries were 102.2% and 105.4% respectively. Iron Iron is determined at the 2483.2 flame.
A wavelength with an air-acetylene
Interferences. The sensitivity is reduced if nitric acid and nickel are present. This effect can be controlled by using a very lean (hot) flame. Reagents. The necessary reagent is a standard solution: obtain commercially or dissolve 1.000 g of iron wire in 50 ml of (1+ 1)nitric acid and dilute to 1 liter with water. 1 ml of solution contains 1mg of iron. Preliminary calibration. Prepare standard iron solutions containing 1-5 Mg/ml using standard iron solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 2483.2 A. Procedure. Transfer an aliquot containing about 0.05 mg of iron to a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.05 mg of iron. Add no iron standard t o the first flask, 0.05 mg of the iron standard to the second flask, and 0.10 mg to the third. Dilute to volume. Aspirate and record the absorbance readings for each sample.
ANALYSIS OF OILFIELD WATERS
80
Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mg Fe x 1,000 = mg/l Fe+’ ml sample
Precision. In a single laboratory using oilfield water samples containing concentrations of 6.3 and 6.8 mg Fe+2/1, the standard deviations were k0.5 and k0.3, respectively. The recoveries were 115.6%and 97%,respectively. copper Copper is determined at the 3247.5 8 wavelength with an air-acetylene flame.
Reagents. The necessary reagent is a standard copper solution: obtain commercially or dissolve 1.000 g of copper metal in a minimum volume of (1 + 1) nitric acid. Dilute 1 liter with 1% (v/v)’ nitric acid. 1 ml of solution contains 1 mg of copper. Preliminary calibration. Prepare standard copper solutions containing 1-5 pg/ml using the standard copper solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 3247.5 8. Procedure. Transfer an aliquot containing about 0.05 mg of copper t o a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 0.05 mg of copper. Transfer equal aliquots containing about 0.05 mg of copper to three 50-ml volumetric flasks. Add no copper t o the first flask, 0.05 mg of the copper standard to the second flask, and 0.10 mg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mgCux 1,000 = mg/l CU+’ ml sample
Precision. In a single laboratory using an oilfield water sample containing a concentration of 3 mg Cu+’ /1, the standard deviation was k0.2. The recovery was 100.5%.
zinc Zinc 1- determined at the 2138.6 flame.
A wavelength with an air-acetylene
ATOMIC ABSORPTION METHODS
81
Reagents. The necessary reagent is a standard zinc solution: obtain commercially or dissolve 0.500 g of zinc metal in a minimum volume of ( 1 +1) HC1 and dilute to 1 liter with 1%(v/v) HCl. 1 ml of solution contains 500 pg of zinc. Preliminary calibration. Prepare standard zinc solutions containing 0.2-1.0 pg/ml using the standard zinc solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 2138.6
a.
Procedure. Transfer an aliquot containing about 10 pg of zinc to a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 10 pg of zinc. Transfer equal aliquots containing about 10 pg of zinc t o three 50-ml volumetric flasks. Add no zinc standard to the first flask, 10 pg of the zinc standard t o the second flask, and 20 pg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under “Lithium” in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
Precision. In a single laboratory using oilfield water samples containing concentrations of 27 and 120 mg Zn+2/1, the standard deviations were +1. The recoveries were 103.5%and 102.3%,respectively. Lead (1) Lead is determined at the 2833.1 A wavelength with an air-acetylene flame.
Reagents. The necessary reagent is a standard lead solution: obtain commercially or dissolve 1.598 g of lead nitrate, Pb(N03)2,in 1 liter of 1%(v/v) HN03. 1 ml of solution contains 1,000 pg of lead. Preliminary calibration. Prepare standard lead solutions containing 2-10 pg/ml using the standard lead solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 2833.1 8. Procedure. Transfer an aliquot containing 100 pg of lead t o a 50-ml volumetric flask. Dilute to volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 100 pg of lead.
82
ANALYSIS OF OILFIELD WATERS
Transfer equal aliquots containing about 100 pg of lead t o the three 50-ml volumetric flasks. Add no lead standard t o the first flask, 100 pg of lead standard to the second flask, and 200 pg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample.
Calculations. See calculations under “Lithium” in the flame spectrophotometric section, and Fig. 3.3, or Table 3.XI: mg Pb x 1,000 = mg/l Pb+’ ml sample Precision. In a single laboratory using an oilfield water sample containing a concentration of 16 mg Pb+*/l,the standard deviation was k2.6. The recovery was 74.8%. Lead (2) Lead is determined by chelating with ammonium pyrollidine dithiocarbamate (APDC) and extracting with methyl isobutyl ketone (MIBK) (Brooks et al., 1967). The organic extract is analyzed by means of atomic-absorption spectrophotometry. Interferences have not been observed in the airacetylene flame.
Reagents. The necessary reagents are methyl isobutyl ketone (MIBK); 0.3M hydrochloric acid; ammonium pyrollidine dithiocarbamate (APDC) (dissolve 1.0 g of APDC in 100 ml of distilled water); bromphenol blue indicator solution (dissolve 0.1 g bromphenol blue in 100 ml of 50% ethanol); 2.5M sodium hydroxide; and lead standard solution. The latter can be bought commercially or made from lead nitrate. The presence of 0.5% nitric acid in the lead standards of low concentrations retards the plating of the lead on the sides of the container. Procedure. Pipet the sample into a 200-ml volumetric flask and adjust the volume to approximately 100 ml with distilled water. Add two drops of the bromphenol blue indicator solution. Adjust the pH by adding 2.5M NaOH by drops until a blue color persists. Add 0.3M HC1 until the blue color disappears. Add 2.0 ml of HC1 in excess. The pH should be 2.4. Add 2.5 ml of the APDC solution and mix. Add 10 ml of MIBK and shake vigorously for 1 minute. Allow the layer to separate and add distilled water until the ketone layer is in the neck of the flask. Aspirate the ketone layer for lead content. Prepare a calibration curve by adding known amounts of lead t o a synthetic brine solution. Calculations: mg Pb (from curve) - mg/l Pb’ ml sample
*
EMISSION SPECTROMETRY
a3
EMISSION SPECTROSCOPY
The basic requirements for all spectroscopic measurements are a source, a dispersion element, and a detector. The source may be an emitter whose emission is to be measured, or it may be a continuum that emits all wavelengths, within a certain range, so that absorption by material in the light path may be measured. In general, emission spectra are concerned with transitions from upper state to lower state electronic levels in atoms and in simple molecular species. Some flames are hot enough to excite upper electronic levels in neutral atoms (un-ionized) and in molecules. Electric discharges produce more vigorous excitation, and a high-voltage spark tends t o increase the ionization of the emitters. In spectrographic analysis the light source first vaporizes and dissociates the sample and second excites the atoms causing them to radiate characteristic spectra. The intensities of the spectral lines of elements excited in a light source are proportional to the concentration of the elements in the sample, thus providing a basis for quantitative analysis. Excitation is mainly thermal in the sources, flames, arcs, and sparks. Temperature is very important in spectrographic analysis because some elements are not easily excited in a thermal source while others are. The ionization potential of the element determines the ease of exciting its spectra. The alkali elements with ionization potentials of 4-5 V are excited in low energy sources while the rare gases with ionization potentials up to 25 V require high temperatures t o be excited. A Bunsen flame gives a temperature of about 1,700'C; an oxyacetylene flame, about 2,700'C; an electric arc, 3,700'C-6,700'C; and an electric spark, about 9,700"C.In the following procedures a plasma arc source was used, capable of temperatures up t o
7,700' C .
The plasma arc was adapted to analytical spectrography by Scribner and Margoshes (1961).The temperature of a direct current arc is increased by thermal-pinch effect. The internal standard method is used in the following procedures, and with this method the intensity of a line of the element present in unknown concentration is measured relative to that of an invariant line of a reference element. With this method the intensity ratio must be highly reproducible. Barium, boron, iron, manganese, and strontium The emission characteristics of barium, boron, iron, manganese, strontium, and lanthanum in 10 solvent systems have been studied (Collins, 1967).The greatest emission enhancement was found in a mixture consisting of 30 ml of water plus 20 ml of 35% n-amyl-alcohol and 65% acetone, as illustrated by Fig. 3.7.Because n-propanol is easier t o work with, it w a s used in the following procedure; however, if additional sensitivity is needed, the n-amyl-alcohol-acetone mixture can be used.
ANALYSIS OF OILFIELD WATERS
84
"'i
35 percent N-Amy1 65 percent acetone
90
Proplonlc Add, Acetone N-Proponoi
w
2 50
a I-
430
2 0
~
t
I
I
I
2
h
I
y
I
I
l
I
~
I
l
I
3 4 5 6 7 8 CARBON, grams per SOml
I
9
f
~
~
i
d
~
I
101I
Fig. 3.7. Relative intensity of lanthanum versus grams of carbon in the solvent aspirated into a plasma arc.
Reagents. The necessary reagents are: Helium. Eastman Kodak D-19 developer. Eastman Kodak rapid fixer and hardener. Stop-bath solution, e.g. 5%acetic acid. Standard spectroscopic stock solutions containing 1 mglml of the following metals (use spectroscopic grade reagents): (1) barium: dissolve 1.4368 g of barium carbonate in a minimum amount of hydrochloric acid and dilute t o 1 liter with distilled water; (2) strontium: dissolve 1.6848 g of strontium carbonate in a minimum amount of hydrochloric acid and dilute to 1 liter with distilled water; (3) boron: dissolve 5.7153 g of boric acid in distilled water and dilute to 1 liter with distilled water; (4) manganese: dissolve 1.5823 g of manganese dioxide in hydrochloric acid and dilute to 1 liter with 6N hydrochloric acid; (5) iron: dissolve 1.00 g of iron wire in aqua regia and dilute t o 1 liter with 6N hydrochloric acid. Internal standard solution: dissolve 2.3455 g of lanthanum oxide in a minium amount of hydrochloric acid and dilute t o 1 liter with distilled water. 1 ml contains 2 mg of lanthanum. Standard solution: prepare a composite standard containing 0.025 mg/ml of manganese, 0.075 mg/ml of iron, and 0.03 mg/ml of strontium by transferring appropriate quantities of the standard spectroscopy stock solutions to a 1-liter volumetric flask. Dilute the resultant mixture t o volume with 6N hydrochloric acid. Synthetic brine solution: prepare a solution containing the following ions, in mg/l: sodium, 32,000; calcium, 4,000; and magnesium, 2,500. Dissolve 73.766 g of sodium carbonate in hydrochloric acid, 9.989 g of calcium carbonate in hydrochloric acid and 2,500 g of magnesium metal in
EMISSION SPECTROMETRY
85
hydrochloric acid. Evaporate these acid solutions to dryness, dissolve the residues in distilled water, combine, and dilute t o 1liter with distilled water. Hydrochloric acid, concentrated. n-Propanol.
Equipment. The necessary equipment includes a spectrograph; a d.c. arc source, 18-A minimum; a plasma arc assembly; a plate-developing machine; a microphotometer; 50-ml volumetric flasks; 10-ml microburets; pipets; and spectrographic plates, Eastman Kodak Type 1-N. Spectrochemical excitation conditions. The conditions which are used to determine barium, boron, iron, manganese, and strontium are as follows: Source, d.c. arc. Current, 18-25 A (keep constant). Voltage, 220 d.c. Pre-exposure time, 5 seconds. Exposure time, 15 seconds. Spectral region, 3200-5200 8,first order and second order. Dispersion, reciprocal linear 8.21 8 / m m first order, 4.02 8 / m m second order or better. Plasma arc assembly. Helium lift gas, 7 liter per minute. Helium tangential gas, 60 liter per minute. Atomizer, Beckman Model 4030 with medium-bore capillary. Arc length above orifice, 7 mm. Full arc length, anode t o cathode, 18 mm. Portion of arc viewed, 2 mm above orifice. Orifice electrodes; lower anode, Ultra Carbon 106 drilled to 3.97 mm in center hole, tapered to 9.53 mm at bottom; center ring, neutral Ultra Carbon 861 drilled to 5.95 mm center hole, tapered to 9.53 mm at bottom. Cathode electrode, vertical position, Ultra Carbon 6.35 mm graphite rod with pointed tip. Slit, 20-p. Filter, 3-step. Spectrographic plate development conditions: 5 minutes in Eastman Kodak D-19 at 2OoC with constant agitation; 30 seconds in stop bath at 3OoC with agitation; 5 minutes in Eastman Kodak rapid fixer and hardener solution at 2OoC with constant agitation; 30 minutes in water rinse with constant fresh supply of water; 30 seconds rinse with distilled water; and 30 minutes in constant air bath to dry. Microphotometer criteria. Slit, 1 2 p wide and 0.5 mm high. Read the background and intensity of the following lines: Ba, 11, 4554.03
ANALYSIS OF OILFIELD WATERS
86
8;B, I, 4995.46 8;Fe, 11, 5198.80 8;Mn, 11, 5152.20 8;Sr, 11,4215.52 8; La, 11, 4086.62 8;and La, 11, 4429.90 8. The background and intensity of the following lines can be read if some of those above are too intense or if more than one line for a given element is wanted: Ba, 11, 4934.09 8; B, I, 4993.56 8;Fe, 11, 4196.74 8; Mn, 11, 5187.46 8; La, 11, 4077.35 8; La, 11, 4123.23 8;Sr, 11, 4077.71 8;and Sr, I, 4607.33 8.
Calibration. A preliminary curve, gamma curve, and calibration curves are needed unless a direct-reading instrument is used. To make a preliminary curve, record an iron spectrum using d.c. arc current excitation at about 4 A. Read the percent transmittance (% 2') of several iron lines at 100%unfiltered portion. (Any filter can be used as long as the 5% T is known.) Plot the 100% unfiltered lines versus the 63.10%filtered lines. The % T of these lines should vary from about 10%T t o 90%T to give a good preliminary curve, shown in Fig. 3.8. After the preliminary curve is plotted, the gamma or emulsion calibration curve is made, as shown in Fig.3.9. There are several methods of establishing a gamma curve. The following example is given: 98 on x-axis set t o equal 0.2, and 96 on the y-axis intersects curve at the same point on the curve that 98 does on the x-axis. The filter factor is now used. In this case, it is 100%/63.10%= 1.585. %?'
98 = 0.2 96 = 0.2
x
1.585
Relative intensity arbitrarily set at 0.2 0.317
Owwit). rotio of filter i s 1.585
100
0 FILTERED, percent
Fig. 3.8. Preliminary curve for emission spectrometry.
87
EMISSION SPECTROMETRY
20-
- 40-
z
P
-
v)
z
-
I
I
I
1
1
1
,
0 R E L A T I V E INTENSITY
Fig. 3.9. Gamma or emulsion calibration curve for emission spectrometry.
A t 96 on the x-axis, find the curve intersection point on the y-axis; in this case, it is 91.
%T 91 = 0.317 x 1.585
Relative intensity 0.502
Repeat above procedure to obtain the following data: 81 = 0.502 x 1.585 0.796 1.262 63 = 0.796 x 1.585 2.000 38 = 1.262 x 1.585 3.17 19.5 = 2.000 x 1.585 9.5 = 3.17 x 1.585 5.024 4.6 = 5.024 x 1.585 7.963 12.621 2 = 7.963 x 1.585 Plot the gamma curve using the above values and plot the values on 3-cycle semilogarithmic paper. Place the 7% T values on the linear portion, usually the x-axis, and place the relative-intensity values on the log portion. The resultant curve should be an inverted S if the linear portion or % T is the x-axis. (Theoretically, only one gamma curve need be plotted for all plates with the same emulsion number.) After the gamma curve is plotted, a calibration curve for each element desired can be plotted, as shown in Fig. 3.10. To do this, spectra are recorded for various concentrations of the element in question. The % T of each of the desired lines is determined, and these % T are referred to the gamma curve to obtain their relative intensities. Ordinarily, internal standards are used t o permit a ratio of the relative intensity of the internal standard line to the relative intensity of the element line to be calculated for each concentration of the element. These ratios are plotted versus the element concentration on 2 x 2-cycle logarithmic paper.
88
ANALYSIS OF OILFIELD WATERS
I '
I
I
0.2
0.4
I I 1 I 0.6 0.8 1.0 2.0 INTENSITY RATIO
I
4.0
t 0
Fig. 3.10. Calibration curve for emission spectrometry.
To obtain data for calibration curves for barium, boron, iron, manganese, and strontium, use size 50-ml volumetric flasks. To one flask add no standard solution; add 1.0 ml t o the second flask; and add 2.5 ml, 5.0 ml, 7.5 ml, and 10.0 ml of standard solution to the third, fourth, fifth, and sixth flasks, respectively. (These aliquots will vary with the sensitivity of your instrument.) Add 2 ml of concentrated hydrochloric acid, 2 ml of internal standard solution, 5 d of synthetic brine solution, 20 ml of n-propanol, and sufficient distilled water t o adjust the final volume t o 50 ml at ambient temperature. For optimum accuracy, prepare duplicate or triplicate samples. Aspirate and burn the samples using the excitation conditions, the development conditions, and the microphotometer conditions described above; plot the curves using the above procedure. The water sample should be adjusted t o a pH of about 1.5 at the time of sampling t o prevent precipitation and adsorption. The sample should be contained in a good quality plastic bottle that has been rinsed first with dilute nitric acid and then with distilled water. Transfer to a 50-ml volumetric flask an aliquot of the sample of sufficient size to provide absolute quantities of the elements which will fall within the calibration curves. The optimum aliquot size will vary from brine to brine; however, equal-size aliquots often can be used for waters with similar specific gravities from the same geologic formation. Add 2 ml of concentrated hydrochloric acid, 2 ml of internal standard solution, 5 ml of synthetic brine solution (or try to approximate the ionic composition of the
EMISSION SPECTROMETRY
89
standard), 20 ml of n-propanol, and sufficient distilled water to adjust the volume to 50 ml at ambient temperature. For optimum accuracy, prepare duplicate or triplicate samples. Aspirate and excite the sample, develop the plate, and read the plate as suggested above. Determine the relative intensity ratios for the following: Ba 4554.03/La 4429.90; B 4995.46/La 4429.90; Mn 5152.20/La 4086.72; Sr 4215.52/La 4086.72; and Fe 5198.80/La 4086.72. Calculations. Refer the calculated ratio to the appropriate calibration curve to determine milligrams of tested ion in the sample. Convert this value to milligrams per liter by use of the following equation: mg from curve x 1,000 = mg/l ml sample The relative intensity ratios for other line pairs can be calculated and used if desired. The precision and accuracy of the method are approximately 2-3% and 4-696, respectively, for strontium and barium; and 5 4 %and 10-1196, respectively, for boron, iron, and manganese. Beryllium Beryllium forms a complex with acetylacetone which can be extracted into chloroform from an aqueous solution. The chloroform extracted is aspirated into a plasma arc, and the beryllium I1 line at 3131.07 A is read. An apparent carbon line at 3036.3 A is used for an internal standard. Reagents. Spectrographic plates, Eastman Kodak Type SA No. 1. Standard beryllium stock solution: dissolve 1.00 g of fused metallic beryllium (spectroscopic grade) in a small amount of 6N hydrochloric acid and dilute t o 1 liter with 1%hydrochloric acid. 1 ml contains 1 mg of beryllium. Standard beryllium solution: prepare a standard by transferring a suitable aliquot of the standard stock solution t o a 1-liter volumetric flask and diluting t o volume with 1%hydrochloric acid. The standard prepared will depend upon the resolution and dispersion of the spectrograph. However, for many instruments, a 0.01 pg/ml solution should be adequate. EDTA solution: dissolve 10 g of disodium ethylenediaminetetraacetic acid and 2 g of sodium hydroxide in water and dilute to 100 ml. Synthetic brine solution: dissolve 80 g of sodium chloride, 30 g of calcium chloride, 10 g of magnesium chloride, 5 g of strontium chloride, and 3 g of potassium chloride in distilled water that is saturated with carbon dioxide and dilute t o 1liter. Hydrochloric acid, concentrated. Sodium hydroxide, 0.5N. Chloroform. Acetylacetone.
ANALYSIS OF OILFIELD WATERS
90
Procedure. The spectrochemical excitations used are the same as those shown in the procedure t o determine barium, etc., with the exception that the spectral region is 2300-3300 8, first order and the slit is 10 p. The plate development conditions are the same as those shown in the procedure t o determine barium, etc., and the microphotometer conditions are the same except that the background and the intensity of only the following lines are read: Be, 11, 3131.07 internal standard line, 3036.3 8;or if the 3131.07-8 line is too intense, the Be, 11, 3130.42-a line can be used. To prevent precipitation and adsorption, immediately acidify the clean, oil-free sample to a pH of approximately 1.5 with concentrated hydrochloric acid. Store the sample for transportation t o the laboratory, in a good quality plastic bottle which previously was washed with dilute nitric acid, rinsed with distilled water, and dried. Transfer an aliquot of the sample estimated to contain 0.01-0.05 pg of beryllium t o a 100-ml beaker, adjust the pH to 0.5 with concentrated hydrochloric acid, adjust the volume to about 30-50 ml with distilled water, boil gently for 5 minutes, and then cool. Add 2 ml of the EDTA solution and adjust the pH of the mixture to 7.0 with 0.5N sodium hydroxide. Add 2 ml of acetylacetone, readjust the pH t o 7.0, mix thoroughly, and allow the solution t o stand for 15 minutes. Transfer the sample to a 125-ml Teflonstoppered, separatory funnel and adjust the volume t o 75 ml with distilled water, add 10 ml of chloroform, and shake the mixture vigorously for 2 minutes. After the phases separate, extract the chloroform phase and centrifuge it. Aspirate the centrifuged extract into the plasma arc using the above excitation conditions. For optimum accuracy, prepare duplicate samples. Develop the plates, make background corrections, and determine the relative intensity ratios for the following lines:
a,
Be 3130.42 Be 3131.07 and 3036.2 3036.3 Determine the concentration of beryllium using a calibration curve prepared by using 0.01-0.05 pg of beryllium standard. This concentration in micrograms can be converted to milligrams per liter by this formula: pg Be (from curve) = mg/l Be+2 ml sample Less than 1 ppb of beryllium can be detected with this method, the precision and accuracy of the method are about 2%and 496, respectively, of the amount present.
Aluminum, Petroleum-associated water containing more than 5 mg/l of aluminum can be analyzed using the same procedure and internal standard that are
MASS SPECTROMETRIC METHODS
91
described above for barium, boron, iron, manganese, and strontium; the can be used. However, if the alumialuminum emission lines at 3082.5 num concentration is less than 5 mg/l, the aluminum should be separated and concentrated from the aqueous phase. This can be done by adjusting the pH of a sample containing up t o 100 pg of aluminum to pH 0.4 with hydrochloric acid, adding 10 ml of a 6% aqueous solution of cupferron, adjusting the pH t o 4.8 with sodium acetate, and extracting the aluminum complex into chloroform. The chloroform phase then is aspirated into the plasma arc using the same conditions and internal standard line that is described above for beryllium.
a
MASS SPECTROMETRIC METHODS FOR STABLE ISOTOPES
The ratios of the stable isotopes of deuterium and hydrogen and of oxygen-18 and oxygen-16 differ in water taken from various sources. These differences are useful in studying the origin of a water, and of studying paleoenvironments if the water is geologically old. The isotopic ratios are measured on a mass spectrometer and are always compared to the ratios found in a standard material because such a comparison proyides greater precision than direct analysis of absolute ratios. Deuterium Friedman and Woodcock (1957) developed a method whereby deuterium is converted t o hydrogen gas by reacting a 0.01-ml sample with hot uranium metal. A mass spectrometer (Friedman, 1953) is used t o compare the deuterium/hydrogen ratio in the emitted gas to the ratio in a standard gas. Replicates agreeing within k0.176 usually are considered satisfactory. The results usually are expressed as deuterium enrichments (+6 values) or depletion (-6 values) relative t o SMOW (standard mean ocean water, with a D/H ratio of 158 x (Craig, 1961b). The standard deviation is about 0.2%, and a sample with a 6 value of -5 has 5% less deuterium than SMOW. Oxygen-18 Epstein and Mayeda (1953) developed a method t o analyze water samples for l 8 0 . A 10-ml sample of water is equilibrated with carbon dioxide at 25OC and an aliquot of the COz is analyzed using a mass spectrometer for l 8 0 . The isotope ratios in the sample are compared to those in a standard material, using the mass spectrometer, which gives a greater precision than direct analysis of the absolute ratios. The standard generally used in SMOW (standard mean ocean water) which is distributed by the National Bureau of Standards (Craig, 1961a). Delta units express the isotopic data as:
ANALYSIS OF OILFIELD WATERS
92
where R is the isotope ratio such as 180/160 or D/H, and the delta values are expressed in per mil like salinity, and &MOW = O%,. COLORIMETRIC METHODS
The instrumental measurement of the absorption of radiant energy at a certain wavelength involves spectrophotometry. The essential components of a spectrophotometer include: (1)Radiant energy source such as a tungsten-filament incandescent lamp for the visible region, while hydrogen or deuterium discharge lamps usually are used for the ultraviolet region. (2) A monochromator, which is a device that isolates a narrow band of the radiant energy. (3) Containers, cells, or cuvettes usually made of glass to hold the solution being analyzed. (4) A detector, which is a device (usually a phototube) that measures the radiant energy passed through the solution. In the application of spectrophotometric analysis the two terms “transmittance” and “absorbance” are important. Transmittance is:
T =-I 2 I1
where T = transmittance; II = radiant energy incident upon the first surface of the sample; and I2 = radiant energy leaving the sample. The term absorbance is defined as: 1 A = -1ogIJ” = lOg1,T or the negative logarithm of the transmittance. In the preparation of spectrophotometric curves of light-intensity ratio plotted against concentration, it is preferable, for convenience, t o use absorbance as the basis of the plot. Under these conditions a system that conforms t o Beer’s law gives a straight-line plot, and the commonly used colorimetric systems that do not conform will usually show only a moderate curvature (Willard et al., 1965). Extreme curvature, when the curve is plotted on the basis of absorbance data, is sometimes a sign that the system is not sufficiently stable for analytical purposes. Semicolloidal suspensions of colored substances often give extreme curvatures. When transmittance data are used for plotting, a curve is always obtained unless semilogarithmic coordinates are used. The modern. spectrophotometers have an absorbance calibration as well as the conventional “percent transmittance”, and it is common practice t o use the absorbance scale. The relations between trans-
COLORIMETRIC METHODS
93
mittance and absorbance plots for potassium permanganate solutions at three wavelengths are illustrated by Mellon (1950, p.95). Several other terms for light absorption are given in the literature and are still found on the printed scales of some photometers. “Optical density” is often used; it is the same as absorbance. Interferences In spectrophotometric determinations, interferences often result from the presence in the sample of dissolved or suspended foreign material that either absorbs radiant energy or reacts with the color reagent to form a complex that absorbs radiant energy. In either case, the absorbance of the sample is decreased. Where the interference results from the formation of an absorbing complex by ions in solution, dilution of the sample can eliminate the interference if the sensitivity of the color reagent for the element sought is sufficiently greater than for the interfering ions. If this is not the case, other methods must be found t o increase the selectivity of the method. Among such methods are: (1)pH adjustment: if pH is an important factor in complex ion formation, its adjustment can favor the formation of the complex of the element desired instead of the interfering ions. (2) Masking: compounds such as EDTA (ethylenediaminetetraacetic acid) are added t o the sample t o form a stable complex with interfering ions, thus preventing their reaction with the color reagent. (3) Solvent extraction: preferential solubility of some ions in organic solvents permits the removal of interfering ions. Another common source of interference in spectrophotometry is the use of color reagents that absorb at the wavelength at which the complex of the element desired is measured. Such interference usually can be reduced or eliminated by the use of a reagent blank. In some samples a significant source of interference results from the presence of natural color. The natural color in water samples often gives appreciable absorbance and requires either compensation or elimination. In some cases it is possible t o select a spectrophotometric reagent of sufficient sensitivity that the absorbance of the constituent sought will exceed the absorbance of the natural color by a large factor. If this factor is 50 or higher the error caused by the natural color is 2% or less. Knowledge of the relative sensitivity of the constituent to be determined relative to the natural color in the sample is necessary before such a factor can be used. If the relative sensitivity is unknown the natural color of the sample should be compensated for or removed. This can be done by determining the absorbance of the test sample versus the blank specified for the procedure. Determine the absorbance of the naturally colored sample versus distilled water. The difference is the corrected absorbance and is used to determine concentration values.
ANALYSIS OF OILFIELD WATERS
94
Iron The spectrochemical procedure will give values only for total iron and will not differentiate ferrous iron from ferric iron. The following procedure can be used t o determine F-+* and Fe+3 in a freshly sampled water (Collins et al., 1961). Reagents and apparatus. Standard iron solution: dissolve 1.00 g of hydrogen-reduced iron in a minimum of hydrochloric acid and dilute to 1 liter with distilled water. This solution contains 1mg/ml of iron. Transfer 1 0 ml of this solution to a l-liter flask and dilute to volume with distilled water. 1ml of this solution contains 0.01 mg of iron. Hydroquinone solution: dissolve 1 g of hydroquinone in 100 ml of distilled water.
IRON, m i l l i g r a m
Fig. 3.11. Plot of the optical density at 522 m p of the ferrous iron complex with 2,2'bipyridine.
COLORIMETRIC METHODS
95
O-phenanthroline or 2,2‘-bipyridine (either reagent can be used, how ever, 2,2’-bipyridine is subject t o less interferences): dissolve 0.5 g of either reagent in 100 ml of distilled water. The solution can be warmed to 60°C t o effect more rapid dissolution. Sulfuric acid, approximately 9N (441.36 g per liter): cautiously pour 270 ml of pure concentrated sulfuric acid into 650 ml of distilled water. Carefully mix the solution, cool, and dilute to 1liter with distilled water. Spectrophotometer capable of measurements at 508 mp or 522 mp, glasselectrode pH meter, 100-ml volumetric flasks, 10-ml microburet, and pipets.
Procedure. Prepare a calibration curve by transferring aliquots of the standard iron solution, containing from 0.02 mg t o 0.20 mg of iron, t o 100-ml volumetric flasks. To separate aliquots, add 5 ml of the sodium citrate solution and determine how much sulfuric acid is necessary to adjust the pH t o 3.5. Add this amount to the aliquots in the volumetric flasks. Add reagents in the following order: 5 ml of hydroquinone solution, 5 ml of 2,2’-bipyridine or O-phenanthroline solution, and 5 ml of sodium citrate. The citrate must always be added last. Convert t o volume with distilled water, mix well, and let stand for 1hour. Prepare a reagent blank in the same manner. Determine the absorbance at 522 mp if 2,2’-bipyridine is used or 508 mp if O-phenanthroline is used. Plot the absorption versus iron concentration on coordinate graph paper. The resulting curve should be linear, as shown in Fig. 3.11. Obtain a clean sample of brine, free of oil. Determine ferrous iron, by following the above procedure, but omit the addition of hydroquinone. To determine dissolved iron, filter the sample and follow the above procedure. To determine total iron, do not filter the sample. The amount of ferric iron can be calculated from the difference. Calculations: 1,000 x mg iron from curve = mg/l Fe+2 or Fe+’ sample volume Concentrating copper, iron, lead, and nickel by ion exchange To determine accurately, using colorimetric methods, copper, nickel, lead, zinc, and cadmium in oilfield brines, they should be separated from interfering ions. Many oilfield brines contain metals in such minute amounts that they must be concentrated before analyses can be made. Concentration methods investigated were ion exchange, electro-deposition, solvent extraction, and evaporation. An ion-exchange method proved t o be the most practical for concentrating copper, nickel, and lead, because it is less time consuming and requires less expensive equipment than any of the other methods studied.
96
ANALYSIS OF OILFIELD WATERS
Acidifying the samples to pH 3.5 with acetic or hydrochloric acid minimizes precipitation and adsorption. If acetic acid is used, 2 ml of formaldehyde per liter of sample should be added to retard mold growth. These precautions will aid in obtaining representative heavy metal analyses; however, to obtain optimum results, the samples should be analyzed as quickly after sampling as possible. If it is necessary to store the samples, they should be stored in a cool, dark place and should not be moved frequently. Light accelerates photochemical reactions, and high temperatures and moving accelerate chemical reactions. Once the seal of the cap of the sample bottle has been broken, the sample should be analyzed immediately. A chelating ion-exchange resin such as Dowex A-1 can be used to separate copper, iron, nickel, and lead from an aqueous solution. Slurry the resin into a plastic column about 36 cm long and 1.7 cm in diameter. Convert the resin to the sodium form by washing with 2 volumes of distilled water, 1 volume being equal to the amount of resin used, followed by 2 volumes of 1 N sodium hydroxide, and then with 10 volumes of distilled water. Because the resin expands more than 100%when changing from the hydrogen form to the sodium form, the column must be backwashed frequently t o reduce compaction of the resin and to prevent shattering of the column. Pass the brine which has been neutralized to pH 7.0 with sodium hydroxide through the column. 2 liters or more probably will be necessary, depending upon the amount of heavy metals present in the brine. Elute the chelated metals with 2 volumes of 2N hydrochloric acid and water effluents t o a small volume; cool and adjust t o a predetermined volume (for example, 200 ml) with water. Use aliquots of this solution for determining copper, iron, nickel, and lead. The resin must be changed back t o the sodium form as soon as the metals have been eluted, because the resin tends to lose its chelating capacity if left in the water-rinsed hydrogen form for longer than a few hours. If this happens, the resin can be regenerated by heating it at 6OoC in a 30-50% sodium hydroxide solution for 24 hours. Once the metals are separated from the brine and concentrated, they can be analyzed using various methods such as atomic absorption spectrometry, flame spectrometry, emission spectrometry, or colorimetry (Collins et al., 1962).
The compound 2,9-dimethyl-1,lo-phenanthroline, assigned the name neocuproine (Diehl and Smith, 1958, p.23), has the following structure:
COLORIMETRIC METHODS
97
This reagent is used t o determine copper because of its nearly specific reaction with cuprous copper. The combining ratio is 2 moles of neocuproine to 1 mole of copper. The increased selectivity of neocuproine for copper is caused by a steric hindrance effect. The cuprous neocuproine compound is formed over a pH range of 3-10 and is bright orange. The compound can be extracted with n-amyl alcohol, isoamyl alcohol, n-hexyl alcohol, or chloroform. The maximum absorption of the compound in isoamyl alcohol occurs at a wavelength of 454 mp. Hydroxylamine hydrochloride can be used t o reduce the cupric ion t o cuprous. Citrate will hold any iron present in solution when the pH is adjusted to between 5 and 6. The chromate ion can cause low results; however, this effect does not occur when iron is present, which is almost always the case with an oilfield brine. The anions such as sulfide, cyanide, periodate, nitrate, t hiocyanate, and ferricyanide can interfere by reacting with hydroxylamine; however, they are eliminated in the ion exchange separation. Reagents. Neocuproine solution: dissolve 1 g of 2,9-dimethyl-l,1 O-phenanthroline in 1liter of ethyl alcohol. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine hydrochloride in 100 ml of water. Isoamyl alcohol, analytical reagent grade. Sodium citrate solution: dissolve 300 g of sodium citrate in 1 liter of water, add 2 ml of the hydroxylamine hydrochloride solution, add 1 ml of neocuproine solution, and extract with 10-ml portions of chloroform until a colorless chloroform extract is obtained. Standard copper solution: dissolve 0.100 g of copper in 5 ml of nitric acid and 5 ml of water by heating gently to dissolve the copper. Add 5 ml of perchloric acid and evaporate to fumes of perchloric acid. Cool, dilute with water, transfer t o a l-liter volumetric flask, and dilute t o volume. Pipet a 100-ml aliquot of this solution t o another l-liter volumetric flask. Dilute to volume with water. This solution contains 10 mg/ml of copper. Sodium acetate.
Procedure, Add 5 ml of 10% hydroxylamine hydrochloride solution and 20 ml of 30% sodium citrate solution to a sample of effluent from the ion exchange column containing 4-150 pg of copper, and adjust the pH of the mixture to between 5 and 6 with 1 g or more of sodium acetate. Extract with a 10-ml portion of isoamyl alcohol. Separate the liquids and discard the alcohol layer. Add 10 ml of 0.1%neocuproine solution and 10 ml of isoamyl alcohol, and shake the mixture vigorously for 1 minute. Let the phases separate and transfer the alcohol layer to a 50-ml volumetric flask. Make additional extractions until the alcohol layer remains colorless. Dilute the combined alcohol extracts t o 50 ml with isoamyl alcohol, mix, and measure the absorbance at 454 mp in a l-cm cell with a spectrophotometer.
ANALYSIS OF OILFIELD WATERS
30
-.#
Calculations. Estimate the amount of copper present by using a calibration curve prepared by using about 10-200 pg of copper: pg Cu (from curve)
ml sample
=
mg/l C U + ~
Nickel Nickel forms a wine-red or brown compound with dimethylglyoxime (Sandell, 1959, p.555). The structure of the chelate on the basis of available evidence is: H3C - C-C
//
/I
H3C - C
=
/I
C
CH3
=
CH,
Dimethylglyoxime gives a nearly specific reaction with nickel that has been oxidized t o its higher valences with an oxidizing agent such as bromine. The wine-red compound is somewhat unstable; therefore, the absorbance measurements should be made within 10 minutes after formation of the nickel dimethylglyoximate. Cobalt and copper also give colored compounds with dimethylglyoxime, but they can be removed by washing the chloroform extract of nickel dimethylglyoximate with dilute ammonium hydroxide. Iron interference is removed by extracting the nickel dimethylglyoximate with chloroform from a solution containing citrate. Palladium, platinum, and gold also give colored compounds when nickel dimethylglyoximate is extracted with chloroform; however, they are removed, if present, by the ion-exchange separation.
Reagents. Dimethylglyoxime solution: dissolve 1 g of dimethylglyoxime in 100 ml of ethyl alcohol. Saturated bromine water. Ammonium hydroxide solution, approximately 4N: add 200 ml of concentrated ammonium hydroxide to 800 ml of water. Standard nickel solution: dissolve 0.100 g of nickel in dilute nitric acid by heating gently. Cool, dilute with water, transfer to a 1-liter volumetric flask, and dilute to volume. Pipet a 100-ml aliquot of this stock solution into another 1-liter volumetric flask and dilute to volume. This solution contains 10 pg/ml of nickel. Hydrochloric acid, approximately 6N: cautiously add 500 ml of concentrated hydrochloric acid to 500 ml of water. Chloroform.
COLORIMETRIC METHODS
99
Ammonium citrate solution: dissolve 1 0 g of ammonium citrate in water and dilute to 100 ml. Hydroxylamine hydrochloride solution: dissolve 1 0 g of hydroxylamine hydrochloride in water and dilute to 50 ml.
Procedure. Add 10 ml of ammonium citrate solution, and 5 ml of hydroxylamine hydrochloride solution to a sample of effluent containing up to 100 pg of nickel, and adjust the pH t o 8 with ammonium hydroxide. Transfer the mixture t o a 125-ml separatory funnel, add 10 ml of dimethylglyoxime solution and 10 ml of chloroform, and shake the mixture vigorously for 1 minute. Let the phases separate and extract the chloroform phase into another 125-ml separatory funnel. Make additional extractions of the sample with 10-ml portions of chloroform until a colorless chloroform extract is obtained. Add 10 ml of 4N ammonium hydroxide solution to the combined chloroform extracts in the 125-ml separatory funnel, and shake the mixture vigorously for 1 minute. Let the phases separate and discard the ammonium hydroxide phase. Acidify the chloroform phase with 1 ml of 6N hydrochloric acid, shake the mixture vigorously for 2 minutes, let the phases separate, and discard the chloroform phase. Add 10 ml of chloroform t o the acid phase, shake the mixture vigorously for 1 minute, and discard the chloroform phase. Adjust the pH of the acid phase to 6.9, transfer it t o a 100-ml volumetric flask, add bromine water until a yellow color persists, swirl the mixture, and allow it to stand for 10 minutes. Add 10 ml of 4N ammonium hydroxide and 10 ml of dimethylglyoxime solution. Swirl to mix, cool to room temperature in an ice water bath, and adjust t o 100-ml volume with water. After 5 minutes determine the absorbance at 445 mp using a 1-cm cell and a spectrophotometer. Calculations. Calculate the nickel concentration in the water by using a calibration curve prepared by using about 10-100 pg of nickel: pg Ni (from curve) = mg/l Ni+ ml sample
’
Lead Dithizone (Sandell, 1959, p. 665) is an excellent reagent for the determination of traces of lead. Lead dithizonate probably has a formula similar to the following: r C6HS
I
100
ANALYSIS OF OILFIELD WATERS
Lead can be extracted from a basic solution with dithizone in chloroform or carbon tetrachloride in the presence of citrate or tartrate, which prevent the precipitation of several metal hydroxides. The optimum pH range for extraction of the lead dithizonate with chloroform is 8.5-11. Cyanide will complex all interfering metals except bismuth, thallium, and stannous tin. Because these metals are separated by ion,exchange, their interference is eliminated. Ferric iron can form a ferricyanide that will oxidize dithizone; however, this reaction can be prevented by adding a reducing agent such as hydroxylamine hydrochloride. Excess of calcium, magnesium, and phosphorus retards the lead dithizonate extraction, but thz ion exchange separation excludes phosphorus as well as much of the calcium and magnesium. The lead dithizonate in chloroform absorbs at 510 mp. The amount of lead in the chloroform phase should not be much greater than 2.5 mg/l for optimum results.
Reagents. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine hydrochloride in water and dilute to 50 ml. Standard lead solution: dissolve 0.100 g of lead in 10-15 ml of nitric acid. Dilute to 1liter volume with water. Pipet a 100-ml aliquot of this stock solution into another 1-liter volumetric flask, add 1 0 ml of nitric acid, and dilute t o 1 liter volume with water. This solution contains 10 pg/ml of lead. Ammonia-cyanide-sulfite solution: add 350 ml of concentrated ammonium hydroxide, 30 ml of a 10%potassium cyanide solution, and 1.5 g of sodium sulfite, t o a 1-liter volumetric flask and dilute to volume with water. Dithizone solution: dissolve 0.01 g of dithizone in 200 ml of chloroform. Chloroform.
Procedure. Transfer a sample of the ion exchange effluent containing up t o 80 pg of lead to a 125-ml separatory funnel, and add 5 ml of hydroxylamine hydrochloride solution, 75 ml of ammonia-cyanidesulfite solution, and 10 ml of chloroform. Shake the mixture vigorously for 1minute, let the phases separate, and discard the chloroform phase. Add 1 ml of 0.005% dithizonechloroform solution, shake the mixture vigorously for 1 minute, let the phases separate, and extract the dithizone-chloroform phase into a 25-ml volumetric flask. If the dithizone-chloroform phase is green or some color other than cinnabar red, three possibilities exist: (1)there is no lead present; (2) there is an oxidizing agent present; or (3) an excess of dithizone has been used. In any event, if the dithizone-chloroform phase is not red, acidify it with 15 ml of 1 : l O O nitric acid, shake the mixture for 1 minute t o transfer the lead to the nitric acid phase, and discard the chloroform. Treat the nitric acid phase with hydroxylamine hydrochloride solution, ammonia-cyanidesulfite solution, and make another dithizone-chloroform extraction using 0.5 ml or less of the dithizone-chloroform solution. If the dithizonechloroform phase still does not turn red, take a larger sample of the effluent. However, if the original dithizone-chloroform extraction did turn red, make
COLORIMETRIC METHODS
101
additional extractions until the dithizone-chloroform phase remains green. Dilute the combined red dithizone-chloroform phases in the 25-ml volumetric flask t o volume with chloroform, mix well, and determine the absorbance with a spectrophotometer at 510 mp. Calculations. Prepare a calibration curve by using aliquots of the standard lead solution containing 10-80 pg of lead:
pg Pb (from curve) = mg/l Pb+2 ml sample zinc Extraction of zinc with dithizone from a weakly ammoniacal solution containing a reducing agent and citrate prevents the precipitation of iron. Extraction of zinc at a pH of 4.75 in the presence of sodium thiosulfate largely eliminates interference from copper, mercury, lead, and cadmium. The zinc dithizonate complex can be broken in 0.02N hydrochloric acid, whereas cupric dithizonate cannot. Lead and cadmium dithizonates will dissociate in 0.02N hydrochloric acid, but only traces of them should be present after the preliminary extractions. More accurate results are obtained by applying a zincon (Platte and Marcy, 1959) method t o the zinc which has been isolated by the dithizone extractions than by making another dithizone extraction of the isolated zinc and using it for absorption measurements. Therefore, the following method is a combination of the dithizone and zincon methods. Traces of any remaining interferences can be complexed. Zinc reacts with dithizone to form a compound similar to:
r
Zinc reacts with zincon: OH 1
CsHS
to form a 1:lblue complex that absorbs at a wavelength of 620 mp.
102
ANALYSIS OF OILFIELD WATERS
Reagents. Standard zinc solution: dissolve 1.00 g of zinc metal in hydrochloric acid and dilute to 1 liter with water. Dilute 1 0 ml of the stock solution to 1liter to prepare a standard containing 1 0 pg/ml of zinc. Sodium citrate solution: dissolve 1 0 g of sodium citrate in water and dilute to 100 ml. Hydroxylamine hydrochloride solution : dissolve 1 0 g of hydroxylamine hydrochloride in water and dilute to 100 ml. Buffer solution, pH 4.75: dissolve 130 g of sodium acetate and 57 ml of glacial acetic acid in water and dilute to 1 liter. Dithizone solution: dissolve 0.1 g of dithizone in a liter of alcohol-free carbon tetrachloride. Extract any alcohol from the carbon tetrachloride by shaking it with distilled water. Keep a water blanket on the extracted carbon tetrachloride when storing it. Potassium cyanide solution: dissolve 1.0 g of potassium cyanide in water and dilute to 100 ml. Buffer solution, pH 9.0: dilute 213 ml of lN sodium hydroxide to 600 ml with water. Dissolve 37.3 g of potassium chloride and 31.0 g of boric acid in water, mix with the sodium hydroxide, and dilute to 1liter. Zincon solution: dissolve 0.13 g of zincon in 2 ml of 1N sodium hydroxide and dilute to 100 ml with water. Chloral hydrate solution: dissolve 10 g of chloral hydrate in water and dilute to 100 ml. Hydrochloric acid, 0.02N:add 1.7 ml of concentrated hydrochloric acid to water and dilute to 1liter. Ammonium hydroxide. Sodium ascorbate. Sodium thiosulfate solution: dissolve 25 g of sodium thiosulfate in water and dilute to 100 ml. Procedure. Add 1 0 ml of the hydroxylamine hydrochloride solution to an aliquot of brine containing up to 200 pg of zinc, mix, add 1 0 ml of sodium citrate solution, and adjust the pH to 8.3 with ammonium hydroxide. Transfer the sample to a separatory funnel, add 3 ml of 0.01% dithizone solution, and shake the mixture vigorously for 1 minute. Let the phases separate and note the color of the dithizone phase. If any zinc is present, the dithizone phase will be red or violet, but not green. If the dithizone phase is green, take a larger aliquot of brine. If the dithizone phase is red or violet, extract it into another separatory funnel containing 1ml of sodium thiosulfate solution and 1 0 ml of pH 4.75 buffer solution. Make additional extractions of the brine solution with dithizone solution until the dithizone remains green, which indicates that all the zinc has been extracted. This is important because the final dithizone phase must be green, not violet. Discard the brine solution and wash the combined dithizone extracts by mixing them vigorously for 1minute with the buffer solution. Let the phases separate, extract the dithizone phase into another separatory funnel con-
COLORIMETRIC METHODS
103
taining 10 ml of 0.02N hydrochloric acid, and shake this mixture vigorously for 2 minutes. Let the phases separate, and extract and discard the carbon tetrachloride phase. Wash the acid phase twice with carbon tetrachloride, extract, and discard the carbon tetrachloride. Transfer the acid phase to a 50-ml volumetric flask and make t o volume with water. Pipet 10-ml aliquots from the 50-ml volumetric flask to two 50-ml Erlenmeyer flasks. To both flasks add 0.5 g of sodium ascorbate followed by, in this order and with mixing, 1 ml of potassium cyanide solution, 5 ml of pH 9.0 buffer solution, and 3 ml of zincon solution. To one sample add 3 ml of chloral hydrate solution, and t o the other (which is the reference solution) add 3 ml of water. Within 2-5 minutes after adding the last reagent, measure the absorbance of the sample versus the reference solution at 620 mp in 1-cm cells with a spectrophotometer.
Calculations. Prepare a calibration using aliquots of the standard zinc solution containing 10-80 pg of zinc, and use the curve t o calculate the amount of zinc in the sample: pg Zn (from curve) = mg/l Zn+* ml sample
Cadmium Cadmium can be extracted from aqueous solutions as cadmium dithizonate into carbon tetrachloride or chloroform. Cadmium dithizonate is extracted more readily into carbon tetrachloride than is zinc dithizonate, but zinc dithizonate is extracted more readily into chloroform than the cadmium compounds. Therefore, because many oilfield brines contain more zinc than cadmium, the cadmium extraction should be made with carbon tetrachloride to insure the best possible separation from zinc. Although citrate and tartrate do not hinder the cadmium dithizonate extraction, they do impede the extraction of lead and zinc. Cadmium dithizonate can be extracted from an alkaline solution containing cyanide and tartrate; the dithizonates of nickel, copper, silver, and tin are not extracted. Most of the interference from iron can be eliminated by oxidizing it with peroxide and filtering. Cadmium reacts with dithizone to form a compound of the type:
104
ANALYSIS O F OILFIELD WATERS
Cadmium dithizonate in carbon tetrachloride absorbs strongly at a wavelength of 620 mp.
Reagents. Standard cadmium solution: dissolve 0.100 g of cadmium metal in hydrochloric acid and dilute t o 1 liter with water. Pipet a 10-ml aliquot of this stock solution into another l-liter volumetric flask and dilute to volume. This solution contains 1pg/ml of cadmium. Ammonium chloride, 1N: dissolve 53.5 g of ammonium chloride in water and dilute to 1liter. Rochelle salt solution: dissolve 100 g of rochelle salt (KNaC4H406 4H20) in water and dilute t o 1liter. Sodium citrate solution: dissolve 100 g of sodium citrate in water and dilute to 1 liter. Hydrogen peroxide, reagent grade 30%hydrogen peroxide. Tartaric acid solution: dissolve 20 g of tartaric acid in 1 liter of water. Store in a refrigerator and discard if any mold is present. No.1 dithizone reagent: dissolve 0.12 g of dithizone in 1 liter of carbon tetrachloride. Store in a refrigerator in a dark bottle. No.2 dithizone reagent: dilute 5 ml of No.1 reagent t o 100 ml with carbon tetrachloride. Store in the refrigerator. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine hydrochloride in 50 ml of water. Prepare fresh weekly. Sodium hydroxide (35%)-potassium cyanide (1%) solution: dissolve 175 g of sodium hydroxide and 0.5 g of potassium cyanide in water and dilute to 1 liter. Ammonium hydroxide 5M: dilute 16.0 ml of concentrated ammonium hydroxide (14.8M) to 50 ml. Sodium hydroxide, 5% solution: dissolve 5 g of sodium hydroxide in water and dilute to 100 ml.
-
Procedure. Filter the brine through Whatman No.4 filter paper (double thickness). Transfer 900 ml or less of the filtered brine to a 2-liter beaker, add 5 ml of 30% hydrogen peroxide, and heat until complete decomposition of the excess hydrogen peroxide is attained. Cool the solution and filter if any precipitate is present. Add 100 ml of ammonium chloride solution, 10 ml of rochelle salt solution, 25 ml of sodium citrate solution, and adjust the pH to between 8 and 8.5 with 5M ammonium hydroxide. Transfer the solution to a liter separatory funnel, add 15 ml of the No.1 dithizone solution, and shake the mixture vigorously for 5 minutes. Let the phases separate and extract the dithizone phase into a 50-ml separatory funnel. Reextract the brine with another 15 ml of No.1 dithizone solution. Separate the dithizone phase into the 50-ml separatory funnel and discard the brine phase. Add 10 ml of tartaric acid solution to the combined dithizone extractions in the 50-ml separatory funnel and shake the mixture vigorously for 2
COLORIMETRIC METHODS
105
minutes. Discard the carbon tetrachloride phase and wash the tartaric acid phase twice with a 3-ml portion of carbon tetrachloride. To the tartaric acid phase add 1 ml of hydroxylamine hydrochloride solution, 5 ml of the 35% sodium hydroxide-1% potassium cyanide solution, and 10 ml of the No. 2 dithizone solution Shake the mixture vigorously for 1minute. Let the phases separate and extract the dithizone phase into another 50-ml separatory funnel. Reextract the aqueous phase with 10 ml of No.2 dithizone solution, and add the dithizone extraction of the previous separation. Wash the aqueous phase with 5 ml of carbon tetrachloride, extract the carbon tetrachloride, and combine it with the two dithizone extractions. Discard the aqueous phase. Add 15 ml of 5% sodium hydroxide solution to the combined dithizone extractions, shake the mixture vigorously for 1 minute, extract the carbon tetrachloride phase, and determine its absorbance at 620 mp in a l-cm cell with a spectrophotometer.
Calculations Run a blank and make appropriate corrections, using a calibration curve prepared by using aliquots of the standard cadmium solution containing 1-7 pg of cadmium: pg Cd (from curve) = mg/l Cd+2 ml sample Phosphate Only orthophosphate will respond to the test. Polyphosphates must be reverted to orthophosphates by boiling with acid (American Petroleum Institute, 1968).
Interferences. Color development in the test is inhibited when the dissolved solids content of the sample is greater than 8% (a specific gravity greater than 1.06) or when the total iron is greater than 50 mg/l. In such cases the sample taken for analysis must be diluted with distilled water so that these limits are not exceeded. Sulfide interferes by giving high results, and should be destroyed by adding potassium permanganate solution t o the acidified sample. Reagents. Hydrochloric acid, concentrated. Reagent No. 1: dissolve 46 g of ammonium molybdate [ (NH), )a Mo, 02, 4H201 in 700 ml of distilled water. The ammonium molybdate used should consist of white crystals without a bluish-green tinge. Add 2.5 ml of concentrated ammonium hydroxide to the solution and dilute t o 1liter with distilled water. Amino solution: dissolve 10 g (about 1 level tablespoon) of amino powder mixture in 100 ml of distilled water. If solution remains turbid, filter. Store solution in a well-stoppered, brown glass bottle and prepare fresh at least every 2 weeks.
106
ANALYSIS OF OILFIELD WATERS
The amino powder mixture is made up by adding 5 g of sodium sulfite and 1.0 g of l-amino-2-naphthol-4-sulfonic acid to a dry mortar. Grind the materials to a fine powder. Transfer the powder t o a large wide-mouthed bottle containing 66.5 g of sodium bisulfite (meta, powder, Naz Sz 0, ) and 35 g of sodium sulfite. Mix well by shaking. If the mix is not uniform, it should be passed through a 20-mesh screen and again shaken in the large bottle. Store mixture in a well-stoppered, wide-mouthed brown bottle. Standard phosphate solution: dissolve in distilled water 0.1335 g of potassium dihydrogen phosphate (KHz PO4 ) which has been dried in an oven at 105°C. Dilute t o 1 liter. 1ml of this solution is equivalent to 0.1 mg sodium metaphosphate (NaP03).
Procedure. Thoroughly shake a freshly drawn sample to disperse the solids and pipet 100-ml aliquot into each of two 250-ml beakers. If the expected concentration of sodium metaphosphate is greater than 10 mg/l, take smaller aliquots diluted t o 100 ml with distilled water. Note: phosphate-free glassware must be used in this determination. The glassware should be soaked in dilute hydrochloric acid, followed by rinsing with distilled water. Add 7 ml of concentrated hydrochloric acid to one of the samples. If it is suspected t o contain sulfide, stir the solution vigorously for a minute to remove as much of the sulfide as possible, then add potassium permanganate solution (8%) dropwise until the solution just turns pink. Boil solution vigorously for 30 minutes while maintaining the volume between 75 and 100 ml by adding distilled water. Cool sample t o a temperature between 70" and 95°F and dilute t o 107 ml with distilled water in a graduated cylinder bearing a mark at the 107-ml level. Add 7 ml of concentrated hydrochloric acid t o the unboiled sample and treat with permanganate as above if sulfide is suspected. Filter both boiled and unboiled samples if turbid. Add 5 ml of reagent No.1 to both samples and mix well. Add 5 ml of amino solution t o both and again mix well. Ten minutes after the amino solution addition, measure the color with a spectrophotometer at a wavelength of 690 mp, after adjusting the meter to 100%transmittance with a proper blank. Calibration curve. Prepare a calibration curve by using aliquots of the standard phosphate solution containing up t o 10 mg/l of sodium metaphosphate. Calculations. Refer the spectrophotometer readings t o the calibration curve (expressed as milligrams of NaP03 versus photometer reading) t o obtain the sodium metaphosphate concentration. The results on the heated sample correspond to total phosphate, whereas, those on the unheated sample correspond to orthophosphate, the difference being polyphosphate, usually expressed as sodium metaphosphate (NaP03):
COLORIMETRIC METHODS
107
where A = mg NaPO, (heated sample), and B = mg NaPO, (unheated sample).
Precision. The precision is about k3% of the amount present. Silica Silicon is the second most abundant element in the earth’s crust and is found in most rocks as the oxide Si02 or as a silicate such as Mg,Si2O5 (OH), . The solubilities of silicate minerals in saline waters are a function of temperature, pressure, pH, Eh, dissolved gases, and other ions in solution. A limited amount of research has been done concerning silicate solubilities (Collins, 1969) in saline solutions. Some investigators believe that most silica exists in solution as H4Si04 (White et al., 1956);others that it exists both in colloidal form and as H4Si04 (Krauskopf, 1956).Hydration of silica gives the following reaction: Si02 + 2H20 + Si(OH), or H4 Si04 A method developed by Schrink (1965)was used to study silicate solubilities in saline waters (Collins, 1969)and it gave satisfactory results. It also has been used to analyze some petroleum-associated waters. The method involves adding 1 ml of a 4% ammonium molybdate solution in 0.75 molar sulfuric acid solution to an appropriate aliquot of the water sample; add 15 ml of 4.5N sulfuric acid; extract for 1 minute with ethyl acetate; and determine the absorbance of the ester extract with a spectrophotometer at a wavelength of 335 mp.
Nitrate nitrogen Nitrate is the most highly oxidized form of nitrogen and is the most stable form in an oxidizing environment. Many fertilizers contain nitrate, and waters will leach the nitrate from soil or rock. Most rocks do not contain much nitrate; therefore, it is unlikely that petroleum-associated waters contain appreciable quantities of nitrate. The nitrate in deep waters also may be depleted through anion exchange (George and Hastings, 1951). Chloride is a serious interference in many of the methods used to determine nitrate nitrogen. Oxidizing or reducing agents such as ferric or ferrous iron also interfere. The Brucine method (Fisher et al., 1958)can be applied to a petroleum-associated water. To determine the nitrate concentration, transfer an aliquot of the sample containing up to 15 pg of nitrate into a 50-ml Erlenmeyer flask, add 15 ml of water, 1 ml of Brucine reagent (2%
108
ANALYSIS OF OILFIELD WATERS
aqueous solution of Brucine hydrochloride), acid, place in a dark area and allow to cool to 30°C. Determine the absorbance of the sample with a spectrophotometer at a wavelength of 410 mp. Arsenic
The determination of arsenic in brines has received little attention despite its toxic relationship t o fish and animals. The arsenic content of sea water was first investigated by Gautier (1903),who found inconsistent variations. He attributed the higher amounts found at great depth to volcanic influences, and the higher amounts found at the surface t o evaporation and disturbances caused by marine animals. Rakestraw and Lutz (1933)and Gorgy et al. (1948)also studied arsenic in sea water. They conclude that 50-60% of the arsenic is in the arsenite form, with 8--10% each of arsenate, dissolved organic arsenic, and arsenic suspended in particulate matter. Smales and Pate (1952)used an activation analysis method t o determine submicrogram quantities of arsenic in sea water. They found an average of 2.6 pg of arsenic per liter, with a range of 1.6-5.0 pg/l. The water analyzed is believed representative for Atlantic Ocean water. The Gutzeit method can be used to analyze a petroleum-associated water for arsenic (Collins et al., 1961). Arsenic is reduced t o arsine with zinc in acid solution. A yellow t o brown stain is produced when AsH3 passes through paper impregnated with mercuric chloride or mercuric bromide. The colorbrown, and ation is produced by A s H ( H ~ B ~-) ~yellow, A s ( H ~ B ~-) ~ As2Hg3 - black. By comparing unknowns with a series of standard papers prepared with known amounts of arsenic, a quantitative estimation can be made. Papers prepared from mercuric bromide can be preserved for several months in a dark, dry atmosphere. Arsenic silver diethyldithiocarbamate method Arsine gas is liberated from arsenic compounds upon the addition of zinc in an acid medium (Stratton and Whitehead, 1962). The arsine gas is passed through a lead acetate scrubber and into an absorbing tube containing silver diethyldithiocarbamate solution. The arsine and the silver diethyldithiocarbamate solution react forming a red color that can be measured spectrophotometrically. Apparatus. Arsine generator, scrubber, and absorber. Spectrophotometer set at the following operating conditions: wavelength - 535 mp; cells - 10 mm; phototube -blue sensitive; and slit width - 0.02 mm. Reagents. Standard arsenious oxide solution: dissolve 1.320 g of As203 in
COLORIMETRIC METHODS
109
10 ml of 1OM NaOH and dilute to 1 liter with distilled water. 1 ml of this solution contains 1.00 mg of A% 0 3 .Dilute this stock standard solution as required. Hydrochloric acid, concentrated, analytical-grade. Lead acetate solution: dissolve 1 0 g of Pb(C2H302)2 * 3 H 2 0 in distilled water and dilute to 100 ml. Potassium iodide solution: dissolve 1 5 g of KI in distilled water and dilute to 100 ml. Store in an amber colored bottle. Silver diethyldithiocarbamate solution: dissolve 1 g of AgS[SN(C2H5 )2 ] in 200 ml of pyridine. Store in an amber colored bottle. Stannous chloride solution: dissolve 40 g of arsenic-free SnC12*2H20 in 1:3 HC1 and dilute to 100 ml with the same acid. Zinc, 20 mesh, arsenic-free.
Procedure. Place a 25-ml sample, or suitable aliquot, containing less than 20 pg of arsenic in a Gutzeit generator. Add to the flask successively, 5 ml of concentrated HCl, 2 ml of KI solution, and eight drops of SnC12 solution. Thoroughly mix after each addition. Allow 15 minutes for reduction of the arsenic to the tervalent state. Insert a plug of glass wool that has been impregnated with the lead acetate solution into the scrubber. Assemble the generating apparatus and add 4 mi of the silver diethyldithiocarbamate solution to the absorber. Glass beads should be added to the absorber until the liquid just covers them. Add 3 g of zinc to the generator and reconnect immediately. Allow 30 minutes for complete evolution of the arsine. Warm the generating flask gently to assure complete evolution of the arsine and then pour the solution from the absorber directly into the spectrophotometer cells. Make the determinations within 30 minutes as the color developed is not permanent.
CuZcuZutions. The quantity of arsenic in the sample is determined from a plot of absorbances of the standards: pg As (from curve) = mg/l As ml sample
Fluoride Because of interferences from large amounts of chloride present in petroleum-associated waters, a standard addition method was developed which is accurate in the presence of large amounts of chloride and sulfate and is more rapid than methods requiring distillation (Collins et al., 1961). Up to 0.01 mg of phosphate in the aliquots taken for analysis can be tolerated. Larger amounts of phosphate than this decolorize the zirconium
110
ANALYSIS OF OILFIELD WATERS
cyanine R complex completely, and distillation is necessary to remove the phosphate.
Reagents. Eriochrome Cyanine R stock solution: dissolve 1.80 g of Eriochrome Cyanine R in 200 ml of distilled water. Zirconyl nitrate stock solution: dissolve 0.40 g of zirconyl nitrate dihydrate in 100 ml of concentrated hydrochloric acid and dilute to 200 ml. Fluoride indicator solution: add 20.0 ml of the Eriochrome Cyanine R solution to about 500 ml of water, stir and add 10.0 ml of the zirconyl nitrate solution, 75 ml of concentrated hydrochloric acid, and 4 g of barium chloride. This mixture is stable for 4 - 6 months. Thiosemicarbazide, powdered solid. Procedure. Measure equal amounts of brine containing less than 0.03 mg of fluoride into each of three 50-ml volumetric flasks. Add lOpg of fluoride to one of the flasks and add 20 pg to another. Add a few milligrams of solid thiosemicarbazide and 25 ml of fluoride indicator solution to each 50-ml volumetric flask. If sulfate is present, it will precipitate as barium sulfate and must be centrifuged out of suspension. Arbitrarily adjust the transmission of the blank (25 ml of fluoride indicator solution made to 50-ml volume with distilled water) at 540 mp to 32% and measure the transmission of the three solutions. Calculations. Using coordinate graph paper, plot the transmission of the standard-addition samples on the y-axis and their concentrations in milligrams of fluoride per liter on the x-axis. Multiply the sample reading at 0 concentration by 2, and from this point on the y-axis, draw a line parallel to the x-axis until it intersects the line plotted. From this point of intersection, draw a line parallel to the y-axis until it intersects the x-axis. This value from the x-axis multiplied by the dilution factor equals the amount of fluoride in milligrams per liter. Fig. 3.3 illustrates this procedure. Iodide
A rapid, accurate method for the determination of iodide suitable for field work utilizes the principle whereby iodide is oxidized to iodine with nitrous acid and extracted into carbon tetrachloride. Hydrogen sulfide will interfere, but it can be removed by acidifying the sample and boiling (Collins et al., 1961).
Reagents. Bromphenol blue: dissolve 0.1 g of bromphenol blue in 100 ml of distilled water. Carbon tetrachloride. Iodide standard solution: dissolve 1.3081 g of potassium iodide in distilled water and dilute to 1,000ml. 1 ml contains 1 mg of iodide.
COLORIMETRIC METHODS
111
Potassium nitrite solution: dissolve 10 g of potassium nitrite in 100 ml of distilled water. Sulfuric acid, 9N.
Procedure. Pipet a sample containing less than 3 mg of iodide into a separatory funnel, and add three drops of the bromphenol blue solution and a few drops of 9N sulfuric acid until the indicator turns yellow. Add 10 ml of carbon tetrachloride and 1 ml of a 10% aqueous potassium nitrite solution, and vigorously mix the combined phases. Extract the carbon tetrachloride phase into a glass-stoppered cylinder. A violet color in the carbon tetrachloride indicates iodine. Repeat the extractions with 5-ml portions of carbon tetrachloride until all of the iodine is extracted. Dilute the combined extracts to 25 ml with carbon tetrachloride and measure the absorbance using a spectrophotometer at a wavelength of 517 mp. Use a calibration curve prepared with standard iodide solutions t o determine the milligrams of iodide in the sample. Calculation: mg I (from curve) x 1,000 = mg/l ml sample
r
Selenium Selenium can be reduced t o the elemental form with sulfur dioxids (Collins et al., 1964), hydrazine, hydroxylamine hydrochloride, hypophosphorous acid, ascorbic acid, and stannous chloride. From hydrochloric acid solutions exceeding 8N,selenium is precipitated free of tellurium when the reducing agent is sulfur dioxide. Both selenium and tellurium are precipitated by sulfur dioxide from 3 t o 5N hydrochloric acid solutions. Traces of nitric acid should be removed before sulfur dioxide reduction. When precipitating selenium, it is important that the temperature of the solution be kept below 30°C because the volatile selenium monochloride easily can form and be lost. A large excess of reducing agent helps to prevent loss of the monochloride. Selenium can be determined semiquantitatively by comparing the color of the red amorphous form, or it can be adjusted to the quadrivalent form, reacted with 3,3’-diaminobenzidine to form the monopiazselenol, and quantitatively determined spectrophotometrically. If sufficient selenium is present, it also can be determined gravimetrically . Selenate (VI) can be reduced to selenite (IV) by heating in concentrated hydrochloric acid. Selenite is the only form that reacts with 3,3’-diaminobenzidine; the reaction is : %
N H2 N
w
- NH2 + H, SeO,
NH2
+
N=
i
SeN
NH2 + 3Ha0 NH2
112
ANALYSIS OF OILFIELD WATERS
Selenium adsorption on glassware can introduce a significant error. Much of this adsorption can be eliminated by treating the glassware with a solution of chlorosilane.
Reagents. Hydrobromic acid, 48%. Selenium, stock solution: dry some selenium dioxide by placing it in a desiccator over phosphorous pentoxide for 24 hours. Dissolve 0.141 g of the dry selenium dioxide in water, add 80 ml of 48% hydrobromic acid, and dilute to 1 liter with water. 1 ml of this solution contains 0.1 mg of selenium. (Note: particles of red selenium may appear in this stock solution after long standing as a result of reduction. When this happens, a new stock solution must be prepared.) Selenium solution: pipet 100 ml of the selenium stock solution into a 1-liter volumetric flask, add 80 ml of 48% hydrobromic acid, and dilute with water. 1ml of this solution contains 0.01 mg, or 10 pg of selenium. Sulfur dioxide selenium free. Hydrochloric acid, concentrated. Sulfuric acid, concentrated. 3,3'-diaminobenzidine hydrochloride: dissolve 0.25 g of 3,3'-diaminobenzidine hydrochloride in 50 ml of water. Prepare a fresh solution each day. Formic acid, 2.5M: dissolve 11.5 g of formic acid in water, and dilute to 100 ml with water. Toluene, spectro-grade. Ammonium hydroxide: dilute 10 ml of concentrated ammonium hydroxide to 100 ml with water. Barium chloride solution: dissolve 5 g of barium chloride in 100 ml of water. EDTA solution, 0.1M: dissolve 37.225 g of disodium ethylenediaminetetraacetate in water and dilute t o 1liter. Procedure. Pipet an aliquot of brine (50 ml or less) into a 100-ml volumetric flask and dilute to volume with concentrated hydrochloric acid. If desired, the detection limit can be increased by first concentrating the brine by careful evaporation after adjusting the pH t o 2 with hydrochloric acid. Mix the solution and allow it t o stand until most of the sodium chloride precipitates. Carefully withdraw 50 ml of the supernatant clear liquor into a 150-ml beaker and add 10 ml of concentrated hydrochloric acid. Heat the mixture to near boiling for 10 minutes. Place the beaker in an ice-water bath beneath an exhaust hood, let the mixture cool t o the temperature of the ice water, and then bubble sulfur dioxide gas rapidly into the solution for about 8 minutes. If a heavy turbidity develops, filter the solution through a micropore filter. Wash the precipitate with 20 ml of cold water if a 30-ml crucible is used, or with 5 ml if a 1.5-ml crucible is used. Take care that no air is
COLORIMETRIC METHODS
113
pulled through the precipitate until the entire filtration and washing process is complete. Transfer the filter and precipitate back t o the 150-ml beaker and add 5 ml of a 1:l mixture of hydrochloric acid and nitric acid. Heat the mixture to near boiling for a few minutes, taking care not t o let the mixture boil violently or to dryness. Examine the mixture carefully t o make sure that all of the selenium has dissolved. Place the beaker containing the mixture in a vacuum desiccator over anhydrous magnesium perchlorate and sodium hydroxide and let the mixture evaporate t o dryness. Dissolve the residue in 5 ml of concentrated hydrochloric acid and heat the mixture to near boiling for a few minutes. Cool the mixture, add 20 ml of water, and filter it through Whatman No.4 filter paper into a 100-ml volumetric flask. Adjust the volume to 100 ml and pipet an aliquot containing 1-100 pg of selenium (IV) from the 100-ml flask into a 100-ml beaker. Add 5 ml of 0.1M EDTA and 2 ml of 2.5M formic acid and adjust the pH to 1.5 with hydrochloric acid. Adjust the volume to about 50 ml with water, add 2 ml of 3,3'-diaminobenzidine solution, mix, and let stand for 30 minutes. Adjust the pH to 8 and transfer the solution to a 1 2 5 4 Teflon-stoppered separatory funnel containing 10.0 ml of toluene. Shake this mixture vigorously for 2 minutes and let the phases separate. Extract the toluene phase, which now contains the monopiazselenol, into a centrifuge tube. Centrifuge briefly t o clear the toluene of water droplets. If a centrifuge is not available, the organic phase can be filtered through a dry filter paper to which has been added 100 mg of anhydrous sodium sulfate. Determine the absorbance of the toluene phase at 420 mp versus a reagent blank. 1-cm cells are used; however, longer path length cells will increase sensitivity.
Calculation. Prepare a calibration curve by plotting log I,JI, which is the extinction or optical density of the solution versus the concentration, using solutions containing known amounts of selenium and treated as previously described. Estimate the amount of selenium from this curve, and calculate as follows: g' Se = mg/l Se-* ml sample
Semiquantitative determination of selenium Pipet a 20- to 50-ml aliquot of brine into a 100-ml volumetric flask and dilute t o volume with concentrated hydrochloric acid. (To increase the detection limit, the brine can first be concentrated by careful evaporation after acidifying it to pH 2 with hydrochloric acid.) Mix the solution and allow it to stand until most of the sodium chloride precipitates. Withdraw an aliquot of the supernatant clear liquor into a small beaker,
114
ANALYSIS OF OILFIELD WATERS
add additional concentrated hydrochloric acid if necessary, and bubble sulfur dioxide gas into the solution for 3 minutes. If the solution remains clear, there is less than 25 pg of selenium present. Filter the solution through a 5-ml micropore filter. Compare the color in the crucible with a series of color standards comprising 3-20 pg of selenium. These cofar standards are prepared with known amounts of selenium and will give the fobwing colorations (in pg of selenium): 3 - very pale yellow; 6 - very pale orange; 10 - pale orange; 15 - orange; and 20 - red orange. Barium Qualitative test This test can be used to detect barium and strontium in an oilfield brine. It is possible to detect barium and strontium individually by using chromate to precipitate the barium. Transfer an aliquot of brine to a test tube, add a few millimeters of 0.5% aqueous sodium rhodizonate solution, stopper the tube, and shake the mixture vigorously. Barium and/or strontium is present if a bright red, a brownish-red, or a yellow-red precipitate forms. The deeper brown indicates barium, while the lighter yellow may indicate strontium. In any event, if a precipitate forms, barium and/or strontium is present. A series of standards can be prepared to help in determining the approximate amounts present. To differentiate between barium and strontium, a few milliliters of a 10% aqueous solution of ammonium chromate can be added to a sample brine 30-60 minutes before the sodium rhodizonate solution is added. The more soluble strontium chromate will react with the rhodizonate while the less soluble barium chromate will not.
GRAVIMETRIC METHODS
Gravimetric methods involve isolating a compound and determining its weight. Their use can involve considerable time because preliminary separations often are necessary to remove interfering elements; e.g., to determine barium as the sulfate, all strontium should be removed before the final precipitation of the barium sulfate. One constituent present in most oilfield waters that has resisted development of a good instrumental method of analysis is sulfate, and perhaps the most accurate method to determine sulfate in oilfield waters is still the gravimetric method. Sulfate Sulfate is precipitated as barium sulfate from an acid solution. The precipitate is baked, cooled, and weighed.
GRAVIMETRIC METHODS
115
Reagents. Hydrochloric acid, concentrated. Barium chloride solution, 10%aqueous. Procedure. Use an aliquot that will produce no more than 100 mg of precipitated barium sulfate. Dilute the aliquot t o 250 ml with distilled water and add 1 ml of hydrochloric acid. If the sample volume itself is larger than 250 ml, add 1 ml of hydrochloric acid per 250 ml of volume. Heat t o boiling and add an excess of hot, 10% barium chloride solution while stirring. Cover the solution and allow it t o stand for about 4 hours at a temperature of about 85OC. Filter through a very retentive filter paper such as Munktells No.OOH or Whatman No.42, and wash with hot water until the filtrate is chloride free. Place the filter plus precipitate in a tared crucible, char slowly without igniting, and bake at 800°C for 1 hour. Place the crucible in a desiccator t o cool and then weigh. Calculation:
Barium Interest in the knowledge of the barium concentration in most petroleum-associated waters has spurred the development of several types of methods to determine barium. Perhaps the most rapid but least accurate method is the turbidimetric method, which measures the turbidity of the sample caused by precipitated barium sulfate after addition of excess sulfate. The gravimetric method also measures precipitated barium sulfate or more preferably barium chromate. It is a more time-consuming method than the turbidimetric method, but will yield more accurate results. The double precipitation as chromate reduces the interference from calcium and strontium.
Reagents. Ammonium chromate solution: dissolve 10 g of ammonium chromate in water, dilute t o 100 ml, and filter. Ammonium chromate wash solution: dissolve 5 g of ammonium chromate in water and dilute to 1 liter. Adjust the pH of this solution to 4.6 with ammonium acetate or acetic and filter. Ammonium acetate solution: dissolve 30 g of ammonium acetate in water, dilute to 100 ml, and filter. Nitric acid, 4N: cautiously add 30 ml of concentrated nitric acid t o 90 ml of water. Ammonium hydroxide, concentrated. Hydrogen peroxide, 30%solution. Procedure. Because iron will interfere if present, it should be removed by
ANALYSIS OF OILFIELD WATERS
116
adding a few drops of hydrogen peroxide to the sample before heating to near boiling, adding ammonium hydroxide dropwise, and stirring until the odor of ammonia is faint but distinct. Heat to boiling to remove excess peroxide and flocculate any precipitate, and then filter out the iron hydroxide. To an aliquot of iron-free filtered water (the water should be filtered even if iron is not specifically removed) containing less than 500 mg of barium and strontium, add acetic acid until the pH is 4.6. Then add 1 0 ml of ammonium chromate and 1 ml of ammonium acetate. Readjust the pH to 4.6. The final volume should be about 200 ml. Boil the mixture for 5 minutes, stirring occasionally. Remove the mixture from the heat, cool it fairly rapidly to room temperature, and allow it to stand at room temperature for 1hour. Filter the solution through a fine porosity filter. Wash the precipitate from the beaker into the filter with the ammonium chromate wash solution. Since a second precipitation is made, it is not necessary to police the beaker. Wash the precipitate on the filter with 50 ml or more of ammonium chromate wash solution or until calcium and strontium are absent. Dissolve the precipitate in 3 or 4 ml of 4N nitric acid. Transfer the dissolved precipitate back to the beaker and repeat the precipitation. The same filter can be used, but make sure that it is acid-free. Dry the second precipitate for 1 hour at llO°C or until it reaches a constant weight.
Calcuhtions. Weigh the barium chromate and calculate barium as follows: mgBaCr04 x 542 ml sample
*
= mg/l Ba+
Precision and accuracy. The precision and accuracy of this method with optimum conditions are 1% and 2% respectively, of the amount of barium in the sample. OTHER METHODS
The approximate concentration of sodium in an oilfield water can be calculated by using a knowledge of the amounts of other major cations and anions in the sample. Likewise, the dissolved solids concentration in an oilfield water can be calculated. Sodium The practice of determining sodium by calculation does not give an accurate sodium value. For example, this value is calculated after determining all of the major common anions plus two or more cations, usually calcium and
OTHER METHODS
117
magnesium. The excess of equivalents per million of anions over cations is assumed to be sodium plus potassium, and this practice includes all the errors of the analysis plus the undetermined ions in the combined sodium plus potassium value. The ions are converted t o milliequivalents per liter (me/l) by dividing each ion concentration (mg/l) by its milliequivalent weight (mglme) to give the milliequivalents per liter for each ion determined. After adding the milliequivalents per liter for both the anions and the cations, the difference is multiplied by the milliequivalent weight of sodium to give the calculated milligrams per liter of sodium. Procedure. The calculation method is demonstrated as follows:
Anions:
chloride (50,000 mg/1)/(35.5 mg/me) = 1,410 me/l sulfate (1,290 mg/1)/(48.0 mg/me) = 27 me/l bicarbonate (204 mg/1)/(61.0 mg/me) = 3 me/l Total anions 1,440 me/l Cations: calcium (5,900 mg/1)/(20.0 mg/me) = magnesium (2,000 mg/1)/(12.1 mg/me) = Total cations
295 me/l 164 me/l 459 me/l
Determination of sodium: (1,440 - 459 me/l) x 23.0 mg/me = 22,600 mg/l Dissolved solids The dissolved solids determination can be used t o estimate the accuracy of the resistivity determination. The specific gravity determination, and the evaporation method can be used t o double check the calculated total dissolved solids. Theoretically, if all the dissolved solids are accurately determined, their sum will equal the weight of the residue left after evaporation of the water. The dissolved solids include all the solid material in solution which is ionized, or which is not ionized but does not include suspended material, colloids, or gases. The residue method involves evaporating a filtered sample t o dryness followed by drying the residue in an oven at 180°C for 1 hour. The cooled residue is weighed and the total dissolved solids are calculated; e.g., if 100 ml of brine is evaporated and the residue weighs 3.0 g, then the dissolved solids equal 30,000 mg/l. The evaporation method is subject to errors when hygroscopic material such as calcium chloride is in the water, as is usually the case in oilfield waters. The calculation method simply involves adding the sum of all the analyzed constituents as follows:
ANALYSIS OF OILFIELD WATERS
118
Constituent Na+
K+ Li+ Ca+ Mg+ c1Br- HC03 S04-2
Total dissolved solids
Concentration(mg/l) 13,500 400 10 2,000 1,200 23,500 500 1,200 1,200 43,500
Spent acid Hydrochloric acid is the oldest and most common solution used in oil-well acidizing (Halliburton Company, 1970)..Many additives and other acids may be used in conjunction with HC1, for example HF and HAc. Normally, 15% HCl is used; however, other strengths are quite common. These acid solutions are pumped into carbonate formations t o dissolve and remove a part of the formation. After reacting with carbonate rocks or being “spent” on the formation, the solutions are returned t o the surface by various means. Often, they are mixed with formation water, and an operator may want t o know when the spent acid has been recovered, or if formation water or a mixture of solution and water is being produced. When 15% HC1 is completely spent on CaC03 or M g C 0 3 , the resulting solution will contain 90,000 mg/l Ca or Ca equivalent. The normal formation water contains only about 10,000 mg/l Ca or Ca equivalent. The procedure is based on these differences. Reagents and equipment. The necessary reagents and equipment include: Calcium carbonate, 10-mesh. NH4OH, reagent. Whatman No.31 filter paper. Plastic funnel. 150-ml beakers. Graduated cylinder, 25 ml. 1-ml syringe or pipette, preferably plastic. 0.5% Eriochrome Black T indicator. 2N NaOH solution. CDTA solution (disodium dihydrogen 1,2-cyclohexanediamine-tetraacetate): dissolve 100.0 g CDTA in 900 ml water and dilute to 1 liter. 1 ml equals 9.0 mg Ca.
OTHER METHODS
119
Buffer solution: 67.5 g NH4C1, 570 ml NH40H made to 1 liter with distilled water. Procedure: (1) Determine the pH of the returned water. If pH is below 4, the presence of HC1 is indicated. (2) Pour 10-15 ml of the sample into a beaker containing 10 g of 10-mesh CaC03. Bring t o a boil, remove from the hotplate, allow to settle for about 5 minutes, and filter. (3) Pipet 1.0 ml of the filtrate into a beaker containing 50 ml of HZO. Heat to boiling, add 1 ml NH40H while stirring, remove from heat, let settle for a few minutes, and filter through Whatman No.31 paper. Wash the beaker and filter twice, using 25.0 ml H20 for each wash. (4)Add 0.5 g Eriochrome Black T indicator t o the filtrate and 10 ml of the buffer solution (pH should be 10). Titrate with standard CDTA solution (1ml = 9.0 mg Ca) t o a permanent clear blue endpoint. Record the milliliters of CDTA used. Refer t o a curve to determine the percent spent acid in the sample. (5) To determine a blank, take 1.0 ml of the formation water through the procedure, starting at step 3 and determine 0% spent acid, or the blank correction. Curve construction It is desirable t o construct a curve tpercent spent acid versus milliliter CDTA) for the determination of spent acid. On rectangular graph paper, plot
Example: I m l formation water = 1.3 ml CDTA
100
-
I ml return water = 5.8 ml CDTA p H r e f w n woter = 6.0 I ml CDTA = 9.0 mg Ca
I
Fig. 3.12. Graph for use in calculating the amount of spent mineral acid in a water sample.
120
ANALYSIS OF OILFIELD WATERS
the blank titration (formation water) as 0% spent acid. Draw a straight line from this point through the intersection of the 100% spent acid and the 10.0-ml CDTA lines as illustrated in Fig. 3.12. This procedure corrects for any Ca+’ or Mg+2 present in the dilution water. In cases where it is impossible t o obtain formation water for the 0%spent acid, a reasonable approximation can be made by titrating 100 ml of the water used for washing and dilution. To this volume of CDTA, add 1.3 ml. This value can then be used for the 0%spent acid point on the plot.
Free HCl When free HC1 is indicated (pH below 4), and it is t o be determined, an additional sample is required. Withdraw 1.0 ml of clear sample. Start with step 3 and follow the procedure. The free HC1 is determined by the difference of the two titrations: % free HC1= (A -B) x 1.5
where A = ml standard CDTA to titrate CaCO, treated sample, and B = ml standard CDTA t o titrate sample.
Acetic acid solutions Generally, acetic acid solutions are mixtures of acetic acid and HC1.
Example: 10% HCI t 5% acetic acid I m l = 7.0 ml CDTA I m l formation waler = 1.3 ml CDTA I ml return water = 4 . 2 ml CDTA I ml CDTA = 9.0 mg Ca Return canlains 51% rpent acid
ml, CDTA
Fig. 3.13. Graph for use in calculating the amount of spent mineral and organic acid in a water sample.
REFERENCES
121
Various proportions of each are common. The determination is complicated by the fact that acetic acid will not completely spend itself on calcium and magnesium carbonates. At a pH of 5-6, considerable free acetic acid is still present in the solution and this necessitates a modification of the procedure. In this case, it is necessary to have a representative sample, or t o prepare a sample of the original acid mixture used on the acid job. Take 10-15 ml of the treating acid and 10-15 ml of the returned water through the same procedure as outlined for HC1. Again a plot is constructed, percent spent acid versus milliliters CDTA. Plot the milliliters CDTA used by the formation water as 0%spent acid and the milliliters CDTA used by the injected acid sample as 100%spent acid as illustrated in Fig.3.13. Connect these points by a straight line. From the curve, determine the percent spent acid in the sample of returned water. Other acid mixtures are sometimes used in oil wells. The handling of these are usually too complicated for a rapid field determination. References American Petroleum Institute, 1968. API Recommended Practice for Analysis o f Oilfield Waters. Subcommittee on Analysis of Oilfield Waters, API RP 45, 2nd ed., 49 pp. Angino, E.E. and Billings, G.K., 1967. Atomic Absorption Spectrometry in Geology. American Elsevier, New York, N.Y., 144 pp. Ballinger, D.G., Booth, R.L., Midgett, M.R., Kroner, R.C., Kopp, J.F., Lichtenberg, J.J., Winter, J.A., Dressman, R.C., Eichelberger, J.W. and Longbottom, J.E., 1972. Handbook f o r Analytical Quality Control in Water and Wastewater Laboratories. National Environmental Research Center, Cincinnati, Ohio, 107 pp. Bogomolov, G.V., Kudelskii, A.V. and Kozlov, M.F., 1970. Ammonium as one of the indications of oil-gas content. Dokl. Akad. Nauk S.S.S.R., 195:938-940 (in Russian). Brooks, R.R., Presley, B.J. and Kaplan, I.R., 1967. APDC-MIBK extraction system for the determination of trace elements in saline waters by atomic absorption spectrophotometry. Talanta, 14:809-816. Burriel-Marti, F. and Ramirez-Munoz, J., 1957. Flame Photometry. American Elsevier, New York, N.Y., 531 pp. Collins, A.G., 1962. Methods of analyzing oilfield waters: flame-spectrophotometric determination of potassium, lithium, strontium, barium, and manganese. US. Bur. Min. Rep. Invest., No. 6047,18 pp. Collins, A.G., 1964. Eh and pH of oilfield waters. Prod. Monthly, 29:ll-12. Collins, A.G., 1965. Methods of analyzing oilfield waters: cesium and rubidium. U.S.Bur. Min. Rep. Invest., No. 6641,18 pp. Collins, A.G., 1967. Emission spectrometric determination of barium, boron, iron, manganese, and strontium in oilfield waters. Appl. Spectrosc., 21 :16-19. Collins, A.G., 1969. Solubilities of some silicate minerals in saline waters. U.S. O f f .Saline Water Res. Dev. Progr. Rep., No. 472, 27 pp. Collins, A.G., Castagno, J.L. and Marcy, V.M., 1969. Potentiometric determination of ammonium in oilfield brines. Environ. Sci. Technol., 3:274-275. Collins, A.G., Waters, C.J. and Pearson, C.A., 1964. Methods of analyzing oilfield waters: selenium and tellurium. U.S.Bur. Min. Rep. Invest., No.6474, 19 pp. Collins, A.G., Pearson, C., Attaway, D.H. and Ebrey, T.G., 1962. Methods of analyzing oilfield waters metallics: copper, nickel, lead, iron, manganese, zinc, and cadmium. US.Bur. Min. Rep. Invest., No. 6087,24 pp.
122
ANALYSIS OF OILFIELD WATERS
Collins, A.G., Pearson, C., Attaway, D.H. and Watkins, J.W., 1961. Methods of analyzing oilfield waters: iodide, bromide, alkalinity, acidity, borate boron, total boron, organic boron, potassium, calcium, magnesium, iron, fluorides, and arsenic. US. Bur. Min. Rep. Invest., No.5819, 39 pp. Craig, H., 1961a. Isotopic variations in meteoric waters. Science, 133:1702-1703. Craig, H., 1961b. Standards for reporting concentrations of deuterium and oxygen-18 in natural waters. Science, 133:1833-1834. Dean, J.A., 1960. Flame Photometry. McGraw-Hill, New York, N.Y., 354 pp. Diehl, H. and Smith, G.F., 1958. The Copper Reagents: Cuproine, Neocuproine, Bathocuproine. G. Frederick Smith Chemical, Columbus, Ohio, 48 pp. Dunlap, H.F. and Hawthorne, R.R., 1951. The calculation of water resistivities from chemical analyses. J. Pet. Technol., 7:17. Epstein, S . and Mayeda, T., 1953. Variation of "0 content of waters from natural sources. Geochim. Cosmochim. Acta. 4:213-224. Fabricand, B.P., Imbimbo, E.S., Brey, M.E. and Watson, J.A., 1966. Atomic absorption analysis of lithium, magnesium, potassium, rubidium, and strontium in ocean waters. J. Geophys. R e s , 71:3917-3921. Fisher, F.L., Ibert, E.R. and Beckman, H.F., 1958, Inorganic nitrate, nitrite, or nitratenitrite. Anal. Chem., 30:1972-1974. Friedman, I., 1953. Deuterium content of natural waters and other substances. Geochim. Cosmochim Acta, 4:213-224. Friedman, I. and Woodcock, A.H. 1957. Determination of deuterium/hydrogen ratios in Hawaiian waters. Tellus, 9:553-556. Furman, N.H., 1962. Standard Methods o f Chemical Analysis. D. Van Nostrand, Princeton, N.J., 6th ed., 332 pp. Garrels, R.M. and Christ, C.L. 1965. Solutions, Minerals, and Equilibria. Harper and Row, New York, N.Y., 450 pp. Gautier, A., 1903. The arsenic content of some biologic materials. Compt. Rend., Acad. Sci. Fr., 137:232. George, W.O. and Hastings, W.W., 1951. Nitrate in the groundwaters of Texas. A m . Geophys. Union Trans., 32:450-456. Gorgy, S., Rakestraw, N.W. and Cox, D.L., 1948. Arsenic in the sea. J. Mar. Res., 7 :2 2-41; Halliburton Company, 1970. Chemical Research and Development. Halliburton Services, Procedures 110.14 and 110.15, unpublished. Herrmann, R. and Alkemade, C.T.J., 1963. Chemical Analysis by Flame Photometry. Interscience, New York, N.Y., 644 pp. Hodgman, C.D., Weast, R.C., Shankland, R.S. and Selby, S.M., 1962. Handbook of Chemistry and Physics. Chemical Rubber, Cleveland, Ohio, 44th ed., 3604 pp. Jones, P.J., 1944. Properties of water found in reservoirs, 111. Oil Gas J., 43( 28):205-209. Krauskopf, K.B., 1956. Dissolution and precipitation of silica a t low temperatures. Geochirn Cosmochirn Acta, 1O:l-26. Latimer, W.M., 1952. Oxidation Potentials. Prentice-Hall, New York, N.Y., 2nd ed., 392 PP. Marsh, G.A., 1951. Portable dissolved oxygen meter for use with oilfield brines. Anal. Chern, 23:1427. Mellon, M.G., 1950. Analytical Absorption Spectroscopy. John Wiley and Sons, New York, N.Y., 618 pp. Mellon, M.G.,. 1956. Quantitative Analyses. Thomas F. Crowell, New York, N.Y., 694 pp. Platte, J.A. and Marcy, V.M., 1959. Photometric determination of zinc with zincon: application to water containing heavy metals. Anal. Chem., 31 :1226-1228. Potter, E.C., 1956. Electrochemistry. MacMillan, New York, N.Y., 418 pp.
REFERENCES
123
Pourbaix, M.J., 1949. Thermodynamics of Dilute Aqueous Solutions. Edward Arnold, London, 136 pp. Rainwater, F.H. and Thatcher, L.L., 1960. Methods for collection and analysis of water samples. U S . Geol. Surv. Water Supply Paper, No.1454, p.70. Rakestraw, N.W. and Lutz, F.B. 1933. Determination of arsenic in sea water. Biol. Bull., 65:397. Ramirez-Munoz, J., 1968. Atomic Absorption Spectroscopy and Analysis by Atomic Absorption Flame Photometry. American Elsevier, New York, N.Y., 315 pp. Robinson, J.W., 1966. Atomic Absorption Spectroscopy. Marcel Dekker, New York, N.Y., 204 pp. Rosin, J., 1955. Reagent Chemicals and Standards. D. Van Nostrand, New York, N.Y., 561 pp. Sandell, E.B., 1959. Colorimetric Determination of Traces o f Metals. Interscience, New York, N.Y., 1032 pp. Schrink, D.R., 1965. Determination of silica in sea water using solvent extraction. Anal. Chem., 37:764-765. Scribner, B.F. and M. Margoshes, 1961. Excitation of solutions in a gas-stabilized arc source. Natl. Bur. Standards Rep., No.7342, 8 pp. Smales, A.A. and Pate, B.D., 1952. The determination of sub-microgram quantities of arsenic by radioactivation, 11. The determination of arsenic in sea water. Analyst, 7 7 :188-195. Stratton, G. and Whitehead, H.C., 1962. Colorimetric determination of arsenic in water with silver diethyldithiocarbamate. J. A m . Water Works Assoc., 54:861-863. Watkins, J.W., 1954. Analytical methods of testing waters to be injected into subsurface oil-productive strata. U.S. Bur. Min. Rep. Invest., No.5031, 29 pp. Welcher, F.J., 1957. The Analytical Uses o f Ethylenediaminetetraacetic Acid. D. Van Nostrand, Princeton, N.J., 356 pp. White, D.E., Brannock, W.W. and Murata, K.J., 1956. Silica in hot-spring waters. Geochim. Cosmochim. Acta, 10:27-59. Willard, H.H., Merritt, Jr., L.L. and Dean, J.A., 1965. Instrumental Methods of Analysis. D. Van Nostrand Co., Princeton, N.J., 4th ed., 250 pp. Wyllie, M.R.J., 1963. The Fundamentals of Well Log Interpretation. Academic Press, New York, N.Y., 3rd ed., 238 pp. Zobell, C.E., 1946. Studies on redox potential of marine sediments. Bull. A m . Assoc. Pet. Geol., 30:477-513.
This Page Intentionally Left Blank
Chapter 4.
INTERPRETATION OF CHEMICAL ANALYSES OF OILFIELD WATERS
Water analyses may be used to identify the source of a water. In the oilfield one of the prime uses of these analyses is to determine the source of extraneous water in an oil well, so that casing can be set and cemented to prevent such water from flooding the oil or gas horizons. In some wells a leak may develop in the casing or cement, and water analyses are used to identify the water-bearing horizon so that the leaking area can be repaired. With the present emphasis on water pollution prevention, it is very important to locate the source of a polluting brine, so that remedial action can be taken. Comparisons of water-analysis data are tedious and time-consuming; therefore, graphical methods are commonly used for positive, rapid identification. A number of systems have been developed, all of which have some merit. Calculating probable compounds The hypothetical combinations of dissolved constituents found in waters are commonly calculated by combining the positive and negative radicals in the following order: calcium magnesium sodium potassium
bicarbonate sulfate chloride nitrate
Calcium is combined with bicarbonate, and if more calcium is available than that consumed by bicarbonate, it is combined with sulfate, chloride, and nitrate until exhausted. Conversely, any excess bicarbonate is combined with magnesium, sodium, and potassium until consumed. Other radicals can and should be added for most petroleum reservoir waters. These include lithium, strontium, barium, iron, borate, phosphate, bromide, and iodide. They can be grouped in the appropriate column and then in the calculations each positive and negative radical is totally combined, the next following radical is combined until both the cations and anions are exhausted. If the analysis is correct, the cations and anions will be present in approximately equivalent amounts. To calculate the hypothetical combinations, the reacting values of the positive and negative radicals or ions are calculated as follows: reacting
INTERPRETATION OF CHEMICAL ANALYSES
126 TABLE 4.1 Reaction coefficients Cation
Anion
Calcium Magnesium Iron Sodium
0.0499 0.0823 0.0358 0.0435
bicarbonate sulfate chloride
Cation (mg/l)
RV
Anion (mg/l)
Ca 4,000x Mg 3,000x Fe 100 x Na 9,400 x
199.6 246.8 3.6 408.9
HC03
0.0164 0.0208 0.0282
TABLE 4.11 Reacting values (RV)
0.0499 = 0.0823 = 0.0358 = 0.0435 =
so4 C1
RV
500 x 0.0164 = 8.2 200 x 0.0208 = 4.2 30,000 x 0.0282 = 846.3
858.9
858.7
TABLE 4.111 Reacting value distribution Ca Ca Ca Mg Fe Na
as as as as as as
calcium bicarbonate calcium sulfate calcium chloride magnesium chloride iron chloride sodium chloride
8.2 4.2 187.2 246.8 3.6 408.9 858.9
values (RV) or equivalents per million (epm) = mg/l of ion x valence of ion/ molecular weight of ion. The term valence of ion/molecular weight of ion is called “reaction coefficient” and the positive and negative ions have values as shown in Table 4.1. Table 4.11 indicates how the results of a water analysis are converted t o reacting values. The reacting values are a measure of the cations and anions dissolved in the water. The 4,000 mg/l of calcium with a reacting value of 199.6 can combine with all the bicarbonate, all the sulfate, and 187.2 epm of the chloride. Magnesium will combine with 246.8,iron with 3.6, and sodium with 408.9 epm of chloride. Thus the reacting values can be considered to be distributed as shown in Table 4.111.
DETERMINING A SOUGHT COMPOUND
127
TABLE 4.IV Combination factors Reaction values given
Compound sought
Combination factor
Ca o r C 0 3 Ca or SO4 Ca or C1 Mg or C 0 3 Mg or SO4 Mg or C1 Fe or C03 Fe o r S 0 4 Fe orC1 Na or C03 Na or SO4 Na or C1
CaC03 CaS04 CaClz MgCO3 MgS04 MgClz FeC03 FeS04 Fa12 Na~C03 Naz SO4 NaCl
50.1 68.1 55.5 42.2 60.1 47.6 57.8 76.0 63.4 53.1 71.0 58.4
TABLE 4.V Hypothetical combinations Ca(HC03 ) to CaC03 CaS04 CaC12 MgCh FeClz NaCl
8.2 x 50.1 4 . 2 68.1 ~ 187.2 x 55.5 246.8 x 47.6 3.6 x 63.4 858.9 x 58.4
= 411 CaC03* = 286CaS04 = 10,390 CaC12 = 11,748 MgClz = =
228 FeCIz 50,160 NaCl
*In mg/l.
Determining a sought compound It is necessary to multiply the reacting value by a combination factor to determine a hypothetical compound. This factor is necessary to convert the reported radical into the desired compound. For example, the factor for converting Ca to CaCO, is 2.50 and the reaction coefficient for Ca is 0.0499. Therefore, the combination factor to convert the reacting value for Ca to CaCO, is 2.50 + 0.0499 = 50.1. Table 4.IV illustrates some combination factors. The combination factors given in Table 4.IV can be used to calculate the hypothetical combinations shown in Table 4.V, using the analysis shown in Table 4.111.
INTERPRETATION OF CHEMICAL ANALYSES
128
Graphic plots Graphic plots of the reacting values can be made to illustrate the relative amount of each radical present. The graphical presentation is an aid t o rapid identification of a water, and classification as t o its type, and there are several methods that have been developed.
Tickell diagram The Tickell (1921) diagram was developed using a 6-axis system or star diagram. Percentage reaction values of the ions are plotted on the axes. The percentage values are calculated by summing the epm’s of all the ions, dividing the epm of a given ion by the sum of the total epm’s, and multiplying by 100. Na
Ca+Mg
Na
ci (a)
CI RV=49.92%
(b)
Ca+Mg
2-
\
So4
h 9 2 ma / I i tar
Fig. 4.1. Tickell (a) and modified Tickell (b) diagram for Gulf Coast water, sample No.1.
Na
(a)
Ca + Mg
Ca+Mg
CI ~ v = 4 9 . 2 9 %
s
(b)
1i07 ma/ lltar
Fig. 4.2. Tickell (a) and modified Tickell (b) diagram for Anadarko Basin water, sample No. 2.
GRAPHIC PLOTS
129
H
so4 C I RV=49.92 %
(a)
so4 5,708 m e / l i t e r
(b)
Fig. 4.3. Tickell (a) and modified Tickell (b) diagram for Williston Basin water, sample N0.3.
No
Ca+Ma
Co+Mg
$
c i\ (b)
so4 1.769 me / liter
Fig. 4.4. Tickell (a) and modified Tickell (b) diagram for Gulf Coast and Anadarko Basin waters, mixed 1:l.
Na
Co+Mg
CI
so4
Na
Ca+Mg
‘7 $.
(b) C I
2870 me / i i t r r
Fig. 4.6. Tickell (a) and modified Tickell (b) diagram for Gulf Coast, Williston, and Anadarko Basin waters, mixed 1 :1 :1.
130
INTERPRETATION OF CHEMICAL ANALYSES
Fig. 4.1.illustrates the Tickell diagram using reaction values in percentage in the diagram on the left, and total reaction values in the diagram on the right. The plots of total reaction values, rather than of percentage reaction values, are often more useful in water identification because the percentage values do not take into account the actual 'ion concentrations. Water differing only in concentrations of dissolved constituents cannot be distinguished. To illustrate differences in patterns for different waters, Fig. 4.1-5 were prepared using the Tickell method. Fig. 4.1 represents a water from the Gulf Coast Basin, taken from the Wilcox formation of Eocene age. Fig. 4.2 is of a sample from the Mer?.mec formation of Mississipian age in the Anadarko Basin. Fig. 4.3 is of sample from a Devonian age formation in the Williston Basin. Fig. 4.4 represents a 1:l mixture of waters of the Gulf Coast and Anadarko Basins, and Fig. 4.5 is a 1:1:1 mixture of all three waters.
REISTLE SYSTEM
Fig. 4.6. Water-analysis interpretation, Reistle system the samples of Fig. 4.1-3.
- sample numbers correspond to
GRAPHIC PLOTS
131
Reistle diagram Reistle (1927) devised a method of plotting water analyses using the ion concentrations as shown in Fig. 4.6. The data are plotted on a vertical diagram, with the cations plotted above the central zero line and the anions below. This type of diagram often is useful in making regional correlations or studying lateral variations in the water of a single formation, because several analyses can be plotted on a large sheet of paper.
St iff diagra m Stiff (1951) plotted the reaction values of the ions on a system of rectangular coordinates as illustrated in Fig. 4.7. The cations are plotted to the left and the anions to the right of a vertical zero line. The end points then are connected by straight lines to form a closed diagram, sometimes called a “butterfly” diagram. To emphasize a constituent that may be a key t o interpretation, the scales may be varied by changing the denominator of the
Fig. 4.7. Water-analysis interpretation, Stiff method - sample numbers correspond to the samples of Fig. 4.1-3.
132
INTERPRETATION OF CHEMICAL ANALYSES
ion fraction usually in multiples of 10. However, when looking at a group of waters all must be plotted on the same scale. Many investigators believe that this is the best method of comparing oilfield water analyses. The method is simple, and nontechnical personnel can be easily trained t o construct the diagrams.
Other methods Several other water identification diagrams have been developed, primarily for use with fresh waters, and they will not be discussed here. The Piper (1953)diagram and the Stiff (1951)diagram were adapted to automatic data processing by Morgan et al. (1966),and Morgan and McNellis (1969).The Piper (1953)diagram uses a multiple trilinear plot t o depict the water analysis, and this quaternary diagram shows the chemical composition of the water in terms of cations and anions. Angino and Morgan (1966)applied the automated Stiff and Piper diagrams to some oilfield brines and obtained good results.
References Angino, E.E. and Morgan, C.O., 1966. Application of pattern analysis t o the classification of oilfield brines. Kans. State Geol. Sum.,Comput. Contrib., No.7, pp.53-56. Morgan, C.O. and McNellis, J.M., 1969. Stiff diagrams of water-quality data programmed for the digital computer. Kuns. State Geol. Sum., Spec. Distrib. Publ., No.43, 27 pp. Morgan, C.O., Dingman, R.J. and McNellis, J.M., 1966. Digital computer methods for water-quality data. Ground Water, 4:35-42. Piper, A.M., 1953. A graphic procedure in the geochemical interpretation of water analyses. US.Geol. Surv. Ground Water Note, No.12, 1 4 pp. Reistle, C.E., 1927. Identification of oilfield waters by chemical analysis. U.S.Bur. Min. Tech. Paper, No.404, 25 pp. Stiff, H.A., 1951. The interpretation of chemical water analysis by means of patterns. J. Pet. Technol., 3:15-17. Tickell, F.G., 1921. A method for graphical interpretation of water analysis. Calif. State Oil Gas Superv., 6:5-11.
Chapter 5.
SIGNIFICANCE OF SOME INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES OF OILFIELD WATERS
In general, the concentrations of the constituents in various natural solids of reservoir rocks must be considered along with the amounts that are found in associated oilfield waters. Some possible chemical reactions between host rock and reservoir water may deplete or enrich the concentration of the constituents in oilfield waters. Another important factor is the solubility of a constituent. The ionic potential, determined by dividing the ionic radius by the valence, influences the solubility of elements. For example, elements with low ionic potential are more likely t o remain in true ionic solution. Elements commonly found in oilfield waters have the following ionic potentials: sodium, 0.95; calcium, 0.50; magnesium, 0.33; chlorine, 1.81; bromine, 1.95; and iodine, 2.16. Apparently the cation (magnesium) and the anion (chlorine) would be the most likely to remain in true ionic solution; however, several other variables occur during diagenesis which lead to depletion or enrichment of constituents in waters.
Lithium Lithium is the lightest alkali metal; it has a distinctly smaller radius, 0.60 8,than the other alkalies and is the smallest of all singly charged cations. It is one of the less abundant elements, and its abundance in the earth’s crust is about 6.5 x wt.% (Fleischer, 1962). Here again, it is an exception because in general, the lighter elements tend to be more abundant than the heavier elements. It is lithophilic in that it tends t o be associated with the silicate phase in rocks (Ahrens, 1965); however, because of its small size, it supposedly cannot replace the abundant alkali metals in mica. It and the other alkali metals exist in a uniform positive one state of oxidation and are inherently ionic. Their chemical behavior depends almost entirely upon electron loss, and their chemistry is simpler than that of any of the other metallic elements (Moeller, 1954). Lithium is potentially toxic to plants (Hem, 1970), yet it is regularly found in plant ashes, which indicates that it normally is present in soil waters (Goldschmidt, 1958). Coal ashes of Neurode, Silesia, contained up to 198 ppm lithium, whereas soils in northeast Scotland contain 30-5,000 ppm. The content of lithium in sediments ranges up to 6 ppm in quartzites and sandstones, up to 15 ppm in calcareous rocks, and up t o 120 ppm in clays and shales.
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
134 TABLE 5.1
Properties of the alkali metals Property
Lithium
Sodium
Potassium
Rubidium
Cesium
Atomic number Nonhydrated radius (A) Hydrated radius (A) Outer electronic configuration Atomic weight Ionization potential (V)
3
11
19
37
55
0.60
0.95
1.33
1.48
3.82
3.58
3.31
-
1s' 2s' 6.939 5.390
2s22p6 3s' 22.990 5.138
3s' 3p6 4s' 39.102
-
4s24p6 5s' 85.47
4.339
1.69
4.176
5s' 5p6 6s' 132.905 3.893
TABLE 5.11 Five relative concentration changes of some dissolved ions during evaporation of sea water and brine* Constituents
Concentrations (mg/l) Sea water
Lithium Sodium Potassium Rubidium Magnesium Calcium Strontium Boron Chloride Bromide Iodide
CaSO4
0.2 2 11,000 98,000 350 3,600 0.1 1 1,300 13,000 400 1,700 60 7 5 40 19,000 178,000 65 600 0.05 2
NaCl
MgS04
11 12 140,000 70,000 23,000 37,000 6 8 74,000 80,000 100 10 1 10 300 310 275,000 277,000 4,000 4,300 5 7
KCI
27 13,000 26,000 14 130,000 0 0 750 360,000 8,600 8
MgC12
34 12,000 1,200 10 153,000 0 0 850 425,000 10,000 8
*Approximate mg/l. Columns headed sea water, CaS04, etc., represent stages in sea water evaporation. For example, sea water contains 0.2 mg/l of lithium, after calcium sulfate has precipitated the residual brine contains about 2 mg/l of lithium, after sodium chloride has precipitated the residual brine contains about 11 mg/l of lithium, the residual brine contains about 12 mg/l of lithium after magnesium sulfate precipitates, 27 mg/l of lithium after potassium chloride precipitates, and 34 mg/l of lithium after magnesium chloride precipitates.
LITHIUM
135
a,
The hydrated radius of lithium is 3.82 as shown in Table 5.1 (Moeller, 1954). The ionic potential is 0.60, and the polarization is 1.67. The polarization is quite high and is a measure of its replacing power in an exchange system. Apparently it can replace strontium, calcium, and magnesium since their polarizations are 1.77, 2.02, and 3.08, respectively. Some surface waters of the volcanic sodium chloride type are enriched in lithium (White, 1957). Lithium from Searles Lake brine is recovered as Li2NaP04 (Brasted, 1957). The content of lithium in oilfield waters is usually less than 10 mg/l but in some Smackover formation waters from east Texas, concentrations up t o 500 mg/l are present. When a brine containing lithium goes through an evaporite sequence, lithium is one of the elements whose concentration does not decrease, as illustrated in Table 5.11, in the liquid phase as various minerals precipitate (Collins, 1970). Fig. 5.1 illustrates the enrichment of lithium as compared t o an evaporite sequence in some subsurface brines from Tertiary, Cretaceous, and Jurassic age sediments. Fig. 5.2 illustrates a similar enrichment for some brines taken from Pennsylvanian and Mississippian age sediments (Collins, 1969a). Possibly lithium was liberated and potassium was depleted by exchange reactions with clay minerals, degradation of lithium containing minerals, or simply a leaching of minerals, primarily silicates, which contain lithium. Lithium substitutes in the structure of several common minerals and forms few minerals of its own. If the minerals in which it has substituted should degrade or break down with depth, the lithium might be resolubilized, thus increasing its concentration in the aqueous phase. White et al. (1963) postulated that because the lithium concentration in magmatic waters is related to volcanic
LITHIUM, mgll
Fig. 5.1. Comparison of the lithium concentrations in some Tertiary (T),Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
136
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
u 10
20
3
LITHIUM, mg/l Fig. 5.2. Comparison of the lithium concentrations in some Mississippian (M) and Pennsylvanian (P) age formation waters from Oklahoma with an evaporating sea water.
emanations, the increase in the lithium content of deeper waters might be related to the same cause.
Sodium The most abundant member of the alkali-metal group is sodium, ranking number 6 with respect t o all the metallic elements. The radius of the sodium ion is 0.95 A, and its geochemistry is controlled to some extent by calcium because of the similarity of their ionic radii. Its abundance in the earth's crust is about 2.8 wt.% (Fleischer, 1962). Table 5.1 shows that its outer electronic configuration is 2s' 2p6 3s' , with a first ionization potential of 5.138 V, indicating that its single outer electron is less firmly held than in the lithium atom with a first ionization potential of 5.390 V. The ionization potential is a measure of the chemical reactivity - the lower the potential, the greater the reactivity. Table 5.1 (Moeller, 1954) also illustrates some qf its other properties. According t o Ahrens (1965),sodium is lithophilic, and many distinctly lithophile elements have valence electrons outside a closed shell of eight electrons. The ionic radius decreases as the charge on the cation increases. Sodium does readily participate in solid solution relationships because its radius is small, making replacement of cations with 30% larger radii difficult. The amounts of sodium in argillaceous sediments and marine shales are about 1,000ppm and 1,300ppm, respectively (Goldschmidt, 1958).
SODIUM
137
Sodium in solution tends to stay in solution; it does not readily precipitate with an anion, and it is less easily adsorbed by clay minerals than are cesium, rubidium, potassium, lithium, barium, and magnesium. The major source of sodium in sea water can be attributed t o the weathering of rocks. Some sodium probably was derived through volcanic activity. The ocean and evaporite sediments contain the bulk of the sodium. Igneous rocks contain appreciably more sodium than sedimentary rocks with the exception of evaporites. Sea water contains about 11,000 mg/l of sodium, as illustrated in Table 5.11. The concentration of sodium increases in brine as it evaporates, t o about 140,000 mg/l, when halite precipitates. Most oilfield waters contain more sodium than any other cation, and most oilfield waters are believed to be of marine origin. Fig.5.3 is a log-log plot of the chloride concentration versus sodium of some subsurface brines taken from sediments of Tertiary, Cretaceous, and Jurassic age. The straight line is a plot of chloride versus sodium concentrations for some evaporite waters, and indicates the enrichment of sodium ions until halite (NaC1) precipitates - at a chloride concentration of about 140,000 mg/l (compared t o that of normal sea water, 19,000 mg/l). The plot of the concentrations of sodium versus chloride for these subsurface brines falls very near the normal evaporite curve, indicating that the concentration mechanism may be related to an evaporite process (Collins, 1970). Fig. 5.4 is a similar plot for some subsurface brines taken from sediments of Pennsylvanian and Mississippian age (Collins, 1969a). Several of these samples are somewhat depleted in sodium which indicates that
SODIUM,
g/l
Fig. 5.3. Sodium versus chloride concentrations for some formation waters taken from Tertiary (T), Cretaceous (C), and Jurassic (J) zge sediments and compared to evaporating sea water.
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
138
500F --/ 200t-
0
/
- Nor ma1 evaoor it e associated b i n e
I-
/"
SODIUM, mg/l Fig. 5.4. Sodium versus chloride concentrations for some formation waters taken from Pennsylvanian (P) and Mississippian (M) age formation sediments and compared t o evaporating sea water.
diagenetic processes, such as ion-exchange or ultra-filtration reactions involving clays and/or carbonates, may operate to deplete the sodium concentration in waters in older sediments. Potassium The second most abundant member of the alkali-metal group is potassium; its abundance in the crust of the earth is about 2.55 wt.% (Fleischer, 1962). Like the other alkali metals, it is lithophilic, and with its large ionic radius (see Table 5.1) it participates in forming solid solutions and forms its own minerals, such as feldspar and mica. The potassium feldspars are resistant to leaching by water, which may account for the low potassium concentrations in many natural waters. Clay minerals readily adsorb potassium, and in illite it is incorporated into the crystal structure in such a manner that it cannot be removed by ion-exchange reactions (Lyon and Buckman, 1960). Potassium is less easily hydrated than sodium, and is more easily adsorbed by colloids; therefore, it is retained in sediments and soils in greater abundance than sodium. It is an essential element t o plants and animals. According to Gol&chmidt (1958),potassium in pulverized potassium feldspars is absolutely unavailable t o plants. The concentrations of potassium in carbonates, sandstones, and shales is about 2,700, 10,700,and 26,600 ppm, respectively (Mason, 1966). Potas-
139
POTASSIUM ~~
.lvv
200
-
- 100 -
1 POTASSIUM,
I I I IIll
g/ I
Fig. 5.5. Potassium versus chloride concentrations for some formation waters taken from Tertiary (T), Cretaceous (C), and Jurassic (J) age sediments and compared to evaporating sea water.
sium concentrates primarily in hydrolysates (clay minerals), such as illite and glauconite, and in evaporites. Table 5.11 illustrates how the concentration of potassium in the aqueous phase increases until sylvite (KC1) precipitates. The concentration of potassium in some subsurface brines usually is depleted with respect to an evaporite-associated sea water. Fig.5.5 illustrates the relation of potassium in some subsurface brines taken from sediments of Terti500 -
,'
-
-Nmal
evaporite curve
-
5--
m POTASSIUM, mg/l Fig. 5.6. Comparison of the potassium concentrations in some Pennsylvanian (P) and Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.
140
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
ary, Cretaceous, and Jurassic ages to an evaporite-associated sea water (Collins, 1970). Fig.5.6 illustrates the same relation for some subsurface brines taken from Pennsylvanian and Mississippian age sediments (Collins, 1969a). The depletion of potassium in subsurface brines might be caused by its uptake by clays. For example, montmorillonite-type clay minerals systematically change to illite with increasing depth of burial, due to thermal diagenesis; and, as a result of this transformation, they lose interlayer (bound) water (Burst, 1969). This change appears t o begin at a temperature above 90°C. (This freed interlayer water can be readily expelled, and its movement probably is important in the first migration stage of hydrocarbons.) Laboratory experiments at elevated temperatures and pressures indicate that montmorillonite loses its interlayer water and transforms into illite in the presence of potassium-enriched water (Khitarov and Pugin, 1966). The structural variations of the expandable minerals in clays apparently are also influenced by the potassium content of the associated waters.
Rubidium Rubidium, like the other alkali metals is lithophilic, and its abundance in the earth’s crust is about 3.0 x wt.%, which is greater than that of lithium (Fleischer, 1962). It tends to be removed from solution more readily than lithium, primarily because of its ability to replace potassium in mineral structures. Table 5.11 indicates that it precipitates from an evaporite along with sylvite to a greater extent than lithium, and it has a high chemical reactivity. The radius of its ion, 1.48 is only about 10% larger than the potassium ion, so it can be accommodated into the same crystal lattices. Because of this, it forms no minerals of its own. Rubidium and cesium occur sympathetically in nature; that is, both are commonly found in amazonite, vorobyevite, and beryl (Goldschmidt, 1958). Rubidium is a member of series NH4-K-Rb-Cs, and members of this series are more similar in their chemical and physical properties than are the members of any other group, with the exception of the halogens. Rubidium concentrates in the late crystallates, particularly those of granitic derivation, and it has a greater tendency t o be adsorbed by clays than has potassium. It is removed from igneous rocks by water leaching and then adsorbed by hydrolysate sediments and soils. Shales contain about 250 ppm of rubidium; deep-sea red clays, about 400 ppm; and some glauconites, about 500 ppm (Goldschmidt, 1958).Sea water contains about 0.12 mg/l of rubidium; subsurface brines contain up t o 4 mg/l. Higher concentrations of rubidium probably can be found in brines associated with rocks containing potassium minerals, such as microcline feldspars, or lepidolite mica.
a,
141
CESIUM
Cesium Cesium is the heaviest alkali metal and also the rarest, with an abundance of about 7 x wt.% in the earth’s crust (Fleischer, 1962). It has an ionic radius of 1.69 8,which is distinctly larger than potassium, and it cannot replace potassium in minerals as easily as rubidium; probably because of this, it forms its own minerals. It is leached from igneous and metamorphic rocks by water during weathering, and is adsorbed by hydrolysate sediments and soils more readily than rubidium or potassium. Its low ionization potential indicates that it has the greatest chemical reactivity of the alkali metals. Cesium and rubidium were discovered in 1860 by Robert Bunsen by use of spectral analysis, a method which he and Kirchhoff invented. Cesium concentrates primarily like rubidium, in marine argillaceous sediments. Some shales contain about 15 ppm; deep-sea red clays, 20 ppm; and glauconite, 15 ppm of cesium (Goldschmidt, 1958). Sea water contains 5x mg/l of cesium, and some subsurface brines contain up to 1mg/l. Beryllium Beryllium is a member of the alkaline earth group in the periodic chart of the elements, but few of its properties are similar t o the more abundant members, such as magnesium, calcium, and strontium. Beryllium, like lithium, is a light element with an atomic weight of 9.012 (Table 5.111; see also Moeller, 1954), and like lithium, it is an exception t o the rule that light elements are more abundant than heavy elements. The earth’s crust contains about 6 x wt.% of beryllium (Fleischer, 1962). In sedimentary rocks, beryllium is restricted primarily to hydrolysates and especially to bauxites enriched in aluminum (Goldschmidt, 1958). Shales contain about 6 ppm, and some coal ashes contain up to 8,000 ppm, although generally only about 4 ppm. The concentration of beryllium in sea
TABLE 5.111 Properties of the alkaline earth metals Property
Beryllium Magnesium
Atomic number 4 Ionic radius (A 1 0.31 Outer electronic configuration 1s’ 2s’ Atomic weight 9.012 Ionization 9.320 potential (V)
12 0.65 2s’ 2p6 3s’ 24.31 7.644
Calcium
Strontium
Barium
20
38
56
0.99 3s2 3p6 4s2 40.08 6.111
1.13 4s’ 4p6 5s2 87.62 5.692
1.35 5s2 5p66s’ 137.34 5.210
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
142
water is about 5 x lo-' mg/l, and some subsurface brines contain 0.02-4.2 mg/l. Since beryllium is highly toxic, waters containing it should be handled with caution.
Magnesium One of the more abundant members of the alkaline earth group of metals, magnesium makes up about 2.1 wt.% (Fleischer, 1962) of the crust of the earth. Magnesium is dissolved during chemical weathering, mainly as the chloride and sulfate. Ferromagnesian minerals in igneous rocks and magnesium carbonate in carbonate rocks are generally considered t o be the principal sources of magnesium in natural waters. Carbon dioxide plays an important role in the dissolution of magnesium from silicate and carbonate minerals. Waters associated with either granite or siliceous sand may contain less than 5 mg/l of magnesium, whereas those associated with either dolomite or limestone may contain over 2,000 mg/l of magnesium. Elements commonly found in oilfield waters have the following ionic potentials: sodium, 0.95; calcium, 0.50; magnesium, 0.33; chlorine, 1.81; bromine, 1.95; and iodine, 2.16. Apparently the cation (magnesium) and the anion (chlorine) would be the most likely to remain in true ionic solution; however, several other variables occur during diagenesis which lead to depletion of magnesium in waters. Depletion of magnesium in some waters probably is a result of the replacement reaction t o form dolomite, CaMg(C0, ) 2 . Whole mountain masses are made of dolomite, which is formed by the regular substitution in the calcite
2oo
t
C J
/
?$
Normal evaporite curve
'so0
500
rpoo
2,000
5ooO
lop00
2 0 m ,
5Q(
MAGNESIUM, mg I I
Fig. 5.7. Comparison of the magnesium concentrations in some Tertiary (T), Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
143
CALCIUM
c Normal evaporite curve
500
M M P
r 20 10
1,000
I 0,000 lO0,OoO MAGNESIUM, mg/l
Fig. 5.8. Comparison of the magnesium concentrations of some Pennsylvanian (P) and Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.
crystal lattice of alternate ions of calcium and magnesium. The large differences in the ionic radii of Ca (0.99A) and Mg (0.65A) are the reason for this diadochy. Magnesium ions in aqueous solution have a large attraction for water molecules and probably are surrounded by six water molecules in octahedral arrangement. This may account for the paucity of magnesium in soils, because the small cation becomes large by hydration. Sodium has a similar reaction, but potassium, which does not, is readily adsorbed by soil colloids. Shales, sandstones, and carbonates contain 15,000,7,000,and 47,000 ppm of magnesium, respectively (Mason, 1966). Subsurface brines contain from less than 100 mg/l t o more than 30,000 mg/l; however, many subsurface brines are depleted in magnesium if compared to a sea water evaporite sequence, (Table 5.11). Sea water contains about 1,300 mg/l. Fig. 5.7 is a plot of chloride versus magnesium for some subsurface brines taken from Tertiary, Cretaceous, and Jurassic age sediments. The position of the normal evaporite curve indicates that all of these waters were depleted in magnesium with respect to this curve (Collins, 1970). Fig. 5.8 is a plot showing similar depletion of some subsurface brines taken from some sediments of Pennsylvanian and Mississippian age.
Calcium The abundance of calcium in the crust of the earth is about 3.55 wt.% (Fleischer, 1962),making it the most abundant of the alkaline earth metals,
144
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
but only in the crust; in the earth as a whole, magnesium is much more abundant. Calcium is dissolved as bicarbonate as a result of chemical weathering of calcium-bearing minerals. Waters associated with limestone, dolomite, gypsum, or gypsiferous shale usually contain an abundance of calcium, but waters associated with granite or silicious sand may contain less than 10 mg/l of calcium. Slight changes in the pH of waters containing calcium bicarbonate will cause calcium carbonate to precipitate, and calcium carbonate is one of the most common deposits found in plugged oilfield lines, equipment, and reservoirs. Precipitation of calcium carbonate in the sea is the prime mode of the origin of limestone. The solubility of calcium carbonate in sea water increases with salinity and increasing partial pressure of carbon dioxide, but it decreases with increasing pH, calcium content, and temperature. The solubility of calcium sulfate decreases with increasing temperature. Shales, sandstones, and carbonate rocks contain about 22,100, 39,100, and 302,300ppm of calcium, respectively (Mason, 1966).Sea water contains 400 mg/l and subsurface brines often contain 2,000-3,000 mg/l, with some as high as 30,000 mg/l. Fig. 5.9 is a plot of chloride versus calcium concentrations for some subsurface waters taken from Tertiary, Cretaceous, and Jurassic age sediments. The amount of calcium in these waters increases with increasing salinity, and the waters from the older sediments appear to contain more calcium. Fig. 5.10 is a similar plot for some subsurface brines taken from sediments of Pennsylvanian and Mississippian age. These samples all appear to be enriched in calcium relative t o the evaporite curve, and the concentration of calcium appears to increase with increasing salinity.
200
Normal evaporite curve
-
\ 0
-
100-
I 1 I I I111
1 500
1 , m
2
p
5poo
lop00
29ooo
CALCIUM, mg/l
Fig. 5.9. Comparison of the calcium concentrations of some Tertiary (T), Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
STRONTIUM
145
&-\-
Normal evaporite curve
M
P P
Ii 201
/
M M 1
00 CALCIUM, mg/ I Fig. 5.10. Comparison of the calcium concentrations of some Pennsylvanian (P) and MEsissippian (M) age formation waters from Oklahoma with an evaporating sea water.
Strontium Strontium, a minor element compared t o calcium and magnesium comprises about 0.03 wt.% of the earth's crust (Fleischer, 1962). Table 5.111 illustrates some of its properties, and it resembles calcium chemically. Strontium has a tendency to work upward during fractional crystallizaticn because of its relatively large radius (Goldschmidt, 1958).It occurs abundantly with potassium in volcanic rocks, alkali rocks, and pegmatites. Dissolved strontium results from water leaching of rocks, and it has been postulated that the strontium in petroleum-associated waters also may be a byproduct of the organic decay processes which originally formed petroleum. Strontium is only a microconstituent in most terrestrial animals, but several species of marine animals contain considerable quantities of strontium in their skeletons (Odum, 1951). Table 5.11 indicates that strontium may reach a concentration of 60 mg/l during sea-water evaporation, and then most of it precipitates with calcium sulfate. The amount of sulfate in the water influences the amount of strontium that remains in solution. Data by Sillhn and Martell (1964)indicate that if the sulfate activity in a water is 100 mg/l, the strontium activity can be about 28 mg/l. Davis and Collins (1971)studied the solubility of strontium sulfate in strong electrolyte solutions and found that 958 mg/l of strontium is soluble in a synthetic brine solution of ionic strength 3.05,
146
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
containing ions of sodium, calcium, magnesium, potassium, chloride, bromide, and iodide. Calcium chloride concentration apparently has a very pronounced effect upon the solubility of strontium sulfate. Celestite and strontianite occur commonly in sediments. Carbonate sediments contain up t o 1,200 ppm of strontium; dolomites, usually less than
“““I
I C -Cretaceous J -Jurassic C
J
c cc 2,000 C cC
T
20
1 0
50
I IIII
100
I
2 a
I IIL
STRONTIUM, mgll
Fig. 5.11. Comparison of the strontium concentrations of some Tertiary (T), Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water. 500
-
-
-
200 -
\
P
Ill
I
50
100 ZOO STRONTIUM, mg/l
500
1,000
Fig. 5.12. Comparison of the strontium concentrations of some Pennsylvanian (P) and Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.
BARIUM
147
170 ppm; and secondary gypsum, up t o 1,100 ppm (Goldschmidt, 1958). Sea water contains about 8 mg/l of strontium, but subsurface brines contain up to 3,500 mg/l. Fig. 5.11 is a plot of chloride versus strontium content for some subsurface brines taken from some Tertiary, Cretaceous, and Jurassic age sediments. Most of these samples were enriched in strontium compared to the evaporite-associated water, and it is possible that a mechanism similar to dolomitization could cause the enrichment. In comparison t o calcium, the strontium appears to be increasingly accumulated; for example, only five samples (from Tertiary sediments) fell within the normal evaporite curve. Fig. 5.12 is a similar plot for some subsurface brines showing similar results taken from sediments of Mississippian and Pennsylvanian age. Barium Barium, like strontium, is a minor element, comprising 0.04 wt.%, of the earth’s crust; it is more concentrated in igneous rocks and less concentrated in sedimentary rocks (Fleischer, 1962). It, like the other alkaline earth metals, is predominantly lithophile. Table 5.111 illustrates some of the properties of barium; its ionic radius, 1.35 A, permits it t o replace potassium, but usually not calcium and even less commonly magnesium. Barium forms more of its own minerals than does strontium. Barium is readily absorbed by colloids, like potassium, and is therefore retained by soils or precipitated with hydrolysates; it is also concentrated in deep-sea manganese nodules (Hem, 1970). Barium dissolves as bicarbonate, chroride, or sulfate during weathering processes, and migrates in aqueous solutions as these compounds. The solubility of barium sulfate increases when hydrochloric acid or chlorides of the alkali or other alkaline earth metals are present in solution. The properties of barium are similar t o those of strontium. Both precipitate through loss of carbon dioxide from a bicarbonate-bearing solution, or as sulfates by the action of sulfuric acid, sulfides, or sulfates. Strontium, however, is less likely t o be absorbed by clays than barium, because its ionic radius is smaller and its ionic potential is higher. Encrustation deposits taken from plugged pipes of waterflood systems for secondary recovery of oil, where barium is present, usually contain barium, calcium, strontium, iron, and traces of other metals. Barium may cause problems in waterflood systems by reacting with the chromate-type oxygencorrosion inhibitors, forming water-insoluble barium chromate. The amount of barium found in sandstones, shales, and carbonates is about 180, 450, and 90 ppm, respectively (Goldschmidt, 1958).Sea water contains about 0.03 mg/l, and subsurface brines may contain more than 100 mg/l; however, many brines contain less than 10 mg/l. Davis and Collins (1971)found that 59 mg/l of barium sulfate is soluble in a synthetic brine with an innic strength of 3.0487, containing sodium, calcium, magnesium,
TABLE 5.IV Properties of aluminum. copper. iron, lead, manganese, and zinc property
Aluminum
Copper
Iron
Lead
Manganese
Zinc
Atomic number Ionic radius (A)
13 0.50
26 0.76(+2) 0.64(+3) . .
82 1.20(+2) 0.84(+4) . .
25 0.80(+2) 0.46c+7 1..
30 0.74
Outer electronic configuration Atomic weight Ionization potential (V)
27 0.96(+1) 0.691+21 . ,
2s22p63s23p1 26.98 5.984
3s23p63d'04s' 63.54 1.723
3s23p63d64s' 55.54 1.165
3s' 3p6 3d54s' 54.938 1.168
3s2'3p63d" 4s' 65.37 9.391
4d'05s'5p64f'5d106s'6p' 207.19 7.415
MANGANESE
149
potassium, chloride, bromide, and iodide ions. Many analyses performed by wet chemical methods indicate rather high concentrations of barium in some subsurface brines. Some of these high results probably should be attributed to strontium plus barium rather than barium only, because satisfactory separation of the two in wet chemical methods is very difficult to accomplish. Manganese Manganese is a member of the VII B group of elements and is well known for its multiplicity of oxidation states. Essentially it is cationic, and the Mn+4 oxidation state usually is found in sediments. Its (+2) ionic radius is 0.80 8,while the ferric iron radius is 0.76 8 (see Table 5.IV); reasonable amounts of interchange in crystal lattices between these two ions are possible. The abundance of manganese is about 0.1 wt.% of the earth’s crust (Fleischer, 1962). Manganese is present in many oilfield brines because it is readily dissolved by waters containing carbon dioxide and sulfate. Except for titanium, manganese is the most abundant trace element in igneous rocks. Nearly all mineral groups of petrological importance contain manganese. During weathering, manganese is dissolved mainly as the bicarbonate. Decomposition of the bicarbonate leads to the formation of M d 4 compounds. In a reducing type of environment Mn+ compounds migrate in aqueous solutions. Mn+ compounds are less mobile, and Mn+4 compounds precipitate from aqueous solutions. In general, manganese remains in solution at a low redox potential and precipitates at a high redox potential. According to Goldberg (1963), manganese oxide nodules on the ocean bottom occur in both shallow water and deep water environments. He attributes these deposits to slow oxidation of dissolved manganese in areas where the waters contact an oxide surface. In most subsurface brines, the manganese is in the reduced form (Mn+*)because the redox potential is low and the pH is less than 7.0. Any in subsurface brines probably would be suspended with particulate matter or complexed by organic compounds, rather than in ionic solution. Shales and carbonates contain about 850 ppm and 1,100 ppm, respectively, of manganese (Mason, 1966). Sea water contains about 0.002 mg/l, and many subsurface brines contain 1.0 to 6.0 mg/l of manganese. Iron Iron is a member of the VIII group of elements and is predominantly siderophile. However, because it has an affinity for sulfur, it is also thiophile; and because it commonly enters into silicate minerals, it is lithophile as well. It is an ubiquitous element, with an abundance of about 5 wt.% of the earth’s crust (Fleischer, 1962).
150
INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES
Iron, cobalt, and nickel possess atomic radii that differ only about 2% or less, so that the crystal chemistry of the three are related. The divalent ions of nickel, magnesium, cobalt, and iron have similar ionic radii; consequently, their chemistries in the sequence of isomorphous crystallization of mixtures are similar. The trivalent ions of iron and cobalt are similar in size, but the high oxidation potential of cobalt prevents much replacement (Goldschmidt, 1958). The solubility of iron compounds in ground waters is a function of the type of iron compound involved, the amounts and types of other ions in solution, the pH, and the Eh. According t o Larson and King (1954), 100 ppm of ferrous iron can stay in solution at pH 8 and pH 7; the theoretical maximum is about 10,000 ppm. The effects of many other ions, plus temperature and pressure differentials, such as those common to oilfield waters, have not been thoroughly studied. When a ground water in which ferrous iron is dissolved contacts the atmosphere, the following reaction can occur: 2Fe2++ 4HCO3- + H20 + 1 / 2 0
2
+ 2Fe(OH), + 4C02
Sandstone contains iron oxide, iron carbonate, and iron hydroxide, and shales and carbonate rocks contain oxides, carbonates, and sulfides of iron. Oilfield waters with characteristic low redox potentials dissolve some iron from the surrounding rock. The iron occurs in such waters at two levels of oxidation, ferrous or ferric. Knowledge of the amount and type of iron compounds in oilfield waters is used to estimate the amount of corrosion that is occurring in the production system, and t o determine the type of treatment required if the water is t o be used for waterflooding. This knowledge also enables determination of the Eh of the in situ water, because the Eh can be calculated from the Fe+2 and Fe+ values. Shales, sandstones, and carbonates contain about 47,200, 9,800, and 3,800 ppm, respectively, of iron (Mason, 1966). Sea water contains about 0.01 mg/l, and subsurface brines contain from traces to over 1,000 mg/l of iron. Copper Copper is a member of the VIII group of elements, and it is characteristically thiophile; the largest concentrations of it are found in various sulfur compounds. The earth’s crust contains about 0.01 wt.% of copper (Fleischer, 1962). Its compounds are dissolved easily during weathering, if the pH of the solution is less than 4.5. Many of the water-soluble copper compounds are salts of organic acids such as acetic, citric, and naphthenic. Much of the copper that is dissolved is precipitated afterward as sulfide. Traces of copper remain in the oceans, but its content is kept low because of the adsorption on, or combination with, marine organisms. Miholic (1947) presented an age
ZINC
151
division for mineral waters based on the presence of heavy metals in waters associated with joints and faults caused by tectonic movements of different geological ages. He placed copper as the predominant heavy metal in the Caledonian Group of the Orogenic Epoch (post-Silurian). Biochemical processes are known to be responsible for enriching a deposit in metals such as uranium, copper, and vanadium; therefore, this classification is restricted to waters of igneous origin. Most shales and carbonates contain about 45 and 4 ppm, respectively, of copper, with sandstones containing less than 1 ppm (Mason, 1966). Sea water contains about 0.003 mg/l, and most subsurface brines analyzed in this laboratory contained from less than 0.5 mg/l up to about 3 mg/l. The solubility of copper generally decreases with decreasing redox potential and increases with increasing redox potential if reduced sulfur is present. Most subsurface oilfield brines have relatively low redox potentials. zinc Zinc is a member of the I1 B group of elements and is predominantly thiophile. Its abundance in the crust of the earth is about 0.013 wt.% (Fleischer, 1962). Its geochemistry results from the similarity of its divalent ionic radius and the radii of Mg+’, Ni+?, Co+’, Fe+’, and Mn+’ (Goldschmidt, 1958). Zinc is dissolved readily as sulfate or chloride from acid rocks, such as granite, during weathering. Conversely, zinc is not dissolved easily from limestone with which it is deposited. Most alkaline waters do not extract zinc; however, a solution of NH,, NH,NO,, and NaC10, can extract and hold small quantities of zinc; the more acidic the water, the greater the amount of zinc extracted. Zinc is precipitated as the sulfide, oxide, carbonate, or silicate. Traces of zinc are found in sea water, but eventually zinc is deposited in carbonated sediments or in bottom muds or sapropels as sulfide. Shales, sandstones, and carbonates contain about 95, 16, and 20 ppm, respectively, of zinc (Mason, 1966). Sea water contains about 0.01 mg/l, and subsurface brines contain traces to more than 500 mg/l of zinc. Mercury Mercury is a member of the I1 B group of elements, which also includes zinc and cadmium. It is relatively abundant for a heavy element, but still must be considered scarce, with an abundance of about 4 x lo-’ wt.% of the crust of the earth (Fleischer, 1962). Most commercial deposits of mercury are of hydrothermal origin and are related to magmatic rocks; the commercial ore is cinnabar, HgS, or the liquid metal itself (Goldschmidt, 1958). Mercury is predominantly thiophile, and its geochemistry is controlled by the fact that it is volatile, with a boiling point of 357”C, and can be reduced to the metal by ferrous iron. Therefore, in a magmatic environment
152
INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES
the temperature and the redox potential control its occurrence. It is transported in hot springs (White et al., 1963). Shales, sandstones, and carbonates contain about 0.4, 0.03, and 0.04 ppm, respectively, of mercury. Sea water contains 3 x lo-’ mg/l, and subsurface oilfield brines contain 0-0.15 mg/l. The samples containing 0.15 mg/l of mercury were found in relatively dilute brines taken from the Cymric and the Rio Bravo oilfields in California. Free mercury is found in the oils produced from these fields, and the ages of the producing formations range from Eocene t o Pleistocene. The mercury content of natural waters has been used t o locate cinnabar deposits (Dall’Aglio, 1968). The amounts of mercury in waters appear t o increase with increasing bicarbonate concentration. Karasik et al. (1965) found that saline waters containing 200,000 mg/l of chloride contain very small amounts of mercury, which suggests that anionic complexes such as HgC14-* may not be important transporters of mercury. Brackish waters containing up t o 3,000 mg/l dissolved solids, up to 400 mg/l of bicarbonate, and the iodide ion sometimes contain up to 10 ppb of mercury, while stronger brines contain 100 ppm 1-100 ppm ppb (most oilfield waters) ppb (some oilfield waters)
Na, C1 Ca, SO4 K, Sr Al, B, Ba, Fe, Li Cr, Cu, Mn, Ni, Sn, Ti, Zr Be, Co, Ga, Ge, Pb, V, W, Zn
They found no relationship between the constituents in the brine and the minerals in the aquifer rocks except for potassium. They postulated that exchange reactions occurred between the clays in the rocks and potassium in the water to control the dissolved potasssium.
ORIGIN OF OILFIELD WATERS
218
TABLE 7.XI Silurian system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent
highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Bicarbonate Sulfate Organic acid as acetic Ammonium
Number of samples
Concentration (mg/l) -
90 89,000 8,400 8 0.4 41,000 12,000 880 15 90 195,000 1,700 30 27 0 3,500
average 37 49,100 1,900 4 0.4 21,000 4,300 7 30 15 30 122,000 520 17 115 830
220 20 0
90 80
8 14 11 2 2 14 12 2 1 10 14 11 10 11 13 9 10
TABLE 7.XII Ordovician system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent
Concentration (mg/l) highest
Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Manganese Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium
Number of samples
average
70 89,100 2,890 6 0.5 39,000 10,900 900 10 80 56 205,600 7 20 70 2,260 60 7,600
20 31,000 990 2 0.2 6,100 1,300 340 6 20 56 62,000 300 25 270 25 1,070
598 26 583
3,300 630
5 20 140
14 16
15 609 15 11 9 609 607 12 10 18 1 609 19
16
RESEARCH STUDIES
21 9
TABLE 7.XIII Cambrian system - highest concentration of a constituent round, average concentration, and number of samples analyzed
_______--__
~-
Constituent
-
highest ._____
Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Boron Chloride Bromide Iodide Bicarbonate Sulfate Organic acid as acetic Ammonium
__
~
Number of samples
Concentration (mg/l) ~
40 43,000 2,000 3.3 0.6 14,500 8,800 360 13 95,000 1,170 40 790 2,600 50 120
_
average _
_
17 23,400 440 3.3 0.6 4,000 1,300 125 7 46,100 520 18 260 1,170 30
60
8 23 10 1 1 23 22 7 5 23 5 3 23 22 3 3
Compared with sea water the 823 brines were enriched in manganese, lithium, chromium, and strontium, and depleted in tin, nickel, magnesium, and potassium. Generally the silicon content varied inversely with the dissolved solids content. This agrees with a study of the solubilities of silicate minerals where Collins (1969b) found that in general the silicon solubilities decreased with increasing concentrations of dissolved salts at ambient conditions. Research studies related to the origin of oilfield brines Tables 7.IV-XIV indicate that the compositions of oilfield brines are not consistent, and that they are not formed by the simple evaporation or dilution of sea water. Oilfield brines are found in deep formations that sometimes contain fresher water nearer surface outcrop areas, in formations containing evaporites or in close proximity to soluble minerals, and in formations close to surface saline waters. The amounts and ratios of the constituents dissolved in oilfield waters are dependent upon the origin of the water and what has occurred t o the water since entering the subsurface environment. For example, some subsurface waters found in deep sediments were trapped during sedimentation, while other subsurface waters have infiltrated from the surface through outcrops.
N ES
0
TABLE 7.XIV Minor elements in 823 oilfield brine samples in United States and Canada*' Number of samples Illinois Basin 22 Louisiana and Texas Gulf Coast 79 East Texas 88 North Texas 24 West Texas and New Mexico 148 Permian only 74 Pennsylvanian only 34 Silurian and Devonian only 15 Ordovician and Cambrian only 21 Anadarko Basin*' 118 Williston Basin, postPaleozoic 25 Williston Basin, Paleozoic 55 Powder River Basin 22 Other Wyoming 28 Colorado 18 California 116 Sea Water Estimated detection limit
Lithium q25 md
Magnesium q75 q25
Manganese
md
q75
q25
Nickel
md
q75
q25
10
15
25
3,000
6,000
8,000
8Op 175p
750p
ND ND ND
ND ND ND
4 ND 15
15 150 3,000
250 250 5,000
550 800 6,000
3.5 8OOp 1 , 8 0 0 ~ 3.3 25 45
>5,OOOp >5,OOOp 90
3 2 3
15 10 10
25 25 20
500 500 500
1,000 1,000 1,000
1,650 2,000 1,500
2OOp 18Op 500p
4
10
25
200
400
560
30p
10 ND
15 10
25 35
500 900
800 1,550
1,000 3,000
150p 400p 600p 5.6
>5,OOOp >5,OOOp
ND
ND
10
10
250
2,000
9Op 300p
450p
18 ND ND ND ND
35 ND ND ND ND 0.1
50 2 45 ND ND
300 10 20 10 35
600 40 100 30 90 1,272
2,000 225 200 300 175
2
5,OOOp >5,OOOp >5,OOOp
300p >5,OOOp
660p 450p 300p 300p 950p lp-lop 1P
1,200~ 2,000~
1,000~ 750p 2,800~
Tin md
q75
q25
md
ND
ND
ND
< 1P < 1P
< lp < < 1P
3p ND 3p ND 150p ND
15p
< lp < 1P < lp < lp < lp
< < <
HC03-2
+ SO4-’
> C1->
Na+ > NO3-
Assuming that this mechanism operates in a shale filtration system, the order of ion concentration on the high brine concentration side would be the same. The ion concentrations on the fresher water side would be the reverse or : NO3-
> Na+ > Cl- > SO4-’
+ HC03-*
> Ca+’ + Mg+’
Other investigators have obtained similar results. For example, Loeb and Manjikian (1965) found a rejection order of SO4-’ > Mg+’ > Ca+’ > Na+ > HC03- > C1- > NO3-. Michaels et al. (1965) found a rejection order of Ca+2 > Li+ > Na+ > K+ for the pressure independent portion of salt transport in cellulose acetate reverse osmosis desalination membranes. This correlates with the size of the hydrated ion radii because calcium is the largest and potassium the smallest. Further, this indicates that the pore size of the membrane is a controlling factor. The data ‘of Larson (1967) showed that sulfate and carbonate scale formed on the high-pressure side of the membrane and if not removed would cause flow to decrease or stop. The pH on the output or fresh-water side of the membrane decreased.
242
ORIGIN OF OILFIELD WATERS
Russell (1933)considered several processes which could produce subsurface brines more concentrated than sea water. He concluded that evaporation of the water by natural gas generally is not important, water evaporation in coarse-grained rocks generally is not important, gravitational settling of dissolved solids is not greatly important, rocks containing considerable amounts of feldspars and other unstable minerals take up large quantities of hydration water, clays adsorb bases and later expel them into solution causing concentration, and osmosis may occur through semipermeable membranes. DeSitter (1947)noted that oilfield waters are altered as a result of two prominent diagenetic phases. During the first phase magnesium, calcium, sulfate, and carbonate precipitate from the original sea water. During the second phase the concentration of magnesium and calcium ions increases along with the concentration of other dissolved solids. He reasoned that the second phase occurred because of filtration through semipermeable shales. The filtration results because of sediment compaction until a semipermeable membrane develops which allows water molecules t o pass through but retards salt ions. Thus, the more concentrated brines are found where sediment compaction and water flow distance were the greatest. This usually occurs in the deepest portion of a basin. McKelvey et al. (1957) forced aqueous saline solutions through ionexchange resins and found that the effluent solutions contained less dissolved salts than the influent solutions. Effluents from cation-exchange resins were found to contain Na/K ratios similar t o those in the influent; however, the Mg/Ca ratios were at first higher than in the influent but with additional squeezing the ratio decreased t o much lower values. They postulated that similar reactions occur during the compaction of sediments to change the concentrations of constituents dissolved in waters. Pressures of 7 kg/cm2 to 105 kg/cm2 were applied t o force sodium chloride solutions through cation-exchange membranes. The results indicated that the membranes desalted the saline solutions, producing a filtrate containing less salt than the influent. This salt filtering effect was attributed t o the electrical properties of the membrane. Milne et al. (1964)determined the filtering efficiencies of sodium chloride solutions by bentonite membranes. The filtration efficiencies were 94% at 140 kg/cm2 and 88% at 703 kg/cm2 with 0.5N sodium chloride. Increased salinity caused less efficient filtration because filtration efficiencies of 94% for 0.W sodium chloride and 66% for 4N sodium chloride at a pressure of 352 kg/cm2 were obtained. A similar mechanism could operate in the subsurface to create concentrated brines. Young and Low (1965)performed an experiment using natural rock and demonstrated that osmotic flow of water through shale and siltstone occurs. The osmotic pressures produced were less than theoretical and they were attributed to microcracks in the natural rock which caused them t o be less effective than a perfect membrane.
RESEARCH STUDIES
243
Bredehoeft et al. (1963) developed a mathematical model t o predict the distribution of ions within a formation. They assumed that a hydrostatic head differential opetates between the margin and center of a geologic basin, producing a water movement upward through confining low permeability beds. If these low permeability beds contain clay membranes to restrict the passage of ions, the waters on the upflow, or more permeable, side become more concentrated in dissolved solids. They theorized that this process produced the concentrated brines found in the Illinois Basin, and that their model added weight t o the membrane theory of brine concentration. A major drawback to the model is the tremendous pressures that are necessary to produce a movement of water upward through confining low permeability beds. Graf et al. (1965) found that isotopic fractionation occurred when waters passed through shale micropores in the Illinois, Michigan, Alberta, and Gulf Coast Basins. Their study did not yield sufficient evidence t o estimate the total fraction of water movement in the basins subsequent t o sediment compaction. The 6 " 0 concentrations in brines did not indicate a direct correlation with ancient oceans. A study of the 6D and 6l80 in formation waters indicated that the water was predominantly meteoric, little exchange or fractionation had occurred to alter the deuterium, but extensive exchange between the water and rock had altered the oxygen (Clayton et al., 1966). They postulated that formation waters in the Gulf Coast Basin lost their original connate water because of sediment compaction and flushing, and that the present water is meteoric water which came in through outcrops. This study was good; however, basic studies concerning the fractionation and exchange of isotopes between water, hydrocarbons, and rocks need to be made. Results of such studies should enable more positive interpretations. A simplistic model was derived t o determine the amounts of fresh water and sea water necessary t o create the brine compositions now present in the Illinois and Michigan Basins (Graf et al., 1966). The model assumes: (1) perfect efficiency of shale ultrafilters; (2) complete bacterial reduction of sulfate with replacement in solution of equivalent bicarbonate; (3) complete removal of bicarbonate and equivalent sodium by shale ultrafiltration; and (4) magnesium reaction with calcium carbonate to form dolomite. The dolomitization reaction furnished more soluble calcium than is possible for the Illinois Basin, so another calculation was made assuming complete loss of magnesium t o clay minerals with no return of calcium. The calculations indicated that less fresh water passed through the rocks of the Illinois Basin than those of the Michigan Basin. These data conflicted somewhat with Clayton et al. (1966) in that they argued that the water molecules now in the Illinois Basin originated as fresh water, while the data of Graf et al. (1966) indicated that too few volumes of fresh water passed through the Illinois Basin t o alter the brine significantly. A study of the hydrodynamics of the Illinois Basin indicated that in
244
ORIGIN OF OILFIELD WATERS
recent times, before pumpage, the differences in vertical head in the deep aquifers were insufficient t o cause upward flow through shale, resulting in ultrafiltration (Bond, 1972). In fact the head differentials were barely sufficient to enable upward flow through an open conduit. Berry (1969) outlined the relative factors that influence membrane filtration in geologic environments. The membrane properties of shales are caused by the electrfcal properties of their clays and organic materials. Clays predominantly are cation exchangers with singly charged SiO- and AlOSi-% sites and minor anion exchanges with replaceable OH- ions. Divalent cations are adsorbed in preference to monovalent cations and sodium is hyperfiltratcd with respect t o lithium and strontium with respect to calcium, because of preferential adsorption of ions with ionic potentials most similar t o the ionic potential of the exchange site. The selectivity of hyperfiltration for the halogens is C1 > Br > I > F because of their substitution for O H in the clays. Thus, in waters concentrated by this process the Ca/Na, Na/Li, Sr/Ca, Cl/Br, Br/I, and I/F ratios should increase. These ratio increases have been found in some brine systems, but by no means in all systems. Billings et al. (1969) found five types of formation waters in the Western Canada Sedimentary Basin and postulated the origin of two of the types. One type of water was formed by selective membrane filtration which produced waters containing high concentrations of dissolved solids. A second type was a mixture of membrane-concentrated formation water and bitterns formed after the precipitation of halite but before the precipitation of sylvite. They theorized that the alkalies were filtered selectively by clayshale membranes, producing a concentrated brine, and that the relative concentration pattern is Rb > K > Na > Li. This pattern is the reverse of what occurs by ion exchange but is similar to the surface mobilities of cations along clay surfaces. A detailed study of the Western Canada Sedimentary Basin, including a determination of the rock volume and pore volume (Hitchon, 1968), the effect of topography upon the fluid flow (Hitchon, 1969a), and the effect of geology upon the fluid flow (Hitchon, 1969b), strongly suggested that thermal, electro-osmotic, and chemico-osmotic forces are operating within the basin to affect the fluid energy gradients. Pressure differentials of about 98 kg/cm*along with salinity differences of 200,000 mg/l between formations in close proximity were found which suggest that chemico-osmotic forces are occurring. Hitchon and Friedman (1969) used chemical analyses and stable-isotope analyses for hydrogen and oxygen for surface waters, shallow ground waters, and deep ground waters in a study of the origin of formation waters in the Western Canada Sedimentary Basin. They postulated that surface waters have mixed with diagenetically altered sea water to form the formation waters. Using mass balance data for the deuterium and dissolved solid contents of the formation waters, they calculated not only how much fresh water is
CONCLUSIONS
245
present in the modified sea water but also observed how it redistributed the dissolved solids t o prQduce salinity variations. They concluded that formation waters result from mixing of surface waters with modified marine or nonmarine water in the subsurface rocks, that exchange of oxygen isotopes between the water and rock caused different water types in different basins, and that formation waters that have passed through shale ultrafilters are more depleted in deuterium. A study of the Surat Basin showed that most of its hydrocarbon accumulations are associated with quasi-stagnant waters. The salinities of these quasi-stagnant waters were higher than were the salinities of the waters in the more dynamic recharge areas. The investigators postulated that these high salinity waters were formed by membrane filtration because of crossformational flow and also that the hydrocarbon accumulations in these quasi-stagnant areas resulted from release of hydrocarbons mobilized by a moving water. The hydrocarbons were released because of the higher salinities of the waters in the quasi-stagnant areas (Hitchon and Hays, 1971). A study of waters in sedimentary rocks of Neogene age in the northern Gulf of Mexico Basin was made by Jones (1969). The hydrologic conditions currently found in these sediments are similar to conditions that previously occurred in older sedimentary basins. Osmotic flow has a dominant influence upon the hydrology of normally and abnormally pressured aquifer systems in the northern Gulf Basin. Jones (1969) found that many forces such as gravity, sediment diagenesis, different water salinities, ionic and molecular diffusion, different electrical potentials of sediments, thermal potentials, pressure, and osmotic membrane filtration affect the hydrology in this basin. Fowler (1970) found that salinity variations within the Frio sands in the Chocolate Bayou field, Brazoria County, Texas, are the result of selective concentration of ions by shales acting as membranes. In this field, pressures seem to reflect the flow paths of the waters, and the greatest changes in pressures are found across shaly sections. Analyses of water samples from this field over a 28-year period indicate decreasing salinity with production time caused by dilution of the original brines by waters squeezed from the shales adjacent to the aquifers. Chilingarian and Rieke (1969) reviewed the processes which can alter the chemical composition of formation waters. They concluded that most of the original water was sea water, and that the concentration process in many cases results from compaction and membrane filtration rather than evaporation. Their experimental results indicated that solutions squeezed out of rocks during compaction progressively decrease in dissolved solids concentrations with increasing depth.
Conclusions The origin of oilfield waters is related to many natural processes. Initially,
246
ORIGIN OF OILFIELD WATERS
meteoric water reacted with weathered rock, soil, and organic matter. The excess waters that did not penetrate the rock or soil caused the rock and soil to erode and channels formed through which the water could move more easily. Forces of gravity caused the water to move from areas of high potential to areas of low potential, and as the waters moved, the concentrations of dissolved solids in them increased. Some of these waters found their way to lakes and the sea. As they entered the lakes or seas their movement slowed, causing some of the suspended particles in them t o deposit. Mixing of the waters with the more saline waters in the sea caused dissolved carbonate and organic compounds to precipitate. Evaporation of the sea and lake waters caused other compounds such as sulfates t o precipitate. The pH of the waters changed slightly because of reactions with the atmosphere, the sediments, and other waters. Each pH change caused precipitation of compounds or dissolution of new compounds. Some of the waters became highly concentrated in dissolved solids in the more shallow marine environments. Evaporites formed in these lagoons, pans, and exposed supratidal sabkhas. Evaporites also formed in deep-water basins when the salinity of the water at the bottom of the basin became sufficiently high. The sediments were buried as additional sediments were deposited on them, and water surrounding the sediment particles also was buried. As the depth of burial increased, the sediments compacted and some of the water was squeezed out. Both the squeezed-out water and the remaining interstitial water reacted with minerals in the sediments t o change the composition of the dissolved solids in the water and the composition of the sediments. Mechanisms that cause the oilfield waters t o differ in composition from water originally deposited with the sediments include ion exchange, infiltrating waters, sediment leaching, mineral formation, sulfate reduction, and ultrafiltration through clay-shale membranes.
References Al'tovskii, M.E., Kuznetsova, Z.I. and Shvets, V.M.,1961. Origin of Oil and Oil Deposits (English Transl. by Consultants Bureau). Plenum Press, New York, N.Y., 107 pp. Anonymous, 1964. Chemistry of the oceans. Chem. Eng. News, 42:12A. Atwater, G.I. and Miller, E.E., 1965. The effect of decrease in porosity with depth on future development of oil and gas reserves in South Louisiana. Presented at Annual Meet., A m . Assoc. Pet. Geol., New Orleans, La., 1965 -Bull. A m . Assoc. Pet. Geol., 49:334. Ault, W.U., 1959. Isotopic fractionation of sulfur in geochemical processes. In: P.H. Abelson (Editor), Researches in Geochemistry. John Wiley and Sons, New York, N.Y., pp.241-259. Baas Becking, L.G.M., Kaplan, I.R. and Moore, D., 1960. Limits of the natural environment in terms of pH and oxidation-reduction potentials. J. Geol., 68 :243-284.
REFERENCES
24 7
Baker, E.G., 1960. A hypothesis concerning the accumulation of sediment hydrocarbons to form crude oil. Geochim. Cosmochim. Acta, 19:309-317. Barnes, H.L. and Czamanske, G.K., 1967. Solubilities and transport of ore minerals. In: H.L. Barnes (Editor), Geochemistry o f Hydrothermal Ore Deposits. Holt, Rinehart, and Winston, New York, N.Y., pp.334-381. Bentor, Y.K., 1969. On the evolution of subsurface brines in Israel. Chem. Geol., 4 :83-1 10. Berry, F.A.F., 1969. Relative factors influencing membrane filtration effects in geologic environments. Chem. Geol., 4:295-301. Billings, G.K., Hitchon, B. and Shaw, D.R., 1969. Geochemistry and origin of formation waters in the Western Canada Sedimentary Basin, 2. Alkali metals. Chem. Geol., 4:211-223. Birch, F., 1954. The present state of geothermal investigations. Geophysics, 19:645+359. Bond, D.C., 1972. Hydrodynamics in deep aquifers of the Illinois Basin. Ill. State Geol. Surv., Circ., No. 470, 7 2 pp. Borchert, H. and Muir, R.O., 1964. Salt Deposits - The Origin, Metamorphism, and Deformation o f Evaporite. D. Van Nostrand, London, 338 pp. Braitsch, 0. and Herrmann, A.G., 1963. Zur Geochemie des Broms in Salinaren Sedimenten, I. Experimentelle Bestimmung der Br-Verteilung in Naturlichen Salzsystemen. Geochim. Cosmochim. Acta, 27:361-391. Braitsch, 0 . and Herrmann, A.G., 1964. Zur Geochemie des Broms in Salinaren Sedimenten, 11. Die Bildungstemperaturen Primarer Sylvin- und Carnallit-Gesteine. Geochim. Cosmochim. Acta, 28 :1081-1 109. Bredehoeft, J.D., Blyth, C.R., White, W.A. and Maxey, G.B., 1963. Possible mechanism for concentration of brines in subsurface formations. Bull. A m . Assoc. Pet. Geol., 47 :257-269. Brod, I.O., 1960. On principal rules in the occurrence of oil and gas accumulations in the world. Int. Geol. Rev., 2:922-1005. Broecker, W.S. and Takahashi, T., 1966. Calcium carbonate precipitation on the Bahama Banks. J. Geophys. Res., 71 :1575-1602. Brongersma-Sanders, M., 1971. Origin of major cyclicity of evaporites and bituminous rocks: an actualistic model. Mar. Geol., 11 :123-144. Bruderer, W., 1956. Les ocbans souterrains fossiles et le phtrole. Assoc. Fr. Technol. Pet. Bull., 120:535-556. Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation t o petroleum migration. Bull. A m . Assoc. Pet. Geol., 53:73-93. Bush, P.R., 1970. Chloride-rich brines from sabkha sediments and their role in ore formation. Inst. Min. Metall. Trans., 79:B137-B144. Butler, G.P., 1969. Modern evaporite deposition and geochemistry of coexisting brines, the Sabkha Trucial Coast, Arabian Gulf. J. Sediment. Petrol., 39 :70-89. Caraway, W.H. and Gates, G.L., 1959. Methods for determining water contents of oilbearing formations. U.S. Bur. Min. Rep. Invest., No.5451, 81 pp. Carpelan, L.H., 1957. Hydrobiology of the Alviso salt ponds. Ecology, 38:375-390. Carpenter, A.B. and Miller, J.C., 1969. Geochemistry of saline subsurface water, Saline County (Missouri). Chem. Geol., 4:135-167. Cartmill, J.C. and Dickey, P.A., 1970. Flow of a disperse emulsion of crude oil in water through porous media. Bull. A m . Assoc. Pet. Geol., 54:2438-2447. Case, L.C., 1955. Origin and current usage of the term "connate water". Bull. A m . Assoc. Pet. Geol., 39:1879-1882. Chave, K.E., 1960. Evidence on history of sea water from chemistry of deeper subsurface waters of ancient basins. Bull. A m . Assoc. Pet. Geol., 44:357-370. Chebotarev, I.I., 1964. Metamorphism of natural waters in the crust of weathering. Geochim. Cosmochim. Acta, 8:22-48, 137-170, 198-212.
248
ORIGIN OF OILFIELD WATERS
Chilingarian, G.V. and Rieke, 111, H.H. 1969. Some chemical alterations of subsurface waters during diagenesis. C h e m Geol., 4:235-252. Clark, S.P. and Ringwood, A.E., 1964. Density disturbance and constitution of the mantle. Rev. Geophys., 2:35-88. Clarke, F.W. and Washington, H.S., 1924. The composition of the earth’s crust. US. Geol. Surv. Prof.Paper, No.127, pp.1-112. Clayton, R.N., Friedman, I., Graf, D.L., Mayeda, T.K., Meents, W.F. and Shimp, N.F., 1966. The origin of saline formation waters, 1. Isotopic composition. J. Geophys. Res., 71 :3869-3881. Cloke, P.L., 1966. The geochemical application of E h - p H diagrams. J. Geol. Educ., 1 4:140-148. Collins, A.G., 1967. Geochemistry of some Tertiary and Cretaceous age oil-bearing formation waters. Environ. Sci. Technol., l :725-730. Collins, A.G. 1969a. Chemistry of some Anadarko Basin brines containing high concentrations of iodide. C h e m Geol., 4:169-187. Collins, A.G., 1969b. Solubilities of some silicate minerals in saline waters. U S . O f f . Saline Water R e s Dev. Progr. R e p . , No.472, 27 pp. Collins, A.G., 1970. Geochemistry of some petroleum-associated waters from Louisiana. U.S. Bur. M i a Rep. Invest., No.7326, 31 pp. Collins, A.G., 1972. Geochemical Classification o f Formation Waters f o r Use in Hydrocarbon Exploration and Production. M.S. Thesis, University of Tulsa, Tulsa, Okla., 63 pp. Collins, A.G., Bennett, J.H. and Manuel, O.K., 1971. Iodine and algae in sedimentary rocks associated with iodine-rich brines. Geol. SOC.A m . Bull., 82:2607-2610. Dapples, E.C., 1959. The behavior of silica in diagenesis. In: Silica in Sediments ( A Symposium) - SOC.E c o n Paleontol. Mineral., Spec. Publ., No.7, pp.36-54. Deffeyes, K.S. Lucia, F.J. and Weyl, P.K., 1964. Dolomitization: observations on the island of Bonaire, Netherlands Antilles. Science, 143 :678+379. Degens, E.T. and Chilingar, G.V., 1967. Diagenesis of subsurface waters. In: G. Larsen and G.V. Chilingar (Editors), Diagenesis in Sediments. Elsevier, Amsterdam, pp.77-502. Degens, E.T., Hunt, J.M., Reuter, J.H. and Reed, W.E., 1964. Data on the distribution of amino acids and oxygen isotopes in petroleum brine waters of various geologic ages. Sedimentology, 3 :199-225. DeSitter, L.Y., 1947. Diagenesis of oilfield brines. Bull. A m . Assoc. Pet. Geol., 31 :2030-2040. Dickey, P.A., 1969. Increasing concentration of subsurface brine with depth. Chem. Geol., 4:361-370. Dingman, R.J. and Angino, E.E., 1969. Chemical composition of selected Kansas brines as a aid to interpreting change in water chemistry with depth. Chem. Geol., 4: 325-339. Dunham, K.C., 1970. Mineralization by deep formation waters - a review. Znst. Metall. Trans., 79:B127-B>36. Eckhardt, F.J., 1958. Uber Chlorite in Sedimenten. Geol. Jahrb., 75:437-474. Egleson, G.D. and Querio, C.W., 1969. Variation in the composition of brine from the Sylvania formation near Midland, Michigan. Environ. Sci. Technol., 3:367-371. Ellis, A.J.’, 1968. Natural hydrothermal systems and experimental hot waterhock interaction: reactions with NaCl solutions and trace metal extraction. Geochim. Cosmochim. Acta, 32:1356-1363. Emery, K.O., 1960. The Sea O f f Southern California: A Modern Habitat o f Petroleum. John Wiley and Sons, New York, N.Y., 366 pp. Emery, K.O. and Rittenberg, S.C., 1952. Early diagenesis of California Basin sediments in relation t o origin of oil. Bull. A m . Assoc. Pet. Geol., 36:735-806.
REFERENCES
249
Evans, G., Kinsman, D.D.J. and Shearman, D.J., 1963 A reconnaissance survey of the environment of the recent carbonate sedimentation along Trucial Coast, Persian Gulf. In: L.M.J.U. van Straaten (Editor), Developments in Sedimentology, 1 . Deltaic and Shallow Marine Deposits. Elsevier, Amsterdam, pp. 129-1 35. Fowler, Jr., W.A., 1970.Pressures, hydrocarbon accumulation, and salinities - Chocolate Bayou field, Brazoria County, Texas. J. Pet. Technol., 22:411-423. Friedman, G.M. and Sanders, J.E. 1967. Origin and occurrence of dolostones. In: G.V. Chilingar, H.J. Bissel and R.W. Fairbridge (Editors), Carbonate Rocks - Origin, Occurrence and Classification. American Elsevier, New York, N.Y., pp.267-348. Friedman, G.M. Fabricand, B.P., Imbimbo, E.S., Brey, M.E. and Sanders, J.E., 1968. Chemical changes in interstitial waters from continental shelf sediments. J. Sediment. Petrol., 38:1313-1319. Garrels, R.M. and Christ, C.L., 1965. Solutions, Minerals, and Equilibria. Harper and Row, New York, N.Y., 450 pp. Gevirtz, J.L. and Friedman, G.M. 1966. Deep-sea carbonate sediments of the Red Sea and their implications on marine lithification. J. Sediment. Petrol., 36:143-151. Ginsburg, R.N. and Lowenstam, H.A., 1958. Influence of marine bottom communities on the depositional environment of sediments. J. Geol., 66:310-318. Grabau, A.W., 1920. Geology of Nonmetallic Mineral Deposits Other Than Silicate, 1 . Principles of Salt Deposits. McGraw-Hill, New York, N.Y., 435 pp. Graf, D.L., Friedman, I. and Meents, W.F., 1965. The origin of saline formation waters, 11. Isotopic fractionation by shale micropore systems. Ill. State Geol. Surv. Circ., No.393, 32 pp. Graf, D.L., Meents, W.F., Friedman, I. and Shimp, N.F., 1966. The origin of saline formation waters, 111. Calcium chloride waters. Ill. State Geol. Surv. Circ. No. 397,60 PP. Grim, R.E., 1952. Clay Mineralogy. McGraw-Hill, New York, N.Y., 396 pp. Ham, W.E. 1962. Classification of carbonate rocks. A m . Assoc. Pet. GeoL, Mem. 1 , 279 PP. Hanshaw, B.B. and Hill, G.A., 1969. Geochemistry and hydrodynamics of the Paradox Basin region, Utah, Colorado, and New Mexico. Chem. Geol., 4:263-294. Helgeson, H.C., 1964. Complexing and Hydrothermal Ore Deposition. MacMillan, New York, N.Y., 128 pp. Hem, J.D., 1970. Study and interpretation of the chemical characteristics of natural water. U.S. Geol. Surv. Water Supply Paper, No. 1473,363 pp. Hiltabrand, R.R. 1970. Experimental Diagenesis o f Argillaceous Sediment. Ph.D. Dissertation, Louisiana State University, Baton Rouge, La., 152 pp., unpublished. Hitchon, B., 1968. Rock volume and pore volume data fo; plains region of Western Canada Sedimentary Basin between latitudes 49' and 60 N. Bull. A m . Assoc. Pet. Geol., 52:2318-2323. Hitchon, B. 1969a. Fluid flow in the Western Canada Sedimentary Basin, 1. Effect of topography. Water Resour. Res., 5:186-195. Hitchon, B., 196913. Fluid flow in the Western Canada Sedimentary Basin, 2. Effect of geology. Water Resour. Res., 5:460-461. Hitchon, B. and Friedman, I., 1969. Geochemistry and origin of formation waters in the Western Canada Sedimentary Basin, I. Stable isotopes of hydrogen and oxygen. Geochim. Cosmochim. Acta, 33:1321-1349. Hitchon, B . and Hays, J., 1971. Hydrodynamics and hydrocarbon occurrences, Surat Basin, Queens land, Australia. Water Resour. Res., 7:658-676. Holser, W.T. 1963. Chemistry of brine inclusions in Permian salt from Hutchison, Kansas. In: J.L. Rau (Editor), Symposium o n Salt. Northern Ohio Geological Society, Cleveland, Ohio, pp.86-103. Hunt, J.M., 1968. How gas and oil form and migrate. World Oil, 163:140-150.
250
ORIGIN OF OILFIELD WATERS
Hunt, J.M. and Jamieson, G.W., 1958. Oil and organic matter in source rocks of petroleum. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp. 735-746. Illing, L.V., 1954. Bahaman calcareous sands. Bull. A m . Assoc. Pet. Geol., 38:l-95. Jones, B.F., Vandenburgh, A.S., Truesdell, A.H., and Rettig, S.L., 1969. Interstitial brines in playa sediments. Chem. Geol., 4:253-262. Jones, P.H., 1969. Hydrology of Neogene deposits in the northern Gulf of Mexico Basin. La. Water Resour. Res. Inst. Bull., GT-2, 105 pp. Kelley, W.P., 1948. Cation Exchange in Soils. Reinhold, New York, N.Y., 144 pp. Khitarov, N.I. and Pugin, V.A., 1966. Behavior of montmorillonite under elevated temperatures and pressures. Geochem. Znt., 3:621-626. Kimura, S . and Souriragan, S., 1967. Analysis of data in reverse osmosis with porous cellulose acetate membranes used. A m . Inst. Chem. Eng., 13:497-503. Kozin, A.N., 1960. Geochemistry of bromine and iodine of formation waters in the Kuybyshev area o n the Volga. Pet. Geol., 4:llO-113. Kramer, J.R., 1963. History of the composition of sea water - liquid inclusions compared with a chemical equilibrium model. Geol. SOC.A m . , Spec. Paper, No. 73, 190 pp. Kramer, J.R., 1969. Subsurface brines and mineral equilibria. Chem. Geol., 4:37-50. Krauskopf, K.B., 1956. Factors controlling the concentrations of thirteen rare metals in sea water. Geochim. Cosmochim. Acta, 9:l-32. Krumbein, W.C., 1951. Occurrence and lithologic associations of evaporites in the United States. J. Sediment Petrol., 21:63--81. Krejci-Graf, K., 1963. Uber Rumanische Olfeldwasser. Geol. Mitt. Hydrogeol. Hydrochem., 2:3 51-392. Kvenvolden, K.A., 1964. Hydrocarbons in modern sediments and the origin of petroleum. Min. Mag., Colo. School Min., 54:24-25. Larson, T.J., 1967. Purification of subsurface waters by reverse osmosis. J. Am. Water Works Assoc., 59:1527-1548. Levorsen, A.I., 1966. Geology of Petroleum. W.H. Freeman, San Francisco, Calif., 2nd ed., 724 pp. Loeb, S. and Manjikian, S., 1965. Six-month field test of a reverse osmosis desalination membrane. Ind, Eng. Chem. Process Design Dev., 4: 207-21 2. Lotze, F., 1938. Wichtigen Lager Stattender “nicht-erze”, III. Steinsalz und Kalisalze Geologie. Borntraeger, Berlin, 936 pp. Mandl, I., 1953. Solubilization of insoluble matter in nature, 11. The part played by salts of organic and inorganic acids occurring in nature. Biochim. Biophys. Acta, 10:540-569. Mandl, I., Grauer, A. and Neuberg, C., 1952. Solubilization of insoluble matter in nature, I. The part played by salts of adenosinetriphosphate. Biochim. Biophys. Acta, 8 :654-663. Manheim, F.T. and Bischoff, J.L., 1969. Geochemistry of pore waters on the Continental Slope of the northern Gulf of Mexico. Chem. Geol., 4:63-82. McAuliffe, C., 1969. Determination of dissolved hydrocarbons in subsurface brines. Chem. Geol., 4:225-234. McKelvey, J.G., Spiegler, K.S. and Wyllie, M.R.J., 1957. Salt filtering by ion-exchange grains and membranes. J. Phys. Chem., 61(2):174-178. Meyerhoff, A.A., 1970. Development in Mainland China, 1949-1968. Bull. A m . Assoc. Pet. Geol., 54:1567-1580. Michaels, A.S., Bixler, H.J. and Hodges, Jr., R.M., 1965. Kinetics of water and salt transport in cellulose acetate reverse osmosis desalination membranes. J. Colloid Sci., 20: 1034-1056. Milne, I.H., McKelvey, J.G. and Trump, R.P., 1964. Semi-permeability of bentonite membranes to brines. Bull. A m . Assoc. Pet. Geol,, 48:103-105.
REFERENCES
251
Moore, C.A., 1969. The occurrence of oil in sedimentary basins. World Oil, 168:69-72. Morris, R.C. and Dickey, P.A., 1957. Modern evaporite deposition in Peru. Bull. A m . Assoc. Pet. Geol., 41:2467-2474. Neruchev, S.G. and Kovacheva, I.S., 1965. The effect of geological conditions on the amount of oil given up by source rocks. Dokl. Akad. Nauk U.S.S.R., 162:913-914. Neumann, H.J. and Jobelius, H., 1967. Detection of emulsifying agents in crude oils and oilfield waters as a contribution to the problem of oil migration. Erdol Kohle Petrochem., 20:622-625. Parker, J. W., 1969. Water history of Cretaceous aquifers, East Texas Basin. Chem. Geol., 4:lll-133. Peake, E. and Hodgson, G.W., 1966. Alkanes in aqueous systems, I. Exploratory investigations on the accommodation of Cz 0-c~3 n-alkanes in distilled water and occurrence in natural water systems. J. A m . Oil Chem. SOC.,43:215-222. Pettijohn, F.J., 1957. Sedimentary Rocks. Harper and Brothers, New York, N.Y., 2nd ed., 718 pp. Philipp, W., Drong, H.J., Fuchtbauer, H., Haddenhorst, H.G. and Jankowsky, W.J., 1963. The history of migration in the Gifhorn Trough ( N W Germany), Sixth World Pet. Congr., Frankfurt/Main, June, 1963, Sect. I, Paper, No. 19, pp. 457-481. Philippi, G.T., 1965. On the depth, time and mechanism of petroleum generation. Geochim. Cosmochim. Acta, 29: 1021-1049. Phleger, F.B. and Ewing, G.C., 1962. Sedimentology and oceanography of coastal lagoons in Baja, California, Mexico. Geol. SOC.A m . Bull., 73:145-181. Pirson, S.J., 1968. Redox log interprets reservoir potential. Oil Gas J., 66:6*75. Plumley, W.J., Risley, G.A., Graves, Jr., R.W. and Kaley, M.E., 1962. Energy index for limestone interpretation and classification. In: W.E. Hem (Editor), Classification o f Carbonate Rocks - A m . Assoc. Pet. Geol., Mem.1, pp.85-107. Pollard, T.A. and Reichertz, P.O., 1952. Core-analysis practices - basic methods and new developments. Bull. Am. Assoc. Pet. Geol., 36:230-252. Quaide, W . , 1958. Claymineralsfrom salt concentration ponds. A m . J. Sci., 256:431-437. Rankama, K. and Sahama, T.G., 1950. Geochemistry. Chicago University Press, Chicago, Ill., 991 pp. Riley, G.A., 1944. The carbon metabolism and photosynthetic efficiency of the earth as a whole. J. A m . Sci., 32:134. Rittenhouse, G., 1967. Bromine in oilfield waters and its use in determining possibilities of origin of these waters. Bull. A m . Assoc. Pet. Geol., 51:2430-2440. Rittenhouse, G., Fulton, R.B., Grabowski, R.J. and Bernard, J.L., 1969. Minor elements in oilfield waters. Chem. Geol., 4:189-209. Rosenqvist, I.T., 1962. The influence of physico-chemical factors upon the mechanical properties of clays. Clays Clay Minerals, 9 :12-27. Ross, C.S., 1943. Clays and soils in relation to geologic processes. J. Wash. Acad. Sci., 33:225-235. Russell, W.L., 1933. Subsurface concentration of chloride brines. Bull. A m . Assoc. Pet. Geol., 17:1213-1228. Schoeller, H., 1955. Geochemie des eaux souterraines. Rev. Inst. F'r. Pet., 10:181-213, 219-246, 507-552. Siever, R., Beck, K.C. and Berner, R.A. 1965. Composition of interstitial waters of modern sediments. J. Geol., 73:39-73. Skinner, B.J., 1969. Earth Resources. Prentice-Hall, Englewood Cliffs, N.J., 149 pp. Sloss, L.L., 1953. The significance of evaporites. J. Sediment. Petrol., 23:143-161. Smith, P.V., 1954. Studies on origin of petroleum: occurrence of hydrocarbons in recent sediments. Bull. A m . Assoc. Pet. Geol., 38:377-381. Stumm, W. and Morgan, J.J., 1970. Aquatic Chemistry. Wiley-Interscience, Div. of John Wiley and Sons, New York, N.Y., 583 pp.
252
ORIGIN OF OILFIELD WATERS
Taylor, D.W., 1956. Fundamentals o f Soil Mechanics. John Wiley and Sons, New York, N.Y., 700 pp. Tenaghi, K. and Peck, R.B., 1968. Soil Mechanics in Engineering Pmctice. John Wiley and Sons, New York, N.Y., 84 pp. Topms, J.S., 1970. Review of knowledge of metalliferous brines and related deposits. Inst. Min. Metall. Trans., 79:B116-B126. Truesdell, A.H. and Jones, B.F., 1969. Ion association in natural brines. Chem. Geol., 4:51-62. Usiglio, J., 1849. Analyse de I’eau de la Mediterrank sur les c6tes de France. Ann. Chim. Phys., 3:92-1 07, 27 :172-191. Valyashko, M.G., 1956. The geochemistry of bromine in halogenesis processes and the use of bromine content as a genetic and prospecting criterion. Geochemistry, 1 :33-49. Van Everdingen, R.O., 1968. Studies on formation waters in western Canada; geochemistry and hydrodynamics. Can. J. Earth Sci., 5:523-543. Van Nostrand Press, 1958.Science Encyclopedia. Princeton, N.J., 2nd ed., pp. 371-1528. Vinogradov, A.P., 1953. The elementary chemical composition of marine organisms. Sears. Found. Mar. R e s , Mem., 11:85,91,107,216. Visher, G.S., 1965. Use of vertical profile in environmental reconstruction, Bull. A m . Assoc. Pet. Geol., 49:41-61. Von Engelhardt, W., 1960. On the chemistry of the pore solution of sediments. Uppsala Univ. Geol. Inst. Bull., 40: 189-204 (in German). Weller, J.M., 1959. Compaction of sediments. Bull. A m . Assoc. Pet. Geol., 43:273-310. White, D.E., 1957. Magmatic, connate, and metamorphic waters. Geol. SOC.A m . Bull., 68:1659-1 682. White, D.E., 1965. Saline waters of sedimentary rocks. In: A. Young and J.E. Galley (Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem.4, pp.342-366. Young, A. and Low, P.F., 1965. Osmosis in argillaceous rocks. Bull. A m . Assoc. Pet. Geol., 49:1004-1008.
Chapter 8. CLASSIFICATION OF OILFIELD WATERS
Classification of waters provides a basis for grouping closely related waters. Because the grouping is chemical, it is dependent upon the dissolved constituents found in the waters. Most of the classification systems developed t o date have considered only the dissolved major inorganic constituents and have ignored the organic and the minor and trace inorganic constituents. Waters as related to the earth are meteoric, surface, and subsurface. Surface waters can be fresh or saline if the amounts of dissolved constituents in the waters are used to classify them. For example, water from melting snow on a mountain top usually will contain small amounts of dissolved mineral matter and can be classified as fresh water, while water in an ocean will contain about 35,000 mg/l dissolved minerals and is classified as saline. Waters found in rivers connecting the mountain stream t o the ocean may contain varying amounts of dissolved constituents and depending upon the amounts can be classified as fresh or saline. In a similar manner, subsurface waters are classified as fresh or saline. Merely classifying a water as either fresh or saline does not provide a very useful classification. The dissolved constituents that are used in many classification systems depend upon the amounts or ratios of sodium, magnesium, calcium, carbonate, bicarbonate, sulfate, and chloride found in the water. The reason for this is that these are the ions that usually are determined or calculated in a water. (Sodium often is calculated from the difference found in the stoichiometric balance of the determined anions and cations.) The amounts and ratios of these constituents in subsurface waters are dependent upon the origin of the water and what has occurred t o the water since entering the subsurface environment. For example, some subsurface waters found in deep sediments were trapped during sedimentation, while other subsurface waters have been diluted by infiltration of surface waters through outcrops. Some waters have been replaced by infiltration water. Also, rocks containing the waters often contain soluble constituents, which dissolve in the waters or contain chemicals which will exchange with chemicals dissolved in the waters causing alterations of the dissolved constituents. The amounts of dissolved constituents found in subsurface waters can range from a few milligrams per liter t o more than 350,000 mg/L This salinity distribution is dependent upon several factors, including hydraulic gradients, depth of occurrence, distance from outcrops, mobility of the dissolved chemical elements, soluble material in the associated rocks, and the exchange reactions.
254
CLASSIFICATION OF OILFIELD WATERS
Portions of three classification systems (Palmer, 1911; Sulin, 1946; Schoeller, 1955) and Bojarski’s (1970) modification of Sulin’s system were applied to about 4,000 formation waters (U.S. Bureau of Mines, 1965). The waters were analyzed by standard methods (American Petroleum Institute, 1968). The results indicated that the classifications are useful in exploration and production problems. Palmer’s classification Palmer (1911) observed that the basic characteristics of natural waters are dependent upon their salinity (salts of strong acids) and alkalinity (salts of weak acids). Salts that cause salinity are those that are not hydrolyzed, while alkalinity is caused by free alkaline bases produced by the easily hydrolyzable salts of weak bases. All positive ions (cations) including hydrogen can cause salinity, but of the negative ions (anions), only the strong acids, (e.g., chloride, sulfate, and nitrate) can cause salinity. Because salinity is dependent upon the combined activity of the cations and anions and is limited by the reacting values of the strong acids, its value is determined by multiplying the total value of the strong acids by two. Alkalinity is caused by free alkaline bases as a result of the hydrolytic action of water on dissolved bicarbonates and other weak acid salts. The alkalinity value is calculated by doubling the reacting values of the bases which exceed the reacting values of the strong acids. The ions that commonly are found in waters comprise three groups: (a) alkalies (sodium, potassium, lithium), whose salts are easily soluble in water and do not cause hardness; (b) alkaline earths (magnesium, calcium, strontium, barium), whose salts cause hardness and many of which are sparingly soluble; and (c) hydrogen, whose salts are acids and cause acidity. Geologists know what “strong alkalies”, “alkaline earths”, “strong acid radicles”, “weak acid radicles”, “ions”, and “reacting values” mean generally. To compare several analyses it usually is easier if they are made on a chemical basis. The. proportions of the various ions do not react in proportion t o the various weights given in milligrams per liter but rather in proportion to their “capacity for reaction”, or “reaction value”. The reacting value of each ion is determined by multiplying the amount of each radicle by weight (mg/l) by its “reaction coefficient”, which is the valence of a radicle divided by its atomic weight. The groups of the ions are determined by summing the reacting values of their members, and according to the predominance of reacting values of the groups, five special properties were designated by Palmer. To determine the special properties, the reacting values of a group of cations or anions are doubled so that the full value of a given special property is considered. The terms “primary” and “secondary” were used t o qualify the general proper-
PALMER’S CLASSIFICATION
25 5
ties of the water; e.g., the principal soluble decomposition products of the oldest rock formations are the alkalies (primary), while more recent rock formations are the principal source of the alkaline earths (secondary), This theory of Palmer’s that the terms primary and secondary are associated with the age of the rock should not necessarily be considered undisputably true, because primary salinity certainly can be acquired from other soluble material than that derived directly from decomposition products of the oldest rock formations. The five special properties of water are: (1) Primary salinity (alkali salinity); that is, salinity not t o exceed twice the sum of the reacting values of the radicles of the alkalies. (2) Secondary salinity (permanent hardness); that is, the excess (if any) of salinity over primary salinity, not t o exceed twice the sum of the reacting values of the radicles of the alkaline earths group. (3) Tertiary salinity (acidity); that is, the excess (if any) of salinity over primary and secondary salinity. (4) Primary alkalinity (permanent alkalinity); that is, the excess (if any) of twice the sum of the reacting values of the alkalies over salinity. (5) Secondary alkalinity (temporary alkalinity); that is, the excess (if any) of twice the sum of the reacting values of the radicles of the alkaline earths group over secondary salinity. Reacting values in percent are used in this system. The percentage values are determined by summing the milliequivalents of all the ions, dividing the milliequivalents of a given ion by the sum of the total milliequivalents, and multiplying by 100. Waters are classified by numerical values of the relationships of anions to the cations, where a , b , and d represent the percentage values of the alkali cations, alkaline earth cations, and strong acid anions, respectively. Any one of the following five conditions may exist: d may be equal t o or less than a , greater than a and less than a + b y equal to a + b y or greater than a + b . Using these conditions, waters are classified into five classes: Classl: d < a 2d = primary salinity 2(a - d ) = primary alkalinity 2b = secondary alkalinity Class 2: d = a 2u or 2d = primary salinity 2b = secondary alkalinity Class 3 : d > a ; d < (a + b ) 2a = primary salinity 2(d - a ) = secondary salinity 2(a + b - d ) = secondary alkalinity
CLASSIFICATION OF OILFIELD WATERS
256
Class 4: d = (a + b ) 2a = primary salinity 2b = secondary salinity Class 5: d > ( a + b ) 2u = primary salinity 2b = secondary salinity 2(d - a - b ) = tertiary salinity (acidity) These five classes of water are found in nature. Examples of the first three classes are various surface waters, sea water and brines represent class 4, while mine drainage waters and waters of volcanic origin fall in class 5 (Palmer, 1911). Rogers (1917, 1919) studied oilfield waters of the San Joaquin Valley, California, and used the classification system of Palmer (1911). He found that generally the surface waters of the San Joaquin Valley possess secondary salinity rather than primary alkalinity, contain more sulfate than chloride, and contain low amounts of bicarbonate. With increasing depth, the subsurface waters decrease in secondary salinity until primary alkalinity becomes evident. Waters above an oil zone often contained hydrogen sulfide, which was attributed t o reduction of sulfates by hydrocarbons, thus decreasing the amounts of sulfate and increasing the bicarbonate in the water, which Rogers called an altered water. Further he found that, in these altered waters in close proximity to hydrocarbon accumulations, chloride becomes relatively and absolutely important because of the residual chloride from the original (ancient) sea water chlorides as compared to waters above the oil zone which often are freshened because of a more hydrodynamic situation. Altered waters, according t o his definition, can have either primary alkalinity or secondary salinity depending upon their amounts of carbonate and chloride, but normal waters have only secondary salinity. Elliott (1953) used the Palmer system t o determine the chemical characteristics of some Paleozoic age formation waters in Texas. He found that all of the waters in the group that he studied (about 70) contained predominant, primary salinity. Many of these waters contained appreciable concentrations of sulfate; one contained 5,800 mg/l sulfate, and many contained more than 2,000 mg/l. The calcium concentration ranged up to 13,000 mg/l while the bicarbonate concentrations ranged up t o 800 mg/l. Ostroff (1967) used the Palmer classification to classify waters from several basins and t o compare this classification system with two other systems. He found that the Palmer system groups some of the constituents together that are not closely related chemically. Furthermore the system does not consider ionic concentrations or saturation conditions related t o sulfate or bicarbonate.
SULIN’S CLASSIFICATION
257
Sulin’s classification Sulin (1946), a Russian geochemist, proposed a classification system based upon various combinations of dissolved salts in the waters. The waters are described according to chemical type, subdivided into group, subgroup, and class. He found four basic environments of natural water distribution: (1) Continental (terrestrial) conditions which promote the formation of sulfate waters. Such conditions supply soluble sulfate constituents t o the water and the genetic type of such a water is “sulfate-sodium”. (2) Continental conditions which promote the formation of sodium bicarbonate waters. The genetic type is “bicarbonate-sodium”. ( 3 ) Marine conditions and the formation of a “chloride-magnesium ” type of water. (4) Deep subsurface conditions within the earth’s crust and the formation of a “chloride-calcium ” type of water. The first two types are characteristic of meteoric and/or artesian waters, the third of marine environments and evaporite sequences, and the fourth of deep stagnant conditions.
Types, groups, and subgroups Water composition is expressed in milligram-equivalents of the separable ions, and the composition is calculated per 100 g of water. The percent of the sum of the equivalents is used t o exclude the degree of water mineralization, and t o compare waters containing different amounts of dissolved solids. The ratio Na/Cl expressed in the percent equivalent form determines the genetic water type. If the value is greater than one, sodium predominates over chloride and the excess sodium can be combined with sulfate or bicarbonate. Therefore waters with a Na/Cl ratio greater than one belong to the bicarbonate-sodium or the sulfate-sodium types. Sulin calculated sodium as the sum of all the alkalies (Li, K, Na etc.) and chloride as the sum of all the halides (Cl, Br, I). The ratio (Na - C1)/S04, if greater than one, indicates that the water is the bicarbonate-sodium type, while if it is less than one it is the sulfatesodium type. Similarly the ratio (Cl- Na)/Na if less than one indicates the chloride-magnesium type, but if greater than one it indicates the chloridecalcium type.
Water classes Subdivision of the groups of waters were made by Sulin (1946) using the Palmer (1911) characteristics, because these characteristics express the dissolved constituents in the waters in a generalized format. For example, the sum of the alkali chlorides and sulfates corresponds t o primary salinity, and the sum of the alkaline earth chlorides and sulfates corresponds to secondary
258
CLASSIFICATION OF OILFIELD WATERS
salinity and the sodium bicarbonate-calcium stage. N o sodium bicarbonate is present in sulfate-sodium, chloride-magnesium, or chloride-calcium types of water; therefore, these types are classified as follows: (1)Class A2 : secondary alkalinity predominates (alkaline earth carbonates and bicarbonates). (2) Class S2 : secondary salinity predominates (alkaline earth sulfates and chlorides). (3) Class S, : primary salinity predominates (alkali sulfates and chlorides). (4) Class S,: tertiary salinity predominates (iron and aluminum sulfates and chlorides and free strong acids). Bicarbonate-sodium type waters contain sodium bicarbonate and are classified as follows: ( 5 ) Class A2 : secondary alkalinity predominates (alkaline earth carbonates and bicarbonates). (6) Class A , : primary alkalinity predominates (alkali carbonates and bicarbonates). (7) Class S, : primary salinity predominates (alkali chlorides and sulfates). (8) Class A, : tertiary alkalinity predominates (iron and aluminum carbonates and bicarbonates). The water classification is expressed by use of a formula representing decreasing values of the Palmer characteristics. For example, S, S2 A 2 indicates that primary salinity is predominant and is followed by secondary salinity and secondary alkalinity. Therefore, the classes are subdivided into subclasses, and class S1 can include the subclass S, S2 A 2 , S1 A2 S2, S1 S2, and S, . Table 8.1 outlines Sulin’s method of water characterization. Table 8.11 briefly outlines the relative values of the coefficients which determine the four genetic types of waters. The Palmer characteristics do not account for the interrelations between chloride and sulfate and between calcium and magnesium. Therefore, Sulin calculated the ratio SO4/C1 and Ca/Mg to establish additional subgroups. The complete water characterization included the following: (a) water formula given in Palmer characteristics; (b) coefficients in percent equivalents for S04/C1 and Ca/Mg; (c) sum of the milligram equivalents per 100 g of water (Z r ) t o illustrate the degree of water mineralization; and (d) the genetic coefficients (Na - C1)/ SO4 and (Ca - Na)/Mg t o determine the water type, and Na/C1 t o determine related genetic types of water.
Hydrochemical indicators of hydrocarbons Sulin (1946) noted that certain properties of subsurface waters were favorable indicators of hydrocarbon accumulations. The bicarbonate-sodium and chloride-calcium types of waters are widely found in oilfields. However, the chloride-calcium type is the more favorable indicator if it has the most characteristic composition plus certain minor or micro constituents. In general, he determined that hydrocarbon accumulations are most commonly
SULIN’S CLASSIFICATION
25 9
TABLE 8.1 Sulin’s method of water characterization _____
Na/Cl
>1
Sulfate-sodium type:
(Na+ - C1-)
so,,-2
Bicarbonate group class A2 calcium subgroup magnesium subgroup Sulfate group class S1 calcium subgroup magnesium subgroup sodium subgroup class Sp calcium subgroup magnesium subgroup Chloride group class S 1 calcium subgroup magnesium subgroup sodium subgroup Na/Cl
CLASSIFICATION OF OILFIELD WATERS
26 0
TABLE 8.11 Coefficients characterizing the genetic types of waters ~~
Type of water
Na+/Cl-
(Na’ - Cl-)/S04-2
(Cl--
Chloride-calcium Chloride-magnesium Bicarbonate-sodium Sulfate-sodium
1
l . As Sulin (1946) noted, if the ratio Na/Cl in epm is greater than 1, the water contains more sodium than chloride and the excess sodium can react with sulfate or bicarbonate ions. Therefore, such waters belong to the bicarbonate-sodium or sulfate-sodium types. If the ratio (Na - C1)/S04 is greater than 1,it indicates an excess of sodium with respect t o both chloride and sulfate. (2) Waters of the sulfate-sodium type with (Na - Cl)/S04 < 1. This ratio, if less than 1, indicates that all of the sodium will react with chloride or sulfate. (3) Waters of the chloride-magnesium type with (Cl- Na)/Mg < 1. A ratio of this type indicates that all of the chloride will react with sodium and magnesiun. Such a water is characteristic of the transition zone between a hydrodynamic area which is becoming more hydrostatic in the deeper part of the basin, and the amount of dissolved bromide increases directly with the (Cl- Na)/Mg ratio. (4) Waters of the chloride-calcium type with (Cl- Na)/Mg > 1. This ratio indicates an excess of chloride with respect t o sodium and magnesium, and the excess will react with calcium. This type of water occurs in deeper zones which are isolated from the influence of infiltration waters and are hydrostatic or almost hydrostatic. Bojarski observed a large variation in the chemical composition in the chloride-calcium type of water and subdivided this type as follows: (a) The first class, chloride-calcium I with Na/Cl > 0.85 characterizes an active hydrodynamic zone with considerable water movement. It is considered a zone of little prospect for the preservation of hydrocarbon deposits. (b) The second class, chloride-calcium I1 with Na/C1 = 0.85-0.75, characterizes the transition zone between an active hydrodynamic zone and a more stable hydrostatic zone of the sedimentation basin, which is generally considered a poor zone for hydrocarbon preservation. (c) The third class, chloride-calcium I11 with Na/Cl = 0.75-0.65 (0.60), characterizes favorable conditions for the preservation of hydrocarbon deposits.It is designated as a fairly favorable environment for the preservation of hydrocarbons. (d) The fourth class, chloride-calcium IV with Na/C1 = 0 . 6 5 4 . 5 0 , is characterized by complete isolation of the hydrocarbon accumulations as well as by the presence of residual waters. I t is considered a good zone for the preservation of hydrocarbons. (e) The fifth class, chloride-calcium V with Na/C1 < 0.50, is characterized by the presence of ancient residual sea water which has been highly altered since original deposition, both in the concentration of dissolved solids and in the ratios of the dissolved constituents. Bojarski considers a zone of this type to be one of the most likely areas where hydrocarbons are accumulated. Additional characteristics of water associated with hydrocarbon accumulations are as follows: (1) iodide > 1mg/l; (2)bromide > 300 mg/l (increasing
262
CLASSIFICATION OF OILFIELD WATERS
iodide and bromide concenbrations may point to a bitumen accumulation); (3) ratio Cl/Br < 350; and (4) SO4 x lOO/Cl< 1. In addition t o indicating the degree of alteration, bromide and iodide as biophile constituents play a decisive role in the classification Bojarski adopted. This followed because of the increased concentration of biophile elements in the waters accompanying a petroleum deposit. The concentration of iodide in the ground waters depends mainly on the organic substances, whereas the’concentration of bromide up t o a certain limit takes place in an inorganic medium, but an increase in bromide must be evaluated as a positive indication. In many waters accompanying petroleum deposits, large amounts of bromide and smaller amounts of iodide were detected, or vice versa. This probably is related t o the type of bituminous substances which absorb the individual biophile elements in different amounts. Chebotarev’s classification Chebotarev (1955),an Australian geochemist, classifies waters on the basis of dissolved bicarbonate, sulfate, and chloride, and he does not consider the acid waters or those that contain free sulfuric or hydrochloric acid. His fundamental assumption is that the anions are independent variables while the cations are dependent. The geochemical types of waters are related t o the products of weathering. Table 8.111 illustrates the cycles and products that are produced by weathering. During the first cycle the igneous rocks are weathered allowing chloride, sulfate, calcium, sodium, silica, and magnesium to go into solution. The second cycle is the weathering of sedimentary rocks with the solution of more of the same products. The third cycle is the weathering of recent drift and yields of the above constituents plus aluminum and iron. Table 8.IV illustrates Chebotarev’s (1955) geochemical classification of subsurface waters. The phase of weathering corresponds t o four phases of the solution and redistribution of the chemical constituents in the earth’s crust and correlates with their relative mobilities. He plotted the relative mobilities of nine chemical constituents using the mobility of chloride as 100%. From this four phases were obtained, namely: (1)chloride and sulfate 100% t o about 58% mobility; (2) calcium, sodium, magnesium and potassium 3% t o about 1.2% mobility; (3) silica about 0.20% mobility; and (4) iron oxide and aluminum oxide, less than 0.05% mobility. The four phases of weathering correspond to the products of weathering shown in Table 8.IV and also to the cycles and products of weathering in Table 8.111. For example the fourth phase in Table 8.IV corresponds with the first cycle in Table 8.111. The genetic types of water shown in the upper portion (A) of Table 8.IV do not correspond directly with the weathering phases since the genetic types overlap the phases. These genetic types are related t o the accumulation products shown in Table 8.111. In the lower portion (B) of Table 8.IV are the
TABLE 8.111 Cycles and types of products of weathering (after Chebotarev, 1955) Cycles of weathering:
b rn
First cycle (orthoeluvium) from igneous and highly metamorphosed rocks
Second cycle (paraeluvium) from sedimentary rocks
Third cycle (Neoluvium) from Recent Drift
residual products
accumulative products
residual products
accumulative products
residual products
(1) chloride-sulphate (chiefly alluvial) (2) calcareous (chiefly colluvial and proluvial) (3)siallitic
(1) detrital
(1) chloride-sulphate
Types of (1) coarse detrital products of weathering (2) calcareous (under vegetation cover) (3)siallitic
(2) siallitic
(supra-calcareous (3) allitic (?)
accumulative products
(1) solonez and (1) chloride-sulphate gypsum bearing (2) calcareous (2) leached supra(2) CaC03 chloride-sulphate calcareous (siallitic) (3) unsaturated siallitic*2 (3) unsaturated siallitic (3) alumino- and ferricalcareous chloridesiallitic*’ siliceous system sulphate
(4) a l l i t i ~ * ~
(lateritic crust of weathering)
*’ A large quantity of silica and much of its calcium and sodium compounds are removed; the aluminosilicates pass gradually into residual aluminosilicic acids, *’ *3
i.e., acids of the kaolin type. The action of carbonated atmospheric water is insufficient to replace the absorbed ions by hydrogen. The accumulation of sesquioxides at the expense of the leaching out of the alkalis, alkaline earth, and silica.
zrrl
is b =iJ
0
z
TABLE 8.IV Geochemical classification of subsurface waters (after Chebotarev, 1955) ( A ) Relationship of the products of weathering to the genetic types of water Presumable phase of weathering
Products of weathering
Genetic types of water
Fourth phase
residual (orthoeluvium and detrital paraeluvium)
bicarbonate (alkaline)
Third and partly second phases
siallitic drift
bicarbonate-chloride (alkaline-saline) chloride-bicarbonate (saline-alkaline)
Second and partly first phases
calcareous accumulation
First phase
chloride-sulphate accumulation
chloride-sulfate (saline) chloride (saline)
(B) Geochemical groups of Waters Major group of water
Class
Genetic types of water
Reacting value in percent
H C 0 3 - + COB- ClBicarbonate
Sulphate
Chloride
SO4-’
Cl-
+
SO4-’
H C 0 , - + CI-
HC03-
I
bicarbonate
>40
40
111
chloride-bicarbonate
30-15
>20
-
-
IV
chloride-sulphate
15-
>20
-
40
-
SO4 > HC03 was found to occur in very high chloride waters and in sea waters, especially when they are saturated with CaS04. If the waters are not saturated in CaS04, the sequence Cl- > HC0,- > S04-2 is predominant, and in low-chloride waters the predominant sequence is H C O ~ -> C T > S04-2.
26 8
CLASSIFICATION O F OILFIELD WATERS
TABLE 8.VI Schoeller’s scheme for classifying petroleum reservoir waters*
Chloride concentration as Ci-
>
Very high if 700 Marine if 420 - 700 High if 140 - 420 Average if 40 - 140 Low if 10 - 40 N o r m a l i f < 10
Sulfate concentration as s04-’
> -
Very high if 58 High if 24 - 58 Average if 6 24 Normalif< 6 Near saturation when J(S04-’)
(Ca+’)
> 70
Bicarbonate plus carbonate concentmtion as H C 0 3 -
+
c03-’
>
Na+ then IBE = (Cl- - Na+)/Cl> Cl- then IBE = ( C T - Na+)/(S04-’
If Na+
Importance o f anions and cations
c1- > SO^' > co3-’ co3-’ > c1-> so4-2 co3-’ > so4-’ > C T
c1- > .co~-‘> so4-’
Na+ Na+
*
> Mg” > Ca+’ > Ca+ ’ > Mg+’
All constituents are calculated in epm.
+ HCO3- + C03-2)
SCHOELLER’S SYSTEM
26 9
Also, in very high chloride waters only the sequence Na+ > Ca+’ > Mg+2 is found. As the C1- decreases, the sequence Na+ > Mg+’ > Ca+’ becomes more frequent. In very high chloride waters SO4-’ > Ca+’, but in less concentrated waters the opposite may occur. The sequence HC03-< Ca+’ always is found in very high chloride waters. In less concentrated chloride waters, either HCO,- < Ca+’ or HC03- > Ca+’ may be found, while in low chloride waters HC03- > Ca+’ is predominant. x Ca+’ to indicate Schoeller used an arbitrary value of 70 for JSO4that a water is saturated with CaS04. (This is not necessarily true because some waters, depending upon their other dissolved constituents, can contain smaller or larger amounts.) He divided waters into four additional types depending upon their amounts of sulfate. Saturation with CaS04 was found to occur only in very high chloride waters. The calcium concentration always is very high - ranging from 150 to 1,100 epm - in high chloride waters which have SO4-’ > 58 epm and usually is less than 150 epm in high chloride waters where SO4-’ < 58 epm. All petroleum waters even if saturated in CaS04 have a low S04/C1 ratio which is attributed t o reduction of sulfates and high concentrations of chloride. The ratio never exceeds one except in low or normal chloride waters. The third subgroup contains three additional types depending upon the amounts of bicarbonate and carbonate in the waters. The preferred formula for this calculation is Y(HCO,- + C03-’ )’ (Ca+’ ) which is proportional to the gaseous pressure of C 0 2 in equilibrium with CaCO, in the water. As the Cl- increases, the tendency is for Ca+’ t o increase and HC03- to decrease; however, because the Ca+’ increases, the product of y(HCO,- + C03-’ )’ (Ca+’ ) does not vary greatly. As the waters move in their subsurface environment their dissolved ions ’have a tendency to exchange with those in the rocks. Two extreme types of adsorption can be noted in addition to intermediate types of adsorption. The extreme types are a physical adsorption or the Van der Waals adsorption with weak bonding between the adsorbent and the constituent adsorbed and a chemical adsorption with strong valence bonds. Both of these adsorptions can act simultaneously. Cations can be fixed at the surface and in the interior of the associated minerals. These fixed cations can exchange with the cations in the water. When the exchange occurs, there is an exchange of bases. With the right physical conditions of the adsorbent, similar exchange can occur with the anions. Some of the formation constituents that are capable of exchange and adsorption are argillaceous minerals, zeolites, ferric hydroxide, and certain organic compounds. Particle size influences rates and capacities, if the solids are clays such as illite and kaolinite. The.rate increases with decreasing particle size. However, if a larger mineral has a lattice, the exchange can easily occur on the plates.
CLASSIFICATION OF OILFIELD WATERS
27 0
The concentration of exchangeable ions in the adsorbent and in the water is important. More exchange will occur when the solution is highly concentrated. Schoeller (1955) used the formula: 1
(a-x)=K
( a E x ) IP
to indicate the relationship that exists between the initial concentration, a , of the cations in milliequivalents in the unreacted water, and x , which equals the final concentration of the cations in milliequivalents in the water after equilibrium or reaction with the rocks. The amounts of cations exchanged by passing from the liquid t o the rock or clay is a - x and the index of base exchange (IBE) = (a - x ) / a . By substitution:
The IBE is used to indicate the ratio between the exchanged ions and the same ions as they originally existed. For example, assume that in the original ' ) , and that when water there were as many equivalents of C1- as (Na+ + K the Na+ and K+ of the water exchanged with the alkaline earths in the rocks alkaline exchange occurred, then: IBE =
C1-
- (Na+ + K+) c1-
and this value is positive if the equivalents are Cl- > (Na' + K ' ) . Theoretically all the halides should be included as C1- and all the alkalies as Na+ or (Na+ + K ' ) . However, when the alkaline earth ions in the water exchange for alkali metal ions on the rocks then: Cl- - (Na+ + K + ) IBE = SO4- + HC03- + NO3and this value is negative if the equivalents are Cl- < (Na+ + K+).The lack of equilibrium between the halides and the alkalies is not always a characteristic of base exchange because sea water has a positive value without the occurrence of base exchange. Negative values usually are observed for water coming from altered crystalline rocks. Waters with an IBE equal t o or greater than 0.129 can be true connate petroleum reservoir waters. Waters with a negative IBE are waters of meteoric origin that have infiltrated into marine sediments. Comparison of petroleum-reservoir waters with other types of subsurface waters revealed that the other waters have most of the same characteristics a much higher SO4- concentration and a lower or Kr. Waters that are in contact with organic matter
SCHOELLER’S SYSTEM
271
(other than petroleum), such as bitumens, lignites, and coals, resemble petroleum reservoir water, but the frequency of a Kr above normal is greater in petroleum associated waters. Waters related - to magmatic reactions commonly possess high concentrations of HC03 . Schoeller’s (1955) study of petroleum reservoir waters indicated that a positive IBE is more frequent as the C1- increases. A negative IBE is more frequent as the C1- decreases, and a negative value is predominant in low and normal chloride waters associated with petroleum. In fact, this characteristic appears specific for petroleum reservoir waters since in other subsurface waters a positive index occurs as frequently as a negative index. Ancient sea water (connate water) deposited with the sediments usually has an IBE > 0.129 and a Cl/Na > 1.17. Intruding meteoric water in sedimentary marine rock has an IBE < 0.129 and Cl/Na < 1.17. Petroleumreservoir waters with an IBE greater than sea water 0.129 also have the characteristics Cl/Na > 1.17, Cl/Ca < 26.8, Cl/Mg > 5.13, Mg/Ca < 5.24; a very high value for $(HCO,-)’ (Ca+*) indicating sulfate reduction; low concentrations of HC03-; and frequent high concentrations of NH4 +. Petroleum-reservoir waters containing infiltrating meteoric water mixed with ancient sea water have an IBE less than sea water, 0.129, and the characteristics Cl/Na < 1.17, the ratio Mg/Ca increases and approaches but never equals 5.24, and the ratios Na/Ca and Na/Mg decrease as the dissolved solids increase.
Gases in petroleum-reservoir waters Schoeller (1955) noted that there should be equilibrium between the free petroleum gases and those dissolved in the water. Considering the solubility of the gases, those in the water should reflect the composition of the petroleum. Components characteristic of petroleum accumulations are ethane, propane, butane, pentane, ethylene, and propylene. Associated components, which may also be present in volcanic waters and in waters in contact with other organic matter such as coal, peat, and lignite as well as in petroleumassociated waters, are favorable components. These are methane, carbon dioxide, organic nitrogen, hydrogen sulfide, helium, radon, and the absence of oxygen. Other components or universal components found in all types of waters are nitrogen and argon. The top waters can contain gases such as Hz S, C 0 2 , and CH4 but because they do not contact the petroleum deposit they are not similar in composition t o the bottom waters. The edge waters are in contact with the petroleum and are characterized by higher amounts of HC03, sulfate reduction, and the presence of H2S , NH4, and small amounts of dissolved hydrocarbons.
27 2
CLASSIFICATION OF OILFIELD WATERS
Oilfield brine analyses About 4,000 oilfield brine analyses were classified. All of these analyses are now in the U.S. Bureau of Mines (1965) open-file report on oilfield brines. The data are on automatic data processing magnetic tape as well as in a file of computer printout sheets listed by State, county, sedimentary basin, formation, etc. These brine data were collected by the Bureau of Mines because the value of oilfield brine analyses in the study of various petroleum-related problems was recognized early in the history of petroleum and natural gas. Before 1928 the Bureau of Mines had indicated in several reports (Ambrose, 1921; Swigart and Schwarzenbek, 1921; Collom, 1922; Mills, 1925; Reistle, 1927) ways in which the analyses could be used. In earlier years, confusion existed because of greatly varying methods of analysis that were used. A paper presenting the methods used by the Bureau of Mines was published by Reistle and Lane (1928). This system of determining the characteristic constituents of oilfield waters and of calculating and reporting results was widely adopted by the petroleum industry. Later the Bureau of Mines cooperated with several interested agencies, and a more detailed report with more modern methods of analyzing oilfield waters was published (American Petroleum Institute, 1968). The Bureau of Mines has an oilfield water analysis laboratory at the Bartlesville Energy Research Center, Bartlesville, Oklahoma. These 4,000 samples were analyzed at this laboratory. Analysis methods The specific gravity of each sample is determined so that a correct aliquot size can be taken for a specific ion analysis. Chloride is determined by titration with silver nitrate, carbonate and bicarbonate are determined by titration with a standard acid, and a pH meter is used to determine the end points. This alkalinity determination should be completed at the time of sampling for accur?te results. However, most of the data for the 4,000 samples were obtained by analysis in the laboratory and were completed within 6 t o more than 48 hours after sampling. Therefore, the alkalinity data cannot be considered absolute but only relative. The calcium was determined by titration of calcium oxalate with permanganate until about 1957, about which time it was determined by complexometric titration such as with disodium ethylenediametetraacetic acid (EDTA) until about 1969; since then, it has been determined by atomic absorption. Magnesium determination has a similar history. I t was precipitated as the pyrophosphate until about 1957, and titrated with EDTA until 1969, from then t o now it has been determined by atomic absorption. Sulfate was determined by precipitation as barium sulfate, and this method still is used. Sodium was determined by calculation from the difference
OILFIELD BRINE ANALYSES
273
between the reacting values of the assumed total anions and cations until about 1960, after which time it was determined by flame photometry or atomic absorption. Several other analyses for dissolved constituents in oilfield brines are now made by the Bureau of Mines. For example, potassium, lithium, rubidium, and cesium are determined by atomic absorption or flame photometry; strontium and barium by atomic absorption; manganese, iron, and boron by atomic absorption or titrimetric methods; and bromide and iodide by titrimetric methods. The precision of the methods is as follows: alkalinity, 2-3% of the amount present; sodium, 2--5% of the amount present; calcium and magnesium, 4--5% of the amount present; sulfate, 1-2% of the amount present; chloride, 1% of the amount present. If sodium is calculated, the precision value reflects the sum of the precision data for the data from which it is calculated plus the undetermined dissolved constituents. The significant figures for the analytical data are all the certain digits and only the first doubtful digit. This number usually is limited to three significant figures except for specific gravity, where four or five are common. It often is recognized that the sampling method is as important as the analytical method. This certainly is true of oilfield brines. Field sampling methods Most of the 4,000 samples were obtained only from wells where reasonable assurance was evident that the formation brine was not contaminated by drilling fluids or by intrusion of water from other formations. Wells were selected on the basis of age, type of completion, and production of fluids. Samples were not taken from some gasfields because of the likelihood of dilution by water condensed from vapor carried up the hole with the gas. Some samples were taken from gas-condensate wells that produced large volumes of brine. In many cases the electrical resistivity measurement was made on the sample at the time the sample was taken, and resulting values were compared with measurements from other samples from the same formation within that field or nearby fields. Obvious discrepancies were eliminated by sampling additional wells. Nearly all samples were withdrawn at production wellheads, and the water was separated from the oil in portable separators. A few samples were taken of brines from formations that did not produce enough water t o permit taking samples at individual wellheads. Such samples were taken from gunbarrels or oil-water separators. Samples were taken in clean, 1-gallon glass jugs that were first rinsed several times with the water sampled and then filled, capped, and labeled. In a few instances samples were obtained by the producer from comparatively isolated small pools or fields and shipped t o the laboratory in 1-gallon polyethylene jugs.
274
CLASSIFICATION OF OILFIELD WATERS
Application of the classification systems Several investigators have applied the classification systems t o determine their usefulness in studies related t o exploration and production of petroleum. Ostroff (1967) concluded that Palmer’s system is less useful than the systems of Sulin and Schoeller. Further he concluded that the Sulin system is more applicable to petroleum formation waters because many such waters contain more than 2,000 epm of dissolved solids, and the Schoeller system tends to lump these highly concentrated waters together. The index of base exchange (IBE) in Schoeller’s system, however, appears to have merit for certain interpretations as does theJ Ca+* x SO4-*. Dickey (1966) concluded that the Palmer system does not correlate very well with the geology of oil reservoirs, that Schoeller’s nomograph is useful, but that the Sulin system appears t o conform better with geology. He also noted that relating water types to geology and flow patterns should be useful in exploration and that distinguishing between a stagnant water and artesian related waters should be highly significant in a new oilfield development area. The Sulin system appears t o be more generally applicable than the other systems in studies of waters from petroleum reservoirs. Because of this and because the Schoeller system appears t o have merit in studying certain types of oilfield waters, it was decided t o apply portions of the Sulin system, the Schoeller system, and Bojarski’s modification of the Sulin system t o a study of brines from various sedimentary basins of the United States that are known to be related to petroleum and natural gas.
Calculations The analyses of most oilfield waters are reported in units of milligrams per liter (mg/l). The conversion of mg/l to equivalents per million (epm) is done using the following formula: mg/l of ion = epm ion atomic weight of ion specific gravity of brine x valence of ion Table 8.VII provides formulas for calculating the epm for many of the common constituents found in oilfield brines. If the constituent is reported in parts per million (ppm), it is not necessary t o divide by the specific gravity of the brine. The sum of the epm’s (Z epm) shown in Table 8.VII are converted to Zr/100 g of water for the Sulin calculations by moving the decimal to the left in the Z epm and calling this Zr. The term s is used t o indicate the percent of equivalent of a given constituent. The percentage equivalent (s) of each ion is determined by dividing the equivalents per 100 g of water by the total equivalent in 100 g of water. For example, if the r for sodium equals
APPLICATION OF THE CLASSIFICATION SYSTEMS
27 5
TABLE 8.VII Formulas for converting milligrams per liter to equivalents per million for constituents commonly found in oilfield waters Lithium Potassium Sodium Magnesium Calcium Strontium Barium Carbonate Bicarbonate Sulfate Chloride Bromide Iodide
mg/l Li+ mg/l K+ mg/l Na+ mg/l Mg+’ mg/l Ca+’ mg/l Sr+’ mg/l Ba+’ mg/l co3-’ mg/l HC03mg/l SO^+ mg/l C1mg/l Brmg/l I-
x x x x x x x x
x x x x x
0.1442/sp. gr.* = O.O256/sp. gr. = O.O435/sp. gr. = O.O823/sp. gr. = O.O499/sp. gr. = O.O228/sp. gr. = O.O146/sp. gr. = O.O333/sp. gr. = O.O164/sp. gr. = 0.0208/sp.gr. = O.O282/sp. gr. = O.O125/sp. gr. = O.O079/sp. gr. = Zepm
epm Li+ epm K+ epm Na+ epm Mg+’ epm Ca+’ epm Sr+’ epm Ba+’ epm c03-’ epm HC03eprnS04-’ epm C1epm Brepm I-
=
* Specific gravity.
200.6 and the Zr for the total equivalents equals 518.8, then 200.6/518.8 x 100 = 38.7, or 38.7 percentage equivalents for sodium. The Sulin classification considers only the macro constituents; if ions such as potassium, lithium, strontium, barium, bromide, and iodide are determined, they should be added t o their associated macro constituents t o be properly considered in the analysis report. For example, when sodium, potassium, and lithium are determined by atomic absorption the total mg/l of each is reported. Therefore, the epm Na + epm K + epm Li should be added to obtain the correct r value for sodium: rNa= rCa= rC1 =
epm Na epmCa epm C1
+
+
+
epm K 10
+
epm Li 10
e p m S r + epmBa 10 10 epm Br 10
epm I 10
The r values then are divided by the Zr and multipIied by 100 to obtain the s value or percentage equivalents. The s values are used to determine the type, class, and other Sulin values of the water as illustrated in Table 8.1, where a = sNa, b = sCa + sMg, and d = sC1+ s S 0 , . The epm values are used to determine the Schoeller characteristics such as the degree of chloridization, the degree of sulfation, IBE, etc., illustrated in Table 8.VI.
27 6
CLASSIFICATION O F OILFIELD WATERS
TABLE 8.VIII Classification of some oilfield waters from 1 0 formations in eight sedimentary basins No.
State
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Ark. Ark. Ark. Texas Texas Texas Texas Texas Texas Texas
31
32 33 34 35 36 37 38 39 40
Formation Basin
Depth
Arbuckle Arbuckle Arbuckle Arbuckle Arbuckle Arbuckle Arbuckle Arb u ck le Arbuckle Arbuckle Lansing Lansing Lansing Lansing Lansing Lansing Lansing Lansing Lansing Lansing Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox
Cent .Kans. Cent .Kans. Cent.Kans. Cent.Kans. Cent.Kans. Cant.Kans. Cent.Kans. Cent.Kans. Cent .Kans. Cent .Kans. Cent. Kans. Cent.Kans. Cent.Kans. Cent .Kans. Cent. Kans. Cent. Kans. Cent .Kans. Cent.Kans. Cent.Kans. Cent .Kans. Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee
1,050 1,091 1,023 1,102 992 1,152 1,174 949 1,195 1,104 928 1,075 966 999 902 1,009 1,063 1,148 1,172 853 1,228 1,731 904 1,539 1,106 1,020 582 1,432 1,97 2 1,865
Nacatoch
E.Texas
360
233.4
9.0
22.4
Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch
E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas
465 373 905 70 1 242 650 283 191 181
463.4 300.4 788.2 481.8 274.0 492.7 295.4 451.5 490.6
39.5 25.9 9.4 7.8 8.1 10.4 8.1 17.6 3.6
59.5 36.8 28.7 17.3 14.7 16.5 14.7 44.7 7.8
>
Concentration (e pm)
634.6 581.6 282.0 639.9 430.3 458.7 473.2 356.4 446.4 577.4 1,759.6 570.4 1,899.9 1,728.2 1,903.8 2,514.8 1,854.7 2,087.2 2,414.5 1,898.4 2,366.5 2,188.4 2,161.5 2,365.3 1,997.1 1,990.0 1,587.2 2,459.0 2,701.4 2,635.5
* Chloride (epm): (VH)C = 700, (MC) = 420-700, (L)C = 10-40, (N)C = 10. Sulfate (epm): (VH) = =
1,115 737 1,297 884 1,417 2,340 2,174 1,943 1,512 1,943 1,897 1,241 1,033 711 2,844 2,519 2,722 2,115 2,289 3,062 1,047 1,750 898 841 1,809 1,144 925 1,442 1,596 1,332 1,061 972 1,060 1,444 2,263 1,312 2,443 1,690 1,315 2,469
Mg+’
1,246.8 90.6 586.0 32.2 1,310.3 107.1 934.9 64.8 1,507.2 138.6 1,642.1 22.5 1,515.1
51.1
1,495.4 89.0 205.3 6.4 1,522.8 90.2 1,971.8 157.2 1,943.8 185.1 1,060.3 108.2 538.6 35.9 1,772.1 127.3 1,861.9 217.4 2,068.5 148.5 1,950.3 139.9 2,020.0 141.2 1,877.5 92.7 1,507.0 44.3 1,263.0 33.4 341.8 4.4 1,060.5 32.4 1,271.9 59.6 809.1 12.7 478.8 12.4 1,451.3 26.7 1,447.3 38.3 1,268.0 30.2 1,124.4 45.0 1,102.9 25.6 1,186.7 59.6 1,733.6 82.7 2,009.8 75.9 1,572.7 88.9 1,721.3 256.7 1,925.9 61.6 1,579.6 60.6 2,153.7 87.7
Ca+’
254.1 86.7 278.5 197.6 504.4 378.5 205.9 448.9 10.0 448.1 669.7 563.6 289.8 75.0 878.8 1,084.7 900.9 726.1 741.8 610.0 161.4 56.3 9.3 58.9 154.9 61.1 23.6 213.3 144.1 81.9 164.7 160.5 164.3 287.3 306.8 224.9 458.0 359.9 252.9 518.9
HCOJ- S04-*
Cl-
0.4 2.8 2.7 1.5 0.5 0.0 4.7 2.7 14.9 1.8 0.0 0.9 0.0 3.0 1.5 1.1 0.6 1.8 0.8 1.0 4.2 8.2 11.3 0.0 1.0 7.3 7.0 3.9 1.6 8.3 2.7 2.0 2.9 0.9 5.2 0.9 0.0 3.9 2.3 0.0
1,594.4 699.0 1,692.0 1,194.9 2,208.7 2,033.8 1,763.1 2,022.2 202.4 2,050.3 2,789.9 2,679.0 1,436.5 643.4 2,773.2 3,165.1 3,111.6 2,802.4 2,909.8 2,576.9 1,705.0 1,342.2 343.5 1,161.4 1,478.2 873.3 507.7 1,685.1 1,629.4 1,367.4 1,330.1 1,288.9 1,411.8 2,108.5 2,358.5 1,890.2 2,433.4 2,352.4 1,893.5 2,765.6
0.0 3.3 .7 1.1 3.3 9.3 3.7 8.2 6.0 8.3 8.0 11.7 21.4 2.9 2.8 2.7 4.8 8.5 4.2 5.4 3.5 1.8 0.6 0.2 4.9 3.7 0.1 0.0 0.1 4.0 0.0 0.0 0.0 0.0 33.6 0.0 0.0 0.0 0.0 1.0
* Chloride (epm): (VH)C 700, (MC) = 420-700, (H)C = 140-420, (A)C = 40-140, (L)C = 10-40, (N)C = 10. Sulfate (epm): (VH) = > 58,(H) = 24-58, (A) = 6-24, (N) - < 6.
300
-
I I
Q-
---- 1------II
200
100
0
0
6
(3 I
I
I
I
I
7
8
9
10
I1
PH
Fig.lO.5. Approximate pH and Eh of waters from unfiltered petroleum producing wells (A); filtered petroleum producing wells ( B ) ;and surface rivers, lakes, and seas (C). (After Vdovykin, 1963.
Fluid mechanics Roach (1965) describes how to apply fluid mechanics t o petroleum exploration. The definition of fluid mechanics as he used it encompasses complete study of subsurface fluids including physical and chemical characteristics, whether hydrodynamic or hydrostatic, and a complete study of the characteristics of the reservoir rock.
Maps Chloride-ion concentrations in water produced from rocks of various ages and depths were mapped in Lea County, New Mexico, using machine mapplotting techniques and trend analyses. Anomalously low chloride concentrations (1,000-3,000 mg/l) were found along the western margin of the Central Basin Platform in the San Andres and Capitan Limestone formations of Permian age. These low chloride-ion concentrations may be caused by preferential circulation of ground water through the more porous and permeable rocks (Hiss et al., 1969). Hanshaw and Hill (1969) studied aquifer systems from: (1) Mississippian age rocks; (2) Pinkerton Trail Limestone; (3) Paradox member of the Her-
HYDROGEOCHEMICAL RESEARCH AND METHODS
321
mosa formLtion; (4)Honaker Trail formation; and (5) Permian age rocks. Recharge in the Paradox Basin occurs on the west flank of the San Juan Mountains and along the west side of the Uncompahgre Uplift. A series of potentiometric surface maps were prepared for the five systems studied. With a few exceptions, most wells in formations above the Pennsylvanian age strata contain fresh to moderately saline water. Much of the strata below the Permian age rocks contained waters with dissolved solids concentrations greater than 35,000 mg/l and some areas favorable for hydrocarbon accumulations. Some of the brines in the Paradox formation contained up to 400,000 mg/l of dissolved solids. Cambrian age strata in much of Colorado is favorable for the accumulation of hydrocarbons. Chemical analyses of water from five Cretaceous aquifers were used to compute ion ratios, which were used in conjunction with structural and stratigraphic information t o interpret hydrologic conditions in the East Texas Basin. Ion ratio comparisons made by maps and diagrams show that the aquifers contain water of distinctive character, and that there are interconnections between aquifers, especially near the Mexia-Talco Fault zone and the Sabine Uplift. A hypothesis is offered that water moves along an unconformity from the Sabine Uplift eastward toward the East Texas oilfield where it enters the Woodbine Sandstone. Ion-ratio maps show the effect of time and of rock composition upon the relative kind and amount of dissolved solids in the water because- of reactions with minerals and organic material in the rocks. The hydrodynamic component of the water in the Woodbine formation from east to west helped form and contain the giant East Texas oilfield (Parker, 1969). Karim et al. (1966) studied three exploratory wells drilled on the east plunge of the Cordillera Isabella, Nicaragua, and all three had gas shows. A stratigraphic cross section and a localized map showing the relationship of magnetic highs obtained from an aeromagnetic survey and results of fluoroanalysis and water-gas surveys are included. The prospect of finding petroleum in coastal northeastern Nicaragua appears fair. Water analysis integrated with the existing knowledge of the geologic framework of an area provides supplementary information t o assist the exploration geologist in solving geologic problems on both a local and a regional scale. Isoconcentration maps showing regional variations in the total solids content of the waters within a given stratigraphic unit are important. Inorganic water analyses data are useful in the correlation of porous zones, and benzene analysis is a promising hydrocarbon exploration tool (Noad, 1966). Maps were prepared delineating: (1) surface outcrop areas of Upper Cretaceous rocks; (2) axial lines of Upper Cretaceous anticlinal structures; (3) zones of water types; (4)zones of water groups; and (5) zones of mixed waters. A diagram of chemical composition of the waters was also prepared. Using these maps, Galin and Plyushchenko (1963) selected areas in Dagestan that are favorable for the accumulation of oil and gas.
322
EXPLORATION FOR PETROLEUM AND GAS
Reviews Gerard and Feug&re (1969) concluded that geochemical exploration techniques are useful in offshore areas. Kroepelin (1967)reviewed geochemical prospecting as applied to petroleum and found that several companies had success ratios up to 59% with it in exploration. Johnson (1970) states that success ratios of about 35% can be attributed to the use of an inorganic technique of prospecting for oil and gas. The method is useful in locating stratigraphic trap accumulations and is based on the postulate that heavy-metal salts concentrate in the soil profile as a result of vertical migration of waters above an accumulation of oil. According to Boyle and Garrett (1970),“geochemical prospecting will play an ever increasing role in the discovery of hidden ore deposits and accumulations of hydrocarbons. The methods need no longer be sold since it has now been generally recognized that they are the only direct approach to the problems of mineral, oil, and natural gas exploration. The methods are direct and have generally proved most successful when applied in conjunction with geological and geophysical exploration techniques.” Geochemical methods applied t o petroleum prospecting have not reached their potential, expecially in the United States. Karaskiewicz (1966)discussed the geomicrobiological and hydrochemical research done by the Polish Petroleum Institute in 1962-63 in the LubelNadbuzan region. Specific methods such as the determination of gas dissolved in water were applied in this area, and 215 water samples were studied from the 9,500-km2 area. Anomalies produced by the bacterial activities which oxidize methane, propane, and butane were determined. Where methane was present, the bacterial microflora activity was lowest; where the biological activity was high, methane was absent. Important aspects of geochemical prospecting are : (1)analytical methods; (2)transport mechanisms of the hydrocarbons; (3)anomalies associated with hydrocarbon accumulations; (4) statistical treatment of the data; and (5) the final result. Significant findings are mapped and interpreted to locate a target area for drilling (Kroepelin, 1967). Case study of the Delaware sand (Bell Canyon formation), Texas, by Visher
(1961) Preliminary work was carried out on the relation of sapropelic material found in both dark shales and laminated silts, and the oil occurring in reservoir rocks. The study indicated that the frequency distribution of the radicals, - benzene, straight chain, CH3, and CH2 -, was identical in both the black shales and the residual crude present in oil saturated reservoir rocks. The entire sequence of Permian rocks from the Bone Springs Limestone through the Lamar Limestone member of the Bell Canyon formation is composed of dark organic sediments, deposited in a reducing environment.
CASE STUDY OF THE DELAWARE SAND
3 23
The total quantity of the sapropelic material available for the formation of oil was not determined, but 1%would be a conservative estimate, indicating that source material in the Delaware Basin is of sufficient quantity t o produce volumes of oil greater than presently discovered. The oilfields which have been discovered in the basin primarily are confined t o the upper porous and permeable sands of the Bell Canyon formation. The present distribution of oil appears t o be controlled by hydrodynamic conditions. Therefore, the tracing of the times and paths of migration is dependent upon reconstructing the paleohydrodynamics.
Potentiometric surface of the upper Delaware sand A potentiometric surface map of the Bell Canyon formation was made from a two-dimensional electric analog model of the central and eastern portions of the Delaware Basin in West Texas (Fig. 10.6). The majority of pressures were bottom-hole measurements from existing fields. Only a few shut-in pressures from drill-stem tests were usable because of the short shutin time commonly used in this area, and consequently, few tests reached true formation pressures. A pressure buildup method should have been used. The total dissolved solids of the formation water range from a low of 90,000 mg/l in the Ford field to over 250,000 mg/l at the South Pyote field. All pressure readings were corrected for effects of varying total dissolved solids. The assumption was made that the concentrations of total dissolved solids are stratified with little mixing. Therefore, the weighted mean is between 50,000 and 120,000 mg/l. The potentiometric surface has a hydrodynamic gradient from west to east with a component of northward flow. In the southeastern portion of the mapped area, the hydrodynamic gradient is reversed because of the influence of the eastern flank of the basin. Stratigraphic traps are formed in areas where linear sand fingers show an updip decrease in permeability and porosity. The change in permeability and porosity between the permeability barrier and adjacent reservoir rocks, however, is not great. The Saber field, for example, has an average porosity of 25% and permeability of 70 md, and the barrier rock an average porosity of 12% and permeability of 3 md. Since in some areas this barrier rock would be considered a possible reservoir, something in addition to these changes in porosity and permeability is necessary to prevent the movement of oil into the barriers. Under equilibrium conditions water flowing through a formation will have a greater pressure gradient across a tight zone than a more permeable zone (see Fig.lO.7). Therefore, the differential pressure in the barrier is greater than in the reservoir, and varies directly with the decrease in permeability between the two. When the oil phase reaches the barrier zone, the pressure gradient increases updip, making i t increasingly more difficult for the oil to
324
I
I
-z-
EXPLORATION FOR PETROLEUM AND GAS
CASE STUDY OF THE DELAWARE SAND
325
246 k g A q cm
Fig.10.7. Relationships of hydrodynamic gradients to permeability; I = decreased hydrodynamic gradient because of increased permeability; 2 = increased hydrodynamic gradient because of decreased permeability; 3= average hydrodynamic gradient; 4 = decreased hydrodynamic gradient.
invade the water-filled rock pores. Finally, the entry pressure of the barrier rock is greater than the invading force of the oil and migration ceases. The downdip hydrodynamic flow increases the efficiency of the trap by reducing the buoyancy effect of the oil. The hydrodynamic enforcement of stratigraphic traps can increase the oil column many times over what it would be under hydrostatic conditions and probably accounts for the development of commercial stratigraphic oil accumulations in the Delaware Basin.
Formation waters The initial approach in the study of stratigraphic problems within the Bell Canyon formation was by the use of formation waters. Over 300 samples of formation water were collected, analyzed, and processed by computer techniques. Data collected in this manner were posted on maps (Fig.10.8-10) and contoured. Several aspects of the waters (relative concentration percentages of SO4,Mg, Ca, and total solids) show systematic variations over the basin. Variation in these parameters is related t o proximity to outcrop and the degree of transmissibility of the formation. The highest sulfate content is near the outcrop belt t o the west; the calcium and total dissolved solids concentrations increase toward more impermeable rocks and areas of low circulation. In the center of the basin, the waters are characterized by very high total dissolved solids, high calcium, and low sulfate content, but in the porous and permeable fingers near the outcrop, salinites are low and sulfate is high. All gradations between these two extremes exist in the Bell Canyon formation. In areas where the sand fingers pinch out very rapidly into lowpermeability sediments, the transition between these two extremes of water composition may take place in a matter of a few well locations. An excellent
3 26
EXPLORATION FOR PETROLEUM AND GAS
example of this is the Saber field (Fig.10.8-10) where a range of waters is evidenced in one homogeneous, continuous sand body. The reason for these rapid changes in formation water compositions may be explained by permeability changes within the Bell Canyon formation. In areas of low permeability there is less circulation, less dilution, and more chance for the maintenance of an equilibrium relation between formation water and sediment. This was substantiated by the distribution of magnesium in the waters. A series of multiple regression analyses was made on the relation of various dissolved ions in the waters to their total dissolved solids. First, all the waters were analyzed as a unit to determine the correlation coefficients and the degree of variability explained by the chosen ions. The second stage was the breakdown of the waters into three arbitrarily defined groups (based principally on salinities) t o see if there were any noticeable changes in either correlation coefficients or degree of explained variance. The only significant change was in the relation of magnesium t o total solids. In those waters containing relatively low concentrations of total dissolved solids, there was a significant positive correlation, but in those with high concentrations of total dissolved solids, there was a significant negative correlation. This indicates that the relative concentration of magnesium decreases in waters of high total dissolved solids. These waters are precisely those that are found in low-permeability, fine-grained, argillaceous rocks in which magnesium would most likely be taken out of waters by diagenetic alteration of clay minerals. The variations found in the formation waters within the Bell Canyon formation can be used as the basis of an exploration technique. Since the composition of formation waters is related t o permeability, and permeability is related t o producibility of reservoir rocks, a workable relation exists between exploration objectives and water compositions. The refining of the maps of the distribution of the composition of the waters aids in defining the distribution of the sand fingers. The updip edges of the permeable sqnd fingers show increased concentrations of total dissolved solids and decreased magnesium which are related to the presence of a “barrier” (or trap) updip from the reservoir sands.
Formation water maps Maps of the total dissolved solids content (Fig. 10.8), the chloride content (Fig. 10.9), and the calcium content (Fig. 10.10) of formation waters were prepared. The inference which may be drawn from these maps is the empirical association of oilfield occurrence versus the iso-mg/l contours of the various constituents. This empiricism shows some remarkable alignments and permits formation-water composition maps to be added to the suite of exploration tools. Some of the subtleties of constituent composition versus rock properties do not lend themselves readily t o mapping techniques but are useful for consideration (i.e., magnesium content versus low permeability
CASE STUDY OF THE DELAWARE SAND I
I
327
328
I
EXPLORATION FOR PETROLEUM AND GAS
!
d W
5
Y
CASE STUDY OF THE DELAWARE SAND
329
33 0
EXPLORATION FOR PETROLEUM AND GAS
and fine-grained rocks, sulphate concentration variation with relationship to outcrop, etc.). It is important to note on each of the constituent maps that the local variations of the iso-mg/l contours are of greatest importance and not the precise value of the contour. For example, on the total dissolved solids map (Fig. 10.8) the concentration in the Ford field is only 50,000 mg/l and ranges in a re-entrant to about 150,000 mg/l, while in the Wheat field, the range is from 150,000 mg/l to nearly 250,000 mg/l. The overall appearance of this map (Fig. 10.8) is a series of fingering expressions. The various oilfields seem to have an occurrence relationship in the transition zone from higher to lower concentration. The Mason, Tunstill, Olds, and Saber fields occur along a transition zone from 250,000 mg/l to 50,000 mg/l. The El Mar and Grice fields are on a transition zone from 250,000 mg/l to 150,000 mg/l. The Two Freds field has a transition zone from 250,000 mg/l to about 150,000 mg/l. The Wheat and also the Ford fields occur in a similar transition zone. This relationship of oil occurrence and differential concentration is of great significance. Even in this limited area it appears that in the block from longitude 103”30’00” to 103”45‘00‘’ and from latitude 30’45’00” to 32’00‘00” there are several places which, from this empirical relationship, have some potentialities. The block immediately south of this, trending southwesterly from the Wheat and also from the Two Freds fields, needs additional study for better delineation. The area east of the Two Freds field lacks adequate control but basically shows the possibility of favorable development. The chloride map (Fig.10.9) has a configuration similar to the total solids map. Again it is not the precise iso-mg/l contour which is of prime concern but the variation in the limited area. This rate of change from higher to lower concentration appears to be a principal key to occurrence. The calcium content map (Fig. 10.10) does not show the prominent fingering, almost pseudodeltaic, effect that the total solids and chloride maps have. Perhaps this is because of the smaller range of values mapped. Some of the high to lower concentration effect is present and in other areas, the Wheat and Two Freds fields, the iso-mg/l closure is developed. This map is less diagnostic than the others; however, considered in conjunction with the other two maps, the coexistence of accumulation and the transition zone, even closure cannot be missed. Formation water maps of other areas Fig. 10.11 is a potentiometric surface map of the Arbuckle formation group in parts of Kansas, Missouri, and Oklahoma drawn by Chenoweth (1964). As ,noted on the figure, the arrows indicate the theoretical direction of the water flow which was inferred from the tilted oil-water interfaces. Detailed pressure data along with reservoir transmissibility data could be used to construct a similar map for determining detailed flow pattern.
331
FORMATION WATER MAPS
/
0
Kilometers 55.60
M
I:l,wom
LEGEND
drrpru indicate ItmoretlcoI dimction of artasion flow. Infarred tmm tilted oil i"l,,t.aCe. in B"ter,Ellswrth, ord Pownee Counties. Kansas. w01.r
Arbuckle b n t . " R e W o n amd'or Gmnite Wash beneath hnnaylvonion. Ellia.Rush and Barton Countlea. Kanroa .
Fig.lO.11. Potentiometric surface map of the Arbuckle formation group in South Kansas, North Oklahoma, and southeast Missouri.
The map shown in Fig.10.12 also was constructed by Chenoweth (1964) and it is a chloride map of the Arbuckle formation group in Kansas and Oklahoma. Note the dilute brines near the outcrop areas, which are diluted by meteoric recharge waters entering the outcrop. As a general rule the trapped petroleum in this group of formations is found associated with the more saline brines, and in the transition areas. Fig. 10.13 is a map that illustrates the variation in salinity at the bottom of the lower Wilcox formation in portions of Texas, Louisiana, Arkansas, Mississippi, and Alabama. The most saline brines occur in the deeper basin areas with the dilute brines nearer outcrop areas. Fig. 10.14 is a similar map which illustrates the salinity variations in the top of the lower Wilcox formation in the same area. Fig. 10.15 illustrates the salinity variations at the base
332
EXPLORATION FOR PETROLEUM AND GAS
Fig.10.12. Map of the chloride concentrations (mg/l) in the Arbuckle formation waters in Kansas and Oklahoma
of the upper Wilcox formation, while Fig. 10.16 illustrates the salinity variations at the top of the upper Wilcox formation. Fig. 10.17 is a salinity map of the lower Tuscaloosa and Woodbine formations, again most of the trapped petroleum is found in areas where the more saline waters occur and in transition areas. Salinity maps are useful as a primary tool in petroleum exploration because they provide information concerning sand fingering, diagenetic changes that affect reservoir and source rocks, and stratigraphic traps. Occurrences of petroleum accumulations often correlate with salinity transition zones, i.e., where the salinity ranges from 50,000 mg/l to 100,000 mg/l.
EEL‘
I
I
S d V N IZIBLVM NOILVNIZIOd
334
EXPLORATION FOR PETROLEUM AND GAS
Fig.10.15. Salinity concentrations in waters taken from the base of the Upper Wilcox formation in portions of Texas, Louisiana, Arkansas, and Mississippi.
Less than 5,600
Fig.lO.16. Salinity concentrations in waters taken from the top of the Upper Wilcox formation in portions of Texas, Louisiana, Arkansas, Mississippi, and Florida.
CONCLUDING REMARKS
335
LEGEND
II
Greater than
im,ooo
mg/l (as NOCI)
5.600 lo 70.000 mp/l
Less than 5.600 mg/I
Fig.lO.17. Salinity concentrations in waters taken from the Woodbine (Dexter) and Lower Tuscaloosa formations in .portions of Texas, Oklahoma, Arkansas, Louisiana, Mississippi, Georgia, South Carolina, North Carolina, and Florida.
Concluding remarks Organic acid salts, petroleum hydrocarbons, and other organic compounds are soluble in water. The ionic composition, the pH, and the Eh of the water influence the solubilities of the organic compounds. The aqueous solubility of petroleum hydrocarbons increases with increasing temperature and pressure, and decreases with increasing water salinity. The aqueous solubility of organic acid salts increases with increasing pH. A mechanism for the migration of petroleum or petroleum precursors, therefore, is water. It is known that petroleum hydrocarbons are generated from organic-rich rocks. The organic material in the petroleum source rocks is transformed by physicochemical reactions into petroleum precursors and/or hydrocarbons which are solubilized by water. The water phase moves the solubilized organics from the source to the reservoir where, because of temperature, pressure, salinity, pH, filtration, or organic salting-out phenomena, the organic phase separates from the water. In the reservoir the petroleum precursors and/or hydkocarbons mature to crude oil and gas, primarily because of temperature and time. Thermal alter-
EXPLORATION FOR PETROLEUM AND GAS
336
ation proceeds both in the fine-grained source rock and in the reservoir at temperatures above 115OC by abiogenic reactions. With increasing temperature the quality of the crude oil improves; however, at higher temperatures the crude oil is destroyed, leaving methane and pyrobitumen. The primary mechanism in the migration of petroleum involves water, therefore, it follows that knowledge of certain characteristics of the water is useful in exploration for oil and gas. The chapters “Classification of oilfield waters,” and “Some effects of water upon the generation migration, accumulation, and alteration of Petroleum” discuss some of these characteristics. Fig. 10.18 illustrates some characteristics related to waters that are likely to indicate an oil or gas accumulation, and some characteristics related t o waters that are likely to indicate a dry reservoir.
/2
-
C
a type water
(a)
Fig.lO.18. Genetic indicators in a water associated with an oil and gas accumulation (a) compared to indicators in a water associated with a dry reservoir (b).
REFERENCES
337
References Anonymous, 1959. Is geochemistry worth the effort. Pet. Week, 23-1-1959:20-24. Anonymous, 1960a. Geochemistry to get full-fledged test as exploration technique. Pet. Week, 2-12- 1960 :34-35. Anonymous, 1960b. 10.9% of U.S. wildcats produce in 1959. World Oil, 1 5 0 : l l l . Baker, D.R. 1962. Organic geochemistry of Cherokee group in southeastern Kansas and northeastern Oklahoma. Bull. A m . Assoc. Pet. Geol., 46:1621-1642. Baker, E.G., 1960. A hypothesis concerning the accumulation of sediment hydrocarbons t o form crude oil. Geochim. Cosmochim. Acta, 29:309-317. Baker, E.G., 1962. Distribution of hydrocarbons in petroleum. Bull. A m . Assoc. Pet. Geol., 46 :76-84. Billings, G.K., 1969. Geochemical petroleum exploration method, U.S. Patent, No.3,428,431. Bond, D.C., 1962. Geochemical process. U.S. Patent, No.3,033,287. Boyle, R.W. and Garrett, R.G., 1970. Geochemical prospecting - a review of its status and future. Earth-Sci. Rev., 6:51-75. Bronston, A., 1947. Geochemical data forecast major field. World Oil, 127:112-116. Buckley, S.E., Hocutt, C.R. and Taggart, M.S., 1958. Distribution of dissolved hydrocarbons in subsurface waters. In: L.G. Weeks (Editor), Habitat o f Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.850-882. Cartmill, J.C. and Dickey, P.A., 1970. Flow of a disperse emulsion of crude oil in water through porous media. Bull. A m . Assoc. Pet. Geol., 54~2438-2447. Chenoweth, P.A., 1964. Potentiometric Surface Map and Salinity Map o f the Arbuckle Formation Group in South Kansas, North Oklahoma and Southeast Missouri. Sinclair Oil Company, unpublished. Cooper, J.E., 1962. Fatty acids in recent-and ancient sediments and petroleum reservoir waters. Nature, 193:7 44-7 46. Cooper, J.E. and Bray, E.E., 1963. A postulated role of fatty acids in petroleum formation. Geochim. Cosmochim. Acta, 27:1113-1127. Cooper, J.E. and Kvenvolden, K.A., 1967. Method for prospecting for petroleum. US. Pa ten t , No. 3,30 5,317. Davidson, M.J., 1963. Geochemistry can help find oil if properly used. World Oil, 157:94-106. Davis, J.B. and Yarbrough, H.R., 1969. Geochemical exploration. U.S. Patent, No.3,457,044. Dunton, M.L. and Hunt, J.M., 1962. Distribution of low-molecular-weight hydrocarbons in recent and ancient sediments. Bull. A m . Assoc. Pet. Geol., 46:2246-2248. Erdman, J.G., 1965. Petroleum - its origin in the earth. In: A. Young and J.E. Galley (Editors, Fluids in Subsurface Environments. American Association of Petroleum Geologists, Tulsa, Okla., pp.20-52. Erdman, J.G., 1967. Geochemical origins of the low-molecular-weight-hydrocarbon constituents of petroleum and natural gases. In: Proceedings 7th World Petroleum Congress, Mexico, D.F. Elsevier, London, 2:13-24. Ferguson, W.S., 1962. Analytical problems in determining hydrocarbons in sediments. Bull. A m . Assoc. Pet. Geol., 46:1613-1620. Feugere, G. and Gerard, R.E., 1970. Geochemical logging - a new exploration tool. World Oil, 170:37-40. Filonov, V.A., 1964. Influence of organic material in oil on the shift of radioactive equilibrium in the watersof oil deposits. Sou. Geol., 6:144-146 (in Russian). Filonov, V.A., 1969. The problem of use of radioactivity of underground waters as an indirect hydrochemical indicator of oil content. Nauch. Tekh. Sb. Ser. Neftegazov. Geol. Geofiz., 3:32-35 (in Russian).
338
EXPLORATION FOR PETROLEUM AND GAS
Forsman, J.P. and Hunt, J.M., 1958. Insoluble organic matter (kerogen) in sedimentary rocks of marine origin. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.747-778. Galin, V.L. and Plyushchenko, V.G., 1963. Hydrogeology of the Upper Cretaceous deposits of Dagestan as related to their oil- and gas-bearing properties. Zzu. Vyssh. Uchebn. Zaued. Geol. Razued., 4:120-217 (in Russian). Gehman, Jr., H.M., 1962. Organic matter in limestones. Geochim. Cosmochim. Acta, 26 :885-899. Gerard, R.E. and Feug&re, G., 1969. Results of an experimental offshore geochemical prospection study. In: P.A. Schenck and I. Havenaar (Editors), Advances in Organic Geochemistry. Pergamon Press, New York, N.Y., pp.355-372. Gutsalo, L.K., 1964. Some regularities of radium distribution in underground waters of middle part of Dnepr-Pon Depression. Geokhimiya, 12:1305-1312 (in Russian). Gutsalo, L.K., 1967. Geochemical relation of radium anomalies in groundwaters to oil and gas deposits. Dokl. Akad. Nauk S.S.S.R., 172:1174-1176 (in Russian). Gutsalo, L.K., 1969. The nature and regular features of distribution of helium anomalies in underground waters adjoining oil and gas deposits, Sou. Geol., 8:112-123 (in Russian ). Gutsalo, L.K. and Krivosheya, V.A., 1965. Certain regularities in the gas-saturation of groundwaters associated with oil-gas-bearing structures in the central DDB (DneprDonets Basin) and their significance for oil-gas prospecting. Geol. Nefti Gaza, 3:51-53 (in Russian). Hanshaw, B.B. and Hill, G.A., 1969. Geochemistry and hydrodynamics of the Paradox Basin region, Utah, Colorado, and New Mexico. Chem. Geol. 4:263-294. Hedberg, H.D., 1964. Geologic aspects of the origin of petroleum. Bull. A m . Assoc. Pet. Geol. 48:1755-1803. Hiss, W.L., Peterson, J.B. and Ramsey, T.R., 1969. Saline water in southeastern New Mexico. Chem. Geol., 4:341-360. Horvitz, L., 1969. Hydrocarbon gqochemical prospecting after thirty years. In: W.B. Heroy (Editor), Unconventional Methods in Exploration f o r Petroleum and Natural Gas. Southern Methodist University, Dallas, Texas, pp. 205-21 8. Hunt, J.M., 1967. The origin of petroleum in carbonate rocks. In: G.V. Chilingar, H.M. Bissel and R.W. Fairbridge (Editors), Developments in Sedimentology, 9B. Carbonate Rocks -Physical and Chemical Aspects. Elsevier, Amsterdam, pp.225-251. Hunt, J.M. and Jamieson, G.W., 1956. Oil and organic matter in source rocks of petroleum. Bull. A m . Assoc. Pet., Geol., 40:477-488. Hunt, J.M. and Meinert, R.N., 1954. Petroleum prospecting. U.S. Patent, No.2,854,396. Johnson, A.C., 1970. How to hunt oil and gas using the inorganic surface-geochemical method. Oil Gas J., 68:llO-112. Karaskiewicz, J., 1966. Forecasting of oil- and gasbearing capacity in the light of geomicrobiological and geochemical investigations. Nafta, 22 (12):19-20 (in Polish). Karim, M, Chilingar, G.V. and Hoylman, H.W., 1966.Northeast Nicaragua has gas and oil indications. World Oil,162:84,86, 91,92,94,96. Kartsev, A.A., Dudova, M. Ya. and Diterikhs, O.D., 1969. Benzene homologs in underground waters and their relation to oil. Geol. Nefti Gaza, 7:41-45 (in Russian). Kartsev, A.A., Tabasaranskii, Z.A., Subbota, M.I. and Moglevskii, G.A., 1959. Geochemical Methods of Prospecting and Explomtion f o r Petroleum and Natural Gas (English transl. edited by P.A. Witherspoon and W.D. Romey) University of California Press, Berkely, Calif., 349 pp. Klemme, H.D., 1972. Geothermal gradients. Oil Gas J . , 69:136,141-144; 70:76-78. Kolodii, V.V., 1969. Origin of some hydrochemical anomalies in petroleum provinces. Nauch. Tekh. Sb. Ser. Neftegazou. Geol. Geofiz., 8:37-40 (in Russian).
REFERENCES
339
Korobov, D.S., 1965. Distribution of trace elements in water and rock of oil deposits in the Saratov-Volgograd region of the Volga and its significance in petroleum exploration. In: Soviet Advances in Nuclear Geophysics (English transl. of Yadernaya Geofizika). Consultants Bureau, New York, N.Y., pp.171-178. Kortsenshtein, V.N., 1964. The estimation of the possible oil-gas presence according to the groundwater analyses and the evaluation of the forecast oil-gas reserves. Dokl. Akad. Nauk S.S.S.R., 158:856-859 (in Russian). Kortsenshtein, V.N., 1965. Estimating the oil and gas potential from data on gas saturation of groundwater under dephased equilibrium conditions. Dokl. Akad. Nauk S.S.S.R., 150:635-638 (in Russian). Kortsenshtein, V.N., 1968. A comparative description of the benzene content in stratal waters of the Mesozoic complexes of South Mangyshlak and of East Caucasia. Dokl. Akad. Nauk S.S.S.R., 180:697-699 (in Russian). Kravchinskii, Z.Ya. 1960. Comparison of chemical characteristics of waters in productive red-colored formations. Geol. NeftiGpza, 12:42-44 (in Russian).. Krejci-Graf, K., 1962. Oilfield watek. Erdol Kohle Petrochem., 15:102-109 (in German). Kroepelin, H., 1967. Geochemical prospecting. In: Latest Developments within the Oil Industry. Proceedings 7 t h World Petroleum Congress. American Elsevier, New York, N.Y., 1B:37-57. Landes, K.K., 1967. Eometamorphism and oil and gas in time and space. Bull. A m . Assoc. Pet. Geol., 51 :828-841. London, E.E., 1964. The degree of groundwater saturation with dissolved hydrocarbons and sulfates as a criterion for the evaluation of oil-gas prospects. Geol. Nefti Gaza, 11:41-47 (in Russian). London, E.E., Zor’kin, L.M. and Vasil’ev; V.G., 1961. Location of gas reserves based on dissolved gas content of formation waters. Geol. Nefti Gaza, 3:35-40 (in Russian). MacElvain, R.G., 1963. What d o near-surface signs really mean in oil finding. Oil Gas J., 61(7):133-136; 61(8):139-146. McAuliffe, C.D., 1966. Solubility in water of paraffin, cycloparaffin, olefin, acetylene, cycloolefin, and aromatic hydrocarbons. J. Phys. Chem., 70:1267-1275. McAuliffe, C.D., 1967. Geochemical method of prospecting for petroleum. U.S. Patent, No.3,345,137. McAuliffe, C., 1969. Determination of dissolved hydrocarbons in subsurface brines. Chem. GeoL, 4:225-233. McLeod, H.O., 19711 Bibliogmphy Related to Research and Geochemical Methods for Petroleum and Gas Exploration. Petroleum Abstracts, 200 pp., unpublished. Noad, D.F., 1966. Water analysis: a key to exploration. Can Petrol., 7:12-14; 7:16-18. Parker, J.W., 1969. Water history of Cretaceous aquifers, East Texas Basin. Chem. Geol., 4:111-133. Peake, E. and Hodgson, G.W., 1966, 1967. Alkanes in aqueous systems, I and 11. J. Am. Oil Chem. SOC.,I, 43:215-222; 11, 44:696-702. Philippi, G.T., 1957. Identification of oil source beds by chemical means. Proc. 20th Int. Geol. Congr., Mexico City, 1957, Sect. III, pp.25-40. Philippi, G.T., 1965. On depth, time, and mechanism of petroleum generation. Geochim. Cosmochim. Acta, 29:1021-1049. Philippi, G.T., 1969. Essentials of the petroleum formation process are organic source material and a subsurface temperature controlled chemical reaction mechanism. In: P.A. Schenck and I. Havenaar (Editors), Advances in Organic Geochemistry. Pergamon Press, New York, N.Y., pp.25-46. Pirson, S.J., 1942. Theoretical and economic significance of geodynamic prospecting. World Petrol., 13:38-42. Pusey, 111, W.D., 1973. How to evaluate potential gas and oil source rocks. World Oil, 176:7 1-75.
340
EXPLORATION FOR PETROLEUM AND GAS
Ransone, W.R., 1947. Geochemical history of the Hardy oilfield, Jones County, Texas. Geophysics, 12: 384-392. Roach, J.W., 1965. How to apply fluid mechanics to petroleum exploration. World Oil, 160:7 1-7 5. Rosaire, E.E., 1939. The Handbook of Geochemical Prospecting. Rosaire, Houston, Texas, 120 pp. Rosaire, E.E., McBermott, E. and Fash, R.H., 1940. Discussion of geochemical exploration. Bull. Am, Assoc. Pet. GeoL, 24:1434-1463. Savchenko, V.P., Vinogradov, V.L. and Yakoylev, Yu. I., 1965. Front- and back-wall effect and its importance in prospecting. Geol. Nefti Gaza, 7:36-40 (in Russian). Schmidt, G.W., 1970. Geochemical prospecting method. US.Patent, No.3,524,346. Schwab, R., 1965. Logging important aspect of hydrodynamic studies. Oilweek, 16:36-37. Serebriako, 0.1. and Tronko, I.V., 1969. Ammonium content in groundwaters of the northwestern Caspian region as an indication of oil and gas. GeoL Nefti Gaza, 9:57-60 (in Russian). Shabarova, N.T., Tunyak, A.P. and Nektarova, M.B., 1961. Study of organic acids in subsurface waters. Geol. Ne f t i Gaza, 11:50-5 1(in Russian). Shvets, V.M. and Shilov, I.K., 1968. On organic matter in underground waters of the southwestern part of the Azov-Kuban Artesian Basin. GeoL Nefti Gaza, 8:46-49 (in Russian). Sikka, D.B., 1963. Mechanisms explaining the formation of radiometric anomalies. Izv. Akad. Nauk S.S.S.R., Ser. GeoL, 6:73-87 (in Russian). Silverman, S.R., 1964. Investigations of petroleum origin and evaluation mechanisms by carbon-isotope studies. In: H. Craig, S.L. Miller and G.J. Wasserburg (Editors), Isotopic and Cosmic Chemistry. North-Holland, Amsterdam, pp.92-102. Simons, H.F., 1940. Scope of soil analysis increased during year. Oil Gas J., 38:54. Sokolov, V.A., Alexeyev, F.A., Bars, E.A., Geodekyan, A.A., Mogilevskii, G.A., Yurovskii, Yu.M. and Yasenev, B.P., 1959. Investigations into direct oil detection methods. Proc. 5th World Petrol Congr., Sect. I , Paper, No.36, pp.667-687. Spencer, D.W. and Koons, C.B., 1970. Studies on origin of crude oil data. Presented at SPE Symp., 44th Annual Meet., Calgary, A l t a , June 22-24, 1970. Stormont, D.H., 1939. Operation of a Gulf Coast field on soil survey information. OilGas J., 38:28-29. Sudo, Y., 1967. Geochemical study of brine from oil and gas fields in Japan. J. Japan Assoc. Pet. Technol., 32: 286-296 (in Japanese). Tissot, B., Califet-Debyser, Y., Deroo, G. and Oudin, J.L., 1971. Origin and evolution of hydrocarbons in Early Toarcian shales, Paris Basin, France. Bull. Am. Assoc. Pet. GeoL, 55:2177-2193. Trask, P.D. and Patnode, H.W., 1942. Source Beds o f Petroleum. American Association of Petroleum Geologists, Tulsa, Okla., 566 pp. Vdovykin, G.P., 1963. Oxidation-reduction potential of formation waters of the northwest Cis-Caucasus and of some surface waters. Pet. GeoL, 7:286--290. Vilonov, V.A., 1962. A feature of the distribution of radioactive elements in the wateroil contact zone. Geokhim. Nefti Neftyanykh Mestorzhdenii, Akad. Nauk S.S.S.R., Moscow, pp.199-206 (in Russian). Visher, G.S., 1961. Petrologic study of the Delaware sand (Bell Canyon formation), Texas. Rep. Sinclair Oil Company, unpublished. Welte, D.H., 1965. Relation between petroleum and source rock. Bull. Am. Assoc. Pet. Geol., 49:2246-2268. Zarrella, W.M., 1969. Applications of geochemistry to petroleum exploration. In: W.B. Heroy (Editor), Unconventional Methods in Exploration for Petroleum and Natuml Gas. Southern Methodist University, Dallas, Texas, pp.29-41.
REFERENCES
341
Zarrella, W.M., Mousseau, R.J., Coggeshall, N.D., Norris, M.S. and Schrayer, G.J., 1967. Analysis and significance of hydrocarbons in subsurface brines. Geochim. Cosmochim. A h , 31 :1155-1166. Zinger, A S . and Kravchik, T.E., 1969. On direct hydrochemical indices in the presence of oil and gas and the role of micellar solubility in the case of hydrocarbon migration in aqueous solutions. DokL Akad. Nauk S.S.S.R., 189:180-184 (in Russian).
This Page Intentionally Left Blank
Chapter 11. GEOPRESSURED RESERVOIRS
The composition of the waters in normally pressured reservoirs often differs from the composition of the waters in geopressured or abnormally high-pressured reservoirs. There are several theories concerning the cause of the geopressured zones; many papers have been written about their occurrence and causes (Burst, 1969; Dickey et al., 1968, 1972; Fowler, 1970; Harkins and Baugher, 1969; Hottmann and Johnson, 1965; Jones, 1969; Powers, 1967;Schmidt, 1973;Wallace, 1969). Knowledge of how to locate geopressured zones is important in drilling operations, because if such a zone is drilled into without adequate preparation, the well may blow out, perhaps causing a fire, loss of the well, loss of the drilling rig, or even loss of life. The usual precaution, if the driller knows of a high-pressure zone, is to increase the weight of the drilling mud; however, the continual use of heavyweight mud is much more expensive than drilling with a lighter weight mud. Drilling rig time is worth about $2,000 per day, and it costs about $44,000 per kilometer to drill a well on land. A drilling barge in the bay can cost from $4,000 to $lO,O.OOper day while a drilling ship plus a full crew costs about $25,000 per day. Considering.the foregoing costs plus the cost for a special crew t o extinguish a fire at an ignited blowing well can be very expensive because the initial fee for the fire extinguishing personnel is about $25,000. Steps, therefore, are taken by the drilling company to assure that an adequate drilling rig is used, that the optimum size borehole is drilled, that the correct weight drilling mud is pumped down, that strong enough casing is inserted into the well, and that blow-out preventers are operative. Geopressure Dickinson (1953)defined abnormally high pressure (geopressure) as any pressure exceeding the hydrostatic head of a column of water (extending from the subsurface tapped stratum to the land surface) containing 80,000 mg/l of dissolved solids. Formations with equal or less pressures are considered normal or subnormal. In the Gulf Coast area the normal pressure gradient is about 0.107 kg m-l, or about equal to 0.21 g cm-3 of drilling mud (Harkins and Baugher, 1969). Normal pressure in the Rocky m-', although excep Mountain region has a gradient of 0.100 kg tions occur in western Montana, the Denver Basin, the Powder River Basin, and the San Juan Basin, mostly in Cretaceous rocks (Finch, 1969). (A gradient of 0.118 kg m-l is normal in the Williston Basin in North
344
GEOPRESSURED RESERVOIRS
Dakota.) Normal pressure is that which is normal for the particular area involved and is related t o the salinity of the reservoir water, rock types, and geologic setting, but in general, it is that pressure exerted by a column of water from the surface t o the observed subsurface formation, which is equal to and will balance the subsurface formation pressure. Abnormally high pressures are those which exceed this normal hydrostatic head. Geostatic ratio Abnormal pressures can be expressed in terms of a “geostatic ratio,” which is the ratio of the observed fluid pressure in a subsurface formation t o the overburden pressure of the overlying sediments. This load at a given m down t o depths of more than depth is approximately 0.231 kg 6,100 m, because the density of rocks changes slowly with depth (Pennebaker, 1968). Any abnormal pressure will therefore have a geostatic ratio in the Gulf Coast area and between between 0.0327 and 0.0703 kg in the Rocky Mountain area. 0.0304 and 0.0703 kg Compaction model The compaction concept was demonstrated using a model consisting of perforated metal plate separated by metal springs in water and enclosed in a cylindrical tube (Terzaghi and Peck, 1948). The springs were used to simulate communication between deposited particles and with the initial pressure upon the upper plate, the springs d o not move because all of the pressure is supported by the water, assuming that water does not escape from the system. Relating the fluid pressure (FP) t o the total pressure (TP) one can derive an equation X=FP/TP to record various formation pressures (X)t o determine the geostatic ratio using the model.
Origin of abnormal pressures Abnormally high pressures in a formation can be caused by compaction. Factors which may cause them are, according to Hottmann and Johnson (1965),“the ratio of shale t o sand thickness, the mean formation permeability, the elapsed time since deposition, the rate of deposition, and the amount of overburden.” These parameters are interrelated in compaction, which is the controlling factor in fluid pressures within subsurface sedimentary environments (Harkins and Baugher, 1969). Dickinson (1953)reports that the fluid pressures within sediments are predominately controlled by two factors; namely: (1)the compression as a result of compaction; and (2)the resistance t o expulsion of water. Compaction begins with sedimentation and deposition of soft muds composed of up to 90% water (Wallace, 1969). In an environment where deposition continues, gradual compaction occurs whereby the muds become clay
ORIGIN OF ABNORMAL PRESSURES
345
minerals and shales. The shales are primarily clay minerals with flat or tabular grain shapes; with additional overburden, the pressure packs the grains closer together, with a resultant expulsion of water from the intervening spaces. In the early stages of compaction, the shale possesses high porosity and permeability, and the expelled water always flows to areas of least resistance and pressure (often porous sand). As the overburden increases, the porosity and permeability of the shale decrease until equilibrium is approached and the pressure in all directions is equal. A t this point, expulsion of additional water is limited. Tectonics, of course, could alter the subsurface environment. Deposition and sedimentation of sand are somewhat different because the sand grains are in contact in the first stage and sand compaction is about complete with deposition. However, reduction of porosity can occur by: (1) solution of the sand grains at contact points; and (2) rearrangement of the grains because of very high pressures. Clay beds separating aquifers are often referred to as semipermeable membranes. Such beds can separate aquifers containing waters of different salinities, causing a hydrostatic head in the direction of the more saline water.
Fig. 11.1. Sand dikes in the Simpson Sand formed by the actiqn of highly pressured subsurface waters forcing the lighter colored sand intrusively into the primary sandstone. The primary sand was formed from white beach sands during Ordovician time.
346
GEOPRESSURED RESERVOIRS
Osmotic pressure can develop, which is dependent upon osmotic efficiency of the clay bed and the differences in salinities of the two aquifers (Young and Low, 1965).According to Jones (1969),stepwise increments of osmotic pressure may develop wiih depth through a series of bedded sands and clays acting as a multistage pump, thus producing the high reservoir pressures in the northern Gulf of Mexico basin. Fertl and Timko (1972)discuss 17 possible causes of abnormally high pressures. They are rate of sedimentation, tectonic activities, potentiometric surface levels, reservoir structures, areal salt deposition, shallow-reservoir repressuring, paleopressures, mud volcanoes, secondary precipitation of cementation constituents, diagenesis of volcanic ash, rehydration of anhydrite, diagenesis of clays, osmosis, permafrost, earthquakes, chemical, thermal chemical, and biochemical effects, and tidal disturbances. Fig. 11.1 illustrates one type of action that results from high pressures, where sand dikes formed by the action of highly pressured subsurface waters forcing the lighter colored sand intrusively into the primary sandstone. The primary sandstone was formed from white beach sands during Ordovician time. Abnormal pressures in the Gulf Coast area In the Gulf Coast area, the abnormal pressure seems t o be related t o rapid deposition of sediments and low regional transmissibility. Fluid pressures are near hydrostatic where there is continuity with normally pressured aquifers and where the sands are sufficiently permeable to dissipate the expelled water from the compacting fine-grained rock. In some of the deep oil and gas wells of the Gulf Coast, the pressure of the interstitial fluids (oil, gas, or water) in kilograms per square centimeter is normally the depth in meters multiplied by 0.107.This is slightly more than the pressure required t o sustain a column of water to the surface. At great depths where the geological section is mostly shale, fluids at abnormally high pressures are found. Sometimes the pressures are very high, approaching 0.2 kg cme2 m-l. Often the increase in fluid pressure is abrupt, taking place in a vertical interval of 30 m or less. In other areas, the increase in pressure is more gradual, extending over 300 m of vertical section. The depth at which the pressure starts t o increase ranges over a wide interval. Abnormal pressures are found at depths as shallow as 1,000m in some offshore fields, and wells in some areas have been drilled deeper than 7,000 m without encountering abnormal pressures. Forty-one formation water samples from gasfields in southwestern Louisiana were obtained and analyzed to determine the relationships of the chemical composition of the waters to normal and abnormally pressured geologic zones (Dickey et al., 1972).The concentration of dissolved solids in the waters from the overpressured zones is generally less than in the normal pressure zones, and this knowledge is significant in electric log interpretation.
ABNORMAL PRESSURES IN THE GULF COAST AREA
347
Fig. 11.2. Slash lines showing the general area in Louisiana where the samples were obtained.
Previous work in the area suggested that the abnormally fresh waters were found in the same part of the section as were the abnormally high pressures (Dickey et al., 1968). The general locations of the wells are shown in Fig. 11.2. They were from the South Lewisburg, Church Point, Branch, South BOSCO,North Duson, Duson, Ridge, and Andrew fields, all in Acadia, and Lafayette Parishes, Louisiana. The water samples were analyzed chemically by using the procedures published by the American Petroleum Institute (1968). The analytical data are summarized in Table 11.1. A subsurface cross section, Fig. 11.3, was constructed in a general northsouth direction showing the stratigraphy and structure across seven oilfields in the area of study (Fajardo, 1968). The initial pressures of the shallower reservoirs are normal. However, below 2,450 m many reservoirs contain fluids with abnormally high pressures. The 4.9-m amplified normal curve was used to recognize the first appearance of abnormal pressures in the shale section. The fluid pressure gradients were estimated following the method described by Hottmann and Johnson (1965). Shale resistivity and fluid pressure gradient versus depth were plotted for 50 wells in different fields of the study area, and of these, 22 are included in the cross section. All of the 41 waters belong t o the chloride-calcium class of Sulin (1946), and none has the composition of meteoric water. The principal cation is sodium, although the concentration of calcium is always high. In some of the more concentrated brines, the calcium concentration is nearly 40,000 mg/l and constitutes over half the reacting value of the sodium. Magnesium is variable in amount, and in two samples it is absent. Chloride is the predominant anion, amounting always t o more than 49.5% of the total reacting values. Sulfate usually is absent and never is present in concentrations greater than 0.5% of the total reacting value. In Fig. 11.3, the top of the section is 2,100 m below sea level. The electric logs indicate the lithology, which is quite sandy down t o a depth of 2,700 m
TABLE 11.I Formation-waters sample locations, constituents found in the waters, shale resistivity (SR),and fluid pressure gradients (FPG) Sample Location o f number of well (S-Twp-R)
1*I 2’1 3*’ 4’1 5*1
6 7 8 9 10
73-105-34E 2l-lOS-03E 2*lOSFo3E 20-10S-aE 17-10S43E O8-lOS43E 25-07S43E 25-07603E
11
12 13 14 15 16 17** 18 19 20 21 22*’ 23 24*’ 25” 26 27 28 29 30 31 32 33*’ 3483 35*3 36” 37 38 39*3 40 41*3
*’ Abnormal pressure.
Depth (m)
Zone
Specific gravity
Concentration (mg/l)
(60°/600F) CI
4,66+4,664 3,643-3,645 3,625-3,627 3.613-3,615 4,060-4,063 3,047-3.049 3,325-3,327 3.293-3.296
1.035 1.062 1.057 L. Camerina 1.059 U. Camerina 1.085 Bolivina-mex Discorbis 1.045 1.060 U. Tweedel 1.069 L. Tweedel U. Nodosaria 1.083 Daigle 1.051 1.065 U. Nodosaria 1.069 Tweedel 1.062 Tweedel 1.128 Struma Frio 1.061 Nodosaria 1.149 Klumpp D 1.090 Frio 1.058 Frio 1.058 Frio 1.057 1.144 Frio 1.090 U. Texana U.Texana 1.092 1.220 Frio 1.202 Frio l.lS3 Frio 1.139 Nodosaria 1.070 Marg howei 1.144 Homeseeker 1.145 Nodosaria A Horn-eker D‘-4 1.088 1.055 Klumpp E 1.089 Brookshire Brookshire 1.089 Brookshire 1.082 Brookshire 1.089 Nodosaria 1.140 Homeseeker 2-D 1.120 1.069 Marg tex 1.082 U. Moicene 1.050 Marginulina Bolivina-mex
U. Camerina
39,000 55,600 56,600 50,000 72.800 33,300 51,700 58,000 50,900 45,300 45,600 57,900 52,600 116,000 50,900 135,000 79,300 46,600 47,900 45,800 125,000 84.500 80,300 201,000 184,000 111,000 119,000 61,600 100,000 109,000 80,000 44,400 77,500 78,000 72,300 75,400 129,000 120,500 55,500 74,400 49,700
HCO3
387 541 826 630 448 503 180 363 507 545 574 586 579 322 330 92 334 741 788 694 135 419 363 0 0 112 76 550 66 73 270 244 171 206 234 203 80 240 539 249 482
SO,
0 407 234 38 tr. 50 0 tr. 33 0 tr. tr. 0 0 67 223 0 60 72 ND 0 122 tr. 352 tr. ti. 0 0 0
SR at
FPG spl. depth (kg cm-2
B
49 62 62 37 43 32 18 26 35 28 29 26 34 38 23 52 48 46 48 44 47 40 45 75 67 42 52 67 42 0 39 77 34 130 36 0 18 0 18 0 33 0 18 0 41 0 43 102 52 0 26 8 8 43
Mg
Br
1
Na
Ca
35 61 52 57 81 37 21 41 70 38 35 56 45 128 43 154 169 40 52 47 64 20 62 213 204 94 117
18 22 21 21 19 16 15 18 23 18
17,800 32,300 34,200 29,500 41,100 19.200 34,800 34,400 27,500
1,070 78 2,210 369 1,380 213 1,380 194 3,850 583 1,390 224 2,730 544 2,020 194 3,050 719 2,310 167 2,950 447 2,660 389 1,570 0 21,600 2.180 1,890 408 15,200 1,270 2,950 447 2,660 1,010 2,180 303 ND ND 15,700 159 7,390 565 3,300 972 38,800 2,140 33,200 5,770 14,300 428 18,400 1,200 3.610 17 18,300 1,090 14,400 700 2,760 35 1,510 447 836 3,530 3,270 972 3,270 564 3,370 894 4.560 0 5,610 564 3,210 136 2,950 855 1,780 141
14
201 71 110 70 81 82 58 162 174 134 40 79 60
22
25 20 26 18 24 74 23 21 22
23 22 20 18 19 24 28 5 21 21 34 30 21 19 18 19 24 38 26 18 35
26,000 24,900 33,300 31,700 49,600 29,600 66,800 46,400 24,700 26,600 ND 61,900 45,800 45,800 80,600 68,900 53,600 52,700 35,200 40,600 51,800 47,800 25,800 44,200 44,400 41,200 42,600 77,800 68,800 32,500 42,800 29,600
*’ Abnormal pressure. but normal chemically. * 3 Samples 33-36 are from the Abheville field or the south and of the area sampled and appear to be in a different chemical family, S-Twp-R = section-township-range; ND = not determined.
K
Li
Sr
518 247 200 204 267 85 208 230 162 134 262 in1 192 427 31 5 813 427 172 166 157 830 324 376 782 640 798 771 137 1,150 631 392 166 236 235 294 232 ND 375 176 264 71
10 7 6 6 6 3 4 5 6 4 4 4 5 9 9 9 10
ND ND ND ND ND P:D ND ND ND ND ND ND ND ND ND ND
6
5 4 15 7 9 17 17 12 18 5 17 15 10 5 2 2 3 2 ND 5 5 2
2
Ba
ND ND ND YD ND ND ND ND ND ND ND ND ND ND ND ND 0 50 0 5 0 5 0 8 0 19 ND ND 0 33 ND ND ND ND ND ND ND ND ND ND N D l ND ND ND 0 110 ND ND 140 97 128 109 265 41 171 102 ND ND ND ND 0 7 195 85 0 4
6’)
NH4 organic acid as acetic 246 295 238 650 178 200
538 301 362 279 210 254 364 218 230 250 202 377 206 96 243 110 142 294 282 295 279 180 222 368 349 195 219 214 258 306 152 295 179 167 160
84 48 96 72 96 120 24 76 24 96 72 96 120 144 48 24 144 120 96 72 48 96 144 48 120 48 24 168 72 96 24 12 96 624 144 48 ND 96 192 432 312
0.77 0.59 0.62 ND 0.38 1.1
ND ND 1.02 0.83 0.97 ND ND 0.90 ND ND 0.60 0.84 ND 1.o
0.94 0.35 0.70 0.50 0.35 1.4 1.o
0.38 1 .o ND ND 0.83 ND ND ND ND 0.95 0.92 0.65 ND 0.57
0.185 0.159 0.157 ND 0.195 0.107 ND ND 0.107 0.107 0.107 ND
ND 0.107 ND ND 0.157 0.107 ND 0.107 0.107
0.191 0.131 0.191 0.203 0.107 0.107 0.193 0.107 ND ND 0.107 ND ND ND ND 0.107 0.107
0.152 ND 0.16X
This Page Intentionally Left Blank
ABNORMAL PRESSURES IN THE GULF COAST AREA
?6 *33a34
4.5
il
30
I
I
1
40
50
60
I 70
357
KEY X AbnzDressure Normal pressure
I
I
I
1
I
80
90
100
110
120
I D
CHLORIDE, g / l
Fig. 11.4. Plot of the depth of the wells versus concentrations of chloride in the formation waters.
-‘-I 2.5
c
0
,
33.
/ KEY X Abnormal pressure/ ?‘Normal presauro
3p
034
0.5 BICARBONATE, p/l
Fig. 11.5. Plot of the depth of the wells versus concentrations of bicarbonate in the formation waters.
GEOPRESSURED RESERVOIRS
358
in the north to 3,050 m in the south. Below this depth, the sands become less abundant and less widespread. The first abnormal pressure as calculated from shale resistivity is indicated by an arrow. The location of a producing horizon from which a water sample was taken is shown by the sample number in a circle. When the water sample was taken from a nearby well, not shown on the section, it was projected onto the section and shown as the sample number inside a square in Fig. 11.3. There is a general tendency for the dissolved salt concentration of the water samples t o increase with depth. This is shown in Fig. 11.4, which shows chloride plotted against depth. Since chloride is the predominant
I
KEY )< Abnormal pressure 0 Normal pressure
25
K 27
X
\
24
\ I
10
I
CALCIUM, g/l
Fig. 11.6. Plot of the depth of the wells versus concentrations of calcium in the formation waters.
359
ABNORMAL PRESSURES IN THE GULF COAST AREA
anion, it serves as an indication of the degree of concentration. The samples of water from abnormally pressured sands are shown as circles in x'es. All of them except 17, 22, 24, and 25 fall below the average concentration line, that is, they are less concentrated than they should be for their depth of burial. Sample 1 especially is much too weak. Bicarbonate, while occurring in much smaller quantities, shows the reverse relation, decreasing in amount with depth, as shown in Fig. 11.5. The waters from horizons with abnormal pressures have more bicarbonate than they should, considering their depth of burial. Calcium increases with depth, as shown in Fig. 11.6. It would be more correct t o say that there are two types of water. Type 1includes waters with less than 5,000 mg/l calcium, all of which are shallower than 3,800 m; type 2 is water with more than 5,000 mg/l calcium, most of which is deeper than 3,500 m. The only minor constituent that indicated a significant change with depth was potassium, and it appears to increase relative to sodium. The abnormally pressured waters seem deficient in potassium for their depth.
Normal pressure x
Abnormal pressu&e-Solution
o N o r m a l pressure- A l t e r e d relict bittern
100 O\&o
0% '0
a0 -
0
\ 0
"\
"
\
O \
-
\ 0
60-
\ "\ .\O
A 0 0
0
0 SODIUM, g/l
Fig. 11.7. Comparison of some brines of a bittern type from the Michigan Basin with some brines from some normal and abnormally pressured reservoirs in Louisiana.
GEOPRESSURED RESERVOIRS
36 0 I50
Normal pressure x Abnormal pressure
I25
I00
/
\
2 5-
.
75 X
0
0
In 50
X
- x
25 X*
Sodium’=mg/l
0
I
0.05
Na
+ 40 mg/l
I
010 BROMIDE,
Ca
I
0.15
I
0.20
!5
g/l
Fig. 11.8. Plot of Na’ versus Br from some brines from normal and abnormal pressured reservoirs in Louisiana.
Four of the waters from high-pressure sands (17, 22, 24, 25) have normal concentrations of dissolved solids for their depth. The other waters from high-pressure sands (1-5, 13, 28, 39, 41) have lower concentrations than normal. They also have less calcium, more bicarbonate, and a higher Cl/K ratio. About 80% of the material in the Gulf Coast shale is clay. Assuming that the waters have reacted with montmorillonite, there should be a direct relationship of calcium t o sodium. Plotting the calcium and sodium data in Table 11.1 plus some data for some brines from the Michigan Basin (as shown in Fig. 11.7) indicate that a relationship of calcium t o sodium does exist in the Gulf Coast waters and that they probably have reacted with montmorillonite. Fig. 11.7 also indicates that the Gulf Coast waters are not an altered relict bittern as are the Michigan Basin brines. In an ion exchange reaction with montmorillonite, 2 moles of sodium are exchanged for 1 mole of calcium, therefore, if salt is redissolved the bromide content in solution should be proportional to the original redissolved solu-
ABNORMAL PRESSURES IN THE GULF COAST AREA
36 1
tion. However, because of the exchange reaction the sodium in solution should be Na + 46/40 Ca or Na'. Fig. 11.8 is a plot of Na' versus Br for the 41 samples. The data scatter to some extent but this can be expected if biogenic derived bromide is present and the presence of iodide indicates that such is the case. Fig. 11.8 indicates that re-solution of salt is a control in these samples. Fig. 11.9 shows further evidence that the Louisiana brines were formed by re-solution of salt. For example, the dashed line in the left portion of Fig. 11.9 is a plot of Na' versus Br of salt dissolved in distilled water, and the solid line just t o the right is a replot of Na' versus Br for the Louisiana brines. The next dashed line t o the right is Na' versus Br for evaporating sea water, and the curved dashed line is Na' versus Br for relict brines from the Michigan Basin. Notable differences in the waters found in the normally and abnormally pressured rocks are evident (Schmidt, 1973). The dissolved solids in the
-Re-solution solt in pure water Southwestern Louisiano brines
\*
I
*\
I
I I
9vaporoting seo water
25
\*
\
h
*\
\
f *\
I I
Sodium I=mg/l sodium
I
t
\
.3:
+%mg/l calcium
2 BROMIDE, g/l
3
m>
4
Fig. 11.9. Replot of Na' versus Br of the Louisiana brines (Fig. 11.8);plus data for relict Michigan brines, evaporating sea water, and resolution of salt. Resolution of salt is an important control for the Louisiana brines
362
GEOPRESSURED RESERVOIRS
normally pressured sandstones range from 600 to 180,000 mg/l, while in the geopressured sandstones the range is from 16,000 to 26,000 mg/l. The dissolvedsolids in the water in the pores of the shales adjacent to normally pressured sandstones are lower than the dissolved solids in the water in the sandstones, but the dissolved solids concentrations are similar in the waters of the adjacent high-pressure sandstones and shales. The concentration order > HC03- > Cl-, and in normally pressured in shale pore water is sandstone water it is C1- > HC03- > S 0 4 - 2 . The temperature gradient in the geopressured zone is about O.8l0C/25 m, while in the normally pressured zone it is about 0.44OC/25 m. This change in temperature gradient is believed t o be related t o the porosity, where a greater porosity causes a decreased thermal conductivity (Schmidt, 1973). The clay mineral composition in the geopressured zone is predominantly a nonexpandable type, while in the normally pressured zone montmorillonite, an expandable type, frequently occurs. This change is believed to be related t o the temperature, and the heat allows the release of water from the clays at temperatures of about 93-104OC. This released water will dilute the pore water and cause the dissolved solids to decrease. The total amount of water released by Gulf Coast shales in geopressured zones is about 13%of the total in the system (Schmidt, 1973). This can be a cause of the lower salinity of the waters found in the geopressured zones. Fowler (1970) studied the Chocolate Bayou field in Texas and evaluated the relationships between geopressure and the migration and accumulation of hydrocarbons. He concluded that faults tend t o act as barriers separating fluid systems in the area; however, cross-formational flow occurs with geopressure causing shale ultrafiltration of the waters. The ultrafiltration produces salinity variations in the waters. Hydrocarbon accumulation in the area is controlled by the hydrodynamic flow. According to Fowler (1970), hydrocarbons are trapped in the upper sands because of slight pressure differentials across fault traps in the West Chocolate Bayou field. However, in deeper strata, abnormal pressures have caused hydrodynamic flow and pressures greater than the displacement pressure in the fault, resulting in no trapped hydrocarbons. In essence then, sands with pressure gradients greater than 0.20 kgcm-' m-' in the Chocolate Bayou field do not contain commercial amounts of hydrocarbons. It also appears that the size of the accumulation may decrease with increasing pressure gradients up t o about 0.16 kg cm-* m-'. The accumulation size may increase with pressure gradients in the range of 0.16-0.19 kg cm-2 m-' and then decreases. Detection of abnormal pressures Estimation of formation pressures from electrical surveys is related to the following assumptions, concerning the origin of abnormal pressures (Foster and Whalen, 1966):
DETECTION OF ABNORMAL PRESSURES
363
(1) Shale porosity is a function of net overburden pressure and normally decreases with an increase in depth. (2) Shales with abnormal pressure will have a higher porosity than normally pressured shales at the same depth, because of the greater amounts of interstitial fluids. (3) Sand bodies (confined by lensing, faulting, etc.) surrounded by shale will have a pressure similar t o those in the shales. Data from acoustic and resistivity logs can be used to establish a shale transit time or shale resistivity versus depth of normal hydrostatically pressured formations. Deviation from the derived curve is used t o determine abnormal pressures (Hottmann and Johnson, 1965). The acoustic log is a function of porosity and lithology; therefore, in any given shale sequence it is primarily a measure of porosity. The acoustic response in normally pressured shales decreases in travel time (velocity increases) with increasing depth. This is the “normal compacted trend”, and the pressures in the shale are normal, or hydrostatic. Deviation from the “normal compaction trend” indicates an abnormally pressured zone. Relating the difference in the travel time of the observed formation pressure ( A T , ) t o a normal formation pressure (AT,) t o the formation pressure gradient (calculated from known depths and pressures of wells in the area), a pressure gradient (AT, -AT,) can be determined. The reservoir pressure can be found by multiplying this gradient by the depth. Fertl and Timko (1970) discuss several methods, using the theory of “departure from the normal” t o detect abnormally pressured zones. Methods they discuss are as follows: (1) Bulk density -this is a measurement of the intensity of back-scattered electrons produced by gamma-ray bombardment; this intensity varies with the bulk density of the rocks surrounding the borehole. (2) Conductivity measurements - measure of an induction log. Electromotive forces set up a current, which is detected by a receiver and recorded. Overpressured shales are noted by greater-than-normal conductivity reading resulting from higher-than-normal water content and porosity. (3) Borehole temperature - geopressured as usually associated with an increase in temperature. (4)Presence of gas in mud - this is not always a good detector, for gas can evolve from formation cuttings, as they come to the surface. One of the best means of obtaining subsurface information, other than drilling, is the use of the reflection seismograph. This geophysical tool is a measure of time between the earth’s surface and various subsurface reflecting horizons. The differences in interval velocities between these different horizons (formations) can be used to obtain a plot of average interval travel time, which varies exponentially with depth. The degree of departure from a “normal” plot of this travel time versus depth is related t o abnormally pressured reservoirs in the Gulf Coast area (Pennebaker, 1968). This departure is noted as an increase in the normally
364
GEOPRESSURED RESERVOIRS
decreasing travel time with depth, because of the undercompacted formations. To measure the formation pore pressure, plots of equal pore pressure gradients are compared, by an overlay, to the abnormally pressured interval travel time depth plots. Forgotson (1969),by experience with wells in the Gulf of Mexico, noted that the presence of high background gas and high trip gas, together with lower than normal shale density, does not necessarily indicate the proximity of an abnormally pressured reservoir. He believes that a minimum of 200% increase in the shale penetration rate when drilling is the best available means to predict abnormal pressures. A recent series of papers explains how downhole temperatures and pressures can affect drilling (Fertl and Timko, 1972;Timko and Fertl, 1972). Methods of detecting abnormal pressures, compensating for them, and evaluating the hydrocarbon potential of geopressured strata are discussed.
References American Petroleum Institute, 1968. API Recommended Practice f o r Analysis of Oilfield Waters. Subcommittee on Analysis of Oilfield Waters, API, RP 45, 2nd ed., 49 pp. Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation t o petroleum migration. Bull. A m . Assoc. Pet. GeoL, 53:73-93. Dickey, P.A., Collins, A.G. and Fajardo, I., 1972. Chemical composition of deep formation waters in southwestern Louisiana. Bull. Am. Assoc. Pet. GeoL, 56:1530-1533. Dickey, P.A., Shiram, C.R. and Paine, W.R., 1968. Abnormal pressures in deep wells of southwestern Louisiana. Science, 160:609-615. Dickinson, G., 1953. Geological aspects of abnormal reservoir pressures in Gulf Coast Louisiana, Bull. A m . Assoc. Pet. GeoL, 37:410-432. Fajardo, I., 1968. A Study of the Connate Waters and Clay Mineralogy. M.S. Thesis, University of Tulsa, Tulsa, Okla., 50 pp. Fertl, W.H. and Timko, D.J., 1970. Overpressured formations, 2. How abnormal pressure-detection techniques are applied. Oil Gas J., 68:62-71. Fertl, W.H. and Timko, D.J., 1972. How downhole temperatures, pressures affect drilling. World Oil, 174(7):67-70; 175(1):47-49; 175(2):36-39, 66;175(4):45-50; 176(2): 47-50. Finch, W.D., 1969. Abnormal pressure in the Antelope field, North Dakota. J. Pet. Technol., 21:821-835. Forgotson, J.M., 1969. Indication of proximity of high pressure fluid reservoir, Louisiana and Texas Gulf Coast. Bull Am. Assoc. Pet. GeoL, 53:171-173. Foster, J.B. and Whalen, H.E., 1966. Estimation of formation pressures from electrical surveys - offshore Louisiana. J. Pet. TechnoL, 18:165-171. Fowler, Jr., A.W., 1970. Pressures, hydrocarbon accumulation, and salinities - Chocolate Bayou field, Brazoria County, Texas. J. Pet. TechnoL, 22:411-423. Harkins, K.S. and Baugher, 111, J.W., 1969.Geological significance of abnormal formation pressures. J. Pet. TechnoL, 21:961-966. Hottmann, C.E. and Johnson, R.K., 1965. Estimation of formation pressures from logderived shale properties J. Pet. TechnoL, 17:717-721. Jones, P.H., 1969. Hydrodynamics of geopressure in the North Gulf of Mexico Basin. J. Pet. TechnoL, 21:803-810. Pennebaker, E.S., 1968. Detection of abnormal pressure formations from seismic field records. Presented at API Southern Dist. Meet., San Antonio, Texas, March 6-8, 1968, API Paper, No. 926-13C.
REFERENCES
365
Perry, D.R., 1969. A Correlation of Reserves and Drive Mechanisms with Reservoir Pressure Gradients on Geopressured Gas Reservoirs in Southwest Louisiana. M.S. Thesis, Southwest Louisiana University, Lafayette, La., 54 pp. Powers, M.C., 1967. Fluid-release mechanisms in compacting marine mudrocks and their importance in oil exploration. Bull. Am. Assoc. Pet. GeoL, 51:1240-1254. Schmidt, G.W., 1973. Interstitial water composition and geochemistry of deep Gulf Coast shales and sandstones. Bull. A m . Assoc. Pet. Geol., 57:321-377. Sulin, V.A., 1946. Waters of Petroleum Formation in the System o f Natural Waters. Gostoptekhizdat, Moscow, 96 pp. Terzaghi, K. and Peck, R.R., 1948. Soil Mechanics in Engineering Practice. John Wiley and Sons, New York, N.Y., 56 pp. Timko, D.J. and Fertl, W.H., 1972. How downhole temperatures, pressures affect drilling. World Oil, 175(5):73-81; 175(6):79-82; 175(7):5*62; 176(1):45-48; 176(4):. 62-65. Wallace, W.E., 1969. Water production from abnormally pressured gas reservoirs in South Louisiana. J. Pet. Technol., 21 :969-982. Young, A. and Low, P.F., 1965. Osmosis in argillaceous rocks. Bull Am. Assoc. Pet. Geol., 49:1004-1008. \
This Page Intentionally Left Blank
Chapter 12. COMPATIBILITY OF OILFIELD WATERS
Waters used for the secondary recovery of oil by waterflooding usually contain a number of inorganic salts and sometimes organic salts in solution. It is common practice t o test the compatibility of the injection water and water in the formation before starting a waterflood operation. Often this test is performed by mixing the injection water with the formation water in a glass container and observing t o determine if a precipitate forms. The precip itate or scale can be analyzed to determine its composition. Waters are compatible if they can be mixed without producing chemical reactions between the dissolved solids in the waters and precipitating insoluble compounds. The precipitated insoluble compounds are undesirable because they can reduce the permeability of a porous petroleum-productive rock formation, plug input wells in waterflood systems, and cause scale formation in water pumps and lines. Some of the more common ions that frequently occur in oilfield waters and that cause precipitation in incompatible waters are: Ca+2,S P 2 , Ba+2, Fe+?,HC03-, and Common reactions are: CaC1, + Na2S04 CaC1, +MgS04 c a w 0 3 )2 CaC12 + 2NaHC03 SrC1, + NaS04 SrClz +MgSO, BaC12 +NaS04 BaCl2 +MgSO, Fe + H2S Fez03 + 6H2S
+= += -b
+= += += += += -b
+=
2NaC1+ CaSO, MgClz +CaS04 C 0 2 + H 2 0 + CaCO, 2NaC1+ C 0 2 + H 2 0 + CaCO 2NaC1+ SrSO, MgC12 + S r S 0 4 2NaC1+ BaSO, MgC12 +BaS04 H2 + FeS 6H,O + 2Fe2 S3
33J-
33J-1
335.
A relatively insoluble compound CA where C is the cation and A is the anion will precipitate from an aqueous solution if:
where ac = the cation activity, a A = the anion activity in the solution, and
SCA = the solubility product of the compound CA. When two salts with a common cation (CAI and CA2) are in equilibrium in a solution, the following will hold:
36 8
COMPATIBILITY OF OILFIELD WATERS
a~~
I ~ A , = &!Al PCA,
If a A l l a A , a~~
> SCA /SCA,,CA,
will precipitate, and CA2 will dissolve if
I ~ A , <SCA,ISCA,.
Deposition of scale in both primary and secondary recovery producing wells and formations is a very costly problem in the petroleum industry. The scale not only restricts production but also causes inefficiency and production equipment failure. Scale deposits are caused by mixing incompatible waters and by environmental changes during the production of well fluids. For example, as production begins, the pressure drops in the vicinity of the wellbore, allowing dissolved gases to escape from solution. The loss of C 0 2 can cause calcium carbonate t o precipitate. The decrease in pressure also can cause the vapor pressure of the brine to increase. The temperature of the brine will decrease because heat energy is required to vaporize the water, causing calcium sulfate t o precipitate. Wellbore and formation damage In several case studies Vetter and Phillips (1970) found that calcium sulfate deposits form in both primary and secondary petroleum production operations. The scale forms within the formation and causes production loss and permanent damage. In many cases damage to the formation cannot be corrected even by fracturing. Research has indicated that sodium carbonate can cause the metathesis of anhydrite and gypsum to calcium carbonate. This might work in a formation that is partially plugged. For example, a water containing sodium carbonate could be injected into the formation and allowed t o react with the scale. An acidified water then could be injected into the formation to remove the carbonate and hopefully clean the formation, allowing recovery of more oil. Potential scale deposition should be predicted as soon as the well begins production, and the correct inhibitor should be added immediately rather than following the common practice which is to pull the tubing after a production decline and find scale on the metal surface. Scaling can occur within the formation and never show up on the tubing. Pressure drops are the primary cause of calcium sulfate scaling within a formation, when the formation brine is saturated with calcium sulfate. Important variables related t o scaling are: (1)Temperature of the formation in relation to solubility of the possible scale former in the fluids passing through it. CaS04 becomes less soluble
WELLBORE AND FORMATION DAMAGE
369
with increasing temperature (Blount and Dickson, 1969), while BaS04 becomes more soluble (Templeton, 1960). (2) Subsurface pressures change for any system, with the highest pressure found while the fluid flows through the formation. The greatest pressure change is at the sand face of the producing well (Vetter and Phillips, 1970), which causes this area t o be where solubility changes are the greatest. Deposition of scale at this point is the most damaging to oil production and the most difficult t o discover or to remedy. Very few data are available on pressuresolubility relations of most scale forming compounds, but CaSO, has been shown to decrease in solubility with decrease in pressure at NaCl concentrations t o 10%(Fulford, 1968). (3) Brine concentration, exclusive of precipitating compounds, also influences scale formation. Most electrolytes in ionic form cause an increase in the solubility of compounds which form scales. The solubility normally increases with increasing electrolyte concentration unless some other solubility equilibrium is reached. This can occur, for example, when BaS04 saturation level is reduced because of increasing amounts of Ca+2 ion in the solution (Davis and Collins, 1971). Other properties of brine known to influence the solubility levels of scale formers are gases in solution, hydrogenion concentration, ion pairs, and dissolved organic chelates (Weintritt and Cowan, 1967). Waterflooding of petroleum reservoirs has been successfully carried out for many years. Large quantities of petroleum are produced through secondary recovery by forcing water (usually a brine) into an oil sand which has become unproductive by primary production methods. However, the efficiency of the operation is often low, and the amount of petroleum remaining in the sand after waterflooding can be as high as 50% of the original accumulation (Shaffer, 1967). One of the reasons such a high proportion of the oil remains unrecovered is because the injection pressures become economically too large to continue forcing water through the sand t o displace the oil. The gradual deposition of solid material precipitating from the water closes the permeable channels and slows the flow at the producing well. Dilution of the water injected into a formation often occurs, and additional makeup water is necessary. The slow mixing of connate (interstitial) waters of the formation or the introduction of water from associated aquifers, both underground and on the surface, contributes to the instability of the injection water. The deposition of scale in wellbores, sand faces, and piping has reduced oil production in many fields. Removal of scale is difficult, often impossible, and methods t o avoid its formation need additional development. Scaling results from the precipitation of a solid from a formation water or from injection water in waterflood operations. The most common causes of scale are: (1) temperature and/or pressure changes t o which the formation water is subjected; (2) dilution with makeup water (in secondary recovery opera-
370
COMPATIBILITY O F 0ILF IELD WATERS
tions) or mixing with other formation water containing incompatible ions; (3) evaporation causing increased concentrations of dissolved solids allowing saturation t o be reached; (4) supersaturation caused by formation water flowing through and dissolving slightly soluble solids. When the composition and temperature of a brine, saturated with CaSO,, remain constant, precipitation will occur if the pressure drops. Scaling is not likely with increasing pressure under the same conditions. However, these conclusions must be modified if the brine is flowing through beds conkaining soluble compounds of calcium or sulfate or if another water source is altering the brine concentration. Because of moderate rise in brine temperature as it travels betwsen the injection wellhead and the bottomhole and the rapid rise in pressure, scaling of CaS0, is not likely in the injection well. The formation of BaS0, scale is worthy of special attention. Most barium compounds are relatively insoluble, and large volumes of brine often are necessary t o cause heavy BaS0, scale. The most unique characteristic of this compound is its crystal growth (Weintritt and Cowan, 1967). It will remain in a supersaturated solution for an unpredictable time and then will precipitate slowly and in a crystal form which has not been duplicated in the laboratory. Some observers attribute this phenomenon t o the requirement of a unique solid crystal acting as a seed to promote BaSO, precipitation. Furthermore, the forming crystal adheres to other larger solids suspended in the solution or attached to the associated solid phase. This causes the scale to occur in larger quantities than if it were pure barite. In the Raleigh field, Smith County, Mississippi, a scale consisting of concentric rings of prismatic barite commonly occurs. The barite prisms are about 0.5 mm in length and contain up t o 1%strontium and lead. The pumping equipment in the Pisgah field, Rankin County, Mississippi, often is plagued with a scale composed of metallic lead containing small fragments of steel. The steel is from the pumping mechanism but the lead must be from the formation water because the amount of dissolved lead ranges up t o 100 mg/l. Maintenance of the wells to remove the lead scale occurs as often as every 10 days. Knowledge of the solubilities of BaS0, and SrS0, in solutions containing NaCl, CaCl,, and NaHCO, needs to be increased t o better understand various precipitation reactions that occur when waters containing these salts mix. Information concerning the effects of heat and pressure upon these reactions is lacking. Solubility of calcium compounds in various salt solutions Frear and Johnston (1929) measured the solubility of calcite in water at saturated with carbon dioxide and obtained an activity of 4.8 x 25OC. Ellis (1963) determined that the solubility of calcite was significantly less in the laboratory salt solutions than in hydrothermal solutions with similar ionic strength.
SOLUBILITY OF CALCIUM COMPOUNDS
371
Stiff (1952) developed a graphic method of predicting the tendency of oilfield waters to deposit calcium sulfate. Diagrams can be used to find the maximum solubility of a salt in waters of similar composition. This information is useful in predicting that a given brine has a scale forming tendency. However, better pressure and temperature data in respect t o their effect on scale formation are needed. Akin and Lagerwerff (1965) studied the solubility of calcite in relation t o ionic strength. The soluble salts used were NaC1, NaHCO,, and CaC12; the ionic strength of the solutions ranged up t o about 0.09 and their data agreed well with the Debye-Huckel theory. They also studied the effect of Mg" and S04-2 and found that the solubility of calcite was enhanced by these ions relative to theoretical values. Ostroff and Metler (1966) determined the solubility of calcium sulfate dihydrate in the system NaC1-MgC12-H20 in 5.50 molal NaCl and 0.340 molal MgC12 admixtures at 28", 38', 50°, 70°, and 90'C. Their results indicate that the solubility increases in the presence of small amounts of MgC12 in NaCl solutions up t o about 2.5 molal NaC1. The MgC12 effect decreases in higher molalities of NaCl until at about 4 mold NaCl a plateau is reached. Shaffer (1967) studied the solubility of gypsum in sea water and sea-water concentrates. He found that the solubility product of gypsum is greater in the highly concentrated brines and also that in these brines it increased with increasing temperature. Glater et al. (1967) developed a method t o measure calcium sulfate scaling thresholds in saline water samples at 100'C. They found a correlation of ionic strength with calcium sulfate solubility, and used a gSaphical method to relate scaling threshold to the concentration of calcium and sulfate ions in saline water. Pytkowicz et al. (1967) measured in situ the pressure coefficient of the aragonitic oolites with pH electrodes, Their results indicate that the pressure coefficient or the apparent solubility must be known to obtain accurate solubility data at high pressures. Fulford (1968) found that the solubility of gypsum or anhydrite increases with pressure because of a small decrease in total volume as the scale dissolves. Subsequently with a pressure drop a supersaturated solution forms and gypsum precipitates. In very concentrated brines this does not occur because the solubility of gypsum in very concentrated brines is less dependent upon pressure. He presented several equations t o calculate anhydrite and gypsum solubilities. Blount and Dickson (1969) determined the solubility of anhydrite in NaCl solutions at 100'-450'C and 1-1,000 bars. They found that anhydrite solubility increased with temperature and NaCl concentrations. Glew and Hames (1970) determined the solubilities cf gypsum, disodium pentacalcium sulfate, and anhydrite in sodium chlor,de solutions. Their results indicated that the solubilities of these comr#ounds decreased in chloride solutions with molalities greater than 3.5.
372
COMPATIBILITY OF OILFIELD WATERS
Vetter and Phillips (1970) included the effects of complicated downhole phase equilibria to develop an improved thermodynamic method to predict deposition of calcium sulfate. According to them the calculated solubility is as accurate as the experimentally determined solubility; however, additional data are needed concerning the solubility of gypsum in brines at high pressures. These data are needed to determine which CaS04 compounds are formed under high pressure in brines. Knowledge of pressure drops either at the wellhead or within the reservoir is important to determine where scale deposition occurs. Solubilities of the sulfates of barium and strontium in saline solutions Neuman (1933) published results of studies of BaS0, solubility in aqueous solutions of potassium, magnesium, and lanthanum as chlorides and nitrates. His data show that BaS04 solubility increases with the increasing complexity of the major solute, and in the order (3, -1) > (2, -1) > (1,-1) of equal molality solutions. Gates and Caraway (1965) analyzed California oil-well scale and found in a BaS0,-type scale significant amounts of strontium along with iron, calcium, magnesium, and some carbonate. Weintritt and Cowan (1967) studied the unique characteristics of BaS04-scale deposition and concluded, “the presence of strontium in barium sulfate scales deposited from oilfield waters appears to be common.” All of the sulfate deposits analyzed contained strontium sulfate in concentrations ranging from 1.2 to 15.9%, and barium sulfate in concentrations ranging from 63.7 t o 97.5%. Templeton (1960) studied the solubility of BaS04 in solutions at 25OC and at sodium chloride molalities between 0.1 and 5.0. He found that at constant ionic strength the solubility of BaS0, increases with increasing temperature, and observed that calcium sulfate exhibits an inverse reaction with increasing temperature. Experimental determination of some solubilities of the sulfates of barium and strontium
A radioisotope-tagged solution of Na2SO4 was prepared from which aliquots were taken (Davis and Collins, 1971). The radioactive isotope was 35S.One aliquot was used to precipitate BaS04 by addition of an excess of BaCl,; a second portion was used to precipitate SrS04 by addition of appropriate equivalents of SrC12.These suspensions were stirred and allowed t o settle. Following a 2 4 t o 48-hour settling period, the precipitates were washed onto a 0.45-pm pore size filtering medium, and the washings were continued until the sulfate ion in the filtrates could not be further reduced. The tagged precipitate was removed from the filter, dried in an oven at 105OC, and transferred t o storage vials. Standard samples of the sulfates were prepared by chelation in a 0.W solution of EDTA. Various strengths of
RESULTS AND DISCUSSION
373
5-80 mg/l of BaSO, and 50-800 mg/l of SrS04 were made and used as reference counting samples for all of the sulfate determinations. A nonionic detergent (Triton X-100) and toluene emulsion (Patterson and Greene, 1965) were prepared, whereby l-cm3 sulfate samples in brine could be counted with greater than 20%efficiency. The emulsion forms a clear gel and permits a homogenous dispersion of the aqueous phase in the fluor with no salting out. Solutions of various salts, such as those usually found in formation waters, were made up in strengths of 0.005-1.77 molal, and tagged solid barium or strontium sulfate was added. The chlorides of sodium, calcium, magnesium, and potassium were prepared, as were solutions of sodium bicarbonate, sodium borate, and potassium bromide. All of the solutions were stored in plastic bottles. To determine sulfate solubility, a 20-cm3 portion of one of the prepared salt solutions was transferred to a small plastic stoppered vial, and 0.1 g of the solid, tagged sulfate was added. This suspension was shaken in a wrist-action type shaker for 72 hours and then allowed t o settle a minimum of 24 hours without opening the vial. The samples were prepared in duplicate to assure equilibrium, and the operation was repeated when better precision was needed. The temperature of the suspension was raised briefly above the stabilized room temperature (25OC k l 0 C ) with a heat lamp during the shaking period, but no change was permitted during the last 24 hours nor during the settling period. When the sample container was opened, it was quickly filtered through a double Whatman No.42 filter paper, and 1 ml was transferred t o a counting vial which contained 12 ml of the Triton emulsion and 7 ml of deionized water. The sample then was counted in a liquid scintillation counter for 50 minutes. The chelated sulfate standards were counted in the same time period. By this method, the correction for radioactive decay could be omitted and the soluble sulfate values determined from a graph of the chelated standards (in mg/l) versus the counts per minute. Barium was analyzed by emission spectroscopy, but adequate precision at levels of 1 mg/l and less was difficult to achieve in the presence of ionic-strength salts encountered in some solutions. Results and discussion of the experimental investigation The values obtained from solubility measurements are shown in Table 12.1. The amounts of the alkaline sulfates which dissolve in other electrolyte solutions are tabulated alongside the total ionic strength of each solution. Ionic strength is the most useful concept yet developed t o include the combined effects of the activities of several ionic species in a solution. Lewis and Randall (1923) state, “in dilute solutions, the activity coefficient of a given strong electrolyte is the same in all solutions of the same ionic strength.” It is defined as s = H m 1 2, ’, where m 1 = the ionic molality, and 2 , = the charge of the ion in solution, the summation being taken over all ions, positive and negative. By definition, the activity of the dissolved species approaches the concentration value (molality) a t infinite dilution.
374
COMPATIBILITY OF OILFIELD WATERS
Since the thermodynamic solubility product Ku = x mso, x y2 and since y equals unity at zero ionic strength, a plot of log mso, versus the ionic strength function would extrapolate to zero concentration where log KuS = log rnso,. Fig. 12.1-3 give plots representing six electrolytes and the values of Ku'h are determined graphically. The value for the sulfate solubilities in pure water was determined experimentally and agreed with values in the literature. In Fig. 12.1, the plot of BaS04 solubility versusds for the six electrolytes is almost identical at low ionic strength, a phenomenon to be expected from the statements above. The extrapolated KuS values of all systems are 1.05 x loe5 (within experimental limits). This fact must be correlated with the nature of the equation defining ionic strength. The square of the ionic valence gives the Mg+' and Ca+' ions four times the numerical weight of the Na+ and K+ ions. Molality values would indicate that the bivalent ions cause increased solubility effects. Because borates are present in many oilfield waters, sodium borate was included t o find differences in sulfate solubility in electrolytes containing a complex ion. As shown in Fig. 12.1 and 3, the solubility deviated from that of monatomic electrolytes, and the relationship described does not hold at higher solubilities of electrolytes containing complex ions. Another ion commonly found in mineral waters is bicarbonate. Many water-bearing zones contain limestone and dolomite which slowly erode in water of low pH. The water carries away carbonates and bicarbonates. In this study, NaHCO, solutions of 0.005-1.0 molal were saturated with tagged
-
10.0
0.0-
KEY -
6.0-
0
-
4.0
-
A
-
0
A
I
0 x
0
2.0 -
X
CoC12 MgC12 No2 84 07 KBr NaCl KCI
__
-
-
0
0 v)
0
m lI.
0
k
1.0
0+
0.8
SO4 Mo Io I i t ies
1
0.6
_I
a 0.4 J 0
H
0.2
0.I
0
0.5
I .o
I .5
2.0
2.5
4 l O N l C STRENGTH
Fig. 12.1. Concentration of saturated B a s 0 4 in strong electrolyte solution.
375
RESULTS AND DISCUSSION TABLE 12.1 Solubility of Bas04 and SrS04 in electrolyte solution Major solute (molality)
5r504
Bas04 major solute system Bas04 (mg/l)
major solute system total ionic strength
SrS04 (mg/l)
total ionic strength
ca Cl2 0.010 0.015 0.020 0.025 0.045 0.050 0.090 0.100 0.136 0.200 0.226 0.300 0.400 0.456 0.500 0.934 1.000 2.000
(5.0)*
6.2
0.03016
7.6
0.06013
11.5
0.15020
15.5
0.30027
17.7
0.60030
16.2 16.3
0.90028 1.20028
16.6
1.50028
(16.3)* (10.8)*
11.3 2.5
3.00019 6.00004
(2.9)*
5.4
0.02979
6.9
0.05982
(17.3)*
214 247
0.0347 0.0504
295 260
0.0815 0.1403
508
0.2819
590
0.4197
757
0.6959
1,152
1.3947
1,942
1.8438
172 203 233
0.0188 0.0344 0.0499
295 394
0.0805 0.1571
422
0.2324
530
0.2866
731
0.7782
1,063
1.5993
M m 2
0.005 0.010 0.015 0.020 0.025 0.049 0.050 0.074 0.099 0.125 0.196 0.254 0.474 0.525 0.902 1.637
*
(44.5)*
9.8
0.14927
13.3
0.29693
18.0
0.58831
25.9
1.42244
32.0 33.2
2.70655 4.93257
See footnote at end of Table.
-
COMPATIBILITY OF OILFIELD WATERS
376 TABLE 12.1 (continued) Major solute (molality)
Bas04 (mg/l)
Na Cl 0.010 0.015 0.020 0.025 0.050 0.086 0.100 0.172 0.200 0.257 0.431 0.500 0.869 1 .ooo 1.771 2.000 KCl 0.010 0.015 0.020 0.025 0.050 0.067 0.100 0.200 0.202 0.338 0.500 0.684 1.000 1.396 2.000
*
5r504
Bas04 major solute system
major solute system total ionic strength
SrS04 (mg/l)
total ionic strength
134 149
0.0129 0.0182
172 199 265
0.0288 0.0543 0.0914
332
0.1756
420 525
0.2667 0.4423
699
0.8840
760
1.7875
144 169
0.0131 0.0185
167
0.0286
375
0.0754
396 502
0.2109 0.3492
742
0.7001
802
1.4139
(5.3)*
3.6
0.01006
(5.6)*
4.2
0.02007
5.4
0.05009
(7.3)*
7.1
0.10012
(11.3)*
10.0
0.20017
14.8
0.50025
(22.3)*
20.2
1.00035
(35.7)*
27.2
2.00047
4.2
0.01007
4.9
0.02008
6.3
0.05011
8.6 11.2
0.10015 0.20019
(3.7)*
(25.8)*
16.8
0.50029
21.6
1.00037
27.2
2.00047
See footnote at end of Table.
BaS04. However, only trace amounts of barium were found in solution, though the sulfate content increased with the amount of NaHCO, in solution. This apparent anomaly can be reconciled by the ionization of the HC03- ion into CO,-*, which in appreciable concentration would reduce the Ba+2 ion concentration according to the solubility product K,, = M B x~ MC03.
RESULTS AND DISCUSSION
377
TABLE 12.1 (continued) Major solute (molality)
(3.6)*
(23.9)*
major solute system total ionic strength
SrS04 (mg/l)
total ionic strength
4.1
0.01007
152 163
0.0133 0.0186
4.7
0.02008 215
0.0467
6.3
0.05011 262
0.0900
8.2
0.10014 320
0.1335
11.0
0.20019 420 509
0.2207 0.4378
669
0.8812
320
0.0462
600
0.1307
690
0.2100
114
0.0025
Bas04 (mg/l) XBr 0.010 0.015 0.020 0.042 0.050 0.084 0.100 0.126 0.200 0.211 0.426 0.500 0.866 1.000 2.000
5r504
Bas04 major solute system
16.2
0.50028
21.8 26.7
1.00037 2.00046
6.0
0.03010
7.8
0.06014
12.9
0.15022
21.0 34.8
0.30036 0.60060
2.5
0.00004
Na2 B4 0 1
0.010 0.013 0.020 0.039 0.050 0.065 0.100 0.200 Pure Water
*
(23.7)* (33.9)*
Parentheses indicate barium ion and sulfate. ion determinations made separately.
The effect of high concentration of CaClz on BaS04 solubility is indicated by the solid curve of Fig. 12.1. At concentrations of 2 molal, the BaS04 solubility has dropped t o values close to that of the compound in pure Hz0. The maximum value is reached between 0.2 and 0.4 molal where the decline begins. The accompanying broken line of Hg. 12.1, which is a plot of Ba+2 + SO4-' ions determined separately, shows the reduced solu-
378
COMPATIBILITY OF 0IL F IE LD WATERS
-1 IONIC
STRENGTH
Fig. 12.2. Concentrations of saturated SrS04 in strong electrolyte solutions of NaCl, KCl, and KBr.
-
- K a = 2.4 x I
I
I
4 IONIC
I
I
I
I
STRENGTH
Fig. 12.3. Concentrations of saturated SrSO4 in strong electrolyte solutions of MgClz, CaClz , and Na2 B4 0,.
RESULTS AND DISCUSSION
379
*
bility of the Ba ion caused by the equilibrium Ca+’ + S04-2 CaSO, ( K s p 10-4). The effect of the ions of strong electrolyte solutions on SrS04 solubility is similar t o that observed when BaS04 solubility was studied. The Na+, K+, C1-, and Br- ions have approximately equal effect, and all determined values fall on a common curve (Fig. 12.2). The increase in sulfate solubility is marked in dilute solutions but reaches a maximum a t concentrations with ionic strength near 1. This is the average value calculated for sea water. When bivalent ions Mg+’ and Ca+2 are used in the strong electrolyte (Fig. 12.3), the SrS04 solubility remains of the same relation t o the total ionic strength as for the monovalent ions. A study of the system SrS04-NaHC03 -H2 0 was limited by the insolubility of SrCO, . The ionization of the bicarbonate to H+ and COSw2would result in the precipitation of any Sr+’ which dissolves and leaves the S04-2 in solution. This relationship is similar t o the BaS04-NaHC03 -H2 0 system and is worthy of special note. That is, when carbonate or bicarbonate waters are diluted or intermixed with waters containing barium or strontium, an unstable solution is formed. Experimental data indicate that maximum sulfate solubility in strong electrolytes begins at an ionic strength of approximately 1.When the principal cation in solution is the Ca+2 ion, sulfate solubility decreases after the ionic strength exceeds unity. Blount (1965), when measuring solubility of CaSO, in the system CaS04-NaClLH2 0, and Lucchesi and Whitney (1962),
< 1.95
TABLE12.11 Sulfate solubilities in synthetic brines ~
~
~~~
Concentration (molal) brine 1
brine 2
brine 3
Br1-
1.2179 0.0250 0.0206 0.0051 1.3019 0.0125 0.0001
1.7399 0.0374 0.0823 0.0051 1.9650 0.0188 0.0000
2.4359 0.0499 0.0411 0.0193 2.6113 0.0250 0.0001
Barium sulfate solubility Bas04 (mg/l) Bas04 (molality) Ionic strength (s)
60 2.57 x D4 1.3600
63 2.70 x D4 2.1038
66 2.83 x 3.0278
Strontium sulfate solubility SrS04 (mg/l) SrS04 (molality) Ionic strength (s)
813 44.26 x lo4 1.3777
922 50.19 x lo4 2.1239
Na+ Ca+’ Mg+’ K+
c1-
,
o4
958 52.18 x lo4 3.0487
380
COMPATIBILITY OF OILFIELD WATERS
when measuring SrS04-NaC1-H2 0 solubility equilibria, found similar maximums. By using ionic-strength calculations in place of weight per unit volume, the predictions of mineral water stability become more accurate and dilutions more feasible. Three synthetic brines were made with salts concentrations in the range of many formation waters and containing the major salts found in these waters. Table 12.11 gives these concentrations and results of a BaS04 and a SrS04 solubility determination. The values found when plotted against the ionicstrength function of the brine fall on the same curve as the barium salt in Fig. 12.1 and the strontium salt in Fig. 12.2. No carbonates were added t o these synthetic brines.
Brine stabilization Efforts to stabilize the brines used in petroleum production have been extensive and successful in many cases, but the complexity of the problem in other cases is reported. Water treating units are considered necessary in waterflooding operations, but none fully satisfy the operator’s apprehension that there may be plugging within the reservoir. Addition of solubilizing, chelating, and clarifying agents to the brine has helped, but economics limit the quantities used. Tests for compatibility of the fluids as they exist in the formation and in the wellbore give erroneous results because the subsurface environment cannot be fully duplicated a t the surface. To aid brine stabilization programs, several studies of the solubilities of various relatively insoluble compounds have been made as previously discussed. Usually the results of these studies are reported as solubility products of various pure compounds (CaS04, BaS04, CaCO, , etc.) in the presence of other ions, dissolved gases, ion pairs, and various sized crystals of the compound under study. Some efforts have been made using mixed cations and anions in solution with the compound under study, including limited study of sulfates in sea water or synthetic sea water (Shaffer, 1967). Various easily measured parameters such as percent chlorides, total solids, and ionic strength have been plotted against solubility product of the potential scale former. Very little correlation suitable for direct field application has been found. For example, the author has measured BaS04 solubility in CaC12, MgC12, NaC1, and other salt solutions using ionic strength as the common property. However, a synthetic sea water containing these compounds and having comparable ionic strength will dissolve double the weight of BaS04 in milligrams per liter with respect to any solution containing a single salt. Fulford (1968) and Vetter and Phillips (1970) proposed useful formulas and graphs t o use in predicting scaling from calcium sulfate. Fig. 12.443 are included for possible use in predicting potential scale problems from calcium sulfate, strontium sulfate, and barium sulfate. The figures are plots of molal solubility versus ionic strength. The advantage of this plot is that the ionic strength of any given water can be calculated from its chemical composition,
BRINE STABILIZATION
381
IONIC S T R E N G T H
Fig. 12.4. Solubility o f CaSO4 versus ionic strength of aqueous solutions (Ostroff and Metler, 1966).
t
y
12-
KEY
1 IONIC S T R E N G T H
Fig. 12.5. Solubility of SrS04 versus ionic strength of aqueous solutions containing CaClz, MgClz, NaCl, KCl, and KBr (Davis and Collins, 1971).
and the solubility of a given compound is a direct function of the ionic strength of the solution. Therefore, a very good approximation of the solubility of a given compound in a given water solution can be made. For example, if a water with an ionic strength of 0.1 contains 0.001 molal of strontium sulfate, it can be assumed that the water is undersaturated with respect to strontium sulfate as illustrated in Fig. 12.5. However, if the water contains 0.003 molal of strontium sulfate it is oversaturated and some treatment should be made if the water is to be reinjected.
382
COMPATIBILITY OF OILFIELD WATERS
01 I a I I ,111 0.01 0.02 0.040.0601 I
I
,
a2
I
I
I IIII
04 060.0 1.0
I
2
,
I , 1 , 1 ,
4
6 8 3
IONIC STRENGTH
Fig. 12.6. Solubility of B a s 0 4 versus ionic strength of aqueous solutions containing CaClz, MgClz, NaCI, KCl, and KBr (Davis and Collins, 1971).
Similar curves can be made using appropriate solubility data for calcium carbonate and for iron compounds. However, it should be noted that application of this technique only gives an estimation of the maximum solubility of a compound in waters of similar ionic compositions. Better data on pressure and temperature and how they affect the solubilities are needed before adequate prediction equations can be developed. Mixing of subsurface waters Mixing of surface and subsurface waters results in solutions which are either saturated or undersaturated with relatively insoluble compounds such as calcium carbonate, calcium sulfate, strontium sulfate, and barium sulfate. These compounds are considered because they often are found in scales formed because of mixing of formation waters. Hydrodynamic potentials caused by differences in elevation, weight of the overlying fluids and rocks, secondary cementation of rock pores (Levorsen, 1967), temperature differences, osmotic pressures, and chemical and physical reactions cause subsurface waters t o move (Hubbert, 1953).Popov and Goldshteyn (1957) described a large hydrodynamic system of descending fresh water and ascending saline water which could mix to form a fresh-saline water mixture. Henningsen (1962)found that recharge waters into Trinity aquifers were two types of water from strata of different lithology and with basinward movement of the waters they mixed to form a third type of water. Mixing of fresh waters with encroaching sea water occurs according t o Kohout (1960),Columbus (1965),and Upson (1966).
MIXING OF SUBSURFACE WATERS
383
Estimating strontium sulfate saturation in waterflood makeup brines (Biles, 19 72) The data in Tables 12.1 and I1 were used by Biles (1972)to estimate the saturation point of strontium sulfate in waterflood makeup waters. According to him, brines used as makeup water for waterflood operations often are more concentrated in dissolved solids than are the single solute samples shown in Table 12.1. However, considering that the sodium concentrations are 96, 93,and 97 mole %, respectively, in the synthetic brines 1, 2, and 3 shown in Table 12.11, it appears reasonable in lieu of experimental data to extend the NaCl data to the higher.concentration range with these data. Fig. 12.7 is a plot of the milligrams per liter of strontium sulfate in solution as a function of the total ionic strength of the solution. The data were taken from Table 12.1 and 11. A smooth curve can be plotted for Fig. 12.7 if the strontium sulfate value at 1.7875 total ionic strength (Table 12.1) is ignored. This curve can be extrapolated for use in estimating the amount of strontium sulfate in milligrams per liter that is likely t o be soluble in more concentrated brines. A similar curve could be plotted for the solubility of barium sulfate. Consider a brine that does not contain the stoichiometric combining weight ratio of strontium and sulfate as shown in Table 12.111.To compare the amount of strontium sulfate apparently at equilibrium in this brine with the solubility data in Tables 12.1 and 11, it is necessary to use another approach. The solubility product of a solute A, Bm is determined by the molalities of the ions composing the solute and their activity coefficients:
"--"I
L
z 0 1.200I4
a
I I 1)
I I 2.0 3.0 TOTAL IONIC STRENGTH
3
Fig. 12.7. Solubility of strontium sulfate versus ionic strength of the solution (J. Biles, written communication, Cities Service Oil Company, Tulsa, Okla., 1972).
COMPATIBILITY OF OILFIELD WATERS
384 TABLE 12.111
Composition of a brine that does not contain a stoichiometric combining weight ratio of strontium and sulfate* Ion Na+
mg/l
s04-'
49,000 220 11,500 2,400 25 1,000 101 106,140 170
Total
170,556
K+ ca+' Mg+' Ba+'
sr+' Fe"
c1-
*
me/l
Molalit y
2,130 6 574 197 < 1 23 4 2,990 4
2.24 0.00589 0.303 0.104 0.00018 0.012 0.00189 3.15 0.00189
Total ionic strength = 3.54;density at 22OC = 1,120 g/l; grams H20/1= 950.
The activity coefficients are determined primarily by the total ionic strength of the solution, and in a solution saturated with the solute A, B, :
If the total ionic strength is unchanged, Y B and ~ [K,,/(YA~ Y B ~ ) are ] constant. Therefore, the concentration of A in equilibnum with a given concentration of B in a saturated solution of A, B, is defined:
Plotting the data in Tables 12.1 and I1 for the solubility of SrS04 in sodium chloride and synthetic brine solutions as the product of the molalities of strontium and sulfate versus total ionic strength, as shown in Fig. 12.8,indicates that the brine in Table 12.111 is undersaturated in SrS04 by d(280 x lo-') - (227 x lo-') molal. This method is in error to the extent that the SrS04 solubility is affected differently by the ions in the brine in Table 12.111 than by the ions in the brines shown in Table 12.11. Nevertheless, this approach is valuable in that a reasonable estimate can be made of the degree of undersaturation of SrS04. Now consider the advisability of mixing the brine shown in Table 12.111 with another brine which contains less dissolved solids and a comparable percentage of cations as sodium and about 1,850 mg/l of sulfate. To determine the solubility of SrS04 in various mixtures of waters, the product of the weighted average molalities of strontium and sulfate was determined for
MIXING OF SUBSURFACE WATERS
240
385
-
200 -
KEY In NaCl solutions A In synthetic brines 0
3 TOTAL IONIC STRENGTH
Fig. 12.8. Solubility of strontium sulfate as a product of the molalities of strontium and sulfate versus the ionic strength of the solution (J. Biles, written communication, Cities Service Oil Company, Tulsa, Okla., 1972). Filled square shows the product of the molalities of (Sr )(SO4 ) in Table 12.111 brine.
6 t J
9 E
a aw.
z
600
0 Ia 500 a
-514
g / kiloliter
3
I-
3 a 2
* 200
0
?? m
100
100 TABLE 12.X BRINE,
percent
Fig. 12.9. Plot of the supersaturation of a mixture of the brine shown in Table 12.111 with a brine containing 1,850 mg/l sulfate versus the Table 12.111 brine in percent (J. Biles, written communication, Cities Service Oil Company, Tulsa, Okla., 1972).
386
COMPATIBILITY OF OILFIELD WATERS
each mixture and compared with the comparable values in Tables 12.1 and I1 and Fig. 12.8. It was determined that a maximum supersaturation of 514 g/kl occurred when the mixture contained 60% of the brine shown in Table 12.111 as illustrated in Fig. 12.9. Mixtures containing less than 9%and more than 97% of the brine shown in Table 12.111 were undersaturated in &SO4 when mixed with a brine containing 1,850 mg/l of sulfate. The same error mentioned in the above paragraph will be present, but the correction would not greatly affect the value obtained using solubility data from Table 12.11. References Akin, G.W. and Lagerwerff, J.V., 1965. Calcium carbonate equilibria in aqueous solutions open t o the air, I. The solubility of calcite in relation to ionic strength; 11. Enhanced solubility of CaC03 in the presence of Mg” and SO4’-. Geochim. Cosmochim. Acta, 29: 343-360. Blount, C.W., 1965. The Solubility o f Anhydrite in the Systems C a S 0 4 - H z 0 and CaS04-NaCl-H2 0 and Its Geologic Significance. Ph.D. Dissertation, University of California, Riverside, Calif., 179 pp. Blount, C.W. and Dickson, F.W., 1969. The Solubility of anhydrite (CaS04) in NaCl-H’O from 100 t o 45OoC and 1 to 1000 bars, Geochim. Cosmochim. Acta, 33:227-245. Columbus, N., 1965. Viscous model study of sea water intrusion in water table aquifers. Water Resour. Res., 1:318-323. Davis, J.W. and Collins, A.G., 1971. Solubility of barium and strontium sulfates in strong electrolyte solutions. Environ. Sci TechnoL , 5:1039-1043. Ellis, A.J., 1963. The solubility of calcite in sodium chloride solutions at high temperatures. A m . J. S c i , 261:259-267. Frear, G.L. and Johnstonb J., 1929. Solubility of calcium carbonate (calcite) in certain aqueous solutions a t 25 J. A m . Chem. SOC.,51:2082-2093. Fulford, R.S., 1968. Effects of brine concentration and pressure drop o n gypsum scaling in oil wells. J. Pet. Technol., 20:559-564. Gates, G.L. and Caraway, W.H., 1965. Oil well scale formation in waterflood operations using ocean brines, Wilmington, Calif. US.Bur. Min. Rep. Invest., No.6658, 28 pp. Glater, J., Ssutu, L. and McCutchan, J.W., 1967. Laboratory method for predicting calcium sulfate scaling thresholds, Environ. Sci Technol, 1:41-52. Glew, D.N. and Hames, D.A., 1970. Gypsum, disodium pentacalcium sulfate, and anhydrite solubilities in concentrated sodium chloride solutions, Can. J. Chem., 48:3734-3738. Henningsen, E.R., 1962. Water diagenesis in Lower Cretaceous Trinity aquifers of Central Texas. Baylor Univ. Geol. Studies, Bull., 3~38. Hubbert, M.K., 1953. Entrapment of petroleum under hydrodynamic conditions, Bull. A m . Assoc. Pet. Geol., 37:1954-2026. Kohout, F.A., 1960. Cyclic flow of salt water in the Biscayne aquifer of southeastern Florida, J. Geophys. Res., 65:2133-2141. Levorsen, A.I., 1967. Geology o f Petroleum (revised by F.A.F. Berry). W.H. Freeman, San Francisco, Calif., 724 pp. Lewis, G.N. and Randall, H.M., 1923. Thermodynamics. McGraw-Hill, New York, N.Y., 723 pp. Lucchesi, P.J. and Whitney, E.D., 1962. Solubility of strontium sulfate in water and aqueous solution of hydrogen chloride, sodium chloride, sulfuric acid and sodium sulfate by the radiotracer method. J. AppL Chem. (London), 12:277-279. Neuman, E.W., 1933. Solubility relations of barium sulfate in aqueous solutions of strong electrolytes. J. Am. Chem. SOC.,55:879-884.
.
REFERENCES
387
Ostroff, A.G. and Metler, A.V., 196:. Soltbility of calcium sulfate dihydrate in the system NaCl-MgC12-H20 from 28 to 70 C. J. Chem. Eng. Data, 11:346-350. Patterson, M.S. and Greene, R.C., 1965. Measurement of low energy beta-emitters in aqueous solution by liquid scintillation counting of emulsions. Anal. Chem., 37 :85 4-85 7. Popov, A.I. and Goldshteyn, R.I., 1967. Hydrologic zoning of hydrostatic systems as a mineralizing factor in the stratal cover of Central Asia. Dokl. Akad. Nauk S.S.S.R., Earth Sci Sect., 17:118-120 (transl.). Pytkowicz, R.M., Disteche, A. and Disteche, S., 1967. Calcium carbonate in sea water at in situ pressures. Earth Planet. S c i Lett., 2:430-432. Shaffer, L.H., 1967. Solubility of gypsum in sea water and sea water concentrates at temperatures from ambient t o 6 5 C. J. Chem. Eng. Data, 12:183-188. Stiff, H.A., 1952. A method for predicting the tendency of oilfield waters to deposit calcium sulfate. AIME, Pet. Trans., 195:25-28. Templeton, C.C., 1960. Solubility of barium sulfate in sodium chloride solutions from 25' to 95'C. J. Chem. Eng. Data, 5:514-516. Upson, J.E., 1966. Relationships of fresh and salty groundwater in the Northern Atlantic Coastal Plain of the United States. U.S. Geol. Surv. Prof. Paper, No.550-C, pp. 2 35-2 4 3. Vetter, O.J.G. and Phillips, R.C., 1970. Prediction of deposition of calcium sulfate scale under down-hole conditions. J. Pet. TechnoL, 22:1299-1308. Weintritt, D.J. and Cowan, J.C., 1967. Unique characteristics of barium sulfate scale deposition. J. Pet. TechnoL, 19:1381-1394.
This Page Intentionally Left Blank
Chapter 13. VALUABLE MINERALS IN OILFIELD WATERS
In the early days of the oil industry, oilfield brines were allowed t o flow by natural drainage into streams until it was noted that some of the once good fishing streams contained less fish. Fur-bearing animals had disappeared in these areas and dead trees and barren soils now bordered these same streams that once had luxurious vegetation. A few years prior t o 1935, litigation pertaining t o pollution of fresh water was taking a heavy toll from oil operators. In certair, older oil producing areas, extensive plots of ground still are barren, with no living vegetation. The litigations against oil operators combined with legislation for fresh-water protection to force better disposal techniques. At first, evaporation ponds were employed; however, usually more brine drained into fresh-water aquifers than evaporated. Until recently a widely employed practice for disposal was the dumping of oil brines into salt-water bodies when they existed nearby. This disposal method was practiced along the Gulf of Mexico and in California. Authorities in these areas insisted that oil separation be highly efficient to prevent damage to fish and oyster populations. Recently the State pollution boards have ruled that oilfield brines can no longer be dumped into surface salt-water bodies. In California excess oilfield waters are being injected into porous subsurface formations as rapidly as the injection systems can be constructed. The Plains States are not only situated in a hard water beit, but seldom have they had an overabundance of usable or surface ground waters. For this reason, State legislatures passed laws for the protection of fresh-water supplies, allowing the return of oilfield brines t o subsurface formations and allowing the repressuring or waterflood of oil properties with salt water. Subsurface brine disposal has since become the common practice. Since the laws were passed to allow subsurface disposal, more legislation has both forced such disposal and set up tight controls for it. A survey of cost data on subsurface injection in 1968 showed that subsurface disposal costs ranged from 6.6 to 19.8 cents per m3. These figures were based on operating costs plus 5-year amortization. Costs vary with the amount of treatment necessary before injection, the number of production wells per injection well, and the costs of drilling injection wells or the depth of the injection formation. The depths of disposal wells normally encountered required no injection pressure. The brines flow readily into the receiving formations under the gravity head alone.
390
VALUABLE MINERALS IN OILFIELD WATERS
Most of the 1.23 billion m3 of saline water that is produced yearly with petroleum is an expense t o oil producers even though some of these waters contain salts yielding valuable elements which might be economically recovered (Angino, 1967). Elements found in some brines in economic concentrations are magnesium, calcium, potassium, lithium, boron, bromine, and iodine. Many of them are recovered by chemical companies from sea water, salt lakes, and subsurface saline waters (Collins, 1966; Brennan, 1966). The recovery of minerals from saline waters dates back to the first time that someone precipitated a compound from a salt solution. Precipitation is the most used separation process employed in separating minerals from sea water or subsurface brines. Research continued on the separation methods which show economic promise in mineral separation from saline waters. The Office of Saline Water, U S . Department of the Interior, supports research aimed at mineral recovery processes to be integrated with freshwater plants. The object of this research is to reduce the cost of the produced fresh water by selling the extracted minerals at a profit. Now consider mineral recovery as a means of reducing the cost of oilfield brine disposal. There are additional advantages t o mineral removal other than profits from the sale of the mineral. For instance, magnesium in sea water causes great expense because of scale formation in fresh-water plants. Recovery of iodine and bromine from oilfield brines
Iodine Iodine consumption in the United States exceeds domestic production. The Dow Chemical Company is the sole domestic producer of iodine. 75% of our domestic consumption is imported from Japan and Chile (Miller, 1965). Chilean nitrate deposits furnish most of the world’s supply of iodine. The United States and Japan obtain iodine from subsurface brines. In Michigan, Dow Chemical liberates iodine from brines by chlorination and blows t h e iodine out with air. Japan recovers iodine from brine by the cuprous iodine, electrolytic, or active carbon methods. In the United States, iodine was discovered in an oilfield brine by C.W. Jones in Louisiana in 1926. The Dow Chemical Company and Jones combined t o produce iodine from a brine well in Louisiana in 1928. At that time, iodine sold at a price between $9 and $11per kg. In 1929 General Salt Company began extracting iodine from oilfield brines in California. General Salt halted operations when Chile cut the iodine price to $3.30 per kg. In 1931 Deepwater Chemical Company began to produce iodine from oilfield brines in California. Deepwater Chemical halted recovery of iodine from brines in the late 1950’s. The Dow Chemical Company moved its iodine recovery operation from Louisiana to the California oilfield brines in 1932. The move was made for two reasons. The first reason was that California brines contained 60 ppm
RECOVERY OF IODINE AND BROMINE
391
iodine as compared to Louisiana’s 35 ppm. Secondly, the Dow Chemical Company was producing the brine in Louisiana from its own brine wells. In California the brine was produced by oil producers, because older wells produced 1 0 m3 of brine for every cubic meter of oil. The Dow Chemical Company used two methods to obtain brine in California. The first was by paying royalties t o oil producers for brines of high quality which were delivered at one pick-up point. The second was from an extensive brine gathering system which Dow built t o collect the brines from independent producing companies. The second method of disposal was done for the producers in lieu of royalties. At one time, Dow operated three iodine recovery plants in California. Only one of the plants utilized a complete iodine recovery process. In 1961 Dow began iodine recovery from Michigan brines at Midland, Michigan. These brines are not oilfield brines and although the Michigan brines contain only 35 ppm, compared to California’s 60 ppm, Dow found the Michigan operation less costly. Oilfield brines of California have two disadvantages. First, the brine source near DOW’Soperation dwindled, and secondly, production costs in California rose. Several economic advantages were available in the Michigan operation. For example, the iodine recovery process was integrated with processes for the recovery of calcium chloride, magnesium hydroxide, magnesium sulfate, bromine, potassium chloride, and magnesium chloride. The Midland operation boosted iodine recovery by using brines which were heated t o 91°C for other extraction processes. The absence of oil in the Michigan brines negated the cost of oil removal. In California, oil removal is necessary t o prevent interference with the oxidation step in the recovery process. The brine feed for the Midland operation is composed of brines produced from various strata in order t o obtain the desired feed for the most economical products.
Bromine Bromine is another element that is recovered from oilfield brines. One plant that is located in Arkansas recovers bromine from the Smackover formation in the Catesville field. The bromine recovery project was originally included as part of the plans to unitize the field in 1956. TABLE 13.1 Bromide recovery economics at Catesville Minimum economical production Designed production Designed brine feed Plant cost Plant payout period
900,000 kg/year 1,800,000kglyear 1,400 m3/day $ 1,000,000 6 years
392
VALUABLE MINERALS IN OILFIELD WATERS
TABLE 13.11 Profitability of Catesville bromide project Return o n investment Profit as sales percent Profit per m3 processed Investment Profit per year
16.7% 14.3% $ 0.346 $ 1,100,000 $ 180,000
Location of a bromine plant at Catesville offered several important advantages such as high bromide content (up to 6,000 mg/l) of the brine, field operation under a single company (unitization), excellent rail and road facilities, low-cost fuel, and regional market outlets. Production of Smackover brine in 1956 was approximately 190 m3 daily from four oil wells. This quantity of brine was not quite economical for a bromine recovery plant. Additional pumping equipment was installed in some of the wells in order t o provide 795 m3 of brine daily for the bromine recovery project. Depleted oil wells later were employed for brine production to raise the plant feed to 1,430 m3 daily. Table 13.1 shows the initial economics associated with the bromine project as reported by Kincaid (1956). The economic data presented in Table 13.1 are based on the bromine prices of 1956. In 1956 the price of bromine was 66 cents per kg. The price fluctuates with supply and demand. The data shown in Table 13.11 were calculated by assuming that all of the bromine is sold at 70 cents per kg, that the total investment is not more than $1.1million, and that the payout time is 6 years (Cox, 1967). Minerals recovered from saline waters
Sodium chloride Minerals are recovered from practically every type of saline water. By far the largest recovery is that of sodium chloride in solar evaporation processes. From the point of view of oilfield brine disposal, where solar evaporation is possible, the cost of disposal is small. The salts recovered, if any, would probably pay for the construction of evaporation pits.
Lithium Lithium is produced from brines by Foote Mineral Company at Silver Peak, Nevada, and by American Potash and Chemical Corporation at Trona, California. American Potash and Chemical Corporation recovers a coproduct lithium sodium phosphate from Searles Lake, California, brines. However, the largest lithium production is from lithium ore mined in North Carolina
MINERALS RECOVERED FROM SALINE WATERS
393
by Lithium Corporation of America. Domestic production of lithium has not been reported since the mid-l950’s, because individual companies do not want t o disclose confidential data. In 1954 about 36,000 metric tons were produced in the United States. The staff of the U.S. Bureau of Mines (1968) reports that both the lithium industry and the government are hampered by restrictions on publishing statistical data on the production and consumption of lithium metal, alloys, and compounds. These restrictions inhibit the determination of requirements, the evaluation of market potentialities, and the planning of future action.
Potassium Tallmadge et al. (1964) report that the commercial recovery of potassium from brines only has been attempted on a pilot plant scale. Precipitation appears the most promising either by the addition of a selective agent specific t o potassium, or by fractional crystallization of saturated brines. Potassium compounds occur in many rocks and minerals, but the commercial sources are limited t o soluble salts in bedded salt deposits and brines. The major deposits of potassium salts in the United States are part of the Permian Salt Basin that underlies parts of Colorado, Kansas, Oklahoma, Texas, and New Mexico, and the Paradox Basin of southwestern Colorado and southeastern Utah. However, commercial beds of potassium minerals have been found only in New Mexico. Commercial operations have been limited t o about 1 4 0 km2 east of Carlsbad, New Mexico. These deposits were discovered by oil well drillers. Commercial recoveries on a limited scale are made from the brines of Searles Lake, California, and Bonneville Flats, Utah.
Rubidium The rubidium-producing industry is very small. During 1958 rubidium production in the United States was only about 100 kg annually, and during that year some new technical-grade rubidium compounds were prepared from alkali carbonate residues of lithium operations. As with many other minerals found in oilfield brines, the production of rubidium is not published because it is withheld as confidential company data. However, with current accumulated stocks and a very small consumption, it is doubtful that the recovery of rubidium from brines would be economical even at $935 per kilogram.
Cesium Cesium, both as a metal and as an industry, is similar to rubidium. The demand for both is small, and the known uses are few. Both cesium and rubidium are obtained commercially from lepidolite, a lithium mineral.
394
VALUABLE MINERALS IN OILFIELD WATERS
Cesium and rubidium are byproducts of the lithium industry, and both are recovered from the residues of the lithium production process are precipitation from solution. The high concentrations of cesium and rubidium in the residues and the fact that the amount therein greatly exceeds demand virtually preclude their removal from oilfield brines on a competitive basis.
Magnesium Magnesium comprised one-third of the value which Collins (1966) attributed to the minerals wasted by oilfield brine disposal, and the price used was that of magnesium metal. In the primary.meta1 form, magnesium commands its highest price. When magnesium is sold as contained in other compounds, its value is less than 2 cents per kg as compared to 77 cents per kg for primary magnesium. Magnesium and magnesium compounds are produced from the following four raw material sources: (1) sea water; (2) dolomite; (3) ores other than dolomite; and (4) evaporite deposits and lake and well brines. In 1963, well brines, bitterns, and sea water combined with calcined dolomite or lime accounted for more than half of the domestic production of magnesium compounds used as chemicals, filters or bases in many industrial products including basic refractories. Magnesium and magnesium compounds are produced and recovered by several companies in the United States. The Dow Chemical Company produces magnesium chloride crystals, magnesium chloride fluxes, and magnesium hydroxide from well brines and calcined dolomite at Ludington, Michigan. The Michigan Chemical Company produces precipitated magnesium carbonate, magnesium hydroxide, and magnesium oxide from well brines and calcined dolomite at St. Louis, Missouri. The Dow Chemical Company produces magnesium chloride, caustic-calcined magnesia, and magnesium hydroxide from sea water and oyster shells at Freeport, Texas. Magnesium compounds are recovered from solution by precipitation of magnesium hydroxide. This method is so economical that a large part of the production of magnesium and magnesium compounds in the United States is derived from sea water. The Dow Chemical Company is the major source of primary magnesium and in 1963 it had a capacity at Freeport, Texas, of 50,000 metric tons per year; at Velasco, Texas, the capacity was 34,000 metric tons. In 1963 the U S . production of primary magnesium was 69,000 metric tons. The Dow process for magnesium recovery from sea water first precipitates magnesium hydroxide. The hydroxide source is calcium hydroxide made from oyster shells. After settling and thickening, a slurry of 17%magnesium hydroxide is attained and neutralized with hydrochloric acid to form a 15% solution of magnesium chloride. After evaporation and dehydration, the resultant 48% magnesium chloride solution is mixed with dried magnesium chloride t o form a paste. The paste is dried t o granules which consist of 74% magnesium chloride, and the granular material is the feed to electrolytic cells
MINERALS RECOVERED FROM SALINE WATERS
395
which produce magnesium metal and chlorine. The chlorine is then converted t o hydrochloric acid which is used in the neutralization step. Shreve (1956) lists three economic factors of importance in the precipitation of magnesium hydroxide. They are: (1) the source of the hydroxide; (2)the dewaterirg procedures used for removal of the magnesium hydroxide from the dilute solution; and (3) the purification of precipitates. The source of the hydroxide is the major economic deterrent factor against the increased use of well brines. Sea water provides the magnesium, and the sea also furnishes the oyster shells for calcium hydroxide production. For well brine feed, dolomite often is employed, and a large source of dolomite must be economically available. Tallmadge et al. (1964) report that waste sodium hydroxide has been tested in Japan, but in most areas, calcium salts are the least expensive sources of the hydroxide. Thus the choice of a raw material must be belanced in cost against plant size and market. While Michigan brines contain four t o five times the magnesium concentration of sea water, the reduced size in necessary equipment for processing the brine does not completely overcome the cost of producing and disposing of the brine. This would appear t o make oilfield brines more attractive than other subsurface brines if a hydroxide source is available at an equivalent expense. Tallmadge et al. (1964) report methods for extracting magnesium from brines by methods other than hydroxide precipitation. However, none appear economically attractive when compared to precipitation unless combined with other processes or products. Among those studied are solar evaporation to produce chloride, use of ion-exchange resins with lime and carbon dioxide or waste liquor from the ammonia-soda process, and electrolysis.
Calcium Calcium production from brines does not appear economical when compared to the source of the world’s calcium consumption. The largest amount of calcium is produced by the mining of mineral deposits (notably gypsum) found extensively throughout the world. Proposals for methods t o recover calcium from brines have been made and are under study, but to compete commercially beyond extremely small, local demands, considerable research is needed.
Mixed salts Mixed salts are precipitated by evaporation of sea water and brines, producing crude separations. The costs of these separations are low compared t o those of highly purified compounds or metals. There are several drawbacks which prevent greater use of this type of recovery. The product does not command a high price, the plant must be at the brine sburce, there must be solar evaporation conditions, and a local market must exist for the majority
396
VALUABLE MINERALS IN OILFIELD WATERS
of the mixed salts. Uses which have been suggested include heat-treating salt baths in the steel industry, raw materials for refractory or catalyst manufacture, and fertilizer components. Precipitation other than by solar evaporation is accomplished by cooling or adding chemical agents. Simple cooling may be all that is necessary for more concentrated brines, but fractional crystallization is necessary for dilute brines such as sea water. Again local markets dictate whether cooling or freezing processes will yield the correct products for a particular area. Adding chemicals t o precipitate a specific product is the most fruitful of the nonsolar evaporation processes. Most of the processes have been aimed at the production of fertilizer. Potassium and magnesium are the minerals in sea water that are most valuable for use in fertilizers. Salutsky and Dunseth (1962) report that metal ammonium phosphates (MAP) containing magnesium, calcium, iron, manganese, copper, and many other trace metals comprise a high-analysis fertilizer. The production of metal ammonium phosphates (MAP) in the United States was started by W.R. Grace and Company in 1960 on a semicommercial scale. The method which Grace used to produce MAP was not disclosed until 1962 after it was patented. The fertilizers are nonburning, long-lasting sources of nitrogen, phosphorus, and various trace metals. Because of their low solubility, MAP’s will not cause salt injury t o seeds or plants. In magnesium ammonium phosphate, practically all of the P z 0 5 is available, and the size of the MAP granules applied to plants determines how long the nutrients will be available. Thus, availability of nutrients can be controlled by granulation and, since growing time varies from crop t o crop, MAP’S can be tailored to a specific crop (Anonymous, 1961). Therefore, fewer applications are necessary with MAP’S than with fertilizers of higher solubility and high nitrification rates. W.R.Grace and Company developed the MAP process for two purposes. First, it is useful t o remove scale-forming materials from sea water before desalination. Secondly, it would yield the valuable, high-analysis fertilizer, magnesium ammonium phosphate. In 1962, W.R. Grace and Company (Anonymous, 1962) reported that the process was ready for the pilot plant. The process is based on phosphate precipitation. To descale sea water and produce high-analysis fertilizer at the same time, wet-process phosphoric acid and anhydrous ammonia are added continuously t o raw sea water. This precipitates the scale-forming elements - calcium, magnesium, iron, and other metals - as metal ammonium phosphates and other phosphates. The precipitated solids are removed by settling, and the descaled sea water is pumped to the saline water conversion plant. The descaled water holds only 1%of the original magnesium and 5% of the original calcium. The slurry of MAP’s is dewatered t o about 35--40% solids by continuous centrifuges and thin it is heated t o 90°C. This converts MAP hexahydrate t o monohydrate. The slurry is filtered, washed, mixed with recycle fines, and granulated. Fig.
MINERALS RECOVERED FROM SALINE WATERS
I
397
I
S e t t l i n g ond thickeninq
Descoled sea w o t e k
t
1 Dehydration
1 Wash
Filtrotion ond wos hing
Gronulotion
Drying
Crushing
I
k
1 Undersize
Screeninq
Oversize
I
Finished product to storoge
Fig. 13.1. Diagramatic flowsheet for producing descaled sea water and fertilizer.
13.1 shows a process flowsheet for producing descaled sea water and fertilizer. Several questions surround the economics of the process. For a plant descaling 3,800 m3 of sea water per day (output: about 10,000 metric tons per year of fertilizer), the fertilizer would have to command a price higher than that of conventional farm fertilizers. The estimate assumes 1962 market prices for raw materials (phosphoric acid and ammonia) and does not take credit for the increased value of the descaled sea water. The cost is just about the same for Grace’s present method of producing MAP’S. Because of its premium quality, MAP can go t o the market as a specialty product. In the phosphoric acid-ammonia process, 2 moles of ammonia per mole of MAP are lost. to ammonium chloride in neutralizing the phosphoric acid. Using disodium phosphate in place of the acid loses no ammonia, and using monosodium phosphate only loses ‘1mole of ammonia. If a cheap method were developed for producing the sodium phosphates, ammonia waste would be reduced. The simplest method for producing sodium phosphates involves
398
VALUABLE MINERALS IN OILFIELD WATERS
the neutralization of phosphoric acid with either dilute sodium hydroxide or soda ash. Caustic soda and chlorine can both be produced from sodium chloride brines.
Chlorine Chlorine is the most abundant element in oilfield brines, and the removal of all chlorides from oilfield brines would virtually desalt the brine. However, present commercial methods for the extraction of chlorine from brines requires evaporation to acquire a saturated brine. Solar evaporation first produces the saturated brine, then electrolytic processes are employed to generate chlorine gas. The largest production of chlorides from brines is from solar evaporation of sea water and salt lake brines. Many oilfield brines are very close t o saturation with sodium chloride at surface temperatures. Subsurface brines are employed as raw materials for chlorine production, but the amount produced is not significant when compared t o surface brine production. Shreve (1956)describes the methods still in use for chlorine production. Caustic soda and chlorine are coproducts in the electrolytic process. Purification of the brine is necessary to produce a purer caustic soda and lessen clogging of the cell diaphragm with consequent increased voltage demand. Calcium, iron, magnesium, and sulfate must be removed. The higher concentrations of magnesium and calcium in oilfield brines cause greater expense in brine purification unless the removed compounds can be sold. Hydrogen, caustic soda, and chlorine are the products of this type of recovery. The hydrogen presents a disposal problem and is frequently made into other compounds such as hydrochloric acid or ammonia or is employed for hydrogenation of organic compounds.
Iodine and bromine Iodine and bromine are two minerals which have closely related processes for recovery from brines. Both are displaced from ions to elements in solution by chlorination and then stripped from solution by air. When bromine is liberated by chlorination, iodine is oxidized to the iodate ion. After bromine is stripped from solution, the iodate ion can be reduced t o free iodine by treatment with ferrous chloride and then stripped by air as was the bromine. The greater portion of bromine production in the United States is from well brines. Slightly over 50% of the domestic production is from well brines, 35% comes from sea water, and the remainder comes from oil well brines and saline lake brines. Substantial expansion recently completed by two producers of bromine from brines should give the industry sufficient capacity for several years t o supply the expected increase in markets (Miller, 1965). The 2-million-kilogram-per-year plant designed for oilfield brines in Arkansas by Michigan Chemical represented 50% of the domestic production
MINERALS RECOVERED FROM SALINE WATERS
399
at that time. In 1963 the domestic production of bromine was about 12 million kg. The recovery of bromine from brines and sea water is accomplished by displacement of the bromide ion with chlorine. The resulting free bromine dissolved in water is then stripped from the solution with air and recovered. A large number of modifications t o the basic process (developed by the Dow Chemical Company) have been proposed and patented from time to time. Tallmadge et al. (1964) report that studies are being carried out on the effect of pH, temperature, organic impurities, chlorine concentration, and foreign ion concentration on the displacement reaction between the bromide ion and chlorine gas. The use of chlorine water rather than gas for the displacement step has been suggested t o reduce the adverse effects of magnesium and calcium interference where more concentrated brines are used. Activated carbon with adsorbed chlorine also has been proposed as a means of carrying out the displacement step. Iodine production in the United States uses oilfield brines and sub-surface brines exclusively. Dow Chemical is the sole domestic producer of crude iodine. Dow extracts iodine from oilfield brines in California and from deepwell brines in Michigan. Roughly 1.1 million kg of crude iodine were imported from Chile in 1963 and about 0.5 million kg were imported from Japan. Japanese production of iodine is almost exclusively from deepwell brines, while Chilean production is from nitrate deposits containing the minerals lautarite and dietzite. Miller (1965) reports that Chilean reserves are in excess of 1 billion tons as a byproduct of the nitrate minerals industry. It was price cuts by Chile that forced all domestic producers except Dow out of iodine production; however, the recent nationalization of the Chilean mines has changed the picture and pushed the price of iodine to about $5.06 per kg. Multiproduct production Tallmadge et al. (1964) report that it is very probable that the most economic system for removal of minerals from sea water may involve two or more recovery steps in some integrated fashion. Only a few multiproduct processes are operated on a commercial scale. The Dow plant in Midland, Michigan, is such a plant. The first step in studying multiproduct processes is t o determine how much of each product can be sold. The second step is to determine the engineering design and production costs for such a plant. Angino (1967) has pointed out that several processes exist that can be set up t o recover elements from petroleum-associated waters. These existent methods and new methods should be utilized and developed to conserve the natural resources dissolved in brines and to aid in the abatement of soil and fresh water pollution. The recovery methods used may include desalination (Christensen et al., 1967), ion exchange (Klein et a1.,'1968), and ion exchange plus other methods (George et al., 1967; Waters and Salutsky, 1968).
400
VALUABLE MINERALS IN OILFIELD WATERS
Fig. 13.2. How gas, oil, and brine are separated after production from subsurface strata.
Fig. 13.2 illustrates a wellhead through which gas, oil, and water are produced from a subsurface formation. Often they are produced as a mixture and it is necessary t o separate them in a tank such as that illustrated and sometimes referred to as a gunbarrel. In this tank the water or brine will settle to the bottom with the oil interface forming over the brine, and the gas will rise to the top. The gas is drawn off the top, the oil is pumped off the top of the water and stored in an oil tank, and the water is siphoned from the bottom into a skimming tank for further oil-water separation. The water is siphoned from the bottom of the skimming tank into a settling pond where additional oil-water separation occurs. The brine or water could be pumped from the settling pond t o a chemical plant for recovery of valuable elements or to an injection well. Fig. 13.3 illustrates a possible scheme for recovery of some elements plus fresh water from the brine. For example, the raw brine could be concentrated by a desalination process which would also yield fresh water. Sulfate then could be taken from the concentrated brine and used t o produce sulfur or sulfur compounds. Next, iodine could be recovered, then bromine, followed by calcium, sodium chloride, and magnesium as suggested in the figure. The remaining sludge could be dried and disposed of as a solid or it could be recycled for additional recovery of elements.
FRESH-WATER PRODUCTION
401
RAW B R I N E
I
1 DESALINATION PROCESS
I-*
PRODUCT F R E S H WATER
CONCENTRATED BRINE
I I I
SULFATE PRECIPITATION
I-*
IODINE RECOVERY
t . BROHINE RECOVERY
PRODUCT SULFUR AND SULFUR COMPOUNDS PRODUCT I O D I N E AND I O D I N E COMPOUNDS PRODUCT BROMINE AND BROMINE COMPOUNDS
PRODUCT CALCIUM
SODIUM CHLORIDE RECOVERY
coMp0uM)s
PRODUCT SODIUM CHLORIDE OR SODA ASH AND CHLORINE PRODUCT MAGNESIUM AND MAGNESIUM COMPOUNDS
---l---
+
SLUDGE CONTAINING CALCIUM, STRONTIUM, BARIUM, MAGNESIUM AND OTHER ELEMENT COHPOUNDS. THE SLUDGE CAN BE D I S P O S E D A S A S O L I D OR RECYCLED FOR CHEMICAL RECOVERY
Fig. 13.3. Diagramatic flowsheet for producing fresh water and valuable elements from brines.
Fresh-water production Dwindling fresh water supplies and polluted supplies have increased research on how t o best obtain fresh water from saline water. Several plants throughout the world produce fresh water from sea water. The price of water for municipal purposes is a highly specific thing. The availability of fresh water and costs of obtaining it vary from place to place. Conventional
402
VALUABLE MINERALS IN OILFIELD WATERS
water supplies range in cost from a few dollars per 1,000 m3 to over $260 per 1,000 m3. The average cost of conventional water supplies in the United States was $100 per 1,000 m3 in 1952. This was chosen as the goal for saline water conversion costs. Several authors have estimated ultimate costs of saline water conversion based on thermodynamic considerations. Dodge and Eshaya (1960) have examined the minimum expected costs for saline water conversion. Prior t o their calculations, other authors reported sea water conversion costs to be ultimately less than $79 per 1,000 m3. Dodge and Eshaya expanded earlier work t o look at departure from isothermal operation, finite product recovery, differential as opposed to single stage operation, and salt
concentration in the feed. They found that $90 dollars per 1,000 m3 is the smallest cost for desalination of sea water. Consider the case for converting brackish water with 5,000 ppm sodium chloride. For converting 50% of the feed t o fresh water, 187 kWhr per 1,000 m3 was the power requirement. For 35,000 pprn sea-water conversion, the power requirement was 1,530 kWhr per 1,000 m3. Both calculations were for 50% recovery of fresh water from feed, where the average power costs used in determining conversion costs are 1.5 cents per kWhr. At this rate, the difference in power costs for sea water over brackish water is $20 dollars per 1,000 m3. Oilfield brines contain up to seven times the concentrations of dissolved salts compared with sea water. Would the power be seven times again as expensive per 1,000 m3? At over $132 for power and $92 for other costs, the cost of obtaining fresh water from oilfield brines probably would be prohibitive when consideration is given to the other sources for feed t o a conversion plant in the same area. An additional factor is that most oilfield brines with their high concentrations are nearly saturated. Removing 50% of the water would in essence leave a precipitated salt. Therefore, since no conversion processes under study deal with saturated brine effluents, it is not technologically feasible to completely desalt oilfield brines at this time. Preliminary economic evaluation The “brine refinery” concept (Collins, 1966) yields a processing plant the size of a large petroleum refinery. The market prices used were for the recovery and sale of the pure elements. The $3 billion in sales from 0.95 billion m3 of brine is the highest sales income possible that would result from recovering and selling the minerals in the form that gives the highest unit price. Consider what is probably the best case of a “brine refinery”, a system that would gather 22 million m3 per year. The cost of gathering and disposing of this brine would be approximately 9.4 cents per m3. The question is whether or not minerals could be sold at a profit such that the disposal expense would be negated or a profit made. First, the minerals t o be
PRELIMINARY ECONOMIC EVALUATION
403
sold must be determined. At 7 ppm lithium, 163 metric tons per year could be produced. This is a large fraction of present consumption and probably would depress the sale price. The same holds true for most other elements of such a refinery. The assumptions lead to a brine refinery that would process 22 million m3 per year and sell $35 million of minerals. Assuming a 15% return on investment and a profit of 15% of sales, the plant would require $35 millioninvestment and yield 23.6 cents per m3 of brine processed. The original disposal operation without mineral recovery was such that only $6 million was invested. The “brine refinery” would turn brine disposal into a profit. But the new investment is six times that for disposal only. It is doubtful that any large oil producer would be interested in a 15%return on investment, and small ones would never gather the cash. Would a chemical company be interested in such an operation since they operate at about a 15% return? Companies that currently remove minerals from brines use brines that are more concentrated in the minerals desired. It is doubtful that a process could combine several less economical operations into a more economical one, and this would probably be true even if the brine was supplied t o a chemical company free of charge. Only in the special case where an oilfield brine contained a concentration very near t o a brine that would be the most economical for separation would the oilfield brine be a best alternate. Therefore, a tax incentive for pollution abatement or some other economic incentive such as price increase of recovered chemicals is necessary.
Other economic factors Table 13.111 illustrates the approximate amount of valuable chemicals per 1 million kg of brine produced from a given depth should contain before it can be considered of economic value at present market conditions. The values shown in Table 13.111 should allow a profit if conventional or better recovery operations are utilized. The marketed end product will influence the selection of the recovery operation as well as the delivered price. The
TABLE 13.111 Dollar value of dissolved chemicals a brine should contain per million kg of brine produced from a given depth Value of dissolved chemicals
Depth of well (m)
$ 462 $ 968 $ 1,430
760 2,130 3,050
404
VALUABLE MINERALS IN OILFIELD WATERS
TABLE 13.W Amount of element necessary in 1 million kg of brine to produce a chemical worth $ 550 at the market Element in the brine
Concentration of element (ppm/106 kg of brine)
Market product
Sodium Potassium Lithium Magnesium Calcium Strontium Boron Bromide Iodide Sulfur
50,000 14,000 170 8,000 11,000 4,000 1,400 1,700 250 5,300
sodium chloride potassium chloride lithium chloride magnesium chloride strontium chloride strontium chloride sodium borate bromine iodine sodium sulfate
price information used to make the approximations was taken from the U.S. Bureau of Mines (1968). Factors that must be considered in evaluating a saline water as an economic ore are the cost of bringing it t o the factory, the cost of the recovery process, and the cost of transporting the recovered products to market. Assuming that a brine is produced only for the purpose of recovering its dissolved chemicals, a prime factor is the cost of pumping the brine. It will cost less to produce the brine from a shallow well than from a deep well. Therefore, neglecting other factors, a brine must contain a certain amount of recoverable chemicals before it can be considered economically valuable, and the farther it must be pumped, the more chemicals it must contain. Today the possibility of recovering elements from brines that are pumped t o the surface is increasingly important because the brines present a pollution hazard if their disposal is improper. Consider the fact that 1 m3 of brine containing 100,000 ppm of chloride is capable of polluting 400 m3 of fresh water so that they are unfit for human consumption. Table 13.IV illustrates the value that chemicals recovered from brines have at the market; however, because the market fluctuates, these values are approximate. The column on the left indicates the elements that are found in petroleum-associated brines, and the second column indicates the concentration that a given brine must contain before it can be used to produce a given amount of chemical. For example, a brine containing 50,000 ppm of sodium will contain sufficient sodium in 1 million kg of brine to produce sodium chloride worth about $550. The data in Table 13.IV indicate that some petroleum-associated waters contain sufficient sodium to establish them as economic for the production of sodium chloride. This is not necessarily true, because factors such as
PRELIMINARY ECONOMIC EVALUATION
405
market demand, ease of recovery, and proximity to market may be discouraging in certain geographic areas. Such factors must be fully considered before startup of a chemical from brine recovery operation. One important goal that should not be discounted nor overlooked is developing a means of ultimately disposing of these brines so that they are not a pollution hazard. Coupling of this goal with the fact that many of these brines contain economic concentrations of several elements should make such recovery operations more attractive. Additionally, several important chemicals can be produced from these elements instead of those shown in the market product column in Table 13.IV. An example is soda ash, which is a basic chemical in many manufacturing processes. Furthermore, the figures shown in Table 13.111 are applicable only if the brine is produced solely for the recovery of its dissolved chemicals. If the brines are pooled from several petroleum production operations, the cost of pumping the brine becomes less, and the necessary amounts of chemicals dissolved in 454,000 kg of brine become less. At the present time, many petroleum-associated brines are injected into subsurface strata, and it is assumed that they are thus disposed of permanently (Crouch, 1964). However, this method of disposal appears subject to question, because in some instances, fresh waters apparently have been polluted by disposal of brines. Subsurface disposal operations are suspected in certain areas as possibly contributing to increased earthquakes and ground tremors (Evans, 1966; Bardwell, 1966). The storage of brines in earthen pits is known to cause pollution of nearby soils and streams. Such ponds which have been abandoned for 10 years still contribute to soil pollution (Bryson et al., 1966). Sound conservation should favor the recovery of valuable elements from brines, and with proper planning, the recovery processes should aid in the ultimate disposal of unwanted brines. Conservation of this type not only will develop new resources, but will benefit the oil producer and the national economy and will aid in abating pollution of soils, potable waters, and streams.
Work necessary f o r an exact preliminary evaluation Aries (1954) spells out the marketing research techniques employed in the chemical industry. There are ways to quickly determine where a market for a product is. Usually these places are currently served by some producer or another. If the competition is located far from the market, then an evaluation of a closer area source is readily made. To find product users, the following methods and approaches are utilized: advertising, company analysis, product analysis, industry analysis, use analysis, and other miscellaneous methods. If new markets must be found, the following types of work are utilized: personal interview, questionnaire, trade analysis, company records, and published sources. Before an economic analysis can be made for a given area, probably several man-months of the listed methods would be
406
VALUABLE MINERALS IN OILFIELD WATERS
required. The product of this type of market study would be a list of elements and compounds that could be sold from a given place. The quantities and prices obtgined would then allow an economic calculation of the production costs. With the quantities, prices, and production costs in hand, it is still not a simple matter t o determine what type of plant t o operate. Regardless of whom the investor might be, he will want to know what return on investment he will get, what risk is involved, and what payout period exists for the project. Depending on the investor, he may want t o limit the plant size by the amount of money he can invest. This does not simply scale down the plant. It may rearrange various ratios of certain products produced in order to give the investor the combination of profits, return on investment, risk, and actual size of investment that he desires.
Locations o f valuable brines Table 13.V lists the approximate geographic locations where subsurface saline waters containing valuable elements are found. The numbers in the left column of the table correspond to the numbered arrow locations on Fig. 13.4. The second column in the Table indicates the age of the geologic strata from which the waters were obtained. These waters are in or near oilproductive sedimentary basins. Concentrations of various elements present in the waters are given in columns 3 through 11 of Table 13.V. These concentrations are representative of one or more subsurface waters from each location; however, the concentrations should not be considered typical of all subsurface waters in an area or stratum. For example, some waters near location 1 from Mississippian age strata may contain 1,000 ppm of bromine, while other waters 80 km away, but from the same geologic strata, may contain 3,600 ppm of bromine. The elemental composition of ocean water is consistent; the composition of subsurface saline waters is inconsistent. Fig. 13.5 is a map showing some areas in the United States where brines containing high concentrations of sodium are found. The solid circles on the figure represent areas where brines containing 75,000-80,000 mg/l of sodium can be found, the open circle represents brines containing 80,000-95,000 mg/l, and the triangle represents brines containing more than 95,000 mg/l. Fig. 13.6 is a map showing some areas in the United States where brines containing high concentrations of calcium are found. On this figure the solid circle represents brines containing 20,000-30,000 mg/l of calcium, the open circle 30,000-50,000 mg/l, and the triangle more than 50,000 mg/l. Fig. 13.7 is a map illustrating some of the areas in the United States where high concentrations of magnesium in brines are found. On this figure the solid circle indicates brines containing 5,000-10,000 mg/l of magnesium, the open circle 10,000-30,000 mg/l, and the triangle more than 30,000 mg/l.
c
PRELIMINARY ECONOMIC EVALUATION
l l
TABLE 13.V Geographic location, geologic age of saline water-bearing strata, and concentration of some of the elements found in the brine's Location* Age of subsurface strata 1 2 3 4 5 6 7 8 9 10 11 12 13
*
Concentration (ppm) lithium sodium potassium magnesium
Mississippian 10 Permian 30 Permian 40 Devonian 100 Cambro-Ordovician ND Pennsylvanian 5 Jurassic 100 Miocene 15 Devonian 25 Devonian 60 * Mississippian Devonian 10 Devonian 90
40 28,000 100 1,000 55,000 2,500 66,000 10,000 8,000 ND 51,000 100 68,000 4,000 73,000 600 74,000 700 14,000 8,000 16,000 9,000 58,000 3,000 72,000 2,000
See arrows in Fig. 13.4. ND = not determined.
10,000 25,000 9,000 5,000 11,000 600 5,000 6,000 5,000 15,000 11,000 5,000 4,000
calcium
strontium boron
sulfur
chloride
bromide
iodide
60,000 100 30,000 40,000 20,000 10,000 30,000 30,000 30,000 70,000 14,000 20,000 35,000
3,000 5 ND 2.000 ND 1,000 ND ND 900 1,500 800 1,000 ND
400 20,000 400 100 350 30 60 60 600 400 20 60 4
179,000
3,200 ND 1,200 700 ND 600 5,000 200 2,000 2,500 1,500 1,300 1,800
40 ND 25 20 ND 1,000 10 20 40 40 40 30 20
40 ND 90 ND ND 10 ND 60 ND 300 ND ND ND
9,000 166,000 198,000 79,000 98,000 175,000 ia4,ooo 182,000 200,000 90,000 143,000 186,000
c
PRELIMINARY ECONOMIC EVALUATION
409
Fig. 13.5. Approximate geographic locations of brines containing high concentrations of sodium.
Fig. 13.6. Approximate geographic locations of brines containing high concentrations of calcium.
VALUABLE MINERALS IN OILFIELD WATERS
410
'
LEGEND 0
A
5,000-10,000mg/l 10,000-30,000 > 30,000
Fig. 13.7. Approximate geographic locations of brines containing high concentrations of magnesium.
Fig. 13.8. Approximate geographic locations of brines containing high concentrations of bromine.
DISPOSAL BRINES
411
Fig. 13.8 is a map illustrating some of the areas in the United States where high concentrations of bromide in brines are located. On this figure the solid circle represents brines containing 1,500-2,000 mg/l of bromide, the open circle 2,000-3,000 mg/l, and the triangle more than 3,000 mg/l. Disposal brines In an attempt to acquire as much information as possible about salt water disposal facilities in the various States, the literature was surveyed and State and Federal agencies and oil companies were contacted. The acquisition of a true compilation of the exact number of disposal facilities, disposal wells, and total number of barrels of brine injected was an impossible task without a large surveillance force; however, the data in Table 13.VI are reasonably representative. To determine the value of the minerals in the brines flowing into these disposal systems, samples were obtained from 40 systems. The samples were analyzed for concentrations of lithium, sodium, potassium, magnesium, calcium, boron, ammonium, sulfate, bicarbonate, chloride, bromide, and iodide. Table 13.VII lists the state, county, subsurface formation, and dissolved solids (DS) from which the brine samples were obtained. The specific gravity of each sample plus the ionic values determined in the laboratory also are given in Table 13.VII. Sample 1is sea water, sample 2 is a brine that contains a high concentration of iodide, and sample 3 is a brine from which bromine currently is extracted.
Worth and value estimates The estimates of the value of a brine are related to the market and the recovery process. The market is dependent upon demand; however, in the following estimates the demand was not considered. According to Christensen et al. (1967): Brine value = market value of products - operating costs exclusive of brines value - fixed costs - profit The maximum value of a brine can be found by letting the “return on investment” equal zero, and the brine worth is: Brine worth = brine value + profit = market value of products ating costs - fixed costs
- oper-
The brine value is less than the brine worth by the amount of profit expected. Although the information necessary to obtahi an accurate calcula-
TABLE 13.VI States where oilfield brines are disposed, total number of subsurface salt water disposal wells (SWDW) and largest disposal facilities State Ah. Alaska Ariz. Ark. Calif. Colo. Fla. Ill. Ind. Kans. KY. La. Mich. Miss. Mont. Nebr. Nev. N. M. N. Y. N. D. Ohio Okla. Pa. S. D. Texas Utah W.Va. wyo.
Total number of SWDW
14* no data 4 421 216 25* 5 no data 325 3,150 60** 1,304 52 2 370 20 0 3 300
-
62 28* 4,900
-
Largest SWDW
SWDW
SWDW
( m3 /day)
2 4700 m3/day
2 1590 m3 /day
191 9,221 25,120 2,544 954
0
5
0 14
0
0
159 3,180 111
0
0 2** 0
2,385
318 1,113 239 135
-
2,703
-
318 318
-
0 0 0 0 0 0 0 0 0 0
0 0
0 0
0 0
0 0 0 4 0 0 0
7,173** 3** 95** 19*
Remarks
143 159
0 0
+ bromine plant effluent
average SWDW is 21 7 m3 /day average SWDW is 32 m3 /day
only 7 producing wells for State
2 most of this is by ponding ponding used number of SWDW permitted from 1950-1971
0 0
* Salt water disposal Systems that may have more than one salt water disposal well per system. * * Approximate.
2P
2
_.
P
U
8
5
-
2--
TABLE 13.VII Analyses of some disposal brines, seawater, and a proven economic brine* Brine State
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
*
Seawater Okla. Ark. Kans Kans. Kans. Kans. Kans Kans. Ark. Ark. Ark. Ark. Ark. Ark. Ark. Ark. N.M. N.M.
N.M. N.M. N.M. Texas Texas Texas Texas Texas Texas Texas Texas Ala. Ala. Ala. La. La. La. Calif. Calif. Ariz. Okla. Okla. Miss.
County
Formation
-
-
Sp. gr.
1.025 Kingfisher Oswego 1.124 Columbia Smackover 1.230 Pawnee Arbuckle 1.036 1.025 Barton Arbuckle 1.012 Butler Hunton 1.034 Ellis Arbuckle 1.020 Pratt LKC-Arb Barton LKC-Arb 1.050 1.046 Palm Graves 1.048 Ouachita L. Graves 1.046 Ouachita L. Graves 1.192 Union Smackover 1.192 Union Smackover Union Smackover 1.199 1.191 Union Smackover 1.162 Columbia Smackover 1.020 Lea San Andres 1.036 Lea Devonian 1.028 Lea Devonian 1.043 Lea Penn 1.039 Lea Devonian Gaines Devonian 1.025 Cherokee Woodbine 1.056 Rusk Petit 1.153 Cherokee Woodbine 1.070 Wood Woodbine 1.065 Wood SubClarkville 1.037 Wood Paluxey 1.076 Hopkins Paluxey 1.010 Mobile Rodessa 1.031 Mobile Rodessa 1.039 Mobile Rodessa 1.052 LaSalle Wilcox 1.064 Calcasieu 1.084 Cameron Miocene 1.076 Fresno 1.026 Kern Kern 1.000 Apache L.Hermosa 1.012 Woods Hunton 1.123 Oklahoma Wilcox 1.151 Wayne Wilcox 1.003
Sp. gr. =specific gravity; DS = dissolved solids,
Constituents (mg/l) Na
Ca
Mg
K
Li
B
NH4
CI
Br
I
SO4
HCOi
DS
10,500 56,250 74,000 14,430 9,850 5,990 16,800 9,400 23,300 19.900 21,100 20,500 64,200 63,900 63,300 64,500 54,500 9,150 18.200 i3:goo 21,600 19,350 12,380 30,000 58,700 36.200 34;OOO 21,200 34,600 5,640 12,180 14,500 18,400 35,600 44,800 42.600 13,600 210 7,760 57,600 68,750 2,880
400 8,300 44,440 2,480 1,450 760 2,630 1,200 4,300 3,500 3,800 3,800 34,500 38,500 36,300 37,300 27,600 1,500 1.850
1,350 260 4,340 700 490 260 690 320 1,300 900 1,030 930 3,950 3,850 4,040 3,895 1,315 500 500 340 770 410 365 345 1,130 690 110 205 970 40 480 550 680 600 230 135 780 5 50 1,640 2,460 10
380 180 4,410 260 75 70 190 105 160 200 160 140 1,845 1,945 1,370 2,000 3,500 245 370 20 150 560 400 105 790 860 550 360 250 50 400 320 460 310 300 200 200 15 20 1.000 980 10
0.17 14 370 20 3 3 10 5 5 5 5 5 160 180 170 165 230 5 10 2 10 10 5 2 35 2 5 2 10 2 5 10 10 5 5 5 1
4.6 18 200 10 10 0 5 3 12 10 16 12 140 150 140 140 160 10 0 0 40 5 5 40 10 0 5 10 20 10 0 12 30 25 10 12 5 0 0 40 0 0
-
19.000 98,300 202,050 32,850 19,460 10,380 30,500 17,YOO 45,100 42,200 43,100 42,400 178,100 180,800 197,600 182,600 150,000 17,800 32,650 24,300 42,600 37,240 22,400 49,100 135,500 61,300 57,100 31,800 68,700 8,350 29,400 37,000 47,540 62,500 74,600 68,900 28,870 330 11,600 115,500 138,600 3,861
65 1,500 5,725 60 50 20 50 50 150 500 400 600 2,450 2,340 4,800 3,390 3,500 50 30 25 210 40 40 270 210 400 370 180 100 70 40 50 30 70 25 30 15 2 20 326 540 3
0.06 1,300 15 10 5 2 2 3 10 10 10 10 5 5 5 10 5 3 0 0 5 2 0 30 30 35 30 35 25 5 10 10 12 20 20 20 20 0 12 150 10 0
3,468 180 220 2,000 2,350 1,400 2,880 1,100 2,270
140 50 95 450 350 60 315 250 260 170 160 60 100 190 200 600 200 1,000 500 600 380 490 590 400 0 300 400 450 300 500 190 160 140 380 140 300 240 130 170 120 85 255
35,308 166,652 335,865 53,290 34,123 18,855 54,072 30,332 76,895 67,439 69,819 68,495 286,420 292,560 345,235 295,955 241,250 32,329 56,428 42,687 69,055 62,137 38,865 89,232 207,045 103.1 57 103,200 55,107 112.185 15,417 49,135 59,742 76,652 101,410 124,115 114,549 45,616 803 21,165 187,096 225,365 7,479
i$oo
2,840 2,400 1,970 8,650 10,320 3.300 10;530 840 6,750 630 5,630 6.750 8,860 1,650 3,960 2,335 1,855 70 1,520 10,120 13,270 50
1
1 10 10 0
300
-
20 30 0 0
0 30 45 40 40 320 260 260 100 50 70 60 0 90 0 100 50 50 40 10 25 40 0 90 80 90 250 25 12 0 0 12 260 140 0
0 0 0
650 440 350 255 190 2,000 2,260 2,000 360 1,630 610 240 270 30 90 0 420 120 710 300 400 0 0 0 0 40 0 350 520 410
c
414
VALUABLE MINERALS IN OILFIELD WATERS
TABLE 13.VIII Formulas for calculating maximum worth, brine worth, and brine value ____
Maximum worth = ( X i ) (market value of compound i)* Brine worth = M.W. - (market cost + fixed charges) Assume: brine worth = M . W . 4 . 7 5 (M.W.) Also assume: brine value = M.W. x 0.1
* X = amount of compound, and i = number of compounds. tion of brine value can be gained only by detailed market research, an estimate can be made by assuming which products will be recovered from the brine, calculating their market values, and relating the total values (or a single product value) back to the brine value after assuming that the brine worth equals a fraction of the total value. Table 13.VIII presents formulas for calculating the maximum worth, brine worth, and brine value. In this study, the maximum worth was calculated TABLE 13.IX Value of assumed recoverable compounds used in calculating brine value Cation
Compound
Cation ($/ton*)
Compound ($/ton)
Na Mg Li Sr K Ba Ca NH4
NaCl (rock) MgClz (99%) MgS04 LiCl (technical) SrC13 KCl BaClz (technical) CaClz NH4 C1
18.09 259.38 346.99 11,515.10 518.89 58.88 284.29 120.64 412.26
7.11 66.14 69.22 1,87 3.91 286.60 30.86 187.39 43.54 138.89
Anion
Compound
Anion ($/ton)
Compound ($/ton)
B
NazB40, * lOHzO NaCl (rock) NazS04 (salt cake) MgS04 NaBr NaI (U.S.P.) NaHC03 CaC03
c1 so4 Br I HC03 (303
*
Metric tons.
519.72 11.72 45.64 86.75 1,135.69 9,114.61 81.24 26.20
55.39 7.11 30.86 69.22 881.84 7,716.10 59.03 15.71
WORTH AND VALUE ESTIMATES
41 5
TABLE 13.X Value of brine constituents* Assumed brine composition (kg/rn3 of brine): Calcium Magnesium Potassium Lithium Boron Sodium Bromide Iodide Sulfate Bicarbonate
23.36 2.25 5.27 0.34 0.31 57.65 1.95 0.04 0.11 145.60
Assumed products (kg): NaCl CaCl2 MgCl2 KCl LiCl Na2 B4 0 7 10H20 NaBr NaI MgS04 NaHCO
146.01 64.73 8.75 10.06 2.05 2.97 2.52 0.04 0.13 0.10
7.ll/ton** at $ 43.54lton at $ at$ 66.14/ton at $ 30.86/ton a t $ 1,873.91/ton at $ 55.39lton at $ 881.841ton at $ 7,716.10/ton at $ 69.221ton at $ 59.031ton
= 1.04 = 2.82 = 0.58 = 0.31 = 3.84 = 0.16 = 2.22 = 0.31 = 0.01 = 0.01
Maximum worth = $ 11.30 Brine worth = $ 11.30-314 (11.30) = $ 2.82/m3 Brine value = $ 11.30 x 0.1 = $ 1.13
* Assuming: 75% of market cost is operating and fixed charges. ** Metric tons. from the market value of compounds that can be derived from the ions found in the brine. This is not necessarily the m d i m u m worth because the bnnes contain some ions other than those determined. For example, the brines probably contain strontium, barium, rubidium, manganese, etc. Table 13.IX gives the values of chemicals (Anonymous, 1971; U.S. Bureau of Mines, 1969) that were used to calculate the brine worth of a sample, as shown in Table 13.X. Note that the brine worth in Table 13.X depends upon the products that are assumed t o be recovered, and that 75% of the market cost is operating and fixed charges. Table 13.XI illustrates the values that were found for brine worth and brine value for each of the brines collected for the study. Also shown in Table 13.XI is a ratio of brine value for a commercial brine (brine 3) over the disposal brine. Only brines 2, 13, 14,15,16, and 17 hadratios of less than 2,
VALUABLE MINERALS IN OILFIELD WATERS
416
TABLE 13.XI Brine worth, brine value, and ratio commercial brine value/disposal brine value Brine
Brine worth ($/m3)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
0.19 3.69 3.86 0.28 0.17 0.10 0.28 0.14 0.39 0.41 0.41 0.82 2.38 2.50 2.99 2.61 2.48 0.18 0.22 0.16 0.33 0.25 0.19 0.52 1.11 0.48 0.64 0.26 0.51 0.10 0.32 0.37 0.53 0.34 0.38 0.33 0.23 0.01 0.17 1.06 0.96 0.02
Brine value ($/m3)
~
_
0.08 1.45 1.55 0.P1 0.07 0.04 0.12 0.06 0.16 0.17 0.17 0.33 0.95 1.00 1.19 1.04 0.99 0.07 0.09 0.07 0.13 0.10 0.08 0.21 0.33 0.19 0.25 0.11 0.05 0.04 0.13 0.15 0.21 0.14 0.15 0.13 0.09 0.01 0.07 0.42 0.39 0.01
_
State
Ratio ~
_
_
19.38 1.07 1.00 14.09 22.14 38.75 12.92 25.83 9.69 9.12 9.12 4.70 1.63 1.55 1.30 1.49 1.57 22.14 17.22 22.14 11.92 15.50 19.37 7.38 4.70 8.16 6.20 14.09 31.00 38.75 11.92 10.33 7.38 11.07 10.33 11.92 17.22 155.00 22.14 3.69 3.97 155.00
~
Sea water Okla. Ark. Kans. Kans. Kans. Kans. Kans. Kans. Ark. Ark. Ark. Ark. Ark. Ark. Ark. Ark. N.M. N.M. N.M. N.M. N.M. Texas Texas Texas Texas Texas Texas Texas Texas Ala. Ala. Ala. La.
La. La. Calif. Calif. Ark Okla. Okla. Miss.
indicating that the majority of the brines probably do not contain enough valuable minerals t o be considered commercial by themselves. However, these six brines may warrant further investigation, it sufficient brine is available.
CONCLUSIONS
417
Brine 1 is sea water, and some chemicals are recovered from sea water. Therefore any brine that is disposed of in large volumes and has a ratio of less than 20 may warrant investigation as a source of minerals because these brines may be considered as polished ores. Brine 40 along with brine 2 may contain commercial amounts of iodine. Although the current market for iodine is attractive, this market may change, depending upon the international political atmosphere. Nevertheless the market for iodine once established should be fairly stable. Brines 38 and 42 cannot be considered brines; in fact, brine 38 is almost potable and with little treatment would be potable. Brine 42 could be used for irrigation and as drinking water for certain types of livestock. Conclusions Some brines contain valuable minerals that if recovered would help pay for part or all of their disposal costs. Recovery of certain minerals and potable water should lower the potential of the disposed brine as a pollutant. Brine value and brine worth formulas should be applied to disposal waters to determine the relative value of their recoverable minerals. References Angino, E.E., 1967. Dissolved salts in oilfield brines - a wasted resource? In: E.E. Angino and R. Hardy (Editors), Proceedings 3rd Forum o n Geology of Industrial Minerals -State GeoL Sum. Kansas, Spec. Distrib. PubL, No.34, pp.120-125. Anonymous, 1961. Grace moves ahead with new fertilizer. Chem. Eng. News, 39:83-84. Anonymous, 1962. Descaling: route to MAP. Chem. Eng. News, 40:52--53. Anonymous, 1971. Current prices of chemicals and related materials. Oi4 Paint, Drug Rep., 200:24-37. Aries, R.S., 1954. Marketing Research in the Chemical Industry. Chemonomics, New York, N.Y., 220 pp. Bardwell, G.E., 1966. Some statistical features of the relationship between Rocky Mountain arsenal waste disposal and frequency of earthquakes. Mountain Geol., 3: 37-42. Brennan, P.J., 1966. Nevada brine supports a big new lithium plant. Chem. Eng., 7 6 :86-8 8. Bryson, W.R., Schmidt, G.W. and O’Connor, R.E., 1966. Residual salt of brine affected soil and shale, Potwin area - Butler County, Kansas. Kansas State Dep. Health, Bull., 3( 1):28 pp. Christensen, J.J., McIlhenney, W.F., Muehlberg, P.E., Hunter, J.A., Heintz, J.A., Jebens, R.H. and Bacher, A.A., 1967. A feasibility study on the utilization of waste brines from desalination plants, I. U.S. Off. Saline Water Res. Dew. Progr. Rep., No.245, 359 PP. Collins, A.G., 1966. Here’s how producers can turn brine disposal into profit. Oil Gas J., 64: 112-1 13. Cox, R.L., 1967. An examination of the feasibility of mineral recovery from oilfield brines. Dep. Chem. Pet. Eng., Univ. Kansas, 41 pp., unpublished. Crouch, R.L., 1964. Investigation of alleged groundwater contamination Tri-Rue and Ride oilfields, Scurry County, Texas Texas Water Comm. Rep., No.LD-O464MR, 16 PP.
418
VALUABLE MINERALS IN OILFIELD WATERS
Dodge, B.F. and Eshaya, A.M., 1960. Thermodynamics of Some Desalting Processes. Advanced Chemistry Series, No. 27. American Chemical Society, Washington, D.C., 246 pp. Evans, D.M., 1966. The Denver area earthquakes and the Rocky Mountain arsenal disposal well. Mountain Geol., 3’:23-26. George, D.R., Riley, J.M. and Crocker, L., 1967. Preliminary process development studies for desulfating Great Salt Lake brines and sea water. U S . Bur. Min. Rep. Invest., No.6928, 34 pp. Kincaid, E.E., 1956. Two moves pay off at Catesville. Oil Gas J., 54:96-98. Klein, G., Cherney, S., Ruddick, E.L. and Vermeulen, T., 1968. Calcium removal from sea water by fixed-bed ion exchange. Desalination, 4:158-166. Miller, W.C., 1965. Bromine. US. Bur. Min. Bull., 630:159-164. Salutsky, M.L. and D u m t h , M.G., 1962. Recovery of Minerals from Sea Water by Phosphate Precipitation. Advanced Chemistry. Series, No. 38. American Chemical Society, Washington, D.C., 199 pp. Shreve, R.N., 1956. The Chemical Process Industries. McGraw-Hill, New York, N.Y., 2nd ed., 1004 pp. Tallmadge, J.A., Butt, J.B. and Solomon, H.J., 1964. Minerals from sea salt. Ind Eng. Chem., 56:44-56. U.S. Bureau of Mines, 1968. U.S. BuMines Minerals Yearbook, Metals, Minerals, and Fuels. Washington, D.C., Vol. 1-11, 1208 pp. U.S. Bureau of Mines, 1969. U S . BuMines Minerals Yearbook, Metals, Minerals, and Fuels. Washington, D.C., Vol. 1-11, 1194 pp. Waters, Jr., O.B. and Salutsky, M.L., 1968. Separating potassium and sodium sulfate from brines and bitterns. U.S. Bur. Min., W.R. Grace Company, U.S. Patent, No.3,402,018.
Chapter 14. SUBSURFACE DISPOSAL
Because many oilfield waters contain appreciable quantities of dissolved solids that are capable of polluting fresh waters and lands, they must be disposed of in some manner. One of the most generally used methods of disposal is injection into a subsurface aquifer. Sedimentary rocks that were deposited in an ancient marine environment are the most likely type to possess the necessary geologic characteristics for injection sites. The technology for the installation of disposal wells for oilfield wastes is highly developed, and a similar technique has been applied to disposal of other types of wastes (Donaldson, 1964). It is believed that about 75,000 wells have been drilled in the United States for salt-water injection and disposal wells by oil companies (Smith, 1970). History of brine disposal operations Wells for the production of salt water were in operation in the United States as early as 1800, 59 years before Col. Drake brought in the first oil well. Salt well operators were not happy with the discovery of oil in some of their operations and not knowing what t o do with the oil, they often moved t o new locations to avoid the “messy nuisance” which spoiled their operation. Today salt water is the messy nuisance that oil-well operators must handle. Early oil operators simply allowed the produced brines t o run off into streams or drain into fresh-water aquifers. Landowners were so ecstatic over the royalty payments that the oil producers could do whatever they desired with the salt water. Landowners simply took their checks and moved into new locations. Then population density increased, fresh water was being polluted, farm land was damaged, and land was more valuable because of its increasing scarcity compared to the population. Therefore, controls on oilfield brine disposal became necessary. The salt-water brine pond was used as an early method of keeping the brines from fresh-water drainage; however, earthen ponds often leak, and this method polluted fresh-water supplies. Sometimes these brine ponds were called evaporation ponds. However, it was found that in the majority of cases, evaporation and rainfall compensated each other, with the brine volume continually increasing (Jones, 1945). In west Texas, evaporation ponds are successful as a means of brine disposal. In the colder climates, winter usually caused a continuous gain in brine pond gage height. These gains were
420
SUBSURFACE DISPOSAL
not overcome in the summer months. In cases where evaporation ponds are well-lined to prevent drainage, the problem of how to handle the salt deposits begins to mount up. Where solar evaporation conditions are favorable, very little salt is needed for the roads when it snows because it seldom snows there. The expense of oilfield brine disposal is the least where salbwater bodies are nearby, but even here some care must be taken. The oil content must be less than 30 pprn for disposal into the oceans in coastal areas, otherwise, the oil collects on the shore and becomes a hazard to oyster and fish life. Along the Gulf of Mexico strict controls by wildlife authorities and oyster and fish industries monitor the brines disposed in the Gulf. A small quantity of oil gives both fish and oysters a bad taste. Subsurface injection Injection may not be the proper word since pressure is not always necessary. The Plains States are in a hard water belt and have the strictest controls on subsurface disposal. Whether or not subsurface disposal expenses should always be considered the 'most expensive means of disposal is debatable. Pressure maintenance by injected brine assists oil production rates. Oilfield brines are also used for waterflooding; therefore, not injecting the brine may decrease production and result in a more expensive method of disposal. Shallow well disposal was one of the first subsurface disposals employed where a shallow well is defined as one using a horizon less than 305 m in depth. A shallow well takes considerable input brine for a time, but increasing pressures become necessary as time goes by and this continually increases disposal costs. Shallow wells are also more likely to allow the injected brine t o reach fresh-water supplies in some areas. When a deep-seated bed is known to be available, it is less costly in the long run to make use of the deeper formation. Deep disposal wells accept immense volumes of brine by gravity, and eliminating the necessity of injection pressures reduces disposal expenses. It often was discovered that clogging occurred near the bore of the injection well and treatment was necessary to reduce clogging. Today treatment plants compose a large fraction of the subsurface disposal operations. Early in the history of subsurface brine disposal, legislation simply allowed any means of disposal. Since then legislation has for all practical purposes forced subsurface disposal and set up tight controls and safeguards for protection of fresh water. In the 1930's, the Railroad Commission of Texas allowed the increase of oil allowable by 1 m3 for every 50 m3 of brine returned to subformations. The increase was fixed at a maximum of 0.8 m3 (Laurence and Leusler, 1958) and this incentive program did a great deal to encourage subsurface disposal of brines in the East Texas field.
PRESENT-DAY TECHNOLOGY
421
Present-day technology in subsurface disposal Several areas of research in the handling of brines were opened because of the subsurface disposal of oilfield brines. The most notable problems encountered were scaling in the salt-water lines, corrosiveness of brines, inefficient separation of oil from the water, and poor well cements. Today, methods to correct most of these problems are available. The scaling problem is a culprit which causes large expenses in brinehandling operations. Morris (1959) describes the method which the East Texas Salt-Water Disposal Company (1953) uses for handling scale formation. Asbestos-cement pipe is used almost exclusively in their salt-water gathering system. This pipe does not deteriorate from salt-water use and has been found quite practical to prevent corrosion. However, accumulation of scale often occurs t o such an extent that the inside of the pipe must be cleaned regularly to maintain injection capacity. Sections of 8-inch lines often are reduced by scale t o 2-inch lines over a period of years. The East Texas Salt-Water Disposal Company (1953) had the maintenance responsibility for 611 km of such pipeline. To accomplish this they had four pipeline-cleaning crews at work 5 days a week cleaning pipelines exclusively. The most efficient method for scale removal was scrapers. The scraper is forced through the pipeline by pump pressure applied against a rubber cementing plug. The plug is followed by a wire brush, and a second type of scraper is used where high temperatures are used to treat water-oil emulsions. The scale deposits in this case are too hard t o be removed by a wire brush, and the scraper used to remove this scale is mechanically operated by a flexible cable. Table 14.1 compares the costs of pipeline cleaning methods (Morris, 1959). Treating oilfield brines before injection is necessary to remove suspended solids which clog the formation. When a brine is brought to the surface in East Texas, the temperature changes from 64OC t o atmospheric. The pressure reduces from an average of 74 kg/cm2 to atmospheric. Carbon dioxide and petroleum gases, which were dissolved in the salt water, are allowed to escape. Oxygen and other elements of the air mix with the elements of the salt water creating many precipitates. TABLE 14.1 Comparative costs of pipeline cleaning methods Pipe size
Costs ($/m)
(cm)
cleaning machine
acidizing
replacement
10.2 15.2 20.3 25.4
0.249 0.174 1.050 1.234
0.958 1.919 4.259 5.899
4.04 5.77 7.64 9.61
422
SUBSURFACE DISPOSAL
Scale-forming precipitates are removed by settling tanks. Elements that will precipitate in the well are chemically precipitated and removed, and oil is skimmed from the settling tanks. Smaller oil particles are removed by using baffles in the skimming process. Aeration is used t o oxidize ferrous compounds and cause them to precipitate for filtration. A high aeration efficiency tends t o reduce the chemical treatment costs. Chlorination is used as an oxidizing agent t o control algae and bacterial growth because algae. and bacterial growths cause plugging problems in the treatment plant as well as in the injection wells and disposal formations. Chemical treatment with hydrated lime and alum removes iron compounds, calcium compounds, and small amounts of hydrocarbon products. Small floc particles which remain in suspension after chemical treatment to cause precipitation are removed by filtration. Where a closed system can be operated, treatment costs are decreased. Exclusion of oxygen prevents oxidation of iron, maintains a low corrosion rate and prevents the growth of aerobic bacteria. Laurence and Leuszler (1958)point out that the maintenance of pressure on the system will hold carbon dioxide in solution and reduce the precipitation of calcium and magnesium carbonates. In some cases a closed system will maintain the stability of the water for reinjection into the producing formation with little or no treatment. Usually a closed system will require only oil skimming and filtration, but in some cases chemicals and bactericides are necessary t o control bacterial growth in a closed system. Economics and oilfield brine disposal Investments and operating costs for oilfield brine disposal systems are difficult to obtain and compare because of differences in accounting systems used by various oil operations. Some operations have brine disposal costs incorporated into oil production costs, others only separate costs of treating and report it as brine disposal costs. Elliston and Davis (1944)reported a survey taken in the early 1940’s. They report investments for 256 systems totalling $4.2 million and operating costs totalling $1.2 million for 86 systems. Disposal costs then range from a few cents per m3 t o 63 cents per m3. The main variables controlling the costs were the amount of treatment necessary, the size of a system, and the well depth necessary t o dispose of the brine. East Texas Woodbine formation disposal averaged approximately 12 cents per m3 injected. Brine from the El Dorado field in Kansas averaged about 6.3 cents per m3. Operations where one disposal well served only eight production wells yielded a cost of 44 cents per m3. A case where one disposal well served 15 production wells gave a disposal cost of only 5 cents per m3. In general, the more brine disposed into one well, the smaller the cost. A small operator or even a major company is not economically justified in installing a deep disposal well if the development limit of his lease or field is
ECON OM ICS
423
only one or two producing wells. Some deep disposal wells show a potential capacity of 1,600-3,200 m3 of brine intake per day under actual test. Therefore, it is evident that it is possible for several operators to use the same disposal well. The largest oilfield brine disposal association is the East Texas Salt-Water Disposal Company. This company serves the East Texas oilfield located in northeast Texas, which is located in parts of five counties, and the company handles approximately 90%of the brine produced in this field. In 1942, with disposal costs averaging in excess of 12.6 cents per m3 and the amount of produced brine increasing, the disposal company was formed by 250 large and small operators. During the second year of operation, costs of disposal dropped t o about 10.7 cents per m3. Table 14.11 shows the history of disposal costs for the East Texas Salt-Water Disposal Company (1953). Table 14.11 shows that although there was a large difference between the costs of labor, equipment, materials, etc., from 1944 through 1958, the company’s cost per m3 of disposed brine changed very little. This probably indicates that in this case, increasing costs have been balanced by technological improvements. The amount of brine disposed per year roughly follows inverse variations t o disposal cost variations. In the data gathered by Elliston and Davis (1944) on disposal system costs, investment amortization was roughly 63%of the total disposal costs. With greater fixed costs than variable operating costs, the unit costs should increase as the amount of brine injected decreases. TABLE 14.11 East Texas Salt-Water Disposal Company’s costs Year
Brine injected (m3)
1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958
88,600 6,454,000 12,606,000 14,034,000 17,546,000 21,666,000 22,932,000 22,715,000 21,535,000 19,458,000 19,310,000 18,320,000 18,514,000 20,O27,000 20,979,000 21,488,000 21,969,000
Total cost before taxes ($/m3 __
0.450 0.108 0.088 0.090 0.089 0.086 0.087 0.091 0.091 0.096 0.089 0.086 0.092 0.084 0.084 0.086 0.087
424
SUBSURFACE DISPOSAL
The East Texas Salt-Water Disposal Company is considered a public utility under the Statutes of Texas, which simply gives the company a monopoly on salt-water disposal in the East Texas field. The company has in excess of $6 billion invested in over 630 km of pipeline& 35 treating plants, 64 collection centers, 212 centrifugal pumps, and 60 injection wells. All of this investment is not in one integrated gathering system, but in several small systems for different producing areas. One small system which services a producing area is made up of three collection centers, three treatment centers, and five injection wells. In the design of a disposal system for a producing area, the location of treatment plants, collection centers, and injection wells must be optimized. Pumping costs and pipeline costs dictate the economic size and location of collection centers and treatment plants are highly automated, requiring little attention. Pipelines for gathering brines are laid t o obtain gravity flow where possible. The unit disposal costs for a particular field are determined by several factors. The more brine produced in an area, the lower the unit disposal costs. An open disposal system requires more treating than a closed one, hence higher costs. An area where gravity flow is attained for most of the brine will yield lower costs. With the costs of brine disposal per barrel remaining constant, oil producers still realize higher disposal costs on a total basis. As an oilfield is produced, the produced brine per m3 of oil increases. While the disposal costs in subsurface brine disposal are to be minimized, the expenditure is not without benefit t o the oil producer. Some produced brines are combined with additional brine-well production for waterflood purposes. Here the brine is injected into the oil-producing formation in order to displace more oil toward the oil wells. Another value of the brine injection is in pressure maintenance. The pressure in an oil reservoir decreases as the oil and brine are produced. This increases the costs of pumping the oil to the surface. Returning the brine to the reservoir formation helps to maintain the pressure. In the East Texas field the pressure dropped from 114 kg/cm2 to 70 kg/cm2 before the subsurface brine disposal program was inaugurated. The brine disposal halted the decreasing reservoir pressure. Now additional brine from other formations is added to the oilfield brine and the reservoir pressure has increased. As Morris (1960) points out, it is estimated that brine disposal in the East Texas field is responsible for the availability of an additional 95 million m3 of oil. An oilfield with 16 million m3 of oil recovery is considered a major field in the oil industry.
Injection well versus disposal well An injection well in an oilfield waterflooding operation is a well into which water or brine is injected to sweep in-place oil out of the formation and into an oil-production well. A primary production oil well producing from a water-drive reservoir, i.e., already under a natural flood, usually is not subjected t o a waterflood operation except possibly peripheral injection.
ACCEPTABLE GEOLOGIC AREAS
425
Water is injected into an oil reservoir that does not possess a natural water drive; this injection water is obtained from water supply wells, and in many cases some of it is recycled. In oilfield waterflooding operations, an injection well is used to introduce water (often brine) into the strata t o be flooded. In such an operation the oil-water interface is kept as uniform as possible to clearly sweep in-place oil out of the formation and to a production well, i.e., to obtain maximum sweep efficiency. The injection pressure and injection rate usually are low and slow during the beginning of a waterflood, and the input well usually is not fractured because channeling is thus less likely. Disposal wells are used in oil-production operations t o dispose of the waters that are produced with petroleum from a natural water-driven reservoir, and some petroleum reservoirs produce large quantities of such brine water. Disposal wells usually are fractured so that the subsurface formation will accept large quantities of fluids at little or no injection pressure; many of these wells initially operate on a vacuum. In a disposal operation, a uniform flood front and absehce of channeling is not required. The primary consideration normally is to put large amounts of fluid into a reservoir at the least possible cost. Acceptable geologic areas Most of the major synclinal basins contain favorable locations for d e e p well injection (Warner, 1967). The most acceptable areas are in porous sedimentary rock strata, e.g., sandstone, limestone, or dolomite. Such strata are found under about 50% of the land area in the United States, mainly in the Central Plains States and southwest coastal areas. Some areas such as the west coast are underlain by complex geologic strata, which has not been satisfactorily studied. Areas where volcanic rocks are present at the surface usually are not acceptable for disposal wells. McCann et al. (1968) reported on possible disposal sites in the New York area of the Appalachian Basin. They found some suitable and some unsuitable areas. Hardaway (1968) reported on the possibilities of waste disposal in a structural syncline in Pennsylvania and found that disposal into certain horizons appeared promising but that additional seismic and well-test data are needed. Similar studies were made by Briggs (1968) for the Michigan Basin; Edmund and Goebel (1968) for the Salina Basin; Garbarini and Veal (1968) for the Denver Basin; and Peterson et al. (1968) for the San Juan Basin. In summary, the most likely acceptable areas are found in the Puget Willimette Valley and Great Valley of California, in the Mid-Continent and Great Plains, on the Gulf coast, in the central and eastern Great Lakes area, and in much of the Mississippi River drainage basin. Areas that are likely to be unacceptable are the Western Mountain ranges; the Ozark, Wichita, Arbuckle Llano Uplift area; the Mexia fault area; the Atlantic Coast area; the
4 26
SUBSURFACE DISPOSAL
Appalachian Mountain area; most of the New England States; and certain areas near western Lake Superior and Lake Michigan.
Geologic maps To conduct a feasibility study of a project in depth, it is desirable to have a suite of maps that show the following (L.R. Reeder, written communication, 1972): (1)surface geology; (2) subcrop at various horizons; (3)structure; (4) tectonics; ( 5 ) convergence; (6)isopachs of various reservoirs; (7) potentiometric gradient of various reservoirs; (8) depth to base of fresh water zones; (9) fresh-water wells; (10)all wells other than fresh-water wells, showing total depths. Suitable disposal zones The development of subsurface disposal operations by the oil industry indicates that almost all types of rocks possess large enough porosities and permeabilities to accept large amounts of fluid under favorable conditions. The injected wastes must be confined t o the disposal formation so that fresh water and other valuable natural resources are protected. Some of the characteristics that are required of an acceptable disposal zone are as follows: (1)The rocks in the disposal strata should have large porosity, permeability, and thickness so that a significantly large volume is available for fluid injection at relatively high rates and at reasonably low pressures. (2) The disposal reservoir should be of large area extent suitable for injection of large quantities of fluid. (3) The reservoir rocks should be uniform and not too heterogeneous to allow calculations concerning the behavior of injection fluids, injection pressures, and possible fluid rock reactions. (4) The injection zone should contain brackish or salty water (a salaquifer). Waters containing more than 1,000 mg/l of dissolved solids are used for domestic, irrigation, and industrial water in some areas (Warner, 1968). (5)The proposed injection zone must be separated from fresh-water zones both laterally and vertically. Such a zone should be vertically below the level of fresh-water circulation and confined vertically by strata that are impermeable. A rule of thumb is the depth at which a confined salaquifer is present; however, this is not always applicable because in some areas salaquifers overlie fresh-water aquifers. ( 6 ) There should be no unplugged or improperly plugged wells penetrating the proposed zone in the vicinity of the disposal well. (7) The fluids t o be injected should be compatible with the rocks in the injection strata and with the fluids in the strata. If they are not, precipitates will form and plug the well. Wastes incompatible with the native fluids can
EVALUATION OF THE DISPOSAL ZONE
427
be injected behind a buffer zone; however, it is difficult t o buffer the rock strata to prevent incompatible fluid and rock reaction (Gabarini and Veal, 1968). (8) The injection zone should have a low internal hydraulic pressure to allow a sufficient margin for injection of fluids without causing hydraulic fracturing of the surrounding strata and to assure a long operating life of the disposal well. (9) The potential injection zone should be surrounded above, below, and laterally by impermeable strata or aquicludes. Many potential zones are surrounded above and below by such strata, and the lateral movement can be monitored in the injection zone. Good seals to prevent fluid movement are provided by anhydrite, clay, gypsum, marl, salt, slate, and unfractured shale. (10) The hydrodynamic gradient, if any, of the proposed disposal formation should be determined so that the path of movement can be calculated. Evaluation of the disposal zone To evaluate a disposal zone requires a detailed study of the geology of the area and maps showing the surface geology, tectonics, surface and subsurface structure and stratigraphy, salinity, potentiometric gradient, and presence of all wells in the area. The information t o construct the maps can be obtained from studies outlined in Table 14.111 (Ross, 1968). Drill cutting samples and cores are taken during drilling. The various logs and drill-stem tests can be run after the entire hole, or a portion of it, has been drilled. Pumping and injectivity tests can be performed through the drill pipe and an open-hole packer before the well is completed or through casing or tubing after the well is completed. TABLE 14.111 Method of obtaining data to evaluate a disposal zone Necessary information
Methods
Fresh-water zones
State agencies, Federal agencies, drill exploitation well, and log electric log, sonic log, radioactive log, core samples drawdown test, pressure buildup test, micrologs, core samples core samples drill-stem tests electric log, sonic log, drilling log, core samples, radioactive log temperature log injection test, differential temperature log, spinner log, radiotracer log
Porosity Permeability Minerals in strata Fluid pressure in strata Subsurface formations and thickness Temperature of strata Injectivity profile
428
SUBSURFACE DISPOSAL
Reservoir transmissibility The reservoir transmissibility can be calculated using knowledge obtained from the following first two items, but it is influenced and may change during disposal operation because of phenomena associated with items shown in (3)through (8). (1)Core analysis data. (2)Reservoir transient tests (Matthews and Russell, 1967). (3)Behavior of fractured reservoirs; opening of natural fractures during pumping (Snow, 1968) (4)Artificially induced fractures (DeLaguna, 1966). (5)Dissolution of rock by injected fluid. (6)Accidentally induced fractures. (7)Plugging from suspended solids; bacteria; corrosion products. (8)Plugging by clay swelling. (9)Plugging by incompatibility of fluids (Ostroff, 1964). Compressibility of rock and water Both the reservoir rock and interstitial fluid are compressible to a very small degree. It is the compression factor that provides the space needed t o inject extraneous fluids into an otherwise full reservoir. Waters and rock in the salaquifer are compressed by the injected waste liquids in an everexpanding cylinder away from the wellbore. Since the rock and water compressibility is of small magnitude, the salaquifer must be of large areal extent t o distribute the pressure buildup. If the formation is confined by faulting, sand pinch-out, or restricted permeability in the region of the disposal well, a very limited area will be available t o compress the formation rock and water, and pressure will build up rapidly or injection rates will decline to a point where the operation becomes impractical. Water compressibility The compressibility of pure water is known to be dependent upon the pressure, temperature, and gas in solution in the water. Note that there is a wide range of compressibilities and that increasing pressures reduce the value, whereas increasing temperatures enlarge it. The compressibility of pure water at 408 atm and 93.3"C is approximately 4.2 x lo-" cm2/dyne. Since with increasing depth higher pressures and temperatures are encountered, it is expected that compressibility will increase, but the magnitude will depend upon the relative increases in pressure and temperqture. A t a given pressure and temperature, the effect of gas in solution in pure water is t o increase the compressibility over that of pure water at the same
EVALUATION OF THE DISPOSAL ZONE
429
pressure and temperature. A graphical method of correction for gas solubility indicates that the effect of gas solubility on the compressibility of water is great, and a reservoir water containing 3.56 m3 of natural gas per m3 will have a compressibility approximately 18%greater than that of pure water at the same temperature and pressure (Amyx et al., 1960).
Rock compressibility The porosity of sedimentary rocks has been shown by Krumbein and Sloss (1963) t o be a function of a degree of compaction of the rock. The compacting forces are a function of the maximum depth of burial of the rock. Sediments which have been buried deeply, even if subsequently uplifted, exhibit lower porosity values than sediments which have not been buried a great depth. Apart from the effect of compaction on grain arrangement, rock minerals are also compressible. Three kinds of compressibility must be distinguished in rocks: (1) rock matrix compressibility; (2) rock bulk compressibility; and (3) bore compressibility (Amyx et al., 1960). The compressibility of each parameter above is the fractional change in volume of that parameter with a unit change in pressure. Data correlated with “net overburden pressure’’ indicate that the pore compressibility is a function of pressure. In summary, pore volume compressibilities of consolidated sandstones are in the order of 7 x lo-” t o 14 x lo-’’ cm2/dyne.
Critical pressures of confining beds Impermeability of overlying and underlying beds is essential. So that the possibility of breakthrough from pressure of injection and pressure from evolved C 0 2 or thermal expansion does not take place, all operating pressures should be below the critical pressures needed to fracture the formations. The value of the critical pressure usually ranges from 0.11 t o 0.33 atm/m of well depth. To design adequate surface and pumping equipment and t o accurately evaluate the hydrologic properties of the disposal formation, injectivity tests should be made. If possible, the test should be made at the critical input pressure; i.e., the point at which the formation begins to fracture. This point will determine the maximum safe injection pressure (McLean, 1969).
Natural and artificial escape routes Joints, faults, fractures from excessive pumping pressures, formation oub crops, and unplugged or poorly plugged wells all represent potential escape routes for injected waste fluids. Many operations can be conducted within the restrictive limits established by the factors above, and some of the factors can be corrected. However, each deserves serious consideration .in conducting a feasibility study for deep-well disposal.
430
SUBSURFACE DISPOSAL
Art i ficial fracturing t o increase permeability Permeability may be increased by artificially fracturing the formation and propping the resulting fractures with silica sand, glass beads, aluminum pellets, or other such agents. Accidentally induced fractures may be produced when critical pressures are exceeded in pumping. Unless these fractures are propped, they will usually close and subsequently heal when pressure is relieved. In any case, the natural tendency will be for fractures to be horizontal when they are developed at depths of less than 305 m and t o be vertical when developed at depths greater than 305 m. However, several sophisticated methods are being tried to direct the fractures below 305 m into the horizontal plane (L.R. Reeder, written communication, 1972). The generally accepted theory is that the attitude of the induced fracture is related t o the regional stress and is oriented in a direction perpendicular t o the least principal stress. Pressure-dista n c e t ime relationship An important but little understood consideration is the pressure effects at various distances from the wellbore for given times and volumes of injected fluids. Where legal situations may develop, or where disposal is conducted in the vicinity of potentially valuable mineral deposits, this factor becomes very important. This information is useful in predicting long-range reservoir performance and design of injection equipment and the effect on unplugged wells in the vicinity. Equations have been developed which describe fluid withdrawal from water wells. These equations can be used t o describe the converse condition; that is, fluid injection into a subsurface formation. The rate at which the pressure increases in a formation and the distance that this higher pressure moves radially out from the injection well can be computed for a specific injection rate from non-equilibrium equations. If equilibrium flow conditions are approached, however, their equations are no longer applicable. As water is continuously pumped into a homogeneous uniform aquifer of infinite areal extent, the pressure radius will increase but at a decreasing rate because of the expanding storage area available (Davis and Dewiest, 1967). Potentiometric levels and gradients should be determined for disposal reservoirs to help analyze and anticipate fluid movement and monitoring methods needed. Depleted oil or gas reservoirs often make ideal reservoirs for the disposal of oilfield brines or other types of liquid waste, because of the volume available as a result of the production of oil and gas. Semipermeable beds Shales and beds of clay at one time were considered impermeable to fluids but it now is postulated that ground waters are transported across these
EVALUATION OF THE DISPOSAL ZONE
431
beds. This postulate assumes that the beds act as semipermeable membranes where the membranes separate waters of different salinity. Transport of water across a shale can result in lower pressure on one side of the shale versus higher pressure on the other side. Assuming that the shale acts as a membrane, the lower pressured side will be the effluent side and will contain filtered or fresher water while the high pressured side will contain the more salty water which will become even more salty as the filtration process proceeds. Because hydrodynamic conditions exist in many ground-water aquifers, it should be a mandatory requirement that the hydrodynamic conditions of the proposed sedimentary disposal aquifer be thoroughly determinzd. Water flow in aquifers usually is determined by use of contour maps cf water elevation in wells plus aquifer permeability and aquifer thickness. This method will not give a true calculation if much water is transported through semipermeable shale or clay beds. Pressure maps are necessary in establishing such transport and the low-pressure aquifer should be used for a disposal site rather than the high-pressure aquifer.
t Injection
Fig.14.1. Cross section of disposal well.
432
SUBSURFACE DISPOSAL
Considerations during drilling and well completion Fig. 14.1 is a cross section of a disposal well. Note that it is cased through the entire fresh-water zone and cemented to a competent horizon below the fresh-water zone. The annulus of the well is filled with an inhibitor fluid under pressure, and the well is equipped with a continuous monitor to detect casing or tubing failure. The well shown in Fig. 14.1 is completely lined with cement to a bottom competent zone. The materials used for tubing, casing, or valves can be carbon steel, plastic, fiberglass, stainless steel, etc., depending upon requirements. Core samples taken during drilling operations should be reacted with the proposed liquid waste to determine what reaction might occur and how to prevent or inhibit the reactions if they are likely to damage the well; for example, to determine what precipitates form and what gases evolve to give pressure increases. Treatment facilities, such as filtration, pH adjustment, and additives, probably will be necessary. For example, the quantity of solids in the fluid and their plugging characteristics with the disposal zone must be determined. If the quantity of suspended solids is excessive, their concentration must be reduced by filtration, settling, decantation, or gas flotation (Amstutz and Reynolds, 1968). A reduction in permeability of the injection horizon and resulting increase in injection pressure or decrease in injection rate can occur as a result of plugging of the pores. Plugging can be caused by suspended solids or entrained gas in the injected fluid, reactions between injected and interstitial fluids, autoreactivity of the waste at aquifer temperature, and pressure and reactions between injected fluids and aquifer minerals. Plugging at or near the wellbore can also be caused by bacteria, mold, and fungi. Selm and Hulse (1959) state that the chemical reactions between injected waste and interstitial water which can cause plugging precipitates to form are as follows: (1)Precipitation of alkaline earth metals, such as calcium, barium, strontium, and magnesium, as relatively insoluble carbonates, sulfates, orthophosphates, fluorides, and hydroxides. (2) Precipitation of heavy metals, such as iron, aluminum, cadmium, zinc, manganese, and chromium, as insoluble carbonates, bicarbonates, hydroxides, orthophosphates, and sulfides. (3) Precipitation of oxidation-reduction reaction products. Additional causes of formation plugging in disposal wells are as follows: (a) partial decomposition or dispersion of salaquifer minerals yielding solid matter in suspension; (b) viscosity increases with an increase of pH; (c) coalescence of gel films at constrictions in pores; and (d) complete gelation of the entire advancing front. These reactions impair the flow the greatest when they occur near the wellbore and the least when they are far removed from the wellbore. Clay minerals occur in sedimentary rocks and are known t o reduce the
FLUID TRAVEL
433
permeability of sandstones to water as compared to their permeability to air. The water permeability of a clay-bearing sandstone decreases with decreasing water salinity, decreasing valence of the cations in solution, and increasing pH of the water. Quartz, feldspars, carbonates, micas and clays, and iron oxides generally constitute the main components in sandstone aquifers. Limestone and dolomite are primarily carbonates but, if impure, may contain as much as 50 percent noncarbonate minerals such as quartz and clay minerals. Quartz, feldspars, and micas are nonreactive except in highly alkaline or acid solutions. Carbonate minerals are soluble in acids. The reaction of carbonate minerals with acid wastes can be beneficial, if no undesirable precipitates form and if the generation of CO, gas does not cause excessive pressure buildup or plugging of the injection zone. Surface treatment or injection of a compatible liquid (buffer liquid) t o move the indigenous brine away from the wellbore before the injection of incompatible waste liquid are two methods of preventing precipitation. If the incompatible fluids do mix later and precipitate solids, they form only at the mixing front and are at such a distance from the wellbore that pore volume is more than ample to contain solids without detectable restriction of injection. Waste waters that are stable on the surface can become unstable at aquifer temperature and pressure. This instability can lead t o polymerization of resin-like materials as suggested by Selm and Hulse (1959).Other reactions, such as the precipitation of calcium carbonate, can occur because of the decreased solubility of dissolved gas at high temperature (Ross, 1968).Case (1970)presents several methods that are useful in handling waters and analyzing problem scales. Organic growths causing operational problems are rare in deep-well disposal projects. Most injected liquids contain sufficient heavy metals and toxic organic or inorganic solutions to provide unfavorable environments for any type of bacterial life. Amstutz and Reynolds (1968)note that fungi can exist and proliferate under a wider range of environmental conditions than can slimes, algae, or bacteria and are more likely to be encountered than the other three types of organisms. He cites an operational problem where fungi growth was encountered in a system disposing of spent sulfuric acid with a pH of 2.8. Problems with organic growths are most likely to appear where liquids are stored or passed through open ponds.
Fluid travel The pressure radius increases but at a decreasing rate when a liquid is pumped into a homogeneous uniform aquifer of infinite areal extent because of the available expanding storage area. Nonequilibrium equations can be used t o compute the distance that the higher pressures move out radially from the injection wellbore and the rate at which the pressures increase
434
SUBSURFACE DISPOSAL
(Wright, 1969; Davis and Dewiest, 1967; Ferris e t al., 1962). Hydraulic gradients and potentiometric levels should be determined for disposal zones to calculate and monitor the injected fluid movement. Hazards of underground waste disposal
Contamination of shallow aquifers Unplugged or poorly plugged wells that penetrate zones into which waste is being injected provide escape routes by which the waste liquid can reach and contaminate shal!ow fresh-water aquifers. This has been common in oilfield experience and is a factor to consider whenever an operation is conducted in the vicinity of old wells. Vertical fracturing caused accidentally by excessive injection pressures or during the process of hydraulic fracturing acts in a manner similar to unplugged wells if the fractures breach the impermeable horizons isolating the injection zone. Surface contamination may occur if there is some malfunction or material failure of surface or well equipment.
Earthquakes The most notable example of a hazard attributed t o deep-well disposal thus far is the Rocky Mountain arsenal well located about 16 km northeast of Denver 1,581 m above sea level, completed September 11, 1961.The well was drilled t o a total depth of 3,671 m, and injection was made into a zone of fractured gneiss from 3,650t o 3,671 m. From March 1962 until February 1966, a volume of 0.625 million m3 was injected at a maximum rate of 1.95. m3/minute and 75 atm (average rate 0.76 m3/minute and 34 atm). The seventh week after injection began, an earthquake of magnitude 1.5 was recorded (April 24, 1962). From April 24, 1962, through August 1967, 1,514 earthquakes were recorded with magnitudes ranging from 0.5 t o 5.3; all were relatively shallow in origin and from an area about midway between central Denver and the arsenal well (Hollister and Weimer, 1968). Present consensus is that the earthquakes are products of a regional stress field of tectonic origin, triggered by the local incremental strain from injection into the arsenal well. In several aspects, however, the stress-strain relationship in the vicinity of the arsenal well seems not t o have been resolved fully. Injection-triggered earthquakes were tentatively identified. State regulations and tax incentives
Regulations Regulations concerning construction and operation of disposal wells are not standard and vary from State to State (L.R. Reeder, written communica-
STATE REGULATIONS AND TAX INCENTIVES
435
tion, 1972). Missouri, Ohio, Texas, and West Virginia have regulations dealing specifically with disposal wells. Apparently no States have regulations which specifically prohibit disposal wells; however, Ohio permits disposal only into the Mountain Simon sand. Texas probably has the most specific and perhaps the most equitable regulations, which are obtained from the Texas Water Development Board. Some of their regulations shown on their Form GW-14 are as follows: “A preliminary report is required before application can be processed. This report should include but not necessarily be limited to the following information: (a) An accurate plat showing location of proposed injection well. (b) A map indicating location of water wells and all artificial penetrations (oil and gas wells, exploratory tests, etc.) of the proposed injection interval(s) in the general area of the proposed injection well. Reasonable diligence shall be used to locate such penetrations. Well and abandonment records for all exploratory oil and gas tests located within the area owned and operated by application should accompany map. (Details within 5 km radius generally acceptable.) (c) Description of local topography and geology pertinent t o injection program. Depth of deepest strata containing fresh water or water of suitable quality for potential beneficial development as determined by well development and/or electrical logs. (Generally required minimum of 91 m of shale between injection zone and base of fresh water.) (d) A detailed description of the chemical, physical, and biological characteristics of the waste to be injected. Complete chemical analyses of all inorganic constituents should be reported in ppm or mg/l. If organic fractions are present, all such constituents should be reported in ppm, mg/l, as individual percentages by weight, or in other appropriate terms. (e) The anticipated average and maximum rate of injection in gallons per minute or barrels per day. Estimated yearly volume of injected waste and anticipated life of project. (Semiannual reports of monthly volumes, injection rates, pressures, cumulative volume, workovers.) (f) Data on completion and operation of proposed injection well: (1)Total depth of well. (2) Casing size, grade, type, weight, and setting depth of all strings; size and type of tubing; name, model, and depth of tubing packer setting. (3) Cement-type and volume of cement t o be used on each casing string and calculated top of cement behind each string. Describe and give percent of all cement additives. Run a cement bond log. (4) Proposed injection interval(s) and perforations. This should include the interval(s) t o be utilized initially and the entire zone requested for future development. (5) Diagramatic sketch of proposed well. (6) Anticipated maximum and average wellhead injection pressures. (7) Description of possible hydraulic fracturing and/or acidizing programs, if anticipated.
436
SUBSURFACE DISPOSAL
(8) Description of proposed injectivity tests. (Logs must be run and submitted. Cement Bond Logs should be specifically required but are not (L.R. Reeder, written communication, 1972). (g) Characteristics of injection interval(s). (1) Lithology, porosity, permeability, temperature. (2) Natural reservoir fluid pressure and equivalent hydrostatic head; fluid saturation and chemical characteristics; and fracture gradient or critical injection pressure. (h) Compatibility of injected waste and formation fluids. (i) Calculated rate of fluid displacement by injected waste and directions of dispersion. (j) Description of program to monitor water quality in fresh-water aquifers. (k) Surface installations. (1) Detailed description of pretreatment process and facilities to be used (include flow diagram if available). (2) Description of type of materials to be used in pretreatment facilities and transmission lines. (3) Description and location of all waste retention ponds, if such are to be used in conjunction with the injection well. In the event an existing well is to be converted to an injection well, applicant should submit a complete electric log, all other logs or surveys performed on the well, and complete casing and cementing data.”
Tax incentives Some states give preferential tax advantages to enterprises utilizing subsurface disposal for water pollution abatement. The following examples were taken from Wright (1969). (1) Connecticut. Exempts pollution control facilities from property tax. (Conn. General Stats. Anno., Sec. 12(81)(51), (Supp. 1966).) (2) Florida. Very limited incentive. Provides lower valuation on control facility for ad valorem tax purposes; no tax on sale of facility, etc. (14A Fla. Stat. Anno., Sec. 403.241.) (3) Georgia. Exempts control facilities from ad valorem tax. (Georgia Code Anno., Sec. 2-5405.) (4) Idaho. Exempts facilities from ad valorem tax. (Idaho Code Anno., Sec. 63-105T.) ( 5 ) Illinois. Provides limited incentive, much like Florida, exempts portion of value of facility from tax on sale of property. (Ill. Anno., Stats., Ch. 120, Sec. 502.) (6) Indiana. Exempts personal property from ad valorem tax when the same is used for pollution abatement. (Ind. Stats., Anno., Sec. 64-241.) (7) Massachusetts. Exempts control facilities from ad valorem tax. Also gives credit against corporate or income tax for portion of cost of facility. (Mass. General Laws Anno., Ch. 59, Sec. 5 (Supp. 1968).)
COSTS OF DISPOSAL SYSTEMS
437
( 8 ) Michigan. Exempts data certified facilities from personal property taxes and taxes on sales of fixtures. (MCLA, Sec. 323.356.) (9) Minnesota. Exempts facilities from ad valorem tax. (Minn. Stats. Anno., Sec. 272.02.) (10) New Hampshire. Provides limited exemption by reduced assessment for property taxes purposes. (Rev. Stats. Anno., Sec. 149: 5-A.) (11) New Jersey. Exempts water or air pollution abatement equipment of devices from ad valorem taxes. (NJSA 54:4-3.56.) (12) New Yorh. Exempts facilities from ad valorem tax. (N.Y. Real Property Tax Law, Sec. 477.) Also provides deduction for expenses associated with pollution control. (N.Y. Tax Law, Sec. 208( 9)(g), 612(h), 706( 9).) (13) North Carolina. Exempts facilities from ad valorem tax. Also gives credit against corporate or income tax. (N.C. General Stats. Sec. 105-296) (Supp. 1967).) (14) Ohio. Exempts facilities from personal property taxes, franchise taxes and sales and use taxes. (Ohio Genl. Code, Sec. 6111.31, Ohio Water Pollution Control Act 1967.) (15) Oklahoma. Provides credit against income tax liability for cost of facility. (82 Okla. Stats., 923.) (16) Oregon. Provides credit against corporate or income tax. (Ore. Revised Stats., Sec. 314.250.) (17) South Carolina. Exempts facilities and equipment from ad valorem tax. (S.C. Stats., Sec. 65-1522(50).) (18) Washington. Provides that owner of facilities may have exemption from ad valorem tax or a credit in a like amount against use of business occupation tax. (RWS 82.04.20.) (19) Wisconsin. Provides deduction for expenses associated with pollution control (Wis. Stats. Anno., Sec. 71.04(26), 71.05(1)(b)5, and 71.05(2B)); exempts facilities from ad valorem tax (Wis. Stats. Anno., Sec. 70.11(21); also provides accelerated depreciation of control works. (Wis. Stats. Anno., Sec. 71.04( ZB).) Costs of disposal systems Rice (1968) gave some investment costs of disposal systems for disposal of oil-associated brines, and at that time he estimated that where large volumes of water from several oil or gas wells are to be disposed of the average cost per well amounted t o $7,900. Pretreatment o f t h e brines such as oil removal by gravity separation, flotation, or filtration adds to the cost of disposal and will vary with the type of operation (Wright and Davies, 1966). Treatment to insure compatibility of the injected waters t o prevent deposition in the injection well and plugging adds t o the cost (Ostroff, 1963). Bleakley (1970) described Shell Oil Company’s salt water disposal operation in its Southern Region Onshore Division. The cost of the total initial installation was $6
438
SUBSURFACE DISPOSAL
million, and they disposed 58,800 m3 of brine per day from 28 fields. Brine disposal costs can represent 2% of net oil sales and up to 25%of total lifting costs (Smith, 1970). Conclusions The storage capacity of potential salaquifers, although considerable in certain areas, is nevertheless limited if long-term continuous disposal is concerned, so that space should be treated as a natural resource. Deep-well disposal has been successful in many cases, and the evidence for questionable operations is inconclusive. However, a high success ratio does not preclude caution in the use of the method. Neither does it relieve the operators of the responsibility of close monitoring of the injected fluids and the reservoir, and the meticulous maintenance of the well facilities. Additional research is needed to develop better methods of evaluating potential disposal reservoirs, of confining and monitoring the lateral movement of fluids in disposal zones, and of determining the rate of mixing of injected fluids with native fluids. Regulations and laws concerning disposal operation should be standardized. References Amstutz, R.W. and Reynolds, L.C., 1968. Is the earth’s crust going to waste, 11. Types of fluids injected and treating procedures Presented at Natl. Pet. Refiners Assoc., MidContinent Regional Meet., Wichita, Kansas, June 12-1 3, 1968. Amyx, J.W., Bass, Jr., D.M. and Whiting, R.L., 1960. Petroleum Reservoir Engineering. McGraw-Hill, New York, N.Y., 610 pp. Bleakley, W.B., 1970. Shell’s SWD meets pollution standards. Oil Gas J., 68:144-146. Briggs, Jr., L.I., 1968. Geology of subsurface waste disposal in Michigan Basin. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f Reservoir Strata A m . Assoc. Pet. Geol., Mem. 10, pp.128-153. Case, L.C., 1970. Water Problems in Oil Production. The Petroleum Publishing Company, Tulsa, Okla., 133 pp. Davis, S.N. and Dewiest, R.J.M., 1967. Hydrogeology. John Wiley and Sons, New York, N.Y., 463 pp. DeLaguna, W., 1966. Disposal of radioactive wastes by hydraulic fracturing. NucL Eng. Design, 3:338-352, 432-438. Donaldson, E.C., 1964. Subsurface disposal of industrial wastes in the United States. U.S. Bur. Min. Inform. Circ., No.8212, 34 pp. East Texas Salt-Water Disposal Company, 1953. Salt-Water Disposal East Texas Field. Petroleum Extension Service, Austin, Texas, 116 pp. Edmund, R.W. and Goebel, E.D., 1968. Subsurface wastedisposal potential in Salina Basin in Kansas. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f Reservoir Strata - A m . Assoc. Pet. GeoL, Mem. 10, pp.154-164. Elliston, H.H. and Davis, W.D., 1944. A method of handling salt-water disposal including treatment of water. Presented at API Meet., Tulsa, Okla.,May, 1944, API Paper, N0.851- 18F. Ferris, J.G., Knowles, D.B., Brown, R.H. and Stallman, R.W., 1962. Theory of aquifer tests, ground-water hydraulics. US. Geol. Sum. Water Supply Paper, No.l536-E, 174 PP.
REFERENCES
439
Garbarini, G.S. and Veal, H.K., 1968. Potential of Denver Basin for disposal of liquid wastes. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f Reservoir Stmta -Am. Assoc. Pet. G e o l , Mem. 10, pp.165-185. Hardaway, J.E., 1968. Possibilities for subsurface waste disposal in a structural syncline in Pennsylvania. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A S t u d y of Reservoir Stmta - A m . Assoc. Pet. GeoL, Mem. 10, pp.93-125. Hollister, J.C. and Weimer, J.C., 1968. Geophysical and geological studies of the relationships between the Denver earthquakes and the Rocky Mountain arsenal well. Colo. School Min. Q., 63(1):1-251. Jones, O.S.,1945. Disposition o f oilfield brines. University of Kansas Press, Lawrence, Kansas, 45 pp. Krumbein, W.C. and Sloss, L.L., 1963. Stratigraphy and Sedimentation. W.H. Freeman, San Francisco, Calif., 2nd ed., 660 pp. Laurence, L.L. and Leuszler, W.E., 1958. ABC's of treating and handling injection water. Pet. Eng., 30:B52-B54, B59. Matthews, C.S. and Russell, D.G., 1967. Pressure Build-up and Flow Tests in Wells. Society of Petroleum Engineers, AIME, 167 pp. McCann, T.P., Privasky, N.C., Stead, F.L. and Wilson, J.E., 1968. Possibilities for disposal of industrial wastes in subsurface rocks on north flank of Appalachian Basin in New York. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f Reservoir Strata - Am. Assoc. Pet. G e o l , Mem. 1 0 , pp.43-92. McLean, D.D., 1969. Subsurface disposal - precautionary measures. Ind. Waste Eng., August 1969: 20-22. Morris, W.S., 1959. Cleaning asbestos-cement pipelines in salt-water disposal service. Pet. Eng., 31:B46-B49. Morris, W.S., 1960. Subsurface disposal of salt Water from oil wells. Water Pollut. Control Fed. J., 32:l-20. Ostroff, A.G., 1963. Compatibility of waters for secondary recovery. Prod. Monthly, 27 :2-4-9. Ostroff, A.G., 1964. Introduction to Oilfield Water Technology. PrenticeHall, Englewood Cliffs, N.J., 412 pp. Peterson, J.A., Loleit, A.J., Spencer, C.W. and Ullrich, R.A., 1968. Sedimentary history and economic geology of San Juan Basin, New Mexico and Colorado. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A S t u d y of Reservoir Strata - A m . Assoc. Pet. G e o l , Mem. 1 0 , pp.186-231. Rice, I.M., 1968. Salt-water disposal in the Permian Basin. Prod. Monthly, 32(3):28-30. Ross, R.D., 1968. Industrial Waste Disposal (Reinhold Environmental Engineering Series). Reinhold, New York, N.Y., 340 pp. Selm, R.P. and Hulse, B.T., 1959. Deep-well disposal of industrial wastes. Proc. 14th Ind. Waste C o n f , Purdue Univ. Eng., Ext. Ser., No.104, pp.566-586. Smith, W.W., 1970. Salt-water disposal: sense and dollars. Pet. Eng., 42:64-65. Snow, D.T., 1968. Fracture deformation and change of permeability and storage upon change of fluid pressure. Colo. School Min. Q., 63:201-244. Warner, D.L., 1967. Deep wells for industrial waste injection in the United States summary of data. US. Dep. Inter., Fed. Water Pollut. Control Adm., Water Pollut. Control Res. Ser. PubL, No.WP 20-10, 45 pp. Warner, D.L., 1968. Subsurface disposal of liquid industrial wastes. In: J.E. Galley (Editor), Subsurface Disposal in Geologic Basins: A Study o f Resenroir Stmta -Am. Assoc. Pet. GeoL, Mem. 1 0 , pp.11-20. Wright, C.C. and Davies, D.W., 1966. The disposal of oilfield waste water. Prod. Monthly, 30:14-17; 22-24. Wright, J.L., 1969. Disposal wells - a worthwhile risk. Presented at 98th Annual Meet., AIME, Washington, D.C., February 16-20 1969, Reprint, 1 5 pp.
This Page Intentionally Left Blank
Chapter 15. SOLUBILITIES OF SOME SILICATE MINERALS IN SALINE WATERS
In a petroleum reservoir the subsurface saline waters are continually in contact with minerals that contain silica. Knowledge of the solubilities of some of these silicate minerals is needed in geochemical studies of sedimentary rocks. To increase this knowledge, a study was made of the solubility of five clay minerals in saline solutions. The solubilities of kaolinite, montmorillonite, nontronite, illite, and serpentine were determined as a function of time in saline solutions at 25OC and 1atm. Hydrothermal solution equipment capable of operating at 25"-600°C and at 1-2,100 kg/cm2 was designed and constructed to study the solubility of a serpentine (verde antique) in saline solutions at elevated temperatures and pressures. Most of the expei-iments were done between atmospheric pressure and the vapor pressure of water at 15OoC(Collins, 1969). Composition and structure of minerals studied Table 15.1 summarizes the approximate chemical compositions of the illite, kaolinite, montmorillonite, nontronite, and serpentine used in the experiments. All samples were natural ones and were obtained from commercial suppliers. TABLE 15.1 Chemical composition of silicate minerals Constituent
Composition (wt.%) illite
kaolinite
montmorillonite nontronite
serpentine
-
SiOz A12 0 3 Fez03 FeO MgO CaO Naz 0
K2 0 Hz O+ Hz 0Ti02 MnO
56.91 18.50 4.99 0.26 2.07 1.59 0.43 5.10 5.98 2.86 0.81
44.82 37.20 0.41 0.07 0.25 0.58 0.40 0.43 12.92 1.76 1.26
-
-
49.91 17.20 2.17 0.26 3.45 2.31 0.14 0.28 7.70 15.77 0.24 0.04
39.92 5.37 29.46 0.28 0.93 2.46
-
7.00 14.38 0.08
-
44.16 0.90 0.27 2.10 40.07 0.02
-
-
7.15 5.01 0.01 0.02
SOLUBILITIES O F SILICATE MINERALS
442
Illite consists of three-layer sheets made up of two layers of silica in tetrahedral coordination, one layer of aluminum in octahedral coordination, and an intersheet of adsorbed potassium. Kaolinite is a double sheet with one layer of aluminum and silica in octahedral coordination and one layer of silica with the silicon in tetrahedral coordination. The chemical structure of montmorillonite consists of two layers of silica in tetrahedral coordination and one layer of magnesium in octahedral coordination with oxygen and hydroxyl in the anion positions. Nontronite belongs to the montmorillonite group and is characterized by its abundance of iron in tetrahedral and octahedral positions; all members of this group swell in water because of introduction of interlayer water in the direction of the C-axis. Serpentine is monoclinic and composed of four molecules of the formula Mg, Si2O5(OH), ;the structure can be chain- or sheetlike. In all silicates, the silicon-oxygen relation is the same; a silicon atom always occurs in the center of four oxygen atoms. This tetrahedron apparently is the fundamental invariable unit of silicate structure. Silicate types differ by the relationship of the tetrahedra in a structure to each other. A serpentine includes at least two distinct minerals, antigorite and chrysotile. Most asbestos are chrysotile. Verde antique is antigorite. The sheet structure of the serpentine is the disilicate type. Such a structure occurs with tetrahedra all in one plane, with each tetrahedron joined to other tetrahedra by three atoms lying in the common plane. Extension of this linkage gives a hexagonal network in the plane. The serpentine minerals are composed of hydrated magnesium silicate layers. These layers may have ordered or disordered stacking arrangements. Oxygen atoms usually are the largest atoms in the structure. They are chiefly responsible, therefore, for the unit cell size. Serpentine minerals can be transformed thermally. Transformation occurs at about 6OO0C and probably proceeds as: 2Mg3Si05(OH),
+
3MgzSi04 + SiOz + Hz0
Serpentines are the magnesium analogs of kaolin. Their basal spacing is 7.2-7.3 8, compared with kaolinite at 7.15 A.
Analytical procedure for dissolved silica 1ml of a 4% ammonium molybdate solution in 0 . N sulfuric acid solution was added to a portion of the aqueous phase; 15 ml of 4.5N sulfuric acid was added, and the mixture was transferred to a separatory funnel. The mixture was agitated for 1minute in the funnel with ethyl acetate. Then the ester was extracted and transferred to a spectrophotometer cell. The absorbance of the ester was determined at 3350 A.
443
SILICATE SOLUBILITIES AT 25OC AND 1ATM TABLE 15.11 Composition of aqueous solutions used in ambient conditions study Solut io n
Molality of added salt
Hz 0
0.23 0.46 0.13 0.25 0.42
HzO + CaClz Hz 0 + CaClz Hz 0 + MgClz Hz 0 + MgClz HzO + NaCl Hz 0 + NaCl Hz 0 + NaHC03 H 2 0 + NaHC03
0.88 0.30 0.61
Silicate solubilities at 25OC and 1 atm Samples of the minerals'(Tab1e 15.1) were ground to 200-mesh size, portions were placed in polyethylene bottles, and each mineral was allowed to react 6 months with the aqueous phases shown in Table 15.11. N o precautions were taken t o exclude CO,; therefore, each phase was presumably saturated with CO, from the atmosphere. Portions of the aqueous phases were removed periodically from the bottles and analyzed for dissolved silica and pH. Fig. 15.1-10 illustrate smoothed solubility curves for dissolved silicon versus time. In general, it appears that as the concentration of dissolved salts in solution is increased, the solubility of the silicate minerals decreases at ambient temperature and pressure.
-
-0 t-
.
0
1
2
1
.03I ,040 ,046 ,044
0
0 -I
Z
.042
.04 I
loo
1,000 HOURS
10
xx)
Fig. 15.1. Silicon concentration as a function of tim: for illite-Hz O,-H2 O-CaClz, -Hz 0-NaHC03 ,-Hz 0-NaCI, and -H2O-MgCl2 at 25 C. a = Hz 0; b = CaC12,0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgClz, 0.13M.
444
-0
a
E"
-
.03I
SOLUBILITIES O F SILICATE MINERALS
I-
"
,048 .046 ,044
0
0 2
.z
,042
.04 I I00
1,000
I o.oO0
HOURS
Fig, 15.2. Silicon concentration as a function of tim% for illite-Hz O,-HzO-CaCI2, -Hz 0-NaHC03 ,-Hz 0-NaCI, and -I& 0-MgC12 a t 25 C. a = Hz 0; b = CaClz, 0.46M; c = NaHC03, 0.61M; d = NaCl, 0.88M;e = MgClZ, 0.25M.
HOURS
Fig. , 15.3. Silicon concentration as a function of time foro kaolinite-H2 O,-Hz 0NaHC03 ,-HzO-NaCI,-Hz 0-CaC12, and -Hz O-MgCl2 a t 25 C. a = Hz 0; b = CaClz, 0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgClz, 0.13M.
c
.04 I 100
1.000
I
I
I I I Ill IC
00
HOURS
Fig. 15.4. Silicon concentration as a function of time forokaolinite-H2 0,-Hz 0NaHC03 ,-HZO-NaCl,-Hz 0-CaC12, and -Hz 0-MgClz at 25 C. a = Hz0; b = CaClz, 0.46M; c = NaHC03, 0.61M; d = NaCl, 0.88M; e = MgClz, 0.25M.
445
SILICATE SOLUBILITIES AT 25'C AND 1 ATM
i
5 . 0 3 4 1
-
!
.
o
3
2
2
In
I
.03I 100
,,
I(
10.000
I.000 HOURS
Fig. 15.5. Silicon concentration as a function of time for montmorillonite-H2 O,-H2OCaC12 ,-H2 0-NaHC03 ,-Hz 0-NaCl, and -Hz 0-MgC12 a t 25'C. a = Hz 0; b = CaClZ, 0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgC12, 0.13M.
.04 . I
0
I00
4
1,000
2
~
10,000
HOURS
Fig. 15.6. Silicon concentration as a function of time for montmorillonite-H2 O,-Hz 0CaClz ,-Hz 0-NaHC03 ,-HZ 0-NaCl, and -Hz 0-MgC12 at 25'C. a = H2 0; b = CaCl2, 0.46M;c = NaHC03, 0.61M; d = NaCl, 0.88M;e = MgCl2, 0.25M.
100
1,000 HOURS
10 100
Fig. 15.7. Silicon concentration a s a function of time for no~tronite-H2O,-H2 a t 25 C. a = HzO; b = CaCl2, 0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M;e = MgClz, 0.13M.
O-CaCl~,-H~0-NaHC03,-HzO-NaCl, and -H20-MgCl2
SOLUBILITIES OF SILICATE MINERALS
446
i00
1,000 HOURS
Fig. 15.8. Silicon concentration a s . a function of time for nontronite-HzO,-Hz O-CaC1~,-H~O-NaHCO~,-H~O-NaCI,and -HzO-MgC12 a t 25'C. a = HzO; b = CaClz , 0.46M; c = NaHC03, 0.61M; d = NaCI, 0.88M; e = MgClz, 0.25M.
.032
f
-
.031 ,048-
l.000 HOURS
IC
Fig.' 15.9. Silicon concentration as a function of time for serpentine-Hz 0,-HzOCaClz ,-Hz 0-NaHC03 ,-HZ 0-NaCI, and -Hz 0-MgC12 a t 25'C. a = Hz 0; b = CaClz, 0.23M; c = NaHC03, 0.30M; d = NaCl, 0.42M; e = MgC12, 0.13M. ,032
-
,031 .048
-
I 100
I
I I I
Ill 1.000
noum
I
1 1 1 Ill1 10
Fig. 15.10. Silicon concentration as a function of time for serpentine-Hz 0,-Hz 0CaClz,-H? 0-NaHC03,-HzO-NaCI, and -HzO-MgCIz a t 25'C. a = H z O ; b = CaC12, 0.46M; c = NaHC03, 0.61M; d = NaCI, 0.88M;e = MgClz, 0.25M.
EXPERIMENTAL EQUIPMENT
447
Experimental equipment Because the silicates, which cause scale deposits in desalination equipment, react and equilibrate slowly with aqueous solutions, pressurized thermal equipment was used t o approach equilibrium more rapidly. No appropriate equipment was commercially available; therefore, the U.S. Bureau of Mines designed and built the equipment shown in Fig. 15.11-14, based on the type used by Dickson et al. (1963). The pressure vessel was machined from sonic-tested stainless steel, and the sample container was machined from pressure-formed Teflon 7. A Teflon filter disk 1.429 cm in 0.d. by 0.318 cm thick (item 7 in Fig. 15.14) was used t o filter the aqueous phase when a sample was collected for analysis.
Fig. 15.11. Hydrothermal equipment; A P pressure gage, 30,000 psi; B = temperature recorder; C = pressure recorder; D = temperature controller; E = rheostat; F = highpressure valves; G = reactor oven; H = reactor rocking mechanism; Z = hydraulic pump; J = hoist; K = Teflon sample container; L = closure piece; and M = sampling valve.
44 8
SOLUBILITIES OF SILICATE MINERALS
EXPERIMENTAL METHOD
44 9
The equipment was capable of operating at 30"-600°C and at ambient to 2,100 kg/cm2. It was designed to agitate the sample continuously by rocking the muffle reactor forward and backward t o 70" from the horizontal. To prolong a run, aqueous phase was added t o the sample container with the recharging assembly shown in Fig. 15.12. Experimental method The solubility of a serpentine (verde antique) was determined at temperatures up to 200°C and at pressures up to 1,055 kg/cm2 in H20-NaC1 and H20-CaC12 phases. Serpentine ground t o 200 mesh was placed in the Teflon container along with several small chunks of the same material. Then the aqueous phase was added, and the container was placed in the reactor. Aqueous phases were prepared with freshly distilled water and 99.99% purity salts. Equilibrium was approached by heating the sample t o 200°C and adjusting the pressure to 1,055 kg/cm2. The sample was equilibrated at these conditions for 48 hours before a portion of the aqueous phase was withdrawn for analysis. The temperature then was dropped 25" or 50°C sequentially while the pressure was held constant, and portions of the aqueous phase were withdrawn every 24 hours and analyzed for pH and dissolved silica. Fundamental equations Fundamental equations that denote thermodynamic relations for the effect of variation of pressure, temperature, and composition of solution upon the solubility relations of chemical compounds emerge from the condition for equilibrium in multicomponent systems derived by Gibbs (1928). At saturation (a condition of equilibrium between solid serpentine and soluof the Mg3Si205(OH), in the solid must tion), the molar free energy, equal the partial mold free energy, G, of the Mg3Si2O5(OH), in solution. A J ~infinitestimal change maintaining equilibrium dG, therefore, must equal dG as shown by:
e,
where u = the molar volume of serpentine; s = the molar entropy of serpentine; = the partial modal volume of serpentine in solution at saturation; S, = the partial molal entropy of serpentine in solution at saturation; m = the total molal concentration of H4Si04 in solution; T = the absolute temperi- bars. ature, and P = the pressure, n The molar free energy, G, is a function of temperature, pressure, and concentration. At saturation, the concentration, m, is also a function of temperature and pressure. As related to the quantities v, and (aG/am)p,T, the effect of pressure
v,
v!,
! d
4'
Fig. 15.13. Muffle furnace with pressure vessel, tubes, and thermocouples in place; 1 = thermocouple; 2 = sampling tube,lined with Teflon spaghetti tubing; 3 = pressure control tube; 4 = insulation; 5 = heating elements; 6 =closure piece; 7 = steel piece for hoist lift; and 8 = front view.
P
?
?
B
I'
Fig. 15.14. Pressure vessel with Teflon sample container in place; 1 = pressure vessel; 2 = thermocouple well; 3 = Teflon seal; 4 = sample tube lined with Teflon spaghetti tubing; 5 = high-pressure tubing; 6 = closure piece; 7 = Teflon filter; 8 = Teflon sample container; and 9 = sample bottle closure pieces.
45 1
FUNDAMENTAL EQUATIONS
on the solubility of serpentine at constant temperature is given by:
where the change in molality with pressure at saturation and constant temperature is (am/aT)T. The effect of temperature on-the solubility of serpentine at constant pressure and the quantities s, S,, and (aG/am)p,T is shown by:
where (am/aT)pequals the change in molality with temperatures at saturation and constant pressure. At saturation and constant composition, the relationship between the effect of pressure variation on the equilibrium tems, and & is given by: perature and the quantities u,
v,,
where (aT/aP), = the change in temperature with pressure necessary t o maintain constant composition of the solution at saturation. &, and (aG/am),,~are unknown quantiIn equations 2, 3, and 4, u, ties. Although values for s are reported in scientific literature, only two of these equations are independent; therefore, a third relationship must be derived t o calculate values of these quantities. The differential molal entropy of solution at saturation (s - K) or ASis the change in entropy that occurs when 1mole of serpentine is dissolved in a large volume of solution saturated with H4Si04. Multiplying AS, by the absolute-temperature gives the differential heat of solution at saturation fi,.If S, is more than s, the entropy of H4Si04 enlarged when serpentine is dissolved. The value of (aT/aP]min equation 4 may bedetermined wjth an equation -_derived from constant molality if AS,, AH,,or AV, is known. To empirically obtain V,, S,, or (aG/am)p,T,however, a value is needed h terms of activity coefficients as shown by:
v,,
where u = the number of ions formed by the dissociation of one H4SiOq molecule, R = the gas constant, T = the absolute temperature in degrees Kelvin, rn = the molal concentration of H4Si04, and y+ = the mean molal activity coefficient of H4 Si04 in solution.
SOLUBILITIES OF SILICATE MINERALS
452
Equations is obtained by combining equations 2 and 5 and can be used t o calculate V , or V , because the value of v is known:
Values for (aln rn/aP)T can be obtained by differentiating empirical equations for the solubility of serpentine as a function of temperature at a series of pressures (see Tables 15.111-X). Evaluation of equation 6 , however, requires a method of determining the mean molal activity coefficient for changes in molality at constant temperature and constant pressure and at the saturation molal concentration. Although activity coefficients can be evaluated in several ways, most methods are based on dilute-solution treatments derived from Debye-Huckel theory. This theory, even when applying the limiting law (Lewis and Randall, 1961) shown in equation 7, is applicable only t o solutes in infinitely dilute solutions: In yf = A 1 z+z-
1 6
(7)
where A = a theoretical parameter and is a function of T and P ; 1z+z-I = t h e absolute of the product of the charges on the solute ions (for H2Si04); and s = the ionic strength of the solution (s equals one-half the summation of the products formed by multiplying the molal concentration mi of each individual ion multiplied by the charge of that ion zi2 and s = '/zZjrnizi2). Equation 8 is a more complete form of the Debye-Huckel expression because it accounts for the average effective diameters of the solution ions and is used to calculate activity coefficients at concentrations t o 0.01 molal:
where B = a theoretical parameter and is a function of T and P , and a = the average effective diameter of the solute ions.
Differentiation of equation 8 yields:
Most of these parameters, except y+, are reported in the literature (Klotz, 1950); the parameter yk for H4Si04 in solutions more concentrated than 0.01 molal is not reported. These values can be obtained graphically (Klotz, 1950) by plotting the logarithm of the molal concentration of H4Si04 in solution versus the square root of the ionic strength of the solution, drawing a smooth curve through the experimental points, and extrapolating to zero
EXPERIMENTAL DATA
453
ionic strength to obtain the logarithm of the square root of the equilibrium constant K,. The y? can be calculated from: yk = Ka'h/m
(11)
Because of limited data, the activity coefficients were not calculated. Experimental data and empirical equations Table 15.111 presents the smoothed data for the amounts of silicon that went into solution when reacting serpentine with aqueous 0.025M calcium chloride solutions at various temperatures and pressures. The coefficients for a, b, c, and d , and the standard deviations u given in Table 15.IV were derived from a least squares fit of the silicon solubilities of serpentine in aqueous 0.025M calcium chloride solutions for a first-degree equation S = a+ bt, a second-degree equation S = a + b t + c t ' , and a third-degree equation S = a + bt + ct2 + d t 3 . Table 15.V gives the smoothed data for the amounts of silicon that went into solution when reacting serpentine with 0.05M calcium chloride at TABLE 15.111 Smoothed molal silicon solubilities from serpentine in 0.025molal calcium chloride solutions at various temperatures and pressures Temperature Pressure ("C)
30 50 76 100 125 150 175 200
176 kg/cm2
352 kg/cm2
703 kg/cma
1,055kg/cm2
0.1045 x 0.1120 0.1150x 0.1075 x 0.9650x 0.7200x 0.4200 x 0.1150 x
0.1735 x 0.1810 x 0.1795 x 0.1705 x 0.1535 x 0.1270 x 0.90oo lo4 0.5500 x lo4
0.2045 x 0.2110 0.2120 0.2050 x 0.1895 x 0.1655 x 0.1350 x 0.10oo
0.2585 x 0.2660 x 0.2670x lo-' 0.2605 x 10" 0.2455 x 0.2210 0.1905 x 0.1590 x lo-'
10-~
lo4 lo4 lo4 lo4
various temperatures and pressures. Table 15.VI contains the coefficients for first, second, and third degree equations derived from a least squares fit of the data in Table 15.V. Table 15.VII presents the smoothed data for the amounts of silicon that went into solution when reacting serpentine with aqueous 0.05M sodium chloride solutions at various temperatures and pressures. Table 15.VIII contains the coefficient for first, second, and third degree equations derived from a least squares fit of the data in Table 15.VII.
454
SOLUBILITIES OF SILICATE MINERALS
TABLE 15.IV Coefficients for empirical equations and resulting standard deviations (a) derived from the data in Table 15.111 Pressure 176 kg/cm2
S=a+bt a b U
352 kg/cm2
703 kg/cm2
0.14567 x 0.22177 x -0.55728 x 10” -0.71175 x 0.16388 x lo4 0.17652 x
S = a + b t + c t2 a 0.86221 x b 0.81657 x c -0.60087 x U 0.21557 x
S = a + b t + ct2 + d t 3 a 0.77886 b 0.11362 c -0.92296 0.93284 d U 0.18318
0.24774 x 0.30126 x -0.61810 x 10” -0.59899 x 0.16566 x lo4 0.16282 x
lo4 0.15754 x lod6 0.77264 x lo-’ -0.64922 x lo-’ lo-’ 0.18634 x lo-’
lo4 lo-’ x lo-’ x lo-’ x lo-’
x x
0.14803 x 0.11370 x -0.10164 x 0.10636 x 0.13391 x
’
1,055 kg/cm2
lo-’ lo-’ lo-’’ lo--’
0.18723 0.78028 -0.61160 0.99021
x
0.18125 0.10095 -0.84257 0.66894 0.56229
x
x
x
lo-*
x
x
x x x
0.24202 x 0.77015 x 10” -5.59881 x lo-’ 0.17255 x lo-’
loW3 lo-’
x x
’
x
lo-’ lo-’
lo4
0.23157 0.11707 -0.10025 0.11693 0.97345
lo-’
x lo-’’ x 10”
TABLE 15.V Smoothed ‘molal silicon solubilities from serpentine in 0.05 molal calcium chloride solutions at various temperatures and pressures Temperature
Pressure
(“C)
30 50 75 100 125 150 175 200
176 kg/cm2
352 kg/cm2
703 kg/cm2
o.1010
0 . 2 0 4 5 ~lov3 0 . 2 1 0 5 ~lov3 0.2025 x 0.1955 x 0.1850 x 0.1720 x 0.1540 x 0.1325 x
0.3165 x 0.3355 x 0.3175 x 0.3170 x 0.2975 x 0.2840 x 0.2735 x 0.2525 x
10-~ 0.1030 x 0.9850 x loW4 0.9300 x 0.8650 x lo4 0.7250 x 0.5650 x lo4 0.3650 x
EXPERIMENTAL DATA
455
TABLE 15.VI Coefficients for empirical equations and resulting standard deviations (a) derived from the data in Table 15.V -.
Pressure 176 kg/cm2
352 kg /cm
703 kg/cm2
S = a + bt
a b (3
0.12328 x lo-’ -0.37430 x 0.76194 x 10-
S = a + bt + ct2 a b C
U
0.95751 x 0.26192 x -0.27826 x 0.12056 x
lo4 lov6 lo-’ lo-’
S = a + bt + ct2 + d t 3 a 0.98227 x lo4 b 0.16698 x 10C -0.18257 x lo-’ d 0.27713 x lo-’ U 0.11573 x lo-’
’
0.23102 x loV3 -0.43229 x 10” 0.80191 x lo-’
0.34775 x -0.42874 x 0.93876 x
0.20276 x 0.21998 x --0.28487 x 0.20391 x
0.32185 x lo-’ 0.16988 x 10” -0.26181 x lo-’ 0.61658 x lo-’
lo-’ lo-’
0.19892 x 0.36685.~10” -0.43253 x lo-’ 0.42652 x lo-” 0.19706 x lo-’
0.29920 0.10385 -0.11373 0.25355 0.53361
x x x x x
lo-’
lo-’ lo-’ lo-’’ lo-’
TABLE 15.VII Smoothed molal silicon solubilities from serpentine in 0.05 molal sodium chloride solutions at various temperatures and pressures Temperature (“C) 30 50 75 100 125 150 175 200
Pressure 176 kg/cm2
352 kg/cm2
703 kglcm’
0.1700 x lo4 0.1950 x lo4 0.2100 100.2200 lo4 0.2200 lo4 0.2100 0.1950 x 100.1700 x lo4
0.3700 0.3950 0.4100 0.4300 0.4200 0;4100 0.4000 0.3700
0.5300 0.5500 0.5600 0.5700 0.5600 0.5500 0.5250 0.5050
x 10x lo4 x lo4 x x lo4 x lo4 x lo4 x lo4
x
lo4
x lo4 x lo4 x lo4 x lo4 x lo4 x lo4 x 10-
1,055 kglcm’ 0.7400 x 100.7700 x lo4 0.7750 x lo4 0.7900 x lo4 0.7650 x lo4 0.7600 x lo4 0.7400 x lo4 0.7150 x lo4
456
SOLUBILITIES OF SILICATE MINERALS
TABLE 15.VIII Coefficients for empirical equations and resulting standard deviations data in Table 15.VII
((7)
derived from the
Pressure
176 kg/cm2
S=a+bt a b U
S = a + bt + ct2 a
b C
U
0.19939 x -0.56601 x 0.18996 x
lo4 lo-’
352 kg/cm2
0.40047 x lo4 0.56299 x -0.13533 x 10- -0.17005 x 0.2Q454 x lo-’ 0.18211 x
lo4
x x x x
lo4 lov6
lo4
x x x x x
lo4
0.13066 x lo4 0.32714 0.15828 x 0.16960 -0.69472 x 10*-0.74119 0.28312 x 0.40924
S = a + bt + ct2 + dt3 a 0.12438 b 0.18236 C -0.93739 d 0.70283 U 0.26986
lo4
0.31912 0.20036 x -0.10512 x lo-’’ 0.89777 x 10” 0.39434
x
703 kg/cm2
x 10”
0.49784 x 0.13354 x -0.65845 x low6 0.38334 x
lo-’ 10-
0.48420 0.18585 -0.11856 0.15266 0.33519
x x x x x
lo-’
lov9 lo4 10”
lo-’
1,055 kglcm’ 0.77924 x -0.19774 x 0.19618 x
lo4 loU7 lo-’
0.71234 x 0.13485 x -0.67625 x 0.71101 x
lo4 low6 low9
0.68299 0.24740 -0.18106 0.32852 0.58769
lo4
’
x x x
x x
lo-’ 10-
’
TABLE 15.IX Smoothed molal silicon solubilities from serpentine in 0.1 molal sodium chloride solutions at various temperatures and pressures Temperature (“C)
Pressure
176 kg/cm2 30 50 75 100 125 150 175 200
0.2350 0.2800 0.3100 0.3300 0.3200 0.2900 0.2700 0.1750
lo4 lo4 x lo4 x lo4
x
x
x x
lo4 x lo4 x lo4
35 2 kg/cm2 0.3200 0.3650 0.4000 0.4150 0.4350 0.4050 0.3600 0.2500
x x x x x x x x
lo4 lo4 lo4 lo4 lo4
703 kg/cm’
1,055 kglcm’
0.5700 0.6100 0.6500 0.6600 0.6450 0.6150 0.5700 0.4950
0.1355 0.1435 0.1475 0.1495 0.1480 0.1450 0.1415 0.1325
x x x x
lo4 lo4 lo4 x lo4 x lo4 x lo4 x lo4
x
x x x
x x x x
lo4 lo4 lo4 lo4 low4 lo4 lo4
457
SUMMARY AND CONCLUSIONS TABLE 15.X
Coefficients for empirical equations and resulting standard deviations (a) derived from the data in Table 15.IX Pressure 176 kg/cm2
S=a+bt a b
a S = a + bt + ct2 a b c
a
703 kg/cm2
0.30724 x -0.27390 x 0.45114 x
lo4 lo4 lo-'
0.39656 x -0.24585 x 0.54718 x
lo4
0.14511 0.34728 -0.16387 0.84998
10-
0.20125 x 0.42680 x -0.19742 x 0.12037 x
lo4 0.48267 x lo4 low6 0.34365 x lo-' -0.16876 x lo-' lo-' 0.65103 x lod
S = a + b t + ct2 + d t 3 0.16424 b 0.27393 c -0.89940 d -0.21411 a 0.80897
a
352 kg/cm2
x
x x
lo-'
x
x 104
0.27529 x
lo-'
low6 0.14290 x x lov9 -0.88707 x x lo-' '-0.82867 x lo-' x 0.65538 x lod x
0.64963 x -0.42217 x 0.46091 x
0.47371 0.37802 -0.20340 0.10033 0.63945
'
lo4 lo-'
1,055 kg/cm2 0.14527 x -0.21139 x 0.56072 x
lo-'
0.12509 0.44518 -0.20395 0.10164
lo-' lo-'
x x x x
x x x x
lo4 0.12176 x lov6 0.57294 x lo-' -0.33271 x lo-' lo-'' 0.37291 x lo-'
x
0.90915 x 10"
'
Table 15.IX contains smoothed data for the amounts of silicon that went into solution when reacting serpentine with aqueous 0.W sodium chloride solutions at various temperatures and pressures. Table 15.X contains the coefficients for first, second, and third degree equations derived from a least squares fit of the data in Table 15.IX. Plots of the silicon solubilities versus temperature are shown in Fig. 15.15. Summary and conclusions Solubility determinations were made of five clay minerals in aqueous saline solutions. In general, the dissociation of silicon as silica from the clay minerals decreased with increasing concentrations of dissolved salts at ambient temperatures and pressures. Solubility determinations of a serpentine mineral in aqueous saline solutions at elevated temperatures and pressures were determined in specially designed hydrothermal equipment. The standard deviations of the experimental data were acceptable within the limits of the equilibria time, and the smoothed data yielded acceptable empirical equations. The equipment proved to be of excellent design and construction. The solubilities of silicon from a serpentine in aqueous salt solutions at various temperatures and pressures can be calculated with the coefficients
SOLUBILITIES OF SILICATE MINERALS
458
- ,031 ,040
-
:-
1
- -
,046/
-
-
=
0
2 .034.-
i
0.025 Molal calcium chloride
-
-
0.05 Molol calcium chloride
x - x y
,031 .048 .O 46
KEY
.O 44 0
352 KQ/Cm2
.042x
.04 I 20
1,055 Kg/Cm2 703 Kg/Cm2 176 Kg/Cm2
40
60 00100
200 20 TEMPERATURE, "C
40
60 00100
200
Fig. 15.15. Molal silicon solubilities from serpentine in aqueous chloride solutions at various temperatures and pressures.
for a first-degree equation S = a + bt, a second-degree equation S = a + bt + c t 2 , and a third-degree equation S = a + bt + ct2 + d t 3 given in Tables 15.IV,
VI, VIII, and X. The first-degree equation can be used to make a rapid calculation with fairly good accuracy; the third-degree equation can be used to calculate a more accurate value consistent with the experimental data. Although equation 11 can be used to obtain approximate activity coefficients, considerably more data are needed for calculating accurate activity coefficient values. The solubility values obtained in a study of five clay minerals in aqueous saline solutions indicated that in general the silicon solubilities decreased
REFERENCES
459
with increasing concentrations of dissolved salts at ambient conditions. To make more specific conclusions, a more detailed study would be necessary. References Collins, A.G., 1969. Solubilities of some silicate minerals in saline waters. U.S. offSaline Water Res. Dew. Progr. Rep., No.472, 27 pp. Dickson, F.W., Blount, C.W. and Tunell, G., 1963. Use of hydrottermal y l u t i o n equipment to determine the solubility of anhydrite in water from 1 0 0 C to 275 C and from 1 bar t o 1,000 bars pressure. A m . J. S c i , 261:61-78. Gibbs, J.W., 1928. The Collected Works of J. Willard Gibbs, 1. Thermodynamics. Longmans Green, New York, N.Y., 353 pp. Klotz, I.M.,1950. Chemical Thermodynamics. Prentice Hall, Englewood Cliffs, N.J., 369 PP. Lewis, G.N. and Randall, M., 1961. Thermodynamics. McGraw-Hill, New York, N.Y., 2nd ed., 723 pp.
This Page Intentionally Left Blank
Chapter 16. ENVIRONMENTAL IMPACT OF OIL- AND GAS-WELLDRIL-
LING, PRODUCTION, AND ASSOCIATED WASTE DISPOSAL PRACTICES*
No detailed studies have been made about how drilling fluids, drilling muds, well cuttings, and well-treatment chemicals may contribute to pollution. Studies of well blowouts and possible development of communication between a fresh-water aquifer and an oil-bearing sand have been made (Vedder et al., 1969) as have studies of possible pollution related to poor production practices (Schmidt and Wilhelm, 1938). The fact that the brines produced with oil and gas can contribute t o pollution is well known (Crouch, 1964; Grandone and Schmidt, 1943; Taylor et al., 1940; Wilhelm and Schmidt, 1935), but no universally satisfactory method of their disposal is available. Disposal of brine by solar evaporation in evaporating ponds has been investigated (Gunaji and Keyes, 1968), but final disposal of the residue salts needs further development. Some brines contain valuable minerals which are economically recoverable, and treatment or disposal of such brines should be coordinated with mineral-recovery processes whenever possible (Collins, 1966). Several publications are available about oilfield brine disposal by subsurface injection into porous and permeable strata (Morris, 1956; Payne, 1966; Rice, 1968); the staff of the East Texas Salt-Water Disposal Company (1953) has prepared a comprehensive report that describes gathering systems, pumps, treatment methods, and injection wells. Subsurface injection of oilfield wastes provides a good method for disposal of potential water pollutants, but the results are not always satisfactory (Donaldson, 1964; Watkins et al., 1960). This disposal method has been blamed as the possible cause of earthquakes, and if a natural disaster, such as an earthquake, occurs, new faults or fractures in subsurface strata may provide communications between the strata containing the waste and fresh-water aquifer (Bardwell, 1966; Evans, 1966; Warner, 1966). Drilling
Drilling fluids and muds The most modern drilling method is the rotary system which requires circulation of drilling fluid for removal of drilled cuttings from the bottom
* Reprinted
with permission from Journal
43~2383-2393 (1972).
Water Pollution Control Federation,
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
462
of the hole t o keep the drill bit and the bottom of the hole clean. The drilling fluids are pumped from ground surface through a drill pipe and bit to the bottom of the hole and returned t o the surface through the annulus between the hole and the drill pipe. The flow of formation gas, oil, and brine into the drill hole is blocked by a fluid-mud column which produces a hydrostatic pressure that counterbalances or exceeds the formation pressures. In certain geological environments, abnormally high-fluid pressures are encountered, i.e., the hydrostatic pressure is greater than 0.107 kg cm-* m-' of depth. When oil or gas wells are drilled into such an environment, there always is the possibility of a blowout unless elaborate precautions are taken and correct drilling muds are used. A situation can develop in uncemented or poorly cemented environments if degradation or sloughing around the casing in a high-pressure zone occurs, allowing the pressured hydrocarbons t o flow along the outside of the casing to a zone of lower pressure; Fig. 16.1 shows how this can happen. Drilling fluids may include gases, liquids, foams, and solids suspended in liquids. Liquid drilling fluids include crude oil, fresh water, and salt water. Most of the solids suspended in
Conductor casing
Surface casing Possible pollution
Oil string casing
Siough o f f area
Fig. 16.1. Manner in which heaving shales or incompetent zones slough off and permit communication of a lower zone with an upper zone.
DRILLING
463
TABLE 16.1 Some constituents used in drilling fluids and muds Quebracho extract Lignosulfonates, calcium and chrome derivatives Acrylonitrites (such as hydrolyzed polyacrylonitrite) Sodium salts of meta and pyrphosphoric acid Natural gums Tannins Molecularly dehydrated phosphates Subbituminous products Protocatechuic acid Barite Lignins (such as humic acids) Bentonite Sugar cane fibers Lime Granular material, such as ground nutshells Corn starch Salt water Soluble caustic/lignin product Carboxy methyl cellulose Crude oil Sulfonated crude oil Oil emulsions Sodium chromate Anionic and nonionic surfactants Organophylic clay Soaps of long-chain fatty acids Phospholipids (e.g., lecithin) Asbestos
liquids are called drilling muds and include the following: suspensiuns of clays and other solids in water (water-base muds); suspension of solids in oil (oil-base muds); oil-in-water emulsions (oil-emulsion muds); and water-in-oil clay emulsion (inverted emulsion muds). Table 16.1 lists some of the compounds in drilling muds (Caraway, 1953; Simpson et al., 1961). Sulfonated drilling muds are prepared by: (1) sulfonating asphaltic crude oil with sulfuric acid, followed by neutralization with sodium silicate and ion exchanging with hydrated lime; or (2) absorbing concentrating sulfuric acid on a porous carrier, e.g., diatomaceous earth, and then sulfonating asphaltic crude oil with acid carrier, followed by partial neutralization with sodium hydroxide and ion exchanging with hydrated lime. The usual asphaltic crude oils that are used yield a 5- t o 7-wt.% carbon residue and have an API gravity in the range of 26'-31'. Some blends may contain an 18' API asphaltic crude oil with a 12-wt.% carbon residue blended with a paraffinic 42' API crude oil with a 0.5-wt.% carbon residue. These muds are usually mixed with oil a t the drilling site and used in the drilling
464
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
operation. As the cuttings plus the used drilling mud are recovered from the well, the drilling mud is usually separated from the cuttings and reused. Some, of course, will be lost because it adheres to the cutting; therefore, some will present a possible water or land pollution hazard (Messenger, 1963). The use of quebracho, starch, and carboxy methyl cellulose in formulating drilling muds has decreased in the last decade, whereas the use of chrome lignosulfonates has increased. The use of lime-treated mud systems has also decreased, whereas the use of low-solid muds, invert emulsions, and chrome lignosulfonate systems has increased. Considerable money is invested in drilling muds, especially in the heavier muds; consequently, they are recovered for reuse. Such muds are primarily used for emergencies, such as lost circulation and high-pressure kicks from both gas and salt water. Many of the used muds are treated with high concentrations of lignosulfonates t o produce a stable mud with specific properties. Possible sources of pollution from drilling fluids and muds are the fluids and muds that may be spilled during drilling, those that may escape into a subsurface fresh-water aquifer, those that cling to the drill cuttings, and those that are not reused. The data in Table 16.1 indicate that several constituents in drilling fluid and mud are capable of polluting water and land.
Chemical treatment of wells Wells are treated with acids t o increase the permeability of the reservoir rocks. This increases fluid flow and increases the recovery of oil and gas; it also improves fluid injection in secondary oil recovery and disposal operations. Hydrochloric, nitric, sulfuric, hydrofluoric, formic, and acetic acids are used. Such treatments produce soluble compounds including calcium chloride, sodium sulfate, sodium fluoride, etc., and in addition, may leave partially spent acids in solution. The volume of acid used t o acidize a well may range from 1.9 t o 1 2 m3, depending upon the amount of acid-soluble strata, the thickness of the horizon being treated, and the calculated productivity of the well (Hurst, 1970). Table 16.11 lists the approximate amounts of hydrochloric, formic, TABLE 16.11 Volume of acids used for oil- and gas-well treatment Acid
Volume (m3 /year)
Hydrochloric Formic Acetic
3.3 7.6 x 3.8 x
lo5 lo2 lo2
DRILLING
46 5
and acetic acid used in the United S treatment. Other pollution problems can develop when the salt-enriched solution plus any unspent acid are withdrawn from the well, because subsequent disposal of these solutions is complicated by their tendency t o form precipitates and their low pH. It also is difficult t o inhibit (Harris et al., 1966) the acidtreatment solution t o prevent corrosion, and when corrosion does occur, the acid solutions and other fluids will escape at the point of pipe failure and pollute the adjacent zone, which may be a fresh-water aquifer.
Corrosion inhibitors According to Hurst (1970), a universal “super” inhibitor has evaded the researcher for 40 years. Such an inhibitor would be useful to prevent steel casing and tubing from corroding as a result of acid treatment of a well. The best available high-temperature inhibitor is a combination of sodium arsenite with an alkyl phenolethylene oxide surfactant, and arsenic-type inhibitors have been used for both low- and high-temperature applications since the 1930’s. Table 16.111 lists some of the inhibitors used in the United States (Cowan, 1970). TABLE 16.111 Types and amounts of inhibitors used in oil- and gas-well treatment ~~
Inhibitor Sodium msenite Imidazoline Abiethylamine Coal tar derivatives Acetylenic alcohol-alkyl pyridine
4.54 5.68 3.18 1.14 1.36
lo5 lo5 lo5 lo5 lo5
~~
TABLE 16.IV Types and amounts of other additives used in oil- and gas-well treatment Additive Lactic acid (44%) Citric acid Alkylaryl sulfonic acid Zirconium oxychloride (20%) Quaternary ammonium derivatives Polymers Gum gum Fluid loss agents Emulsion preventers
2.61 x 9.08 x
2.27 x 1.14 9.08 4.54 2.61 x 8.17 2.04
lo5 lo3 lo5 lo5 lo4 lo4 lo6 lo5 lo5
466
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
Other additives To reduce friction, reduce loss, sustain permeability, prevent emulsions from forming, and prevent precipitation, additives are added to the oil- or gas-well systems. Table 16.IV lists some of the compounds used for these purposes and the approximate amounts used in 1year.
Possible pollution from petroleum An opening or cylindrical hole from the ground surface to a subsurface oil-t o gas-bearing formation is a well. Such an opening usually is lined with a metal pipe cemented in place, and production equipment is fastened t o the cased hole to regulate and control oil or gas withdrawal rates. Before drilling
Possible break
Conductor casing Pass i b Ie po I Iu t ton
Surface casin
Fig. 16.2. Probable manner whereby a well blowout can develop communication between an upper sand and a lower sand.
PRODUCTION
467
a well, some knowledge of the geologic formations to be penetrated is useful, as is knowledge of the approximate depth of the target petroleum-bearing zone. This information is needed so that the appropriate diameter, length, and type of tubular goods can be selected in planning the well. Most States have laws requiring the setting of surface casing to protect the fresh-water subsurface sands from invasion by brines and hydrocarbons from deeper horizons. Therefore, a minimum of two strings of casing - the surface casing and the oil-string casing - will almost always be required. Additional strings of casing may be required if heaving shales are found while drilling, if abnormal pressures are encountered, or if a zone of lost circulation is found. Each additional string of casing requires more capital and increases the cost of the well. If appropriate precautions are not taken in planning, drilling, and completing an oil or gas well, disastrous consequences can occur. For example, during drilling operations or when pulling the drill pipe, a well may blow out if adequate mud pressure is not maintained. Such a situation may develop if the mud line is accidentally broken or if the well casing is not properly cemented to competent zones. Fig.16.2 illustrates what might occur if fluid from a high-pressure well escapes into an incompetent zone and develops communication of a lower hydrocarbon-bearing horizon with an upper sand. Production
Possible pollution from petroleum Crude oil in excessive amounts is detrimental to vegetation; oily wastes on surface waters can cause a fire hazard, can be deleterious to fish life, and gradually will combine with particulate matter, sink, and thus pollute the bottom of the stream or lake. Further, crude oil has destructive effects on fowl that may swim in the polluted water and may damage the surrounding flora and the surrounding beaches. Mercury concentrations in excess of 20 ppm are present in some crude oils. In essence then, it can be assumed that excessive amounts of produced crude oil that finds its way to surface lands or waters will cause deleterious pollution. The composition of the crude oil that pollutes the water or land will determine the extent and type of pollution. For example, some heavy crude oils possess a specific gravity of about 1, contain about 5-wt.% sulfur, and have an overall minimum boiling point of about 27OoC. Conversely, some light crude oils contain virtually no elements other than carbon and hydrogen, have 0.8 or less specific gravity, and distill below 27OoC. The major nonhydrocarbons in crude oils are basic and nonbasic nitrogen and sulfur compounds and acidic and nonacidic oxygen compounds. Usually the nonhydrocarbons are more highly concentrated in the heavier portions of the crude oils. In an overall classification, most crude oils can be classified as naphthenic paraffinic, or intermediate; the naphthenic type usually is the heaviest, the paraffinic the lightest.
46 8
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
Once the crude oils escape upon land or water, they are subjected to evaporation, oxidation, solution, dispersion, and utilization by microorganisms. The lighter crude oils will evaporate more readily than will the heavy ones. The lower hydrocarbons, e.g., methane and benzene, though relatively insoluble in water, will be more soluble than the higher molecular weight hydrocarbons; the crude oils containing sulfur compounds probably will oxidize less rapidly than will those containing metallo compounds. Crude oils, when spread on salt water, such as the sea, will quite rapidly form highly stable water-in-oil emulsions, as was exhibited in the Torrey Canyon disaster. This type of emulsfon forms thick blobs of oil which are fairly resistant t o dispersal, oxidation, and bacterial reactions. The reason that this type of emulsion forms with salt water has not been clearly established. A means of readily reverting such emulsions t o an oil-in-water type would be desirable for quick dispersal (Dean, 1968). Emulsions of petroleum and brine or mixtures of crude oils and sand that are difficult to break can be found on surface disposal ponds. Should these ponds overflow, the surrounding land or surface streams will be polluted. Crude oil also may escape from leaky connections, improperly plugged wells, improperly cased and cemented wells, holes in lines or storage tanks, or as a result of an accident. Burning of the petroleum or emulsions, or both, that enter brine ponds can contribute to air pollution, and all of the petroleum will not be completely consumed by the fire. Oil production may produce pollution in onshore or offshore areas from blowouts of the wells, dumping of oil-based drilling muds and oil-soaked cuttings, or losses of oil or brine in production, storage, and transportation. Over 320,000 km of pipelines operating at pressures to 70 kg/cm2 are used throughout the country and in offshore areas. Any rupture or accidental puncture of any of these lines results in pollution.
Possible pollution from natural gas Blowouts of natural gas wells will contribute to pollution, especially if the natural gas contains appreciable quantities of hydrogen sulfide. Many gas wells contain enough hydrogen sulfide t o pollute any fresh water they may contact. Such contact may develop if a well is faulty and communication between the gas zone and an upper fresh-water zone occurs. Brines associated with hydrogen sulfide-bearing gas zones also will contain appreciable quantities of the sulfide.
Possible pollution from oilfield brines Waters associated with petroleum in subsurface formations usually contain many dissolved ions. Those most commonly present in greater than trace amounts are sodium (Na+), calcium (Ca+2), magnesium .(Mg+*), potassium
PRODUCTION
469
(K+), barium (Ba+’ ), strontium (Sr+’), ferrous iron (Fe+’ ), ferric iron (Fe+3), chloride (Cl-), sulfate (SO4-’ ), sulfide (S-’ ), bromide (Br-’ ), bicarbonate (HC03-), and dissolved gases, such as carbon dioxide (CO’), hydrogen sulfide (H,S), and methane (CH,). The stability of petroleumassociated brine is related to the constituents dissolved in it, the chemical composition of the surrounding rocks and minerals, the temperature, the pressure, and the composition of any gases in contact with the brine (Fulford, 1968). Scale inhibitors are added to waters and brines t o prevent the precipitation reactions. Some of the chemicals used in these inhibitors are listed in Table 16.V. TABLE 16.V Chemicals used in scale inhibitors Ethylenediamine tetraacetic acid salts Nitrilotriacetic acid salts Sodium hexametaphosphate Sodium tripolyphosphate Sodium carboxymethyl cellulose Aminotrimethylene phosphate
Knowledge of the oxidation state of dissolved iron in brines is important in compatibility studies. Brines in contact with the air will dissolve oxygen, and their Eh generally will be from 0.35 t o 0.50 mV. Brines in contact with petroleum in the formation normally will have an Eh lower than 0.35 mV, as will waters in contact with reducible hydrocarbons (Hem, 1961). Any change in the oxidation state of brine containing dissolved iron may result in the deposition of dissolved iron compounds. The sediments or precipitate formed from brines can cause environmental pollution directly or indirectly. For example, if the produced brines are stored in a pond, the sediments may cause soil pollution; if the brines are injected into a disposal well, the sediments may plug the face of the disposal formation, resulting in the necessity to increase injection pressures which may rupture the input system. The amount of salt water or brine produced from oil wells varies considerably with different wells and is dependent upon the producing formation and the location, construction, and age of the well. Some oil wells produce little or no brine when first produced, but as they are produced, they gradually produce more and more brine. As some wells become older, the produced fluids may be more than 95% brine; or for each cubic meter of oil coming t o the surface, 100 m3 or more of brine also is produced. The produced brines differ in concentration but usually consist primarily of
470
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
sodium chloride in concentrations ranging from 5,000 t o more than 200,000 ppm; the average probably is about 40,000 ppm. For comparison, note that sea water contains about 20,000 ppm of chlorides. 1 m3 of brine containing 100,000 ppm of chloride will raise the chloride content of 400 m3 of fresh water above the maximum recommended for drinking water. Petroleumassociated brines may escape and contact fresh water or soil in different ways. For example, t o protect the upper fresh waters from the deeper mineralized waters that might rise in the drilled well, the upper portion of the well is sealed by a string of cemented surface casing. If a well has insufficient surface casing, an avenue may be provided for the escape of brines if they are under sufficient hydrostatic head t o cause them to rise in the hole to the surface or to the level of fresh water sands. Handling the tremendous volume of brine produced simultaneously with petroleum is hazardous. Basically, the problem is to handle and dispose of the brine in such a manner that it does not contact soil or fresh water and cause detrimental pollution. Currently, some produced brines are being discharged into approved surface ponds, whereas most brines are returned underground for disposal or to repressure secondary oil or gas recovery wells. The discharge of brines to any surface drainage is strictly prohibited in most States. Potential water and soil pollution problems are associated with both disposal methods. For example, if the surface pond is faulty, the brine will contact the soil and various chemical reactions will occur between the soil and the brine. Sometimes the brine will pass through the soil, reappear at the surface, and produce scar areas; sometimes it will pollute the soil and leaching will pollute surface streams or shallow subsurface aquifers.
Residual salt concentrations beneath or near abandoned unsealed disposal ponds Unsealed surface ponds used for the disposal of oilfield brines have polluted fresh surface waters, potable groundwaters, and fertile land. Because of chemical and physical phenomena and dispersion, the movement of soluble pollutants from these pqnds is complex. For example, the soluble pollutants move slowly in relation to the soil-water flow rate, and dispersion effects a displacement which causes the contaminated zone to grow. The Kansas State Department of Health studied the soils beneath and near and old unsealed brine disposal pond that had been abandoned for 10 years. During its use, the pond received more than 29,000 metric tons of salts, and most of those soluble salts probably escaped by soil leaching and downdrainage and penetrated below the underlying limestone formation. Eleven test holes were drilled into the soil and shale beneath and adjacent t o the pond, both above and below the natural drainage slope. Chemical analysis of the test hole core samples indicated that more than 430 tons (about 1.4%of the original) of soluble residual salt still remained t o be leached out of the
DISPOSAL
471
soil and shale in the pond area. This amount of soluble or leachable salt remaining in the area indicates that the return of the subsurface water and soil t o their prepollution level is a very slow process and may take several decades. Network pollution zones appear to form where formation fracture conjugates occur. Leaching appears to be entirely dependent upon the flushing mechanism provided by meteoric water. The cation concentrations in the clay minerals were evaluated by X-ray diffraction techniques to trace cation transportation rates. Chloride analysis was selected as the most useful single means of detecting the presence of oilfield brine pollution, but the associated cation concentration should also be determined t o formulate a more complete picture. Cation adsorption studies are apparently useful in differentiating brine-polluted soil and shale, clay mineral studies provide the information on the environmental characteristics of the pollution media, and cation exchange information aids in explaining the apparent differential transportation rates of ions in brine seepage solutions (Bryson et al., 1966; Siever, 1968). Disposal
Subsurface disposal A problem associated with subsurface brine disposal is casing leaks in the disposal well, which could allow the brine to enter fresh-water aquifers. Fig. 16.3 shows how an improperly designed disposal well and a leaky oil well can pollute a fresh-water aquifer. Erroneous geologic knowledge of the subsurface formation into which the brine is being pumped presents another problem. Brine usually is pumped into a subsurface formation that contains similar brine; however, exact knowledge of the faulting and fracturing of such a subsurface formation is difficult t o discern. Because the brine is pumped into the formation, bottomhole pressure must not exceed 0.23 kg cm-’ m-l of overburden, or the hydraulic pressure may cause fracturing and in time, the wastes may migrate t o a fresh-water zone. Petroleum-associated brines from two different formations may form precipitates if they are mixed. For example, with a well used for disposal of brines produced from several producing oil wells, it is imperative that precautions be taken in mixing and treating the brines before injection. If the brines are incompatible and inappropriate precautions are taken, there is a possibility that deposits will form and filter out on the face of the injection formation, thus reducing the permeability. The quantity of deposits formed from incompatible brines depends on ions present. The more common deposits resulting from reactions of incompatible brines are gypsum (CaS04 * 2 H’O), anhydrite (CaS04), aragonite (CaCO,), calcite (CaC03), celestite (SrS04), barite (BaS04), troilite (FeS), and siderite (FeCO, ). Subsurface brine disposal can be categorized as confinement or containment; confinement is the placement of brines in a horizon where any move-
472
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
Fig. 16.3. Routes by which salt water can enter fresh water wells from faulty oil or disposal wells.
ment can be controlled or monitored, while containment is the placement that precludes the movement of the brines out of a formation or zone. Note that containment cannot be used for an unlimited supply of brine, but that confinement necessitates the monitoring of the migration of the brines. The necessary knowledge to define the hydrodynamics of brines injected into subsurface environments is expensive to obtain, and much of the necessary fundamental knowledge of subsurface formations is not available. Formations into which brines are often pumped for disposal are called salaquifers, and these zones consist of permeable sedimentary rock. Some information needed before such a zone can be used for disposal operations is: How big is the zone? If the brine migrates in the zone, might it reappear in another zone or perhaps migrate to the surface? What mechanisms control movement in a given salaquifer or perhaps out of it? What steps are necessary t o assure containment or confinement of the brine within the salaquifer? It is difficult, if not impossible, to develop adequate knowledge concerning how or where escape channels may occur from a salaquifer. Test drilling is the only known method that can provide such knowledge, and the drilling is expensive, as is the subsequent evaluation (Drescher, 1965).
DISPOSAL
473
Joint ownership of disposal systems by several companies helps minimize installation and maintenance costs. Brines can be gathered through common lines and accumulated at a central location, so that one disposal well serves many producing wells. Investment costs primarily depend upon the brine characteristics from the producing formations, the receptivity of the disposal formation, and the condition of the surface soil for gathering-line installation. Where the brine production is relatively small, a complete system can be installed for less than $500 per producing well, but if large volumes of brine are produced, the disposal well may be able t o service only a few producing wells, and the cost may be $8,000 or more per well. In 1968 the operating costs in representative fields in the Permian Basin amounted t o 0.42 mill/m3 for 6.04 million m3 of brines (Research Committee, Interstate Oil Compact Commission, 1968; Rice, 1968). In 1967 in Texas, there were 41,000 active oil wells, and about 6,900 active gas wells from which more than 0.8 million m3/day of brine was produced. That amount of brine contains approximately 66 x lo6 kg of salts, and that amount of daily produced salt can pollute 98 million m3 of fresh water to the point that it would not be acceptable as drinking water. Waterflooding of oil sands was begun in Bradford field, Pennsylvania, in 1907, and was developed into a systematic operation after 1934. Considerable care must be exercised in using this method to recover oil, e.g., there is a danger that the reinjected brine will migrate to fresh-water streams. Subsurface disposal of oilfield brines, as well as industrial wastes, is being increasingly used to replace surface disposal (Enright, 1963; Research Committee, Interstate Oil Compact Commission, 1960). The ideal conditions for formations used for such disposal are large areal extent, high permeability and porosity, overlying and underlying aquicludes, low internal pressure, salaquifer, compatible fluids, no unplugged wells open t o an outcrop, and uniformity. The reservoir used for disposal must be tremendous in size, and even though the amount of fluid that can be injected is large it is ultimately limited. Many things are not known about what happens within a formation used for disposing of wastes. For example, many wastes are low in pH and apparently no studies have been made of how the pH changes with time within the formation. Conceivable the acid can react with the rock and perhaps break out. It is known that most accidental fractures of the formation or the overlying aquiclude will be horizontal if the well is no deeper than 300 m. However, if the disposal well is deeper than 450 m, the fracture orientation is likely to be vertical, and vertical fractures can, if large enough, cause communication with an upper zone. Salt water under pressure will attempt to escape from any type of confinement. The salt water may escape through fractures becausa of a mechanical failure within the individual well system, through an old drill hole that penetrates the injection zone, or through a natural fault system caused by a recent earthquake.
474
ENVIRONMENTAL IMPACT OF PETROLEUM PRODUCTION
Recovery o f valuable elements before disposal Elements found in some brines in economic concentrations are magnesium, calcium, potassium, lithium, boron, bromine, and iodine. Many of them are recovered by chemical companies from sea water, salt lakes, and subsurface saline waters (Brennan, 1966; Collins, 1970). Factors which must be considered in evaluating a saline water as an economic resource are the cost of bringing it to the factory, the cost of the recovery process, and the cost of transporting the recovered products t o market. Assuming that a brine is produced only for the purpose of recovering its dissolved chemicals, a prime factor is the cost of pumping the brine. It will cost less t o produce the brine from a shallow well than from a deep well. Therefore, disregarding other factors, a brine must not only contain a certain amount of recoverable chemicals, but it must be available in large quantity before it can be considered economically valuable, and the farther it must be pumped, the more chemicals it must contain. References Bardwell, G.E., 1966. Some statistical features of the relationships between Rocky Mountain arsenal waste disposal and frequency of earthquakes. Mountain Geol., 3: 37-42. Brennan, P.J., 1966. Nevada brine supports a big new lithium plant. Chem. Eng., 73: 86-88. Bryson, W.R., Schmidt, G.W. and O’Connors, R.E., 1966. Residual salt of brine affected soil and shale, Patiwin areas Buller Co., Kansas. Kansas State Dep. Health Bull., 3( 1): 28 pp. Caraway, W.H., 1953. Quebraco in oil well drilling fluids. Petrol. Eng., 25:B81-83, B86, B88, B89, B92. Collins, A.G., 1966. Here’s how producers can turn brine disposal into profit. Oil Gas J . , 64: 112-1 13. Collins, A.G., 1970. Finding profits in oil well waste waters. Chem. Eng., 77: 165-168. Cowan, J.C., 1970. Some secondary properties of chemicals used for mineral scale inhibition. Div. Pet. Chem., A m . Chem. SOC.Meet., Houston, Texas, February 22-27, 1970, Preprints, pp. F47-F57. Crouch, R.L., 1964. Investigations of alleged groundwater contamination, Tri-Rue and Ride oilfields, Scurry County, Texas. Texas Water Comm. Rep., No. L.D.-0464-MR, 1 6 PP. Dean, R.A., 1968. The chemistry of crude oils in relation to their spillage on the sea. In: J.D. Carthy (Editor), Proceedings Symposium Field Studies Council, Biol. Eff. Oil Pollut. Luttoral Communities, London, pp. 1-6. Donaldson, E.C., 1964. Subsurface disposal of industrial wastes in the United States. U.S. Bur. Min. Inform. Circ., No. 8212, 34 pp. Drescher, W.J., 1965. Hydrology of deep-well disposal of radioactive liquid wastes. In: A. Young and J.E. Gallup (Editors), Fluids in Subsurface Environments - A m . Assoc. PetroL GeoL, Mem. 4 , pp.399-406. East Texas Salt-Water Disposal Company, 1953. Salt- Water Disposal East Texas Field. Petroleum Extension Service, Austin, Texas, 116 pp.. Enright, R.J., 1963. Oil field pollution. Oil Gas. J., 61:76-87.
REFERENCES
475
Evans, D.M., 1966. The Denver area earthquakes and the Rocky Mountain arsenal disposal well. Mountain GeoL, 3:23-26. Fulford, R.S., 1968. Effects of brine concentration and pressure drop on gypsum scaling in oil wells. J. Pet. TechnoL, 20:559-564. Grandone, P. and Schmidt, L., 1943. Survey of subsurface brine-disposal systems in western Kansas oilfields. U S . Bur. Min. Rep. Invest., No.3719, 20 pp. Gunaji, N.N. and Keyes, Jr., C.G., 1968. Disposal of brines by solar evaporation. U.S. O f f . Saline Water Res. Dev. Progr. Rep., No.351, 213 pp. Harris, O.E., Henrickson, A.R. and Coulter, A.W., 1966. High-concentration hydrochloric acid aids stimulation results in carbonate formations. J. Pet. TechnoL 18:1291-1296. Hem, J.D., 1961. Stability field diagrams as aids in iron chemistry studies. J, Am. Water Works Assoc., 53:211-232. Hurst, R.E., 1970. Market for completion and stimulation chemicals. Div. Pet. Chem., A m . Chem. Soc., Meet., Houston, Texas, February 22-27, 1970, p.l5(12)F9 (abstract). Messenger, J.U., 1963. Composition, properties and field performance of a sulfcnated oil-base mud. J. Pet. TechnoL, 15:259-263. Morris, W.S., 1956. Salt waters disposal from the engineering viewpoint. Presented to the Res. Comm., Interstate Oil Compact Comm., Dallas, Texas, May 31, 1956. Payne, R.D., 1966. Salt water pollution problems in Texas. J. Pet. Technol., 18: 1401-1407. Research Committee, Interstate Oil Compact Commission, 1960. Production and Disposal of Oilfield Brine in the United States and Canada. The Interstate Oil Compact Commission, Oklahoma City, Okla., 95 pp. Research Committee, Interstate Oil Compact Commission, 1968. Subsurface Disposal of Industrial Wastes. The Interstate Oil Compact Commission, Oklahoma City, Okla., 109 PP. Rice, I.M., 1968. Salt water disposal in the Permian Basin. Prod. Monthly, 32:28-30. Schmidt, L. and Wilhelm, C.J., 1938. Disposal of petroleum wastes on oil producing properties. U.S. Bur. Min. Rep. Invest., No.3394, 36 pp. Siever, R., 1968. Establishment of equilibrium between clays and sea water. Earth Planet. Sci. Lett., 5:106-110. Simpson, J., Cowan, J.C. and Beasley, Jr., A.E., 1961. Some recent advances in oil-mud technology. Presented at 36th Annual Meet., AIME, Dallas, Texas, October 8-1 1 , 1961, SOC.Pet. Eng. Paper, No. 150, 16 pp. Taylor, S.S., Holliman, W.C. and Wilhelm, C.J., 1940. Study of brine disposal systems in Illinois oilfields. U.S. Bur. Min. Rep. Invest., No.3534, 20 pp. Vedder, J.G., Wagner, H.C. and Schollhomer, J.E., 1969. Geologic framework of the Santa Barbara channel retion. U.S. GeoL Sum. Prof. Paper, No.679, pp.1-11. Warner, D.L., 1966. Subsurface injection of liquid wastes. In: N.E. Grosvenor, J.D. Haun and D.T. Snow (Editors), Natural Gas, Coal, Groundwater: Exploring New Methods and Techniques in Resources Research. University of Colorado Press, Boulder, Colo., pp. 1 07-1 2 5. Watkins, J.W., Armstrong, F.E. and Heemstra, R.J., 1960. Feasibility of radioactive waste disposal in shallow sedimentary formations. NucL Sci Eng., 7:133-143. Wilhelm, C.J. and Schmidt, L., 1935. Preliminary report on the disposal of oilfield brines in the Ritz-Canton field, McPherson Co., Kansas. U.S. Bur. Min. Rep. Invest., No.3297, 20 pp.
This Page Intentionally Left Blank
REFERENCE INDEX*
Agaev, A.A., 185 Ahrens, L.H., 133, 136 Akin, G.W., 371 Alexeyev, F.A., 308 Alkemade, C.T., 59 Al’tovskii, M.E., 211 Ambrose, A.W., 272 American Petroleum Institute, 19,25, 27, 47,105,254, 272,347 Amstutz, R.W., 432,433 Amyx, J.W., 429 Andrews, L.J., 179 Angino, E.E., 65,132,226, 390, 399 Anonymous, 179,194,302,308,396, 415 Aries, R.S., 405 Armstrong, F.E., 461 Attaway, D.H., 108, 109, 110 Atwater, G.I., 209 Ault, W.U., 208 Aumeras, M., 161 Baar, C.A., 163 Baas Becking, L.G.M., 170,208 Bacher, A.A., 399,411 Bailey, N.J.L., 303 Baker, D.R., 311, 312 Baker, E.G., 211, 296,298,299,311 Ballinger, D.G., 25,43 Bardwell, G.E., 405,461 Barnes, H.L., 224 Bars, E.A., 429 Bass, Jr., D.M., 429 Baugher,III, J.W., 343, 344 Bazilevich, Z.A., 162 Beasley, Jr., A.E., 463 Beck, K.C., 207,294 Beckman, H.F., 105 Beerstecher, Jr., E., 301 Bennett, J.H., 227
Bentor, Y.K., 225 Bernard, J.L., 157, 216, 217 Berner, R.A., 207 Berry, F.A.F., 182 Berry, J.W., 186, 244 Biles, J., 383 Billings, G.K., 65,244, 317 Birch, F., 212 Bischoff, J.L., 223 Bixler, H.J., 241 Black, A.P., 186 Bleakley, W.B., 437 Blount, C.W., 369, 371, 379, 447 Blyth, C.R., 243 Bogomolov, G.V., 43 Bohon, R.L., 179 Bojarski, L., 254,260, 282 Bond, D.C., 244, 314 Bonoli, L., 182 Booth, R.L., 25 Borchert, H., 228 Bordovskii, O.K., 165,184 Boyle, R.W., 322 Braitsch, O., 163,227 Brannock, W.W., 107 Brasted, R.C., 183 Braus, H., 183 Bray, E.E., 296, 297, 311 Bredehoeft, J.D., 243 Brennan, P.J., 390,474 Brenneman, M.C., 303 Brey, M.E., 65,207 Briggs, Jr., L.I., 425 Brod, I.O.,211 Broecker, W.S., 202 Brongersma-Sanders, M., 204 Bronston, A., 308 Brooks, R.R., 82 Brown, R.H., 434 Bruderer, W., 210
*Only page references to text pages are made in this index. References to pages containing bibliographic details have been omitted. These details are given at the end of each chapter.
478 Bryson, W.R., 405, 471 Buckley,S.E., 12, 169, 181, 313 Buckman, H.D., 138 Burges, A., 186 Burmistov, D.F., 162 Burnam, C.W., 185 Burriel-Marti, F., 54 Burst, J.F., 140, 209, 240, 294, 343 Bush, P.R., 224 Butler, G.P., 205 Butt, J.B., 393, 395, 399 Califet, Y., 182 Caraway, W.H., 207, 297, 372, 463 Carpelan, L.H., 203 Carpenter, A.B., 226 Cartmill, J.C., 210, 298, 311 Case, L.C., 195, 433 Castagno, J.L., '43 Chapman, G., 182 Chave, K.E., 207, 240 Chebotarev, I.I., 226, 262, 265, 267, 289 Chenoweth, P.A., 330, 331 Cherney, S., 399 Chilingar, G.V., 207, 245, 321 Christ, C.L., 50, 167, 198 Christensen, J.J., 399, 411 Christman, R.F., 186 Clark, S.P., 194 Clarke, F.W., 197 Claussen, W.F., 179 Clayton, R.H., 243 Cloke, P.L., 168, 208 Coggeshall, N.D., 12, 181, 314 Collins, A.G., 15, 27, 29, 37, 43, 54, 60, 61, 63, 83, 94, 96, 107, 108, 109, 110, 111, 135, 137, 140, 143, 145, 156, 1 5 7 , 1 6 9 , 1 7 1 , 204, 219, 226, 227, 229, 232, 233, 238, 297, 346, 369, 372, 3 9 0 , 3 9 4 , 4 0 2 , 4 4 1 , 4 6 1 , 4 7 4 Collom, R.E., 272 Columbus, N., 382 Conolly, J.F., 180 Cooper, J.E., 184, 297, 311, 315 Corbett, C.S., 293 Coulter, A.W., 465 Cowan, J.C., 369, 370, 372, 463, 465 Cox, D.L., 392 Craig, H., 91 Crocker, L., 399 Crouch, R.L., 4 0 5 , 4 6 1 Czamanske, G.K., 224
REFERENCE INDEX Dall'Aglio, M., 1 5 2 Dapples, E.C., 208 Davidson, M.J., 308 Davies, D.W., 437 Davis, J.B., 185, 301, 302, 313 Davis, J.W., 145, 147, 369, 372 Davis, S.N., 430, 434 Davis, W.D., 422, 423 Dean, J.A., 54, 80 Dean, R.A., 468 Deffeyes, K.S., 203, 204 Degens, E.T., 181, 182, 207, 210 DeLaguna, W., 428 Deroo, G., 309 DeSitter, L.Y., 242 Dewiest, R.J.M., 430, 435 Dickey, P.A., 1 , 205, 210, 226, 274, 288, 298, 311, 343, 346 Dickinson, G., 343, 344 Dickson, F.W., 369, 371, 447 Diehl, H., 96 Dingman, R.J., 132, 226 Disteche, A., 371 Disteche, S., 371 Diterikhs, O.D., 308, 314 Dodge, B.F., 402 Donaldson, E.C., 419, 461 Drescher, W.J., 472 Dressman, R.C., 25 Drong, H.J., 209 Dudova, M.Ya., 308, 314 Duffy, J.R., 180 Dunham, K.C., 224 Dunlap, H.F., 32, 3 5 Dunseth, M.G., 396 Dunton, M.L., 310 Dutoit, M.M.S., 1 8 6 Duursma, E.K., 178 East Texas Salt-Water Disposal Company, 421,423,461 Ebrey, T.G., 94, 9 6 Eckhardt, F.J., 238 Edmund, R.W., 1 7 7 , 4 2 5 Egleson, G.D., 226, 234 Eichelberger, J.W., 25 Eley, D.D., 177 Elliott, Jr., W.C., 256 Ellis, A.J., 227, 370 Elliston, H.H., 422, 423 Emery, E.M., 184 Emery, K.O., 201, 207, 293 Engelbrecht, R.S., 1 8 3 Enright, R.J., 473
REFERENCE INDEX Epstein, S., 91 Erdman, J.G., 309,310 Eshaya, A.M., 402 Ettre, L.S., 185 Evans, D.M., 405,461 Evans, E.D., 296 Evans, G., 205 Evans, M.E., 177 Ewing, G.C., 203 Fabricand, B.P., 65,207 Fajardo, I., 346, 347 Fash, R.H., 308 Ferguson, W.S., 311 Ferris, J.G., 434 Fertl, W.H., 346, 363,364 Fettke, C.R., 2 Feugere, G., 310, 322 Filonov, V.A., 317,318 Finch, W.D., 343 Fisher, F.L., 107 Fleischer, M., 133, 136, 141, 143, 145, 147,149,150,151,152,153,156, 158,159,161,162,164 Forgotson, J.M., 364 Forsman, J.P., 310 Foster, J.B., 362 Fowler, Jr., W.A., 245,343, 362 Frank, H.S., 177 Frear, G.L., 370 Friedman, G.M., 202,207,234 Friedman, I., 91,243, 244 Fuchtbauer, H., 209 Fulford, R.S., 369, 371, 380, 469 Fulton, Jr., R.B., 154, 216,217 Furman, N.H., 44 Gaida, K.H., 294 Galin, V.L., 321 Garbarini, G.S., 425,427 Garrels, R.M., 50,167,198 Garrett, R.G., 322 Garrison, A.D., 2 Gates, G.L., 207,297,372 Gautier, A., 108 Gehman, Jr., H.M., 310 Geodekyan, A.A., 308 George,’D.R., 399 George, W.O., 107 Gerard, R.E., 310, 322 Gevirtz, J.L., 202 Gibbs, J.W., 449 Ginsburg, R.N., 202 Glater, J., 371
479 Glew, D.N., 371 Goebel, E.D., 425 Goldberg, E.D., 149 Goldschmidt, V.M., 133,136,138, 140, 141,145,147,150,151,155,158 Goldshteyn, R.I., 382 Goncharov, Yu.,152 Gordon, W.C., 3 Gorgy, S., 108 Grabau, A.W., 205 Grabowski, R.J., 157,216, 217 Graf, D.L., 243 Grandone, 461 Grauer, A., 211 Graves, Jr., R.W., 201 Greene, R.C., 373 Grim, R.E., 230, 232 Griswold, W.T., 1 Gullikson, D.M., 297 Gunaji, W.N., 461 Gutsalo, L.K., 313, 318,319 Haddenhorst, H.G., 209 Halliburton Company, 74,118 Ham, W.E., 201 Hames, D.A., 371 Hanshaw, B.B., 224, 320 Hanson, W.E., 181 :Hardaway, J.E., 425 Harkins, K.S., 343, 344 Harris, O.E., 465 Hastings, W.W., 107 Hawthorne, R.R., 32, 35 Hays, J., 245 Hedberg, H.D., 178, 309 Heemstra, R.J., 461 Heintz, J.A., 399,411 Helgeson, H.C., 224 Hem, J.D., 133,135,147,155,159,161, 168,170,469 Hemley, J.J., 152 Henningsen, E.R., 382 Henrickson, A.R., 465 Herrmann, R., 59, 227,228 Hill, G.A., 224, 320 Hiltabrand, R.R., 239 Hiss, W.L., 320 Hitchon, B., 244, 245, 296, 297, 299 Ho, A., 186 Hocutt, C.R., 12,169,181,357 Hodges, Jr., R.M., 241 Hodgman, C.D., 34 Hodgson, G.W., 180,210,296,299,311 Hoerr, C.W., 183
480 Holliman, W.C., 461 Hollister, J.C., 434 Holser, W.T., 163, 205 Hood, D.W., 183,184 Horvitz, L., 308 Hottmann, C.E., 343, 344, 347, 363 Howell, J.V., 1 Hoylman, H.W., 321 Hubbert, M.K., 382 Hulse, B.T., 432,433 Hunt, J.M., 178,205, 210, 296, 298, 310 Hunter, J.A., 399,413 Hurst, R.E., 464, 465 Ibert, E.R., 107 Illing, L.V., 202 Imbimbo, E.S., 65,207 Jamieson, G.W., 205, 310 Jankowsky, W.J., 209 Jebens, R.H., 399,413 Jeffery, L.M., 183, 184 Jeltes, R., 181 Jobelius, H., 211 John, L.M., 183 Johnson, A.C., 322 Johnson, R.K., 343, 344 Johnston, J., 370 Jones, B.F., 223, 225 Jones, O.S.,419 Jones, P.H., 245,298,343,346,347,363 Jones, P.J., 32 Kabot, F.J., 185 Kaley, M.E., 201 Kaplan, I.R., 82, 170, 208 Karasik, M.A., 152 Karaskiewicz, J., 322 Karim, M., 321 Kartsev, A.A., 297,308,314 Kawai, K., 165,179 Kazmina, T.I., 153 Keefter, R.M., 179 Kelley, W.P., 230 Kellog, M.W., and Company, 172,173 Keyes, Jr., C.G., 461 Khitarov, N.I.,140, 232, 240 KidwelI, A.L., 178,298 Kimura, S.,241 Kincaid, E.E., 392 King, R.M., 150 Kinsman, D.D.J., 205 Klein, G., 399 Klemme, H.D., 309
REFERENCE INDEX Klotz, I.M., 352 Knopf, A., 155 Knowles, D.B., 434 Knudsen, M., 172 Kwrner, W.E., 184 Kohout, F.A., 382 Kolodii, V.V., 316 Koons, C.B., 311 Kopp, J.F., 25 Korobov, D.S., 317 Kortsenshtein, V.N., 313,314 Kovieheva, I.S., 210 Koyama, T., 165 Kozin, A.N., 233 Kozlov, M.F., 43 Kramer, J.R., 225, 226, 240 Krause, H.R., 183 Krauskopf, K.B., 107 Kravchik, T.E., 314 Krejci-Graf, K., 232, 317 Krivosheya, V.A., 313 Kroepelin, H., 322 Kroner, R.C., 25 Krouse, H.R., 303 Krumbein, W.C., 204,205,429 Kudelskii, A.V., 43 Kuznetsov, S.I., 301, 302 Kumetsova, Z.I., 211 Kvenvolden, K.A., 205, 315 Lagerwerff, J.V., 371 Lamontagne, R.A., 182 Landes, K.K., 309 Lane, A.C., 3 Lane, E.C., 272 Larson, T.E., 150 Larson, T.J., 241 Latimer, W.M., 29, 167 Laurent, P., 183 Laurence, L.L., 420,422 Leobourg, M., 11 Leuszler, W.E., 420,422 Levorsen, A.I., 197, 295,382 Lewis, G.N., 373,452 Lichtenberg, J.J., 25 Linnenbom, V.J., 181 Litchfield, C.D., 183 Lochte, H.L., 185 Loeb, S., 241 Loleit, A.J., 425 London, E.E., 313, 317 Long, F.A., 180 Longbottom, J.E., 25 Lotze, F., 204
REFERENCE INDEX Louis, M., 182 Low, P.F., 242, 346 Lowenstam, H.A., 202 Lucchesi, P.J., 379 Lucia, F.J., 203,204 Lutz, F.B., 108 Lyon, T.L., 138 Mandl, I., 211 Manheim, F.T., 223 Manjikian, S., 241 Manuel, O.K., 227 Marcy, V.M., 43, 101,122 Margoshes, M., 83 Marsden, S.S., 165,179 Marsh, G.H., 47 Martell, A.E., 145 Mason, B., 138, 143, 144, 149,150, 151, 153,158,159,161,163,165 Matthews, C.S., 428 Mayeda, T.K., 91, 243 Maxey, G.B., 243 McAlister, J.A., 11 McAuliffe, C.D., 180, 181,210, 311,314, 315 McBain, J.W., 183 McBermott, E., 308 McCann, T.P., 425 McCutchan, J.W., 371 McDevit, W.F., 180 McElvain, R.G., 318 McIlhenney, W.F., 399,413 McIver, R.D., 295 McKelvey, J.G., 242 McLean, D.D., 429 McLeod, H.O., 307 McNellis, J.M., 132 Meents, W.F., 243 Meinert, R.N., 310 Meinschein, W.G., 298 Mellon, M.G., 35 Merritt, Jr., L.L., 92 Messenger, J.U., 464 Metcalfe, L.D., 184 Metler, A.V., 371 Meyer, H.W.H., 185 Meyerhoff, A.A., 212 Michaels, A.S., 241 Middleton, F.M., 183 Midgett, M.R., 25 Miholic, S., 150 Miller, E.E., 209 Miller, J.C., 226 Miller, W.C., 390, 399
481 Mills, R. van A., 2, 272 Milne, I.H., 242 Mitgarts, B.B., 153 Moeller, T.,133, 135,136, 141 Mogilevskii, G.A., 297 Moore, C.A., 202 Moore, D., 170, 208 Morgan, C.O., 130 Morgan, J.J., 199 Morris, R.C., 205 Morris, W.S., 421,424,461 Mousseau, R.J., 12,324 Muehlberg, P.E., 399, 411 Muir, R.O., 228 Mun, A.I., 162 Munn, M.J., 1, 2 Murata, K.J., 107 Myagkov, V.F., 162 Nagy, B., 180 Nalco Chemical Company, 186 Namoit, A.Y., 177 Natural Gasoline Association of America, 186 Naumor, G.B., 168 Nektarova, M.B., 185, 316 Nemethy, G., 177 Neuberg, C., 211 Neruchev, S.G., 210 Neuman, E.W., 372 Neumann, H.J., 211 Noad, D.F., 11, 321 Nordby, H.E., 186 Norris, M.S., 12, 314 Nutter, B.P., 11 O’Conner, J.T., 183,474 O’Conner, R.E., 405 Oden, S., 185 Odum, H.T., 145 Ostroff, A.G., 169,256,274,371, 428, 437 Oudin, J.L., 309 Ovchinnikou, N.V., 165 Packman, R.F., 186 Page, H.J., 186 Paine, W.R., 343, 347 Palmer, C., 254,256, 257 Parker, J.W., 224, 321 Pate, B.D., 108 Patnode, H.W., 310 Patterson, M.S., 373 Pavolva, G.A., 165
482 Payne, R.D., 461 Peake, E., 180,296,311 Pearson,C.A., 108,109,110,111 Peck, R.B., 206, 294, 344 Pennebaker, E.S., 344, 363 Peterson, J.A., 320,425 Peterson, J.B., 425 Pettijohn, F.J., 197 Philipp, W., 209 Philippi, G.T., 295, 296, 300, 309,310 Phillips, R.C., 368,369, 372, 380 Phleger, F.B., 203 Piper, A.M., 132 Piper, T.J., 183 Pirson, S.J., 169, 208, 308 Pirsson, L.V., 155,168 Platte, J.A., 101 Plumley, W.J., 201 Pollard, T.A., 207 Popov, A.I., 382 Posner, A.M., 183 Potter, E.C., 27 Pourbaix, M.J.N., 30, 167 Powers, M.C., 294,343 Prescott, J.M., 183 Presley, B.J., 82 Price, L.C., 178, 296 Privasky, N.C., 425 Pugin, V.A., 140, 232, 240 Pusey, 111, W.D., 311 Pytkowicz, R.M., 371 Pyushchenko, V.G., 321 Quaide, W., 205 Querio, C.W., 226,234 Rae, A.C., 182 Rainwater, F.H., 31 Rakestraw, N.W., 108 Ralston, A.W., 183 Ramirez-Munoz, J., 54, 67 Ramsey, T.R., 320 Randall, M.H., 373,452 Rankama, K., 197 Ransone, W.R., 308 Reed, W.E., 210 Reeder, L.R., 426,430,434,436 Reichertz, P.O., 207 Reistle, Jr., C.E., 131,272 Research Committee, Interstate Oil Compact Commission, 473 Rettig, S.L., 223 Reuter, J.H., 181,182,210 Reynolds, L.C., 432,433
REFERENCE INDEX Rice, I.M., 437,461,473 Rich, J.L., 2 Rieke, 111, H.H., 245 Riley, G.A., 206 Riley, J.M., 399 Ringwood, A.E., 194 Risley, G.A., 201 Rittenberg, S.C., 207 Rittenhouse, G., 157,216, 217, 229 Roach, J.W., 320 Robinson, J.W., 65 Robinson, L.R., 183 Rogers, G.S., 256 Rogers, W.B., 1 Rosaire, E.E., 308 Rosenqvist, I.T., 207 Rosin, J., 23 Ross, C.S., 197 Ross, R.D., 426,427,433 Rubio, F.E., 303 Ruddick, E.L., 399 Russell, D.G., 428 Russell, W.L., 242 Sahama, T.G., 197 Salutsky, M.L., 396, 399 Sandell, E.B., 98,99 Sanders, J.E., 207,234 Saraf, D.N., 180 Savchenko, V.P., 316 Scheraga, A.A., 177 Schilthuis, R.J., 1, 3 Schmidt, G.W., 294, 314,343,361, 362, 405 Schmidt, L., 461,471 Schoeller, H., 232, 254, 267,271 Schollhomer, J.E., 461 Schrayer, G.T., 12, 314 Schrink, D.R., 107 Schwab, R., 317 Schwanenbek, F.X., 272 Scribner, B.F., 83 Selby, S.M., 34 Selm, R.P., 432,433 Serebriako, O.I., 317 Sestini, F., 186 Shaborova, N.T., 185, 316 Shaffer, L.H., 369,371,380 Shankland, R.S., 34 Shaw, D.R., 166,244 Shaw, T.I., 166 Shearman, D.J., 205 Shilov, I.K., 316 Shimp, N.F., 243
REFERENCE INDEX Shiram, C.R., 343, 347 Shiskina, O.V., 165 Shreve, R.N., 395 Shvets, V.M., 211, 316 Siegel, A., 182 Siever, R., 207,471 Sikka, D.B., 318 Sillen, L.G., 145, 166 Silverman, S.R., 311 Simons, H.F., 308 Simpson, J., 463 Skinner, B.J., 193 Sloss, L.L., 429 Slowey, J.F., 184 Smales, A.A., 108 Smith, G.F., 96 Smith, H.M., 303 Smith, N.O., 180 Smith, P.V., 211 Smith, W.W., 419,438 Snow, D.T., 438 Sokolov, V.A., 308 Solomon, H.J., 393, 395, 399 Souriragan, S.,241 Spencer, C.W., 425 Spencer, D.W., 311 Spiegler, K.S., 242 Ssutu, L.,371 Stallman, R.W., 434 Stead, F.L., 425 Steelink, C., 186 Stiff, H.A., 131,132,371 Stormont, D.H., 308 Stratton, G., 108 Stumm, W., 199 Subotta, M.I., 297 Sudo, Y.,316 Sugawara, K., 165, 166 Sulin, V.A., 254, 257, 258, 347 Swigart, T.E., 272 Swinnerton, J.W., 181 Tabasaranskii, Z.A., 297 Taggart, M.S.,12, 169,181,313 Taguchi, K., 296 Takahashi, T., 202 Tallmadge, J.A., 393, 395, 399 Taylor, D.W., 206 Taylor, S.S., 461 Templeton, C.C., 369, 372 Terada, K., 165, 166 Terzaghi, K., 206, 294, 344 Tickell, F.G., 128 Thatcher, L.L., 31
483 Timko, D.J., 346, 363, 364 Tissot, B., 309 Tollin, G., 186 Tooms, J.S., 227 Torrey, P.D., 1, 2 Trask, P.D., 310 Trelease, S.F., 160 Tronko, I.V., 317 Truesdell, A.H., 223, 225 Trump, R.P., 242 Tunnell, G., 447 Tunyak, A.P., 185, 316 Ulrich, R.A., 425 Upson, J.E., 382 U.S. Bureau of Mines, 254, 272, 393, 404, 41 5 Usiglio, J., 203 Valyashko, M.G., 162,227 Vandenburgh, A.S., 223 Van Everdingen, R.O., 223 Van Nostrand Press, 211 Vasil’ev, V.G., 313 Vasileuskaya, A.Ye., 152 Vdovyking, P., 319 Veal, H.K., 425,427 Vedder, J.G., 461 Veldink, R., 181 Vermeulen, T., 399 Vetter, O.J.G., 368, 369, 372, 380, 461 Vilonov, V.A., 319 Vinogradov, A.P., 316 Vinogradov, V.L., 228,316 Viher, G.S., 200, 322 Von Engelhardt, W., 240, 294 Wagner, H.C., 461 Wallace, W.E., 8,343, 344 Walton, G., 183 Wangersky, P.J., 184 Waring, G.A., 135 Warner, D.L., 425,426,461 Washington, H.S.,197 Water, C.J., 111 Waters, Jr., O.B.,399 Watkins, J.W., 47, 108, 109, 110,461 Watson, J.A., 65 Weast, R.C., 34 Weaver, C.E., 294 Weimer, J.C., 434 Weintritt, D.J., 369, 370,372 Welcher, F.J., 40 Weller, J.M., 206
484 Welte, D.H., 297, 298, 309 Weyl, P.K.,203,204 White, D.E., 107,135,162,194,195, 206,232,240 White, W.A., 3, 243 Whitehead, H.C., 108 Whiting, R.L., 429 Whitney, E.D., 379 Wilhelm, C.J., 461 Willard, H.H., 92 Williams, J.A., 302, 303 Williams, P.M.,184 Wilson, A.L., 186 Wilson, D.F., 182 Wilson, J.E., 425 Winter, J.A., 25 Winters, J.C., 302,303
REFERENCE INDEX Witherspoon, P.A., 180,182 Woodcock, A.H., 91 Wright, C.C., 437 Wright, J.L., 434 Wyllie, M.R.J., 32, 242 Yakoylev, Yu.I., 316 Yarbrough, H.R., 313 Yasenev, B.P., 308 Young, A., 244, 346 Yurovskii, Yu.M., 308 Zarrella, W.M., 12,181, 314 Zinger, A.S., 180, 314 Zobell, C.E., 30, 167,302 Zorkin, L.M., 313
SUBJECT INDEX
Abiogenic, 205, 293 Abnormal pressure, 344, 359 -, detection, 343 Accumulation of petroleum, 2,298 Accuracy, in analyzing methods, 21 -, of measurements, 23 Acetic acid solutions, 120 -, in oil-well acidizing, 120 Acidified samples, 23 Acidity, 37, 156 Acid treatment, 464 Additives, 463-466 Aerobic bacteria, 169, 208, 213, 302, 304,314 Alabama, 333 Algae, 165 Aliquot size, 41 Alkali metals, concentration during evaporation, 134 -, properties, 134 Alkaline earth metals, concentration during evaporation, 134 -, properties, 141 Alkalinity, 37, 155,254 Alkanes, 298,309 Allochthonous origin, 293 Alteration of hydrocarbons, 299 Aluminum, abundance, 155 -, atomic absorption method, 65 -, constituent of oilfield waters, 155 -, emission spectroscopy method, 90 -, properties, 148 Amazonite, 140 American Petroleum Institute, 19, 25, 27, 254, 347 American Society for Testing and Materials, 19 Amino acids, 182 -, chromatographic techniques, 182 Ammonium nitrogen, 157 -, concentration and economical profit, 413 -, determination by titrimetric methods, 43
Ammonium pyrollidine dithiocarbamate, 82
Anadarko Basin, 128,129,226, 297 Anaerobic bacteria, 169 Analytical method, choosing of, 20 Anhydrite, 159,471 -, solubility, 372 Antigorite, 442 Apatite, 158 Aquifer, 225, 244 -, contamination, 434 Aragonite, 471 Arbuckle formation, 51, 330 -, chloride concentration, 332 -, potentiometric surface map, 331 Arkansas, 333,335 Aromatic hydrocarbons, 314 Arsenic, colorimetric methods, 108 -, constituent of oilfield waters, 158 -, diethyldithiocarbamate method, 108 -, Gutzeit method, 108 -, occurrence, 108 Artificial fracturing, 430 Athabasca tar sands, 303 Atomic absorption methods, 65-82 Authigenic deposition, 201, 206 Autochthonous origin, 293 Bacterial alteration of petroleum, 301 Bacterial reduction, 50, 235 Barite, 471 Barium, 141,147 -, atomic absorption method, 65,77 -, colorimetric methods, 114,115 -, emission spectroscopy method, 83 -, flame spectrophotometric method, 63 -, gravimetric method, 115 -, properties, 141, 171 -, qualitative test, 114 -, recovery, 415 Barium sulfate, 367, 370,372 -, concentration and ionic strength, 374 -, scale, 370 -,solubility, 372, 375-377, 379, 382
486 Bathyal-abyssal deposits, 201 Bell Canyon formation, 322, 327-329 -, calcium concentration, 329 -, chloride concentration, 328 -, dissolved solids content, 327 -, potentiometric surface map, 323, 324 Bell Creek field, Wyoming, 303 Benzene, 315 -, solubility, 179 Benzene method for prospecting, 181, 314 Beryl, 140 Beryllium, atomic absorption method, 65 -, constituent of oilfield waters, 141,142 -, emission spectroscopy method, 89 -, properties, 141 Bicarbonate, concentration and depth, 357 -,- and economic profit, 413 Bioconcentration, 228 Biogenic origin, 205, 293 Biological degradation of hydrocarbons, 301 Biological weathering, 197 Biosphere, 193 Bischofite, 162 Bittern-type brines, 359 Boiling point, 172 Bolivar Coastal field, 303 Borate boron, titrimetric method of boron, 37 Borehole temperature, 364 Boron, 21, 37 -, abundance, 153 -, concentration and economic profit, 404,413 -,- by evaporating, 154 -, - and geologic age, 408 -, constituent of oilfield waters, 153 -, emission spectroscopy method, 83 -, properties, 171 -, recovery, 415 Bradford Sand, 1, 2 Brine value, 414,416 Brine worth, 414,416 Brines (see also Oilfield brines and Oilfield waters), 13, 27, 117, 159,160, 181, 213, 269,319,348,380,395,427,471 -, analyses, 272 -, classification, 225, 254, 257, 260,267 -, commercial, 415,416 -, composition, 384 -, concentrations of elements, 395 -, containing high bromine, 391
SUBJECT INDEX Brines (continued)
-,- high magnesium, 394
-,- high sodium chloride, 392 -, disposal, 411-417, 471 -, evaluation, 427
-, operations, 389,395 -, ponds, 389,470 -, refinery, 390,395, 396,402,403
-, stabilization, 380,381 -, state regulations, 434
Bromide, 45, 162,227, 360,408, 410,411
-, concentration, 361
-,- in brines, 162,163
-,- and economic profit, 404,413
-,- by evaporating, 163, 164
-, - and geologic age, 408 -, properties, 171 -, recovery, 41 5 -, recovery from Catesville, 391, 392 -, seaweed and coral, 228 -, sodium chloride relation, 162 -, titrimetric methods, 45 Bromine, 162,391,398 abundance, 164 locations with high concentration, 410 occurrence, 163 recovery, 391 Buffers, 28 -, definition, 198 Bulk density, 363 Burner height device, 67
-, -, -, -,
Cadmium, 152 abundance, 152 atomic absorption method, 65 colorimetric method, 103 properties, 153 sphalerite as carrier, 153 Calcite, 194,198, 201, 208,370, 371, 461,471 -, solubility, 370 Calcium, 40, 73,140, 143, 283-289, 370, 395 -, abundance, 144 -, atomic absorption method, 65,72,75 -, complexometric method, 40 -, concentration, 289,361 -,- in Bell Canyon formation waters, 329 -,- and depth, 358 -,- and economical profit, 404,413 -,- by evaporating, 143,144 -,- and geologic age, 408 -, constituent of oilfield waters, 141,143
-, -, -, -, -,
SUBJECT INDEX
487
Calcium (continued)
Chloride, concentration (continued)
-, locations with high concentrations, 409 -, properties, 141,171
-,- and relationships to other elements,
-, recovery, 415 -, solubility, 370 Calcium carbonate, precipitate, 144, 171
-, solubility, 202
Calcium sulfate, solubility, 144, 381 Calculated resistivity, 35 Calculating probable compounds, 125 Caledonian Group, 151 Calibration curve in flame spectrophotometric methods, 53 Caliche evaporite deposits in Chile, 161 Cambrian, 213 -, concentrations of elements, 219 Capillary cell method, 182 Capitan Limestone, 320 Carbonate, 146,152 -, depositional environments, 201 -, recovery, 415 Carbonated waters, dissolved elements, 198 Carbon dioxide, 50, 142,155,170, 208, 299 -, determination, 50 Carboniferous age, 185 Carnallite, 162, 228 Case histories, geochemical, 308 Catesville field, 391 Celestite, 471 Cell preparation, resistivity measurements, 34 Central Basin platform, 320 Cesium, 141,393 -, abundance, 141 -, constituent of oilfield waters, 134,141 -, flame spectrophotometric method, 59, 61 -, properties, 134 Chebotarevk classification, 262, 267 Chelating agents, 40,82,96 Chemical analysis, 125 Chemical treatment of wells, 464 Cherokee Group, 311 Chile caliche evaporite deposits, 161 Chloride, 44 -, concentration, 238, 239 -, - in Arbuckle formation waters, 332 -,- in Bell Canyon formation waters, 328 -,and depth, 357 -,- and economic profit, 413 -,- and geologic age, 408
229-239 properties, 171 -, recovery, 415 -, titrimetric method (Mohr), 44 Chlorine, 161,398 -, abundance, 161 Chlorinity, 24 Chocolate Bayou field, 245, 362 Chromatographic techniques, 181-185 Chromium, atomic absorption method, 65 Chrysotile, 442 Cinnabar, 151,152 Classification of oilfield waters, 253, 276 Classification systems, applications, 274289 Clastics, 200 -, depositional environments, 200 Clay minerals, 140,209, 230, 240, 345, 430,433,441,442 -, authigenic, 201 Cleaning pipelines, 421 Coal ashes, 133 Colorado, 323 Colorimetric methods, interferences in, 93 Combination factor, 127 Compaction, 206,294 Compaction model, 344 Compatibility of oilfield waters, 367 Completion of disposal wells, 432 Composition of minerals, 441 Composition of oilfield waters, 213 Compressibility of rock and water, 428 Concentrating by ion exchange, 95 Concentrating brines, 240 Concentration change during evaporation, 134, 204,227,229,231-234, 238 Concentration ratios in brines, 236, 237 Concentration versus proximity to an oil accumulation, 315 Confining beds, critical pressure of, 429 Connate water, 3, 169, 194, 270, 271 Contamination of shallow aquifers, 434 Continental slope drill hole, 223 Copper, 150 -, abundance, 150 -, atomic absorption method, 65,80 -, colorimetric method, 95,96, 150 -, properties, 148 Cordellera Isabella, Nicaragua, 323 Core samples analysis, 310 Corrosion inhibitors, 465,469
-,
488 Cretaceous age, 135, 154, 163,215, 227, 229 -, lithium concentration, 135 -, magnesium concentration, 143 -, potassium concentration, 139 -, sodium concentration, 137 -, strontium concentration, 146 Cretaceous age rocks, brines from, 237 Cretaceous system, concentration of elements, 213 Critical pressure, 429 Cymric fields, California, 152 Decarboxilization, 185 Deep well injection, 420 -, acceptable geologic areas, 425 Delaware Basin, 329 Delaware sand, 322, 323 Deltaic deposits, 201 Density, 172 Denver Basin, 343 Deposition, organic matter, 205 -, silica, 206 Depositional basin, 200 Depositional environments, 200, 201,203 Depth, 226 -, versus concentration, 357, 358 Description for water sample, 16 Deuterium, mass spectrophotometric method, 91 Devonian age, 234 Devonian deposits, 317 Devonian system, 185, 213, 217, 261 Dexter formation, 335 Diagenesis, 133, 207, 208, 232, 245, 267, 346 Diagenetic water, 194 Dilution technique, 41 Disodium 1,2-~yclohexanediaminetetraacetic acid, 40 Disposal, 471 Disposal brines, 411 Disposal costs of brines, 422 Disposal systems, cost of, 437 Disposal well, 412,424,425,432 -, cross section, 431 Disposal zone, 426,427 -, evaluation, 427 Dissolved gases, 12 Dissolved solids, 117,216, 284-286, 323, 325,362,410,419 - in Bell Canyon formation waters, 3,327 Dnepr-Donets Basin, 313,318
SUBJECT INDEX Dolomite, 142, 194, 203, 234, 239 Dolomitization, 195,203, 204, 208, 234, 238 Dolostone deposits, 234 Dow Chemical Company, 391,399 Dowex A-l,96 Drilling, 463 -, disposal wells, 432 Drilling fluids, 461,463 Drilling muds, 8,170,343, 461, 463,468 Drill-stem test, 8, 12, 181 Earthquakes, 434 East Texas Basin, 224,232,272,283288, 321 East Texas field, 420,424 East Texas Salt-Water Disposal Company, 423 Economic value of brine, 403 Eh, 14,19,29, 166, 170, 199 Eh, unfiltered and filtered petroleum producing wells, 320 Eh/pH plot, 159,168,170, 199 Electric log, 153, 341 -, cross section of southwest Louisiana, 349-352 Elements, minor, 220 Emission spectroscopy, 83 -, calibration curve, 88 -, emulsion calibration curve, 87 -, gamma curve, 87 Eocene age, 104,114,130, 152, 167,185, 212 Eolian deposits, 160 Epm (equivalents per million), 24, 274 Equivalents per million (epm), 24, 126, 269,275 Erosion, 198 Escape routes, 429 Evaluation, economic, 402 Evaporites, 137, 223, 238, 239 -, basin, 203 -, depositional environments, 203 Exchange reactions, 211 Fatty acids, 183,315
-, chromatographic techniques, 184 - in sea water, 184 Feldspar, 138,140,242,433 Ferromagnesian, 142 Fertilizers, 397 Fertilizer production, flowsheet for, 397 Field sampling methods, 273
SUBJECT INDEX Filtered petroleum-producing wells, pH and eH values, 320 Flame spectrophotometric methods, 53 Florida, 333-335 Flow diagram of solubility equipment, 448 Flow line, 13 Flowsheet for descaled sea water and fertilizers, 397 Flowsheet for fresh water and valuable elements, 401 Fluid mechanics, 320 Fluid travel, 433 Fluoride, colorimetric method, 109 Fluorine, abundance and properties, 161 Fluorite, 161 Fluorspar, 124, 161 Ford field, Texas, 323 Formation damage, 368 Formation interval tester, 10 Formation pressure, 12 Fossil water, 3 Freezing point, 172 Fresh-water conversion, 401 Fresh-water production, flowsheet for, 401 Fresh-water well, 471,472 Fulvic acid, 185 Gas analysis, 12 Gas chromatographic methods, 147, 181, 182,184,185 Gases in petroleum, reservoir waters, 271 Gas/oil ratio, 11 Gasoline Association of America, 186 Gaslwater relationships, 256 Generation and migration, 295 Genetic indicators, 336 Genetic types of waters, 260 Geochemical case histories, 308 Geochemical methods, exploration, 307 Geochemistry versus geologic environment, 266 Geologic maps, 426 Geopressure, 343 Geopressured reservoirs, 343 Geopressured zones, 343, 362 Georgia, 335 Geostatic ratio, 344 Gibbsite, 155 Glass electrode, 27 Graphic plots, 128 Graphite, 295
489 Gravimetric methods, 24, 114 Green River formation waters, alkalinity, 156 Groznyv oil district, 247, 302 Gulf Coast area, 128, 129, 181, 343, 346, 360 Gulf Coast Basin, 243 Gulf Coast shales, 179, 240, 360 Gypsum, 194, 203,204,235,317,471 Hackberry field, Louisiana, 304 Halite, 162, 163, 194,203, 227-229 Heaving shales, 462 Holocene age, 165 Honaker Trail formation, 321 Host rock, 133 Hot brines, 227 Humic acids, 182,184,185,186 -, chromatographic technique, 185 Hydrocarbons, 12, 178, 194, 211,298, 310 -, accumulation, 178,298 -, alkanes, 309 -, alteration, 299, 301 -, aromatic, 298, 309 -, bacterial attack, 302 -, biological degradation, 301 -, compaction, 294 -, containing nitrogen, 182 -, diffusion, 180 -, generation-migration, 295, 296, 362 -, hydrochemical indicators, 258 -, maturation, 295 -, migration, 299, 314, 316 -, in natural gas, 180 -, origin, 57 -, in petroleum, 181 -, in recent sediments, 244 -, in sedimentary rocks, 245 -, solubility, 178, 179,296 -, water washing, 300 Hydrochemical anomalies, 308, 316 Hydrochemical indicators, hydrocarbons, 258 Hydrodynamic gradients, 295,325 Hydrodynamic potential, 382 Hydrodynamic zones, 266 Hydrogen chloride, free, 120 Hydrogen sulfide, 51,156,159,309,468 -, method for determination, 51 Hydrogeochemical exploration, 311 Hydrogeochemical research and exploration 313
490 Hydrolysate, 139, 141, 171
-, rocks, 195,197 -, sediments, 141 Hydrolysis reactions, 15 Hydrosphere, 193 Hydrothermal equipment, 447 Igneous rocks, average composition, 196 Illinois Basin, 243 Illite, 138-140, 209,230, 239, 240, 269, 441,443,444 Index base exchange (IBE), 267,270, 271,283 Inhibitors, 465 -, chemical, 465 -, corrosion, 465,469 Injection of subsurface brines, 420 Injection well, 424 Instrumental methods, 20 Interior regions of the earth, 193 Internal standard solution in emission spectroscopy, 84 Interstitial water, 3, 194, 206, 207, 209 Iodate, 166 Iodide, 45, 110,226 -, colorimetric method, 110 -, concentration and economic profit, 404,413 -,- by evaporating, 166 -,- and geologic age, 408 -, properties, 171 -, recovery, 390,415 -, seaweed and coral, 228 -, titrimetric method, 45 Iodine, 164, 390, 398 -, concentration by algae, 165 -, recovery from brines, 390 Iodoargyrite, 164 Iodoembolite, 164 Ion association, 225 Ion exchange, 230 -, concentrating by, 95 Ionic potential, 133, 142,171 Ionic radii, 171 Ionization interferences, 66 Iron, 149 -, abundance, 149 -, atomic absorption method, 65, 79 -, colorimetric method, 94,95 -, emission spectroscopy method, 83 -, properties, 148 Isotopic fractionation, 243
SUBJECT INDEX Jurassic age, 135, 154, 163, 227, 229 lithium concentration, 135 magnesium concentration, 143 potassium concentration, 139 sodium concentration, 137 strontium concentration, 146 Jurassic system, 213, 214 Juvenile water, 3, 195
-, -, -, -, -,
Kainite, 162 Kansas brines, 226 Kaolinite, 155, 184, 209, 230, 269, 441, 443,444 Kazhim stratigraphic well, 185 Kerogen, 205, 309, 311 Lacustrine deposits, 200 Lanthanum, 83,84 Lea County, New Mexico, 320 Lead, 152 -, abundance, 152 -, atomic absorption method, 65,81, 82 -, colorimetric method, 95,99 -, dithizone method, 99 -, ion exchange, 99 -, isotope ratio, 318 -, properties, 119,148, 152 Lepidolite, 140 Limestone, 51, 197,201, 235, 320 -, dolomitization of, 204 Liquid exchange, chromatography, 182 Lithium, 133,392 -, abundance, 133 -, atomic absorption method, 68 -, concentration and economic profit, 404,413 -,- by evaporating, 135,136 -,- and geologic age, 408 -,- in Mississippian and Pennsylvanian age formation waters, 136 -,- in Tertiary, Cretaceous and Jurassic age formation waters, 135 -, constituent of oilfield waters, 133 -, flame spectrophotometric method, 54 -, properties, 133,134,171 -, recovery, 415 -, toxicity, 133 Lithology, 225 Lithophile, 147 Lithosphere, 193 Locations of valuable brines, 406 Louisiana, 135, 142, 144, 146, 163,231, 333,335,349-352, 359-361
SUBJECT INDEX Magnesium, 40,142,283-289, 394
-, abundance, 142 -, atomic absorption method, 65,71, 74 -, complexometric method, 40
-, concentration, 289,238 _ , _ and economic profit, 404,413
-,- and geologic age, 408 -, -,in Mississippian and Pennsylvanian age formation waters, 143 - ,_ ,in Tertiary, Cretaceous and Jurassic age formation waters, 142 -, constituent of oilfield waters, 141,149 -, locations with high concentration, 410 -, properties, 141,171 -, recovery, 394,415 -, specific gravity versus concentration,
72 Magnesium ammonium phosphate, 396 Manganese, 40, 149 -, abundance, 149 -, atomic absorption method, 65,78 -, emission spectroscopy method, 83 --,flame spectrophotometric method, 61 -, properties, 148 Maracaibo Basin, Venezuela, 303 Mass spectrophotometric methods, 91 Maturation, 295 Maximum worth, 414 Membrane-concentrated brines, 240 Membrane effect, 240 Metals, 65 Meteoric water, 194, 227, 253, 267, 270,
271,289 Methane, 178, 181, 295, 300, 316 -, measurement of, 12,180,181 -, solubility, 180 Mercuric iodide in brackish water, 152 Mercury, 151 -, abundance, 151 -, atomic absorption method, 65 -, properties, 151 Mexia-Talco Fault, 321 Mica, 140, 232,433 Michigan, 361 Michigan Basin, 205,226, 243, 359, 360 Microcline, 140 Microphotometer criteria, in emission spectroscopy, 85 Micropipet, 41 Microsyringe-evaporating flask, 186 Miersite, 164 Migration, 2, 293, 295 Milligram per liter (mg/l), 25, 269, 275
491 Mineral-acid acidity, 37 Minerals, formation, 234 -, recovered from saline waters, 392 Minor elements, 220 Miocene age, 165 Mississippi, 333, 335 Mississippian age, 135, 145,154, 164,166,
213
-, lithium concentration, 136 -, magnesium concentration, 143
-, potassium concentration, 139 -, sodium concentration, 138 -, strontium concentration, 146 Mississippian system, 216 Mixed salts, 395 Mixing of subsurface waters, 382 Molecular hydrogen, 180 Montmorillonite, 140, 155,184, 209, 238,
240, 294,360,441,445 Mud filtrate, 11 Nalco Chemical Company, 186 Naphthenic acids, chromatographic technique, 185 National Bureau of Standards, 23 Natural gas, 178,307,309, 310 -, deposits, 178 Neogene age, 245 Nernst equation, 167 New Mexico, 224 Nickel, 98 -, atomic absorption method, 65 -, colorimetric method, 95,98 -, ion exchange, 95 Nitrate nitrogen, colorimetric method, 107 Nitrogen containing hydrocarbons, 182 Nitrogen-free organic compounds, 178 Nodules of manganese oxide, 149 Nontronite, 441, 442,445,446 Normal pressure, 359 North Carolina, 335 Ohm meter, 32 Oilfield brines (see also Brines and Oilfield waters), 25, 193,219, 273, 389, 422 -, analysis, 25, 272 -, disposal, 389, 420 -, economics, 422 -, field sampling methods, 273 -, origin, 193 -, pollution, 468 Oilfield waters (see also Brines and Oilfield brines), 215, 226, 367,372, 374, 389
492 Oilfield waters (continued) altered, 242, 243 analysis, 272 classification, 253 compared t o sea water, 227 -, concentration of elements, 217, 238 -, incompatibility, 367 -, physical properties, 133 Oil in water, 186 -, determination, 186 Oklahoma, 136, 139, 143, 145, 146, 157, 164,166,335 Optical density, 94 Ordovician age, 213, 218, 345 Organic acids, 186,216, 311,312,316 -, chromatographic technique, 186 Organic compounds, 156,178,184-186, 188,205,317,433 -, deposits, 293 -, in oilfield brines, 188 -, in petroleum-associated waters, 186 -, in saline waters, 177 -, in subsurface water, 185 Organic matter, deposition, 205 -, in sea water, 178 Origin of brines, 219 Origin of oilfield waters, 193 Osmotic pressure, 173 Oxygen, 47,91,158 -, method for "0 -, solubility, 157 -, titrimetric method, 47 -, Winkler method, 47
-, -, -, -,
Paleozoic age, 160 Palmer's classification, 254, 256 Paradox Basin, Utah, 224, 321 Paraffins, 310 Parts per million (ppm), 24, 25 pE, 166 Pedosphere, 193 Pennsylvanian age, 135,145, 154, 164, 166,238 -, lithium concentration, 136 -, magnesium concentration, 143 -, potassium concentration, 139 -, sodium concentration, 138 -, strontium concentration, 146 Pennsylvanian system, 213, 215 Permeability, 207, 303,325, 345,430 -, artificial fracturing, 430 -, calculations, 11 Permian age, 213, 214, 320, 322
SUBJECT INDEX Permian Basin, 473 Petroleum, 195,210,211, 244, 295, 299, 310,315 -, accumulation, 178,225, 298,309,310, 313 -, alteration, 293, 299, 304 -, compaction, 294 -, degradation, 301 -, exploration, 307,320 -, generation, 210, 212, 219, 295,297, 310 -, migration, 211,295, 297 Petroleum Abstracts, 307 pH, 10,14,15,27,37,168-170,198, 207 -, temperature and, 28, 29 -, unfiltered and filtered petroleumproducing wells, 320 pH meter, performance characteristics, 29 Phenol, chromatographic technique, 181 Phosphate, 105 -, colorimetric method, 105,106 Phosphorus, 158 Photosynthesis, 165,206 -, by algae, 165 Pinkerton Trail Limestone, 320 Plasma arc, 83,84 Plate development in emission spectroscopy, 85 Playa deposits, 223 Pleistocene age, 150 Plugging of formations, 432 Polarization, 171 Pollution, 466-468 -, of uppersand by well blowout, 466 Porosity, 207,303,345 Porosity reduction, 345 Potassium, 138, 393 -, abundance, 138 -, atomic absorption method, 70 -, concentration and economical profit, 404,413 -,- by evaporating, 139 -,- and geologic age, 408 -,-, in Pennsylvanian and Mississippian age formation waters, 139 -, -, in Tertiary, Cretaceous and Jurassic age formation waters, 139 -, constituent of oilfield waters, 134, 138 -, depletion, 140,240 -, flame spectrophotometric method, 58 -, properties, 134,171 -, recovery, 415
SUBJECT INDEX Potentiometric surface map, 331 Powder River Basin, 303, 343 Precipitates, 432 Precision, 20 Precursors, 210, 212, 216, 293, 297, 311 Preliminary evaluation, marketing research, 405 Pressure equalizer, 1 2 Pressure-distance-time relationship, 430 Pressure head, 11 Pressure relationships, 433 Pripvatsky Depression, 317 Produced water, 4 Production, 467 Quality control, 19, 20 Quartz, 194,201, 209,433 Quartzite, 155 Quebracho, 464 Radioactive anomalies, 318 Radioactive compounds, 317 Radium, 318 -, anomalies, 318 -, concentration, 317 Radium/uranium ratio, 317-320 Reacting values, 126 Reaction coefficients, 126 Reagent chemicals, 23 Reagent solutions, 23 Recovery, 4, 390, 392,415,474 Redox potential (Eh, pE), 29, 157, 166 Reef carbonates, 202 Regressive marine deposits, 200 Reistle diagram, 131 Replacing power of ions, 230 Reporting analytical results, 25 Reservoir transmissibility, 428 Residual salt concentration, 470 Resistivity, 32, 33, 35, 317 Reverse exchange, 233 Reverse osmosis, 240, 241 Rio Bravo fields, California, 152 Rocky Mountain area, 156, 343 Rodessa formation, 284, 286-288 Rounding-off numbers, 26 Rubidium, 59, 140, 393 -, abundance, 140 -, constituent of oilfield waters, 139, 140 -, flame spectrophotometric method, 59, 61 -, properties, 134 -, standard-addition technique, 6 1
493 Saber field, 323, 326, 330 Sabkha sediments, 224 Salinity, 24, 226, 254, 257 -, concentration, 333-335 Salt water disposal, 4 12 Salting out effect, 179 Sample container, 16 Sample treatment, 22 Sampling, 8 -, methods, 273 Sand dike, 345 Sandstone, 152,158, 201, 345 San Juan Basin, 224, 343 San Juan Mountains, 321 Saturated hydrocarbons, 314 Scale, 368, 370 Scale inhibitors, 469 Schoeller’s system,.267, 268 Searles Lake, 135 Sea water, 137, 143, 149, 152, 158, 194, 215,227,245,392, 394,402 -, average composition, 195 -, composition, at dolomitization or bacterial reduction, 235 -, -, at gypsum precipitation, 234 -, fatty acids in, 184 Sedimentary basins, 212,406 Sedimentary rocks, 11,51,135,140,147, 195,210,245 -, average composition, 196 Sediment compaction, 206 Sediment diagenesis, 207 Seleniferous vegetation in the U.S.,160 Selenium, abundance, 160 -, colorimetric method, 111 Sensitivity for metals in atomic absorption methods, 65 Separation of gas, oil and brine, 400 Serpentine, 446,449,453,458 -, silicon solubility from, 453-458 Shales, 141,152, 158,179, 206, 240, 245, 360,430 Shallow aquifers, contamination of, 434 Siderite, 471 Significant figures, 25 Silica, 107 -, abundance, 156 -, analytical procedure when dissolved, 442 -, colorimetric method, 107 -, deposition, 206 -, spectrophotometric method, 107 Silicate, 149, 240,443
494 Silicate (continued)
-, chemical composition, 441 -, solubility, 441 Silicon, concentration, 443-446
-, solubility, 458 -,-, from serpentine, 453-458
Silurian age, 213, 218 Silver, atomic absorption method, 65 Simpson Sand, 345 Sloughing, 462 Smackover formation, 12, 135, 230, 232, 234,238,239,391 -, brines, 233, 235,236, P38,240, 392 -, -, concentration ratios, 236 Smackover Limestone water, drill-stem test, 12 Sodium, 136, 283, 284, 286, 287, 289, 360 -, abundance, 137 -, atomic absorption method, 65,68 -, concentration, 289, 361 -,- and economic profit, 404,413 -,- and geologic age, 408 -,-,in Pennsylvanian and Mississippian age formation waters, 138 -,-,in Tertiary, Cretaceous and Jurassic age formation waters, 137 -, constituent of oilfield waters, 134, 136 -, determination by calculation, 116 -, flame spectrophotometric method, 57 -, locations with high concentrations, 409 -, properties, 134,171 -, recovery, 415 Sodium chloride, 33,392 -, bromide relation, 163 -, resistivity, 33 Solids, dissolved (see also Dissolved solids), specific gravity and concentration, 36 -, -, in brines and sea water, 413 Solubility, 144, 178, 296, 370, 372, 375377,382,441 Solubility equipment, flow diagram, 448 Source rocks, 309 South Carolina, 335 South Caspian Basin, 316 South Pyote field, Texas, 323 Specific gravity, 35,410 -, versus concentration for magnesium solution, 72 -, chloride solution, 45 Specific heat, 172 Spent acid, 118 Sphalerite, 153
SUBJECT INDEX Stable-isotopes analysis, 15 Standard-addition technique, 50, 55, 57, 60,61 Standard solutions, 23 State regulations, 434 Stiff diagram, 131 Stinkfluss, 161 Stratigraphic interval, 11 Stratigraphic problem, 325 Stratigraphic traps, 312, 323,325, 326 Strontium, 62,76, 83, 145,385 -, abundance, 145 -, atomic absorption method, 76 -, concentration, 239 -,- and economic profit, 404 -,- and geologic age, 408 -, -, in Pennsylvanian and Mississippian age formation waters, 146 -, -, in Tertiary, Cretaceous and Jurassic age formation waters, 146 -, constituent of oilfield waters, 141,145 -, emission spectroscopy method, 83 -, flame spectrophotometric method, 62 -, properties, 141, 171 -, recovery, 415 Strontium sulfate, 385 -, concentration, 145 -, concentration and ionic strength, 378 -, saturation in waterflood makeup brines, 383 -, solubility, 145,370, 372,375-377, 379,381,383,385 Structural trap, 312 Structure of minerals, 441 Subsurface brines, 143, 159, 225, 315 420 -, analyses, 321 -, classification, 216, 234 -, contents, 159,313-315 -, hydrochemical anomalies, 316 -, maps, 320 -, mixing, 382 -, properties, 214 -, sampler, 9,224 Subsurface disposal, 419,421, 426, 471 Subsurface waters, 7,9,135, 253, 289, 382 Sugars, chromatographic technique, 181 Sulfate, 114,385 -, concentration and economic profit, 413 -, determination, 53 -, gravimetric method, 114
SUBJECT INDEX Sulfate (continued)
-, properties, 171 -, recovery, 415 -, reduction, 208,317 -, solubilities in synthetic brines, 379 Sulfide, 51
-, determination, 52,53 Sulfur, 52, 159
-, abundance, 159 -, concentration and economic profit, 401
-,- and geologic age, 408 -, determination, 52 Sulin’s classification, 257, 259 Suiyosei-ten’nengasu, 178 Surat Basin, 245 Surface tension, 173 Suspended solids, 31 Sylvania formation, 226 Sylvite, 139, 163, 228 Synthetic brines, 27,383
Tar mats, 299 Tax incentives, 436 Temperature, 15, 29, 212 Temperature gradient, 212 Ternary diagrams, 288 Tertiary age, 154, 163,212, 213, 227,229 -, lithium concentration, 135 -, magnesium concentration, 142 -, potassium concentration, 139 -, sodium concentration, 137 -, strontium concentration, 146 Tertiary age rocks, brines from, 237 Texas, 333-335 Thermal conductivity, 172 Thermodynamic equations, 449 Thiosulfate, determination, 53 Tickell diagrams, 128 Titrimetric analysis, 24, 37 Toluene, 181,314 Transgressive marine deposits, 200 Transmissibility, 428 Traps, 299, 312, 326 -, stratigraphic, 299, 312,323 -, structural, 312 -, types, 247 Treatment facilities, 420,421 Troilite, 471 Turbidity, 31 Tuscaloosa formation, 335
495 Uinta Basin, 157, 210, 212 Uncompahgre Uplift, 321 Underground waste disposal, 434 Unfiltered petroleum producing wells, pH and eH values, 320 Units for water analysis, 24 Unsaturated hydrocarbons, 314 Unstable properties, 15 Upper Cretaceous, 321Uranium, 318 Utah, 157 Valuable brines, 406,407 Valuable elements, 474 -, production, flowsheet for, 401 Value of brine, constituents, 415 Value of dissolved chemicals versus depth,
403 Value estimate, 414 Vapor pressure, 172 Viscosity, 173 Valence, 171 Volcanic waters, 58 Volga region, 180 Vorobyevite, 140 Waste disposal, 471 Waste disposal well, 472 Water, compatibility, 12, 290 -, compressibility, 428 -, properties, primary, 214, 261 -, washing, 300 Water analysis, 24, 130, 321 -, interpretation, 130, 131 Water and hydrocarbons, 311 Water exchange coefficient, 265 Waterflood makeup brines, 383 Waterflooding, 4, 369, 420,473 Waters, composition, 280 -, connate, 3, 194 -, diagenetic, 194 -, formation, 195 -, fossil, 3 -, interstitial, 3, 194 -,juvenile, 3, 195 -, meteoric, 194 -, subsurface (see also Subsurface waters),
297 Weathering, 197,198, 200,262 -, cycles and types of products, 263 Well blowout, pollution by, 466 Wellbore, 368 Wellhead, 14,15
496 Wellhead sampling, 13 Well treatment, 13 -, pH for determining of, 170 Wilcox formation, 331-334 Williston Basin, 129 Woodbine (Dexter) formation, 335 Worth estimate, 414 X-ray diffraction, 201, 471
SUBJECT INDEX Zinc, 80, 101, 151
-, abundance, 151 -, atomic absorption method, 65, 80 -, colorimetric method, 101
-, ion exchange, 151 -, properties, 148