Processing and Properties of Compound Semiconductors SEMICONDUCTORS AND SEMIMETALS Volume 73
Semiconductors and Semim...
24 downloads
687 Views
16MB 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
Processing and Properties of Compound Semiconductors SEMICONDUCTORS AND SEMIMETALS Volume 73
Semiconductors and Semimetals A Treatise
Edited by R. K. Willardson CONSULTING PHYSICIST
12722 EAST 23gD AVENUE SPOI~ANE, WA 99216-0327
Eicke R. Weber DEPARTMENT OF MATERIALS SCIENCE AND MINERAL ENGINEERING UNIVERSITY OF CALIFORNIA AT BERKELEY
BERKELEY, CA 94720
Processing and Properties of Compound Semiconductors SEMICONDUCTORS AND SEMIMETALS Volume 73 Volume Editors
R. WILLARDSON WILLARDSON CONSULTING
H. S. NALWA Scientific Advisor
ACADEMIC PRESS A Harcourt Science and Technology Company
San Diego San Francisco London Sydney Tokyo
New York
Boston
This book is printed on acid-free paper. Q Compilation copyright © 2001 by ACADEMIC PRESS All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher's consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2001 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0080-8784/01 $35.00. Explicit permission from Academic Press is not required to reproduce a maximum of two figures or tables from an Academic Press chapter in another scientific or research publication provided that the material has not been credited to another source and that full credit to the Academic Press chapter is given. The articles in this book were selected from the Academic Press multi-volume work titled Handbook of Surface and Interfaces, edited by Hari S. Nalwa and are uniquely arranged to focus on the processing and properties of compound semiconductors.
Academic Press A division of Harcourt, Inc. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com
Academic Press Harcourt Place, 32 Jamestown Road, London NW1 7BY, UK http://www.academicpress.com International Standard Book Number: 0-12-752182-8 International Standard Serial Number: 0080-8784 PRINTED IN THE UNITED STATES OF AMERICA 01 02 03 04 05 EB 9 8 7 6
5
4
3
2
1
Contents LISTOF CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
Chapter 1 Introduction . . . . . . . . . . . . . . . . . . . . . .
1
S. J. Pearton 1. INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 . WETETCHING REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 2 12
Chapter 2 Gallium Arsenide Heterostructures . . . . . . . . . .
15
Eric Donkor 1.
INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 . 1. Growth of GaAs Heterostructures . . . . . . . . . . . . . . . . . . . 1.2. Material Characterization . . . . . . . . . . . . . . . . . . . . . . GROWTHA N D PROPERTIES OF GAAS . . . . . . . . . . . . . . . . 2 . CRYSTAL 2.1. Crystal Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Impurities and Deep Levels . . . . . . . . . . . . . . . . . . . . . . 2.3. Crystal Structure and Lattice Properties . . . . . . . . . . . . . . . . . 2.4. Electronic and Electrical Properties . . . . . . . . . . . . . . . . . . 3. GROWTHA N D MATERIAL PROPERTIES OF GAASHETEROSTRUCTURES . . . . . . . 3.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Critical Thickness of Strained-Layer Quantum Wells . . . . . . . . . . . . 3.3. Heterostructures of the Type Ill-VIGaAs . . . . . . . . . . . . . . . . 3.4. Heterostructures of the Type IIIx-IIIi-x-V/GaAs . . . . . . . . . . . . . 3.5. Heterostructures of the Type III-Vx-Vi-x /GaAs . . . . . . . . . . . . . . 3.6. Heterostructures of the Type (III.-IIIi-.),-IIIi-,V/GaAs . . . . . . . . . . 3.7. Heterostructures of the Type IIIx-IIIi-x-V,-Vi-, /GaAs . . . . . . . . . . . 4 . PHYSICAL PROPERTIES O F GAAs-BASEDQUANTUM W E L L STRUCTURES AND SUPERLATTICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Quantum Wells Energy Levels . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V
15 17 21 22 22 24 24 21 30 30 31 32 35 46 41 41
48 48 50 55
vi
CONTENTS
Chapter 3 Growth and Optical Properties of GaN
. . . . . . . .
63
Annamraju Kasi Viswanath 1.
INTRODUCTION . . . . . . .
2.
GALLIUM NITRIDE AND ITS GROWTH ON DIFFERENT SUBSTRATES . . . . .
.
.
.
.
.
.
.
.
.
2.1.
Sapphire Substrate . . . . . . . . . . . . S i C Substrate . . . . . . . . . . . . . . . Z n O Substrate . . . . . . . . . . . . . L i G a O 2 Substrate . . . . . . . . . . . . . M g A l 2 0 4 Substrate . . . . . . . . . . . . M g O Substrate . . . . . . . . . . . . Si Substrate . . . . . . . . . . . . G a A s Substrate . . . . . . . . . . . 2.9. G a N Substrate . . . . . . . . . . . . . . 2.10. Lateral Epitaxial O v e r g r o w t h . . . . . . . . .
.
.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 2.8.
3.
.
.
.
.
.
63 67 67 76 77 77 77 78 78 79 80 81 84 97 99
.
. . . . . . . . .
.
.
. .
. .
. .
.
.
. . .
4. 5. 6.
LINE WIDTH AND QUANTUM BEATS IN GAN . . . . . . . . TIME-RESOLVEDSPECTROSCOPY OF GAN EPILAYERS . . . . . p-GaN . . . . . . . . . . . . . . . . . . . . . . . n-GaN . . . . . . . . . . . . . . . . . . . . . . .
7.
OPTICAL PUMPING AND LASING IN G A N EPILAYERS AND H E T E R O S T R U C T U R E S . .
8.
GAN QUANTUM WELLS . . . . . . . . . . . . . . . . . . . . . . . REFERENCES .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
110 121 126 134
.
Chapter 4 SiGe/Si Processing . . . . . . . . . . . . . . . . . . .
151
D. Y C. Lie and K. L. Wang 1.
INTRODUCTION .
2.
S I G E / / S I MATERIAL PROPERTIES AND PROCESSING CHALLENGES . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2.1. S i / S i G e Heterostructures: Lattice M i s m a t c h a n d B a n d g a p E n g i n e e r i n g .
.
2.2. M a t e r i a l s G r o w t h . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. C h a r a c t e r i z a t i o n Techniques f o r S i / S i G e Heterostructures . . . . . . . . 2.4. G e n e r a l P r o c e s s i n g C h a l l e n g e s to the Fabrication o f S i / S i G e D e v i c e s . REFERENCES .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Chapter 5 Advances in Quantum Dot Structures . . . . . . . . .
.
.
.
151 153 153 157
164 174 192
199
S. Kim and M. Razeghi 1.
INTRODUCTION .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2.
PHYSICAL PROPERTIES .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2.1. D e n s i t y o f States . . . . . . . . . . . . . . . . . . 2.2. E n e r g y States . . . . . . . . . . . . . . . . . . . 2.3. Optical A b s o r p t i o n a n d Transition in Q u a n t u m D o t s . . 2.4. D e v i c e s B a s e d on Z e r o - D i m e n s i o n a l Q u a n t u m Structure
. . . .
. . . . . . . .
STATE OF THE ART .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
REFERENCES .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3.
.
.
.
.
.
.
.
199 201 201 202 204 207 209 212
vii
CONTENTS
Chapter 6 Wet Etching of I I I - V Semiconductors Walter
P
. . . . . . . . .
Gomes
1.
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.
SEMICONDUCTOR ELECTROCHEMISTRY: BASIC PRINCIPLES AND
3.
215
215
EXPERIMENTAL METHODS . . . . . . . . . . . . . . . . . . . . . . . . .
218
2.1. 2.2. 2.3. 2.4.
REFERENCES . . . . . . . . . . . . . . . . .
218 220 228 233 238 238 239 241 243 244 246 246 253 257 259 266 268 275 275 284 292 292 293
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTENTS OF VOLUMES IN THIS SERIES . . . . . . . . . . . . . . . . . . .
297 301
Semiconductors . . . . . The S e m i c o n d u c t o r - L i q u i d Electrochemical Reactions Electrochemical Reactions
. . . . . . . . . . Solution Interface . at S e m i c o n d u c t o r s in at S e m i c o n d u c t o r s in
. . . . . . . . . . . . . . . . . . . . . . Indifferent Electrolytes . R e d o x Electrolytes .
.
.
SEMICONDUCTORS .
.
TYPES OF ETCHING REACTIONS . . . . . . . . . . . . . . . . . . . . . .
3.1. 3.2. 3.3. 3.4. 4.
. . . . . . . . . . . . . . . . . . . . SOME SOLID-STATE AND ELECTROCHEMICAL DATA ON III-V
5.
KINETICS AND MECHANISMS OF ETCHING REACTIONS AT I I I - V SEMICONDUCTORS .
5.1. 5.2. 5.3. 5.4. 5.5.
(Photo)Electrochemical Etching . . . . . . Photoetching . . . . . . . . . . . . . . Electroless Etching . . . . . . . . . . . Chemical Etching . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
(Photo)Electrochemical Etching . . . . . . . . . . . . . . Photoetching . . . . . . . . . . . . . . . . . . . . . . Electroless Etching . . . . . . . . . . . . . . . . . . . Chemical Etching . . . . . . . . . . . . . . . . . . . . Electroless a n d C h e m i c a l Etching Occurring in Parallel . . .
6.
MATERIAL-SELECTIVE ETCHING . . . . . . . . .
7.
ETCH MORPHOLOGIES AND PROFILES . . . . . . .
. . . . .
. . . . . . . . . . . . . . . .
. . . . . .
7.1. Etch M o r p h o l o g i e s at M a c r o s c o p i c Size Surfaces 7.2. Profile Etching . . . . . . . . . . . . . . 8.
CONCLUSIONS . . . . . . . . . . . . . . . . ACKNOWLEDGMENT . . . . . . . . . . . . . .
This Page Intentionally Left Blank
List of Contributors Numbers in parentheses indicate the pages on which the authors' contribution begins. (15), Department of Electrical and Computer Engineering, University of Connecticut, Storrs, Connecticut, USA
ERIC DONKOR
P. GOMES (215), Laboratorium voor Fysische Chemie, Universiteit Gent, Belgium
WALTER
S. KIM (199), Center for Quantum Devices, Electrical and Computer Engineering Department, Northwestern University, Evanston, Illinois, USA D. Y. C. LIE (151), Communications Research and Development Center (CRDC), IBM Microelectronics, Encinitas, California, USA S. J. PEARTON (1), Department of Materials Science and Engineering, University of Florida, Gainesville, Florida, USA M. RAZEGHI (199), Center for Quantum Devices, Electrical and Computer Engineering Department, Northwestern University, Evanston, Illinois, USA ANNAMRAJU KASI VISWANATH (63), Center for Materials for Electronics Technology, Ministry of Information Technology, Pune 411 008, India K. L. WANG (151), Department of Electrical Engineering, University of California, Los Angeles, California, USA
This Page Intentionally Left Blank
SEMICONDUCTORS AND SEMIMETALS, VOL. 73
CHAPTER
1
Introduction S. J. Pearton DEPARTMENT OF MATERIALS SCIENCE AND ENGINEERING, UNIVERSITY OF FLORIDA, GAINESVILLE, FLORIDA, USA
1. INTRODUCTION
1
2. WET ETCHING
2
REFERENCES
12
1.
Introduction
Compound semiconductors have a wide variety of applications, including visible and infrared laser diodes and light-emitting diodes (e.g., InGaAsP/InP, GaAs/AlGaAs, InGaAlP, InGaN/AlGaN) for displays, information storage and communications, metal-semiconductor field effect transistors, heterojunction bipolar transistors, and high electron mobility transistors for high speed data transmission networks, microwave amplifiers, low-noise amplifiers, and wireless communication; see, for example, [1]. The total percentage of the microelectronics market occupied by compound semiconductor devices and circuits is ^^5%, but they do fill important niches unavailable to Si. There are a number of challenges when processing compound semiconductors, including the relatively high vapor pressures of the group V elements compared to the group III elements, the high diffusivities of many acceptor dopants, and the difficulty in forming highly reliable ohmic and rectifying contacts. Many of the above-mentioned device structures are based on lattice-matched heterostructures, such as GaAs/AlGaAs, GaAs/InGaP, InP/InGaAs, InP/InAlAs, and InGaAsP/InP, and it is necessary to develop highly selective, as well as nonselective, etch processes for these different materials, as well as to be able to maintain the stoichiometry of the layers during the various process steps. Much effort has been devoted to achieving lattice-matched compositions to avoid the introduction of threading dislocations that degrade the electrical transport and optical qualities of devices subsequently fabricated. To some extent the InGaN/AlGaN system represents an exception, since highly luminescent lightemitting diodes (LEDs) and laser diodes have been demonstrated [2, 3]. For LEDs the resultant reliability is sufficient for commercial applications, but the 1 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved. ISBN 0-12-752182-8 ISSN 0080-8784/01 $35.00
2
S. J. PEARTON
high dislocation density in heteroepitaxial material limits the lifetime of laser diodes, where the much higher current density leads to metal migration that shorts out the p-n junction. In material grown on quasi-GaN substrates, this mechanism is absent [4], and the laser diodes have much longer lifetimes. In the following sections, we will cover the main processing steps for groups III-V compound semiconductors, including wet and dry etching, ion implantation for doping or isolation, and ohmic and Schottky contact formation. Examples will be given for each of the main III-V materials, namely GaAs, InP, and GaN, and their related ternary and quaternary alloys. In addition, the effect of atomic hydrogen incorporated into these materials unintentionally during growth and processing will be discussed.
2.
Wet Etching
Typically the wet etching of III-V materials involves the use of an oxidant to oxidize the surface, followed by dissolution of a soluble reaction product [5-8]. The resultant etching tends to be basically isotropic in nature, proceeding as shown in the schematic of Figure 1. This illustrates the selective etching of layer 1 from layer 2, and the undercutting of a mask on layer 1. In the case of III-V compounds, differential etch rates for crystallographic directions containing predominantly one or the other elements can lead to a degree of anisotropy and different side wall shapes [7]. The etch rate may be limited by the diffusion of the active etchant species to the semiconductor surface or by the out-diffusion of the soluble product [5]. In this case the etching is termed diffusion-limited, and its characteristics include a square root dependence of etch depth on etch time, an activation energy 20,000\
\
•
3,000 210
8,800
150
-v,
\
V
\
'^
°/.9^
>30,000\ 7,300
(Unit = A/min )
11
10
1
I
1
Ea= 54.4±5KJ/mol
®
7 6A ^ . 5 ^ ^ 0 . 5 ^ ^ HCl (12M) 2.8
I
2.9
I
3.0
\
3.1
I
3.2
I
3.3
I
3.4
I
3.5
1000/T(1/K) FIG. 8. Arrhenius plot of etch rate of AlGaP in HCl solution.
3.6
1
INTRODUCTION
9
of the etchant, however, in that wherever gold-based metalUzation is exposed to the etch solution it will be corroded [10]. Sometimes gold-based emitter metallization is preferred as the etch mask for emitter mesa definition in selfaligned processing. With a similar KI/I2 solution at an adjusted pH value, the reverse etch selectivity of GaAs over AlGaAs can also be achieved. The differential etch rates of GaAs and Gaj.^Al^As as a function of the pH value provide selectivity. The etching selectivities increase with the increase of Al content from 8:1 for AlQ2Gao8As to 20:1 for Alo44GaQ56As. The recommended use of this etch is for a solution containing 0.3-M KI and 0.04-M I2 in a solution buffered to pH 8.7-9.3. The preferred buffer is a solution of borax with NaHC03 added to obtain the desired pH. NaHC03 is considered essential for this etch. The K2Cr207/H3P04/H20 solution is a general etching solution for Al that contains III-V semiconductors, such as InAlAs or GaAlAs. The etch rates of GaAlAs rise rapidly as a function of the Al content, from ^^50 A min~^ at 10% Al to 320 A min~^ at 50% Al. Although this solution showed only a small selectivity of Ga^.^^Al^As over the AlGaAs/GaAs material system, the etch rate is quite slow (400 A min~^ at 26 °C) and can be easily handled and controlled in the emitter mesa definition process. Relatively little success has been obtained in developing wet etch solutions for III-V nitrides [11, 12]. For AIN, a number of different solutions have been reported for amorphous or polycrystalline material. For example, hot (tm), and very long wavelength (2.0-30-/im) lasers [14]. An alternative classification scheme for heterostructure lasers is based on the nature of carrier confinement [15]. This leads to single quantum well lasers, multiple (including double heterojunction) quantum well lasers, separate confinement quantum well lasers, which provide different confinement for carriers and photons, and graded-index separate confinement quantum well lasers. Gallium arsenide heterostructure-based photodetectors [16, Chap. 11; 17] transform electromagnetic radiation into electrical signals. They offer high
2
GALLIUM ARSENIDE HETEROSTRUCTURES
17
speed, high sensitivity, high quantum yield, and low intrinsic noise performances. Gallium arsenide heterostructure photodetectors that rely on interband transitions in quantum well structures are often designed for the visible region, whereas those that utilize intersubband transitions are designed to operate in the mid and far infrared region. The most common heterostructure-based electrooptic modulator is the selfelectrooptic effect device (SEED) [18]. The SEEDs operate on changes in the optical absorption induced by an electric field normal to the growth direction of the heterostructure [19], according to the Franz-Keldysh effect [20, 21]. The quantization effect in quantum well structures causes the resulting absorption spectrum to have large discrete steps. The wavelengths of these absorption steps shift when the quantum well structure is placed under the influence of an electric field. Furthermore, distinctive peaks exist in the absorption spectrum of quantum wells. These peaks also shift upon the application of an electric field. This peak shift is called the quantum confined Stark effect, which is the basic phenomenon underlying the operation of the more common heterostructure-based electrooptic modulators.
1.1.
7.7.7.
GROWTH OF GAAS HETEROSTRUCTURES
Metal-Organic Chemical Vapor Deposition Growth
Metal-organic chemical vapor deposition (MOCVD) has evolved as a major technology [22-25] for the growth of GaAs heterostructures. The layers are grown by transporting different precursors or reactants in the vapor phase, under controlled pressure, into a reactor (vertical, horizontal, or barrel) chamber [26, 27] that holds the semi-insulating GaAs substrate. Alkyls of the group III metals and hydrides of the group V elements are typical precursors. Tables I and II list some common group III and group V precursors and their properties. Three techniques have been introduced to improve the MOCVD growth processes: plasma-enhanced (PE-MOCVD) [28], atomic layer epitaxy (ALE) [29], and low-temperature (LT-MOCVD) [30] methods. In the PE-MOCVD growth, the plasma induces precracking of the precursors into the various constituents, thereby facilitating the controlled incorporation of the different chemical elements in the alloy formation at lower temperatures. In ALE growth, the substrate is separately and alternatively exposed to the precursors containing the group III and group V elements. This alternation allows for better control of the growth process at the monolayer level. In LT-MOCVD growth, the growth rate of the layers is less dependent on temperature. Thus the growth rate can be held constant during the process. This enables layers to be grown with uniform thickness and composition, and interfaces have abrupt compositional changes. Key parameters that affect the MOCVD growth characteristics include the substrate orientation and preparation, flow rate of the precursors, growth temperature.
18
ERIC DONKOR
TABLE I PROPERTIES OF COMMON GROUP III METALS FOR THE EPITAXIAL GROWTH OF SOME I I I - V ALLOYS ON GAAS SUBSTRATE
Precursor TMAl TEAl DMAIH TIBAl TMGa TeGa TIBGa TMGa—TMP TMGa—TEP TMGa—TMAs TMIn TEIn Cpin, {C,U,ln) EDMIn TMIn—TMN TMIn—TMP TMIn—TEP TMIn—TMAs ^\og(p[tOTT]) =
Melting point (°C)
Vapor pressure"
Boiling point (°C)
a
biK)
p(torr)/r (°C)
126 186(207) 154
8.224 10.784 8.92
2134.83 3625 2575
4.3 -15.8 -82.5
55.8 143
8.501 9.172 4.769
1824 2532 1718
56-57 32-35 23-24 88 -32
135.8 184
10.520
3014
2.2/0; 9/20 0.5/55 2/25 0.5/37 64.5/0; 66.0/0; 178/20 3.4/20; 16/43 0.078/18 0.13/0; 1.71/20 0.008/0; 0.04/20 1.3/0; 5.3/20 0.3/0 1.7/20
15 -52.5
Sublimes at 50 (0.01 mm) 94-96 43-45 33-36 28-29
0.85/17 0.003/0; 0.03/20 0.04/0; 0.22/20 0.0004/0; 0.0003/20 0.27/0; 1.2/20
a-b/T.
pressure, and source material purity. The effect of substrate temperature during growth is different for different combinations of source material. Commercial MOCVD reactors are housed in a cabinet that provides for a continuous flow of air through the system by a negative exhaust. The major MOCVD subassemblies are (a) gas system, (b) temperature controller system, (c) electrical control system, (d) heating system, (e) purifier system, and (f) leak detection and safety system. The gas system includes metal-organic (and other gaseous) sources and mass flow controllers. Stainless steel tubes are used to transport sources to the reactor chamber. There is a separate temperature bath and controller for each of the metal-organic sources. Hydrogen, nitrogen, and helium are the most common carrier gases used in MOCVD growth processes. The carrier gases are highly purified to eliminate impurities. The most common impurities in a hydrogen carrier gas are oxygen and water vapor. Therefore, the MOCVD purification system usually includes an oxygen remover and a hydrogen purifier (palladium). The palladium is contained in a cell heated to about 400-425 °C . At this temperature, only hydrogen diffuses through the palladium tube and all other elements are blocked.
2
19
GALLIUM ARSENIDE HETEROSTRUCTURES TABLE II
PROPERTIES OF COMMON GROUP V METALS FOR THE EPITAXIAL GROWTH OF SOME I I I - V ALLOYS ON GAAS SUBSTRATE
Precursor P PH3 TMP TEP IBP TBP
Melting point (°C)
Boiling point (°C)
Vapor pressure'' a
biK)
1/260 -84 -88(-85) -20 4
38 127 78 54
7.7627 8.035 7.578 7.586
1518 2065 1648 1539
0.5/55 46.5/50 112/25 141/10 1/370
-87.3
-62.5 50 140
7.3936
1456
7.532 7.339 7.243 8.47 7.7068
1443 1680 1509 2410 1697
238/20 5/20 15.5/37 176/0 40/20 81/10 1.8/20
7.628
1816
AS
ASH3 TMAs TEAS DMAs DEAs TBAs PhAsH2 TMSb TESb TMBi
p (torr)/r (°C)
-1
36.3 102 65
-87.6 -98 -107.7
80.6 160 110
4/25 27/20
Mog(/7[torr]) = a - Z 7 / r .
The metal-organic sources are contained in stainless steel cylinders equipped with a temperature controlled bath. Commercial temperature control baths are capable of temperature control, using a thermocouple element, over a wide range from about - 2 0 to 100 °C at ±0.01% accuracy. Precisely controlling the temperature regulates the partial pressure of the source. Electronic mass flow controllers control the exact amount of carrier gas flowing through the bubblers and at the same time maintain a constant vapor pressure of the source. A heating system heats the substrate during growth. The heater can raise the temperature of the substrate to about 1000 °C. Heating can be achieved by radiofrequency (rf) induction heating, radiative heating, and resistance heating. Radiofrequency heating is common in large commercial systems. In this type of heating, the substrate susceptor is inductively coupled to the rf coil. In radiative heating, the heat energy from a resistance element is transformed into radiant energy. The susceptor absorbs the radiant energy and converts it back to heat energy. In the resistance heating method, current flows through an electrically conductive layer mounted to the susceptor to generate the required heat energy. Safety precautions include gas leak detectors for the carrier gas, the purifier, and toxic gases. Commercial MOCVD are computer controlled, and can be used to define the growth process and to monitor reactor conditions.
20 7.7.2.
ERIC DONKOR
Molecular Beam Epitaxy Growth
Figure 1 shows the basic components of a multichamber ultra high vacuum molecular beam epitaxy (MBE) system [31]. It consists of growth chamber, wafer loading and preparation chamber, and analytical chamber. The growth chamber, which usually has a 450-mm inner diameter, holds a removable source flange that provides support for effusion cells, their shutters, and their liquid nitrogen (LN2) shroud. The cells are oriented such that their beams converge on the substrate in the growth position. The beam sources are thermally isolated from each other by an LN2 cooled radial vane baffle, which prevents chemical cross-contamination. A second cryopanel surrounds the substrate such that the stainless steel bell jar is internally lined by this cryopanel for the purpose of reducing contamination due to outgassing of the chamber walls. The sample preparation and loading chamber is connected to the growth chamber via large diameter channels and isolation valves. The analytical chamber allows for in situ and postgrowth surface analysis without exposing the sample to outside environments. The substrate surface must be prepared prior to epitaxial growth. Preparation of the surface entails preloading chemical treatment as well as in situ cleaning processes. The goal of the preloading chemical treatment is to provide either a clean chemical oxide layer of a few angstroms thick on the GaAs substrate
GROWTH CHAMBER
Transfer rod Electron analyser (AES.LEELS.ESCA)
Bectron diffraction gun 10-50 ktV
WAFER LOADING AND PREWVRATION CHAMBERS
FIG. 1. Schematic diagram of an MBE system showing the growth chamber, the wafer loading and preparation chambers, and the analytical chamber for surface analytical studies. Reproduced with permission from K. Ploog, "Springer Proceedings in Physics" (G. Lelay, J. Derrien, and N. Boccara, Eds.), Vol. 22, p. 10. ©1987 Springer-Verlag, New York.
2
GALLIUM ARSENIDE HETEROSTRUCTURES
21
surface or hydrogen passivation of the GaAs surface. An approach for GaAs surface preparation was described by Drummond et al. [32]. The first step is polishing of the substrate surface to mirror finish. One approach is to use pellon cloth saturated with Br—CH3OH. This is followed by wet chemical cleaning of the substrate in a 3 : 1 : 1 solution of H2SO4 : H2O2: H2O for about 10 min to remove organic contamination. The wafers are then degreased, etched again in H2SO4 : H2O2: H2O, and rinsed in deionized water. Next the substrate is loaded into the airlock high vacuum growth chamber. The pressure of the growth chamber is typically 10"^ torr. The growth chamber is equipped with liquid nitrogen shrouds that surround the effusion cells and the substrate, and line the walls of the chamber. The purpose of the shrouds around the effusion cells is to prevent excessive heating of the chamber, especially around the ovens, which can lead to contamination due to desorption of unwanted residues. The shrouds that surround the substrate and line the walls condense residual gases and water vapor, especially those desorbed by the substrate heater, and remove any fluxes. An ionization gauge situated behind the substrate block, which is mounted in the growth chamber, monitors the As flux and a high-energy electron diffraction (HEED) system. The shrouds are filled with liquid nitrogen well before growth is initiated and are kept on until the furnace temperatures are reduced after growth. Once the chamber has been chilled, the effusion cells are turned on and stabilized to the desired temperature with a dc current, controlled by a temperature controller (W-5%ReAV-26%Re) thermocouple. Likewise, the substrate surface is heated to about 630 °C. The HEED pattern is closely monitored to ensure that the native oxide layer is completely desorbed off the substrate at a precise temperature, say 580 °C. The substrate temperature is subsequently raised or lowered to the growth temperature. Once the desired growth temperature is reached, the shutters for the appropriate effusion cells are opened, followed by opening of the main shutter to initiate epitaxial growth. The initiation of the thin film is monitored by the HEED pattern. A commercial MBE facility is shown in Figure 2. The system is designed for the growth of GaAs-based heterostructure devices, optoelectronic devices, and integrated circuits.
1.2.
MATERIAL CHARACTERIZATION
The Hall effect is the most basic characterization technique for evaluating semiconductor heterostructures. It is used to determine the type and concentration of elements in a sample. It can also allow for determination of the electrical activation energy for donors and acceptors in GaAs and its alloys by analyzing ionized impurity scattering and its effect on carrier mobility. The measurement techniques are well established and widely reported in the literature
[33].
22
ERIC DONKOR
FIG. 2. A commercial MBE system used for the growth and analysis of GaAs heterostructure materials and devices at the University of Connecticut.
Characterization of heterolayers based on techniques that give structural information include transmission electron microscopy, X-ray diffraction, scanning tunneling microscopy, and those that give information on the band structure. The latter comprise optical measurements such as photoluminescence excitation spectroscopy, absorption and reflectivity, and electrical measurements such as Hall and Shubnikov-de Haas measurements. Etching has been employed to characterize the structural and chemical inhomogeneities in semiconductors [34-36]. The structural and chemical defects are observed indirectly from their effects on the etching mechanism. Etching also has been used to study defects such as dislocations and staking faults [37]. The major drawback of this technique is that it resorts to destructive testing.
2. 2.1.
Crystal Growth and Properties of GaAs
CRYSTAL GROW^TH
The MOCVD growth of GaAs results from the reaction between group III alkyls (e.g., TMGa, TEGa) and group V hydrides (e.g., ASH3, TEAs). The
2
GALLIUM ARSENIDE HETEROSTRUCTURES
23
reaction takes place in a reactor chamber that contains a semiinsulating GaAs substrate placed on a heated carbon susceptor. Typically growth temperatures range between 575 and 700 °C, and depend on the precursors used. Highest purity material, using TMGa and ASH3, has been achieved [21, 22] in the temperature region from 600 to 650 °C. The chemical reactions that lead to the formation of GaAs from metal-organic precursors is a complex process that is not fully understood. The reactions break down into gas-phase reactions and surface reactions [38, 39]. The gas-phase reactions involve multiple pyrolytic decompositions of organometallics and reactions with hydrogen radicals [4042]. The surface reactions entail decomposition, surface adsorption, and desorption. Volatile by-products are removed from the surface by adsorption, by colliding gas-phase radicals, and by bimolecular surface recombination. One major problem with the group III sources is the unintentional doping of the layers with carbon [43-45], making the alloy effectively p type [46]. However, GaAs grown from TEGa and ASH3 turn out to have lower carbon contamination. The ratio of group V to group III also affects the quality of the alloy growth. At low V : III values the GaAs is p type, having high carbon concentrations. The carbon concentration decreases with increasing V : III ratio, and at a critical ratio the GaAs material becomes semi-insulating. The substrate orientation also affects the introduction of carbon. The (111) and (311) orientations show lesser carbon incorporation, but (100) is the preferred substrate orientation because it yields the best surface morphologies and cleaves easily. The MBE growth of GaAs occurs through the interaction of Ga atoms and As2 (and/or AS4) molecules impinging on a GaAs substrate [47, 48]. Most recent developments use gas sources and are variously known as gas-source molecular beam epitaxy, chemical beam epitaxy, or metal-organic molecular beam epitaxy (MOMBE). In this case the growth occurs between TMGa (or TEGa) and ASH3. The sticking coefficient of Ga atoms must approach unity, with condensation of As2 via bonding with Ga [49, 50] for growth to occur. The sticking coefficient of As2 increases as the ratio of Ga flux to AS2 flux ((/>Ga/^As2) increases, reaching unity when (t)Q^ = 24>j^^^. This situation implies that stochiometric formation of GaAs can occur for (J)Q^ < 24>^^^ and the excess AS2 is lost by desorption. The layer by layer growth of GaAs by MBE or MOCVD results in surface reconstruction, that is, a surface that is different from the "native" surface of the material. The surface may reconstruct so that surface atoms care share bonds. This reconstruction results in a two-dimensional symmetry with periodicity differing from that of the underlying atoms of the GaAs crystal. The surface atoms may also relax, that is, change the bond angles, but not the number of nearest neighbors, to seek new equilibrium positions. The reconstructed GaAs surface has variety of structures. These are the (1 x 1) structure of the (110) face, (2 x 2) structures of the (111) and ( i l l ) faces, and a series of structures [c(4 x 4), c(2 X 8), c(8 X 2), p{\ X 6), p{A x 6), etc.]. The letter "p" indicates that the unit cell is primitive and "c" indicates that the unit cell has an additional scatter in the center.
24
ERIC DONKOR
The GaAs surface structure has been studied using a variety of techniques such as reflection anisotropy spectroscopy (RAS) spectra for MBE- and MOCVD-grown (001) GaAs [51, 52]. The RAS method is based on the fact that a cubic material such as GaAs is optically isotropic in first-order reflectivity. Thus anisotropic reflectivity originates from the surface with different symmetries. A more common method for surface measurement is the low-energy electron diffraction (LEED) method. In LEED, electrons of well defined energy and direction diffract from the crystal surface. The low-energy electrons are scattered mainly by individual atoms on the surface and produce a pattern of spots on a fluorescent screen. The spots in the pattern correspond to the points in the two-dimensional reciprocal lattice.
2.2.
IMPURITIES AND DEEP LEVELS
A number of elements are electrically active impurities in GaAs and produce shallow donor or acceptor levels [53]. Deep levels due to impurities or lattice defects [54] also exist. Table III gives a summary of some of the impurities, their activation energies, and their diffusion in GaAs. The most common dopants for MBE-grown GaAs are Be for p type and Si, Ge, and Sn for n type. Beryllium acts as an acceptor in MBE-grown GaAs [55]. Abrupt doping levels can be achieved due to the low diffussivity of Be in MBE-grown GaAs [56]. At substrate temperature exceeding 550 °C and at high doping levels above 5 X 10^^ cm"-^, the surface morphology degrades [57] and the diffusion of Be is enhanced [58, 59], resulting in degradation of the doped epitaxy. On the other hand, lowering the substrate temperature to 500 °C lowers the diffusivity of Be and acceptor levels of 2 x 10^^ cm~^ can be achieved [60]. Silicon is the most commonly used n-type dopant in MBE-grown GaAs. It is incorporated on Ga sites under As-stabilized conditions and yields n-type material. Germanium is an amphoteric dopant and it can be used to prepare either por n-type films, depending on the growth condition [61, 62]. Germanium acts as an acceptor on As sites and as a donor on Ga sites. The site substitution depends critically on the As: Ga flux ratio and on substrate temperature. Figure 3 [31, p. 27] gives doping concentrations for Si, Be, Ge, and Sn in MBE-grown GaAs, as a function of temperature for a constant growth rate of 1 /jum/h [31].
2.3.
CRYSTAL STRUCTURE AND LATTICE PROPERTIES
The GaAs crystal structure has been studied and reported extensively [63]. It has a zincblende crystal structure with a lattice constant, QQ, that is temperature dependent as shown in Figure 4 [64]. The nearest-neighbor configurations are such that each Ga species is surrounded by four As species and vice versa, with a nearest-neighbor bond length of r^ = {V^UQ/A) = 2.44793 A at 300 K,
2
GALLIUM ARSENIDE HETEROSTRUCTURES
25
TABLE III ACTIVATION ENERGIES OF IMPURITIES AND THEIR DIFFUSION IN GAAS
Element
Activation energy (meV)
Shallow donors S Se Te Si Sn Ge Pb C
6.0^5.89^ ,5.845'-,,5.87^ 6.0^5.85^ ,5.812^,5.789' 3.0« 5.808\5.799^ , 5.839^ 5.817^ 5.908^5.949^ ,5.888^ 5.773^ 5.937^
Shallow acceptors Zn Cd Li Ge Mg Be
24^30.7^ 2^,34.7^^ ly 80^40.4^ 12«,28.8^ 28^
«S. M. Sze and J. C. Irvin, Solid State Electron. 11, 599 (1968). ^C. M. Wolfe et al., Conf. Ser. Inst. Phys. 33b, 120 (1977). '^M. Ozeki et al., Conf. Ser. Inst. Phys. 45, 220 (1979). '^ A. G. Milnes, Electron. Electron Phys. 61, 63 (1983). "U. Kaufmann and J. Schneider, Electron. Electron Phys. 58, 81 (1982).
and a bond angle of 109.47°. Gallium arsenide cleaves most readily on {110} family planes, but can also cleave on {111} planes and between (111) and (011). Of the eight planes in the {111} family, four {111A} planes contain only Ga atoms and four {1115} contain only As atoms. These two planes have different chemical activity and behavior [65, 66]. The elastic properties of GaAs include compliance and second- and third-order moduli. The small-stress second-order moduli have only three independent components [67]. The shear modulus, bulk modulus. Young modulus, Poisson ratio, isotropy ratio, Cauchy ratio, and Bom ratio are determined from the second-order moduli with the use of the formulae [63, p. 3] indicated in Table IV. The speed of nondispersive or (long-wavelength) bulk acoustic waves can be expressed in terms of the second-order moduli and the crystal density [63, p. 3], and is given for the high-symmetry [100], [110], and [HI] directions as Hsted in Table V. The room temperature phonon dispersion curve reported by Waugh and Dolling [68] is represented graphically in Figure 5. The data are the wave vectors along the [100], [110], and [111] directions.
26
ERIC DONKOR
TEMPERAIURe C O 1100 KXK) 900 800
10 07
0.8
09
700
1.0 1.1 1000/T (K-M
600
500
1.2
1.3
FIG. 3. Room temperature carrier concentration in MBE-grown GaAs as a function of effusion cell temperature for Si, Ge (n-type), Be, and Sn. The data were obtained from Hall effect and capacitance-voltage measurements at constant growth rate, substrate temperature, and AS4 : Ga flux ratio. Reproduced with permission from K. Ploog, "Springer Proceedings in Physics" (G. Lelay, J. Derrien, and N. Boccara, Eds.), Vol. 22, p. 10. ©1987 Springer-Verlag, New York.
FIG. 4. Variation of lattice constant versus temperature for stoichiometric, Ga excess, and As excess MBE-grown GaAs. Reproduced with permission from O. Madelung, "Data in Science and Technology: Semiconductors," p. 104. ©1991. Springer-Verlag, New York.
2
GALLIUM ARSENIDE HETEROSTRUCTURES
27
TABLE IV FORMULAE FOR ACOUSTIC AND MECHANICAL PROPERTIES OF GAAS
Parameter
Formula Cu
Shear modulus
C,2
2
c
Bulk modulus
Xic
5 , = -^
Young Modulus along [100]
3 Y, = (i—^-[\n{\+b)(mVmo)^Af,V 1+^ b . '•^^-'^'%'-'T'
(2.7)
n' = n + {No-N^-n)^^
(2.8)
A^,. = « + 2A^^
(2.9)
Here T is the temperature in kelvin, e is the electronic charge, and ^* is the Callan effective charge (and e""/e = 0.20), M is the reduced mass of the cell that equals 5.92 x 10^^ kg, 11 = 5.05 x 10"^^ m^ is the volume of the primitive cell, 0 is the polar phonon temperature, z = 0/T, G{z) is a screening factor [60], n is the electron density, k^ is the static dielectric constant, and A/^, and A^^ (per cubic centimeter) are the donor and acceptor densities, respectively.
3. 3.1.
Growth and Material Properties of GaAs Heterostructures
INTRODUCTION
Epitaxial layers of GaAs-based heterostructures are binary, tertiary, quaternary, or quinary alloys of III-V compounds. Binary systems are of the type III^-Vi_^, where III and V imply elements from group III and V, respectively. Ternary systems are of the type III^-III,_^-V and III-V^-Vi_^. Quaternary systems are of three main types and quinary systems are of two types. The three types for the quaternary system are III-(V^-V,_ J^-Vi_^., III^-IIIi_^-V^,-Vi_^, and (III^-IIIi_^)^-IIIi_^-V, and for the quinary system, the two types are (III;,IIIi_J^-III,_^-V,-Vi_' and III,-III,_,-(V,-V,_,),-Vi_^, where 0 < JC < 1, 0 < y < 1, and 0 < z < 1. Heterostructures may be lattice-matched or latticemismatched based on the relative lattice constant of the constituent alloys. In lattice-matched structures, the lattice constants of constituent alloys are practically equal. As a result, interface defects are eliminated and a device grade crystalline structure is formed. The difference in lattice constant of a latticemismatched structure is accommodated by a combination of coherent strain and misfit dislocations at the heterointerfaces. Misfit dislocations are defects that severely degrade material properties, especially if the epilayers have thickness in excess of 1000 A. However, there exists a critical layer thickness below which the energy of the lattice mismatch at a heterointerface is totally accommodated by strain. The mismatch layers can be strain-relieved (i.e., elastic or coherent strained) [76, 77]. Coherent-strained (or pseudomorphic) structures have negligible misfit defects and are, therefore, used as active media for electrical and
2
31
GALLIUM ARSENIDE HETEROSTRUCTURES
optical device applications. Strain-relieved layers produce defects to accommodate stress relief. They are electrically and optically inactive, and serve mainly as substrate materials [78-80]. Strain can affect the confinement of electronic states [81] by inducing large internal electric fields [82-84]. Furthermore, strain alters band structure and modifies transport and optical properties of the strained-layer heterostructure. Therefore, strained-layer structures offer a variety of material and physical properties that cannot be obtained with lattice-matched systems.
3.2.
CRITICAL THICKNESS OF STRAINED-LAYER QUANTUM WELLS
The lattice mismatch, / , between a substrate and an epitaxial layer is given by
/ =
a^c.
•OL,
epi
(3.1)
-epi
where a^^^ and a^^^ are the lattice parameters for the substrate and epilayer, respectively. The lattice mismatch is generally accommodated by a combination of in-plane coherent strain ^u and misfit dislocation 5: / = £„ + §. Under appropriate growth conditions the lattice mismatch is compensated for by distortion of the lattice of the epilayer without formation of misfit dislocations or clusters. That is, the strain of the epilayer film equals the mismatch, f = s^. This growth mode continues only up to a critical film thickness, h^, which is a function of the lattice mismatch and the growth temperature. Two models for determining the critical thickness are owing to Matthews and Blakeslee [85] and People and Bean [86], respectively. In the Matthews and Blakeslee model, the critical strain is given by SM
=
2h(l-\-p)C0S\\_
27T{lJLf-\-fl,)
In
Ph
(3.2)
Here /x^ and /jLf are the shear moduli of the substrate and the epitaxial film, respectively, v is the Poisson ratio, b is Burger's vector, A is the angle between the slip direction and the direction in the film plane (which is perpendicular to the line of intersection of the slip plane and the interface), and h is the thickness of the epitaxial layer. The angle of inclination between Burger's vector and the dislocation fine [87] is S^^ and /3 is the core energy parameter, which is j8 = 1 for metals and j8 = 4 for semiconductors [88]. Setting / = Sn gives the critical thickness as h =
Z?(l-i^cos2©db)
f^s
477/(1 +^')C0SA fljr-^fX^
m)
(3.3)
For a MQW consisting of n pairs of wells (thickness h^ and strain s^) and barriers (thickness /z^ and strain e^), the critical thickness can be expressed in
32
ERIC DONKOR
~wz: 7iAs\
V'-. \ V
GaP
2.0 C o XJ
-AlSb
1.6
AlAs
\ { \v
- ^ AlSb
GaAs" — InP
inpyv
0.8
1.033
'^v^
-InAs^/
./GaAs
'GaP
y^'AlAs
JnSb/^]S^'AlP 0
II 1 5.5
1 II 1 5.6 5.7
J 5.8
\
JcaSb
P> i«^«>wH GaSb"
0.4
_ 0.775
^aAs
1.2 O
_ 0.620
'**''***
l_
X'5.9 InP
xjnAs ^ AlSb GaSb 1 ll 1 1 6.0 ' 6.1 6.2 InAs
_
1.550
\ ~
3.100
v ^v^^^lnSbX
>>--^
T
J
6.3
1 6.4
I
'6.5 InSb
Lattice Parameter (A)
FIG. 7. Lattice parameters, band gaps, and emission wavelengths of binary III-V compounds. Reproduced with permission from A. Zunger and S. Mahajan, "Handbook on Semiconductors" (S. Mahajan, Ed.), Vol. 3b, p. 1403. ©1992 North-Holland, Amsterdam.
terms of an effective strain e* and total thickness h* as h: = e
=
b{l-vcos'e,,) 87rfi*(l H-i^)cosA K + hu
(3.4) (3.5)
Figure 7 is a composite graph of wavelength versus lattice constant, and energy bandgap versus lattice constant for III-V compounds [89]. The figure suggests a vast number of binary, ternary, quaternary, and quinary alloys that can be grown on GaAs. However, the criteria for choosing any compound is dictated by the desired energy band and band offset. Compounds that form lattice mismatches of less than 2% with GaAs can form either lattice-matched or strained-layer coherent epitaxy films. Although attempts have been made to grow epitaxial layers with lattice mismatches between 2 and 7% with GaAs, the quality of such materials degrades due to the high density of misfit dislocations.
3.3.
3.3.1.
HETEROSTRUCTURES OF THE TYPE I I I - V / G A A S
InP/GaAs Lattice-Mismatched System
Heteroepitaxial InP has been successfully grown on (001) surfaces [90-92], as well as (111) surfaces [93] of GaAs-oriented substrate. Typical MOCVD growth uses TMIn and PH3 (diluted in hydrogen) as source materials [94]. Growth temperature falls within 570-680° C. In one such growth [95], the V : III ratio
2
33
GALLIUM ARSENIDE HETEROSTRUCTURES
was 80, with flow rates of 8.8 x 10"^ and 7.0 x 10"^ mol/min for TMIn and PH3, respectively. Growth at both low pressure [92] and atmospheric pressure [93] has been demonstrated. The InP/GaAs heterostructure has a lattice mismatch that gives rise to strain along the growth axis, £^^, and a biaxial strain in the interfacial plane, e^. *±GaAs •
*InP
«Tr
880 nm, the PL bands involve transition to impurity or defect states [92]. As the temperature increases, the transitions associated with the light hole (Ih) increase. As the layer thickness decreases (below 2 /xm), the band-edge emission broadens and shifts toward the red end of the spectrum due to band-edge recombination at low temperatures [92]. The PL intensity has been found to increase with the flow rate of the reactants during growth [95]. The mobility and carrier concentration of InP grown on GaAs both depend on the 0.002
InP on GaAs • Model • XRD A PL a.
B
0.001
0.000
Layer Thickness [\un] FIG. 8. Layer thickness dependence of the biaxial compressive strain for heteroepitaxial InP on GaAs substrates. Results show X-ray diffraction, PL measurements, and the theoretical model. Reproduced with permission from D. J. Olego, Y. Okuno, T. Kawano, and M. Tamura, J. AppL Phys. 71, 4492 (1992). © American Institute of Physics, New York.
34
ERIC DONKOR
A(1^19tV)
T=10K
^.^ =)
P = lW.ciiiM
"*>'(d — *' •
1
•4-*
*
.^
Here the index i = w,b stands for the well and barrier regions, respectively. The Schrodinger equation then becomes
( That is.
^2
^2
^2^2
h^k^\
-2 f
(4.10)
where
Pl^^-^iK-EJ
V„<E^
CL
2.6
Photon energy (eV) (a)
335
340
345
3.50
3 55
photon energy (eV) (b)
FIG. 28. (a) Photoluminescence spectra at selected temperatures for Mg-doped GaN. The magnesium concentration is less than 1 x 10'^ cm"\ DAP is the zero-phonon donor-acceptor pair transition. DAP-ILO and DAP-2L0 are the first and second phonon replicas, (b) Exciton region PL spectrum at 12 K. Reprinted with permission from A. K. Viswanath et al., J. Appl. Phys. 83, 2272 (1998).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
103
GaN:Mg Annealed Mg
/
/
B
CL
y ^ '^ 10
1 - 1 0-93 spon p 100
Excitation density (MW/cm^)
FIG. 57. Peak intensity as a function of excitation densities. The intensity of the spontaneous emission peak (open squares) increases almost Hnearly with excitation power. The lasing peak intensity (open circles) exhibits a strong superlinear increase as the pump density is raised. The solid line represents a linear fit to the experimental data. Reprinted with permission from S. Bidnyk et al., Appl Phys. Lett. 73, 2242 (1998).
They compared these results with those of GaN epilayers and GaAs/GaAlAs MQWs. Strong quantum confinement of excitons in the GaN wells was revealed. The exciton-LO phonon interactions in AlGaN barriers were noted to be enhanced. Localized excitons at low temperatures and free excitons at higher temperatures were identified. An increase in the lifetime of excitons up to 60 K was observed and indicates radiative recombination. Cingolani et al. [504] studied MQWs grown by MBE. The structure of the sample is shown in Figure 58. In the photoluminescence spectrum (Fig. 59), A, B, and C excitons and their phonon replicas were observed. From the intensity dependence of the luminescence (Fig. 60), the assignments of the transitions were confirmed. The temperature dependence of the spectra is shown in Figure 61. The n—\ band persists up to room temperature, whereas the extrinsic emission bands are ionized above 120 K. The thermal shift of the « = 1 band follows the thermal shift of the bulk GaN energy gap. Morkoc and co-workers [505] fabricated a 50-A/50-A GaN/Al^Gai_^N {x = 0.07) multiple quantum well by MBE. Dry etching was used to pattern an array of microdisks of approximately 9-ixm diameter and 50-)Ltm spacing. Picosecond time-resolved spectroscopy was used to study the emission dynamics. Strong enhancement of the intrinsic exciton transition quantum efficiency was observed in the microdisk compared to the MQW. It was proposed that microdisk structures are very suitable for vertical cavity surfaceemitting lasers.
3
129
GROWTH AND OPTICAL PROPERTIES OF GAN
AI,„Ga,„N
nrff"ff¥W
132 eV
3 907 eV
E = 3 692 eV
•L,— m, "0.2
fT^ • OS
L,» 10 iim
U, « 200 nm
n\«1.1
nv«0.l7
U, * 5 nm
L, » 100 nm
FIG. 58. Band lineup of the GaN/Al^Gai_^N separate confinement heterostructure. The main electronic parameters are indicated in the figure (the effective masses are in unit of AMQ). The ternary alloy gap was calculated by linear interpolation between the binary GaN and AIN constituents. The thick arrows indicate the calculated excitonic energies. Due to the relatively narrow barriers, the n = 1 state exhibits a minibandwidth of about 1.5 meV, which is neglected in the calculations. Reprinted with permission from R. Cingolani et al., Phys. Rev. B 56, 1491 (1997).
A theoretical investigation on the effect of biaxial strain on the valence bands in GaN quantum wells was made by Niwa et al. [506] using the tight binding method. Shan et al. [507] measured picosecond time-resolved luminescence of photoexcited carriers in GaN/AlGaN double heterostructures. Radiative and nonradiative recombination dynamics were studied. The diffusion constant for the minority carriers in the AlGaN barrier layers was estimated. Zeng et al. [508] investigated the effects of well thickness and Si doping on the optical properties of MQWs of GaN/AlGaN. Quantum confinement was observed in quantum wells (QWs) of thickness less than 40 A. The exciton lifetimes in MQWs with well thickness less than 40 A increase linearly with temperature up to 60 K. It was also observed that Si doping improves the quality of the crystals. Resonant Raman scattering in GaN/AlGaN single quantum wells was reported by Behr et al. [509]. The frequency of the A^ LO phonon in QW was found to be close to that in bulk GaN when the width of the well is narrow. Mair et al. [510] observed optical modes within the PL spectra of microdisks in GaN quantum wells. Application of these materials to fabricate microdisk cavity lasers was suggested. Grandjean et al. [511] fabricated GaN/AlGaN quantum wells by molecular beam epitaxy. The quantum well width could be controlled up to monolayer scale. Figure 62 shows the PL spectra of the sample. Quantum confinement effects can be clearly seen in the PL spectra. The QW transition
130
A N N A M R A J U KASI VISWANATH
BE.":;
n=2
f? A
E
f\/\B
3
y
B-A m c o £ C
o
V)
•-o.
/Hi
to
E
IXJ
*^;— . i.^
^1.^.
Energy (eV) FIG. 59. Photoluminescence (full dots) and photoluminescence excitation (thick continuous line) spectra recorded at 10 K under 0.3-mW/cm~^ excitation intensity, n = 1 and n = 2 indicate the transition energies calculated in Fig. 58. The photoluminescence spectrum is deconvoluted by independent Gaussians (dashed lines), revealing the presence of the A, B, and C excitons and their phonon sidebands ( L 0 A , L 0 B ) , of bound exciton to acceptor impurity (BE^), and of conduction band to acceptor recombination (B-A). The total calculated spectrum, accounting only for the transitions from the ground level to the donor-acceptor band, is displayed by the thin continuous line that interpolates the data points. The detection energy for the PLE experiments was fixed at 3.42 eV. Reprinted with permission from R. Cingolani et ai, Phys. Rev. B 56, 1491 (1997).
energy increases as the thickness decreases. Each QW exhibits a clearly resolved emission peak, which is due to the relatively narrow PL line widths, ranging between 20 and 30 meV. However, luminescence from the AlGaN barriers is not detected, perhaps due to strong capture of the excitons by the QWs. The variation of QW energy as a function of thickness is shown in Figure 63. When the thickness of the well is decreased, the intensity of the QW emission vanishes. This observation was explained as due to thermal escape of carriers from the QW to the barrier due to insufficient confinement. The electric field dependence of electron mobility in GaN QWs was theoretically investigated by Zakhleniuk et al. [512]. For temperatures below 200 K, the electron mobility was controlled by acoustic phonons. At higher temperatures, polar optical phonons were found to determine the mobility. Zeng et al. [513] investigated the time-resolved PL in GaN/AlGaN MQWs with very high excitation that is necessary to obtain lasing threshold. Under these conditions, they found that the carrier distributions are characterized by plasma temperatures. Leroux et al [514] grew high-quality MQWs of GaN/AlGaN by MBE. They examined the quantum confined Stark effect by temperature-dependent luminescence and reflectivity. The temperature dependence of the PL energies is shown in Figure 64. An electric field strength
3
GROWTH AND OPTICAL PROPERTIES OF GAN ' ' """1
1
131
•]
10°
A^ 1 ! /
/
7
/
3.6
BE
A
LO, O B-A a n=1 o n = 2 •!
3.7
Energy (eV)
FIG. 60. Intensity dependence of the luminescence spectra recorded at 4 K. The excitation power increases from bottom to top: 2.5 W/cm~^ 7.5 W/cm~^ 30 W/cm~^ 40 W/cm~^ and 70 W/cm"^ Inset: Intensity dependence of the integrated emission of the different luminescence bands. « = 1, the phonon replica, and /i = 2 are linear over 3 orders of magnitude, whereas BE^ and B-A exhibit a clear saturation above 20 W/cm~^. Reprinted with permission from R. Cingolani et ai, Phys. Rev. B 56, 1491 (1997).
of 450 KV/cm was determined. They concluded that the origin of the electric field is predominantly spontaneous polarization effects rather than piezoelectric effects in the well. Suziki et al. [515] calculated electron scattering rates in GaN/AlGaN quantum wells. The intersubband scattering time was found to be dominated by LO-phonon scattering and was estimated as 100 fs. Im et al. [516] studied the piezoelectric fields in GaN/ AlGaN quantum wells by time-resolved PL. A reduction in oscillator strength was observed that was attributed to piezoelectric fields. An increase in luminescence decay time with increasing well width was observed along with a red shift of the emission peaks. Recombination dynamics of free and localized excitons in MQWs of GaN/AlGaN was reported by Lefebvre et al. [517]. The decay times of excitons in both the wells and the barriers was found to be -^330 ps at 8 K. Spectral distribution of lifetimes was attributed to localization of carriers by potential fluctuations that arise due to alloy disorder and well width and depth variations. The radiative lifetime of free excitons in the low-temperature limit was estimated as 2.4 ps, which is much smaller than that for GaAs/GaAlAs MQWs. Suziki and lizuka [518] reported on the effect of the built-in field caused by the piezoelectric effect and the spontaneous polarization inherent in nitride quantum wells on the intersubband transition. Grandjean et al. [519] measured the PL of GaN quantum wells for
132
A N N A M R A J U KASI VISWANATH
3.51 h
• > 3.50 h 10 K . \ 3L ^
•,V:t5 3-»H
c o *S)
0
50 100 150 200 250 300 Temperature (K)
E
lij
-^^. 3.6
3.7
3.8
Energy (eV) FIG. 61. Temperature dependence of the luminescence spectra recorded under power density of 50 W/cm~^. The n = 1 band persists up to room temperature. Inset: Thermal shift of the n = 1 band. The continuous curve is the best fit to the Varshni law. The dashed and dotted curves are the thermal shifts obtained for bulk GaN by different authors. Reprinted with permission from R. Cingolani et al, Phys. Rev. B 56, 1491 (1997).
9MLs /\
(a)
T = 9K
D (0 13 MLs /
5) c
GaN A
0)
\
/
5MLs
c
I
(b)
c 0)
c 8 E 3 O x:
7 MLs 15MLJ ; f^ 11 MLs
J
CL
3.40
3.45
3.50
\
I
,...v
3.55
3 MLs (20
3 60
3.65
3.70
Photon energy (eV)
FIG. 62. 9-K photoluminescence spectra of GaN/Alg nGaoggN QWs that correspond to (a) sample A and (b) sample B. Note the GaN buffer layer emission at 3.478 eV. Reprinted with permission from N. Grandjean and J. Massies, Appl Phys. Lett. 73, 1260 (1998).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
Al
oil
^3.75
^
Ga
0 09
133
N
370
0)
S
3.60
c E
3.55
GaN 0
2
4
6
8
10
12
14
16
QW thickness (ML)
FIG. 63. Transition energy of GaN/Alo nGooggN QWs as a function of the well width [one monolayer (IML) = 2.59 A]. Open and closed symbols correspond to samples A and B, respectively. The bandgap of AIQ uGaoggN is taken from the literature. Reprinted with permission from N. Grandjean and J. Massies, AppL Phys. Lett. 73, 1260 (1998).
different Al contents. Strong internal electric fields were found to be present and to have a linear relationship with Al content. The GaN/AlGaN quantum wells were characterized by photoreflectance spectroscopy [520]. Gil et al. [521] characterized long-lived oblique excitons in GaN/AlGaN MQWs in picosecond time-resolved spectra.
Al,,Ga,,N/GaN QWs
QW(5 MLs)
> 0)
0
SO
100
150
200
250
300
Temperature (K)
FIG. 64. Temperature dependence of the PL energies of sample A. The closed squares are freeexciton energies deduced from reflectivity. Reprinted with permission from M. Leroux et al, Phys. Rev. B 58, R13, 371 (1998).
134
ANNAMRAJU KASI VISWANATH
References 1. S. Strite, M. E. Lin, and H. Morkoc, Thin Solid Films 231, 197 (1993). 2. H. Morkoc, S. Strite, G. S. Gao, M. E. Lin, B. Sverdlov, and M. Bums, J. Appl. Phys. 73, 1363 (1994). 3. S. N. Mohammad, A. A. Salvador, and H. Morkoc, Proc. IEEE 83, 1306 (1995). 4. H. Morkoc, in "Semiconductor Heteroepitaxy" (B. Gil and R. L. Aulombard, Eds.), pp. 238-249. World Scientific, Singapore, 1995. 5. T. D. Moustakas, J. L Pankove, and Y. Hamakawa, Eds., "Wide Bandgap Semiconductors," Vol. 242. Materials Research Society, Pittsburgh, PA, 1992. 6. M. Razeghi and A. Rogalski, J. Appl. Phys. 79, 7433 (1996). 7. S. Nakamura and G. Fasol, "The Blue Laser Diode." Springer-Veriag, Beriin, 1997. 8. R. F. Davis, Proc. IEEE 79, 702 (1991). 9. S. Nakamura, Solid State Commun. 102, 237 (1997). 10. B. Gil, Ed., "Group III Nitride Semiconductor Compounds." Clarendon, Oxford, 1998. 11. M. A. Khan, J. N. Kuznia, A. R. Bhattarai, and D. T. Olson, Appl. Phys. Lett. 62, 1786 (1993); M. A. Khan, A. Bhattarai, J. N. Kuznia, and D. T. Olson, Appl. Phys. Lett. 63, 1214 (1993). 12. A. Gustafsson, M. E. Pistol, L. Montelius, and L. Samuelson, J. Appl. Phys. 84, 1715 (1998). 13. G. Popovici, H. Morkoc, and S. N. Mohammad, in "Group III Nitride Semiconductor Compounds" (B. Gil, Ed.). Clarendon, Oxford, 1998. 14. R. Dingle, D. D. Sell, S. E. Stokowski, and M. Ilegems, Phys. Rev. B 4, 1211 (1971). 15. B. Monemar, Phys. Rev. B 10, 676 (1974). 16. J. L Pankove, J. E.Berkeyheiser,H.PMaruska, and J. Wittke,5o//J 5rflr^CommM«. 8,1051 (1970). 17. S. Nakamura, M. Senoh, and T. Mukai, Jpn. J. Appl. Phys. 32, L8 (1998). 18. S. Nakamura, T. Mukai, and M. Senoh, Appl. Phys. Lett. 64, 1686 (1994). 19. S. Nakamura and T. Mukai, Jpn. J. Appl. Phys. 31, L1457 (1992). 20. S. Nakamura, T. Mukai, M. Senoh, and N. Isawa, Jpn. J. Appl. Phys. 31, L139 (1992). 21. S. Nakamura, M. Senoh, S. I. Nagahama, N. Isawa, T. Yamada, T. Matsushita, H. Kiyoku, and Y. Sugimoto, Jpn. J. Appl. Phys. 35, L74 (1996). 22. S. Chichibu, T. Azuhata, T. Sota, and S. Nakamura, / Appl. Phys. 79, 2784 (1996). 23. Y. Kawakami, Z. G. Peng, Y. Narukawa, S. Fujita, S. Fujita, and S. Nakamura, Appl. Phys. Lett. 69, 1414 (1996). 24. S. Chichibu, A. Shikanai, T. Azuhata, T. Sota, A. Kuramata, K. Horino, and S. Namamura, Appl. Phys. Lett. 68, 3766 (1996). 25. A. Shikanai, T. Azuhata, T. Sota, S. Chichibu, A. Kuramata, K. Horino, and S. Nakamura, J. Appl. Phys. 81, 417 (1997). 26. S. Nakamura, S. Masayuki, S. Nagahama, N. Iwasa, T. Yamada, T. Matsushita, Y. Sugimoto, and K. Hiroyuki, Appl. Phys. Lett. 70, 1417 (1997). 27. A. Alemu, B. Gil, M. Julier, and S. Nakamura, Phys. Rev. B 57, 3761 (1998). 28. M. Julier, J. Campo, B. Gil, J. P Lascaray, and S. Nakamura Phys. Rev B 57, R6791 (1998). 29. S. Chichibu, H. Okumura, S. Nakamura, G. Feuillet, T. Azuhata, T. Sota, and S. Yoshida, Jpn. J. Appl. Phys. 36, 1976 (1997). 30. S. Chichibu, T. Mizutani, T. Shoida, H. Nakanishi, T. Deguchi, T. Azuhata, T. Sota, and S. Nakamura, Appl. Phys. Lett. 70, 3440 (1997). 31. S. Nakamura, Jpn. J. Appl. Phys. 30, 1620 (1991). 32. H. Amano, K. Hiramatsu, and I. Akasaki, Jpn. J. Appl. Phys. 27, L1384 (1988). 33. K. Naniwae, S. Itoh, H. Amano, K. Itoh, K. Hiramatsu, and I. Akasaki, J. Cryst. Growth 99, 381 (1990). 34. K. Hiramatsu, T. Detchprohm, and I. Akasaki, Jpn. J. Appl. Phys. 32, 1528 (1993). 35. H. Amano, M. Kito, K. Hiramatsu, and I. Akasaki, Jpn. J. Appl. Phys. 28, L2112 (1989). 36. D. Volm, K. Oettinger, T. Streibl, D. Kovalev, M. Ben-Chorin, J. Diener, B. K. Meyer, J. Mejewski, L. Eckey, A. Hoffmann, H. Amano, I. Akasaki, K. Hiramatsu, and T. Detchprohm, Phys. Rev. B 53, 16,543 (1996).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
135
37. C. Wetzel, D. Volm, B. K. Meyer, K. Pressel, S. Nilsson, E. N. Mokhov, and P. G. Baranov, Appl Phys. Lett. 65, 1033 (1994). 38. D. Kovalev, B. Averboukh, D. Volm, B. K. Meyer, H. Amano, and I. Akasaki, Phys. Rev. B 54, 2518 (1996). 39. B. K. Meyer, D. Volm, A. Graber, H. C. Alt, T. Detchprohm, A. Amano, and I. Akasaki, Solid State Commun. 95, 597 (1995). 40. T. Detchprohm, K. Hiramatsu, K. Itoh, and I. Akasaki, Jpn. J. Appl. Phys. 31, L1454 (1992). 41. D. Volm, T. Streibl, B. K. Meyer, T. Detchprohm, H. Amano, and I. Akasaki, Solid State Commun. 96, 53 (1995). 42. M. Drechsler, D. M. Hofmann, B. K. Meyer, T. Detchprohm, H. Amano, and I. Akasaki, Jpn. J. Appl Phys. 34, LI 178 (1995). 43. I. A. Buyanova, Mt. Wagner, W. M. Chen, B. Monemar, J. L. Lindstrom, H. Amano, and I. Akasaki, Appl. Phys. Lett. 73, 2968 (1998). 44. S. Heame, E. Chason, J. Han, J. A. Floro, J. Figiel, J. Hunter, H. Amano, and I. S. T. Tsong, Appl. Phys. Lett. 74, 356 (1999). 45. W. Shan, T. J. Schmidt, X. H. Yang. S. J. Hwang, J. J. Song, and B. Goldenberg, Appl Phys. Lett. 66, 985 (1995). 46. W Shan, T. J. Schmidt, R. J. Haunstein, J. J. Song, and B. Goldenberg, Appl Phys. Lett. 66, 3492 (1995). 47. W. Shan, T. Schmidt, X. H. Yang, J. J. Song, and B. Goldenberg, J. Appl Phys. 79, 3691 (1996). 48. W Shan, R. J. Haunstein, A. J. Fischer, J. J. Song, W. G. Perry, M. D. Bremser, R. F. Davis, and B. Goldenberg, Phys. Rev. B 54, 13,460 (1996). 49. W. Shan, B. D. Little, A. J. Fischer, J. J. Song, B. Goldenberg, W. G. Perry, M. D. Bremser, and R. R Davis, Phys. Rev. B 54, 16,369 (1996). 50. A. J. Fischer, W. Shan, J. J. Song, Y C. Chang, R. Homing, and B. Goldenberg, Appl Phys. Lett. 71, 1981 (1997). 51. W. Shan, A. J. Fischer, S. J. Hwang, B. D. Little, R. J. Haunstein, X. C. Xie, J. J. Song, D. S. Kim, B. Goldenberg, R. Horing, S. Krishmamkutty, W. G Perry, M. D. Bremser, and R. F. Davis, /. Appl Phys. 83, 455 (1998). 52. M. Tchounkeu, O. Briot, B. Gil, J. P Alexis, and R. L. Aulombard, J. Appl Phys. 80, 5352 (1996). 53. B. Gil, O. Briot, and R. L. Aulombard, Phys. Rev B 52, R17,028 (1995). 54. B. Gil and A. Alemu, Phys. Rev B 56, 12,446 (1997). 55. B. Gil, S. Clur, and O. Briot, Solid State Commun. 104, 267 (1997). 56. G. D. Chen, M. Smith, J. Y Lin, H. X. Jiang, M. A. Khan, and C. J. Sun, Appl Phys. Lett. 67, 1653 (1995). 57. M. Smith, G D. Chen, J. Y Lin, H. X. Jiang, M. A. Khan, and C. J. Sun, Appl Phys. Lett. 67, 3295 (1995). 58. G D. Chen, M. Smith, J. Y Lin, H. X. Jiang, S. Wei, M. A. Khan, and C. J. Sun, Appl Phys. Lett. 68, 2784 (1996). 59. M. Smith, G. D. Chen, J. Y Lin, H. X. Jiang, M. A. Khan, C. J. Sun, Q. Chen, and J. W. Yang, J. Appl Phys. 79, 7001 (1996). 60. M. Smith, J. Y Lin, H. X. Jiang, and M. A. Khan, Appl Phys. Lett. 71, 635 (1997). 61. A. K. Viswanath, J. L Lee, S. Yu, D. Kim, Y Choi, and C. H. Hong, J. Appl Phys. 84, 3848 (1998). 62. A. K. Viswanath, J. L Lee, D. Kim, C. R. Lee, and J. Y Leem, Phys. Rev. B 58, 16,333 (1998). 63. A. K. Viswanath, J. L Lee, C. R. Lee, J. Y Leem, and D. Kim, Solid State Commun. 108, 483 (1998). 64. A. K. Viswanath, J. I. Lee, C. R. Lee, J. Y Leem, and D. Kim, Appl Phys. A 67, 551 (1998). 65. C. F. Li, Y S. Huang, L,. Malikova, and F. H. Pollak, Phys. Rev. B 55, 9251 (1997). 66. Y Li, Y Lu, H. Shem, M. Wraback, M. G. Brown, M. Schurman, L. Koszi, and R. A. Stall, Appl Phys. Lett. 70, 2458 (1997).
136
ANNAMRAJU KASI VISWANATH
67. M. O. Manasreh, Phys. Rev. B 53, 16,425 (1996). 68. N. V. Edwards, S. D. Yoo, M. D. Bremser, T. W. Weeks, Jr., O. H. Nam, H. Liu, R. A. Stall, M. N. Morton, N. R. Perkins, T. F. Kuech, and D. E. Aspnes, Appl Phys. Lett. 70, 2001 (1997). 69. S. Kim, I. R Herman, J. A. Tuchman, K. Doverspike, L. B. Rowland, and D. K. Gaskill, Appl. Phys. Lett. 67, 380 (1995). 70. A. Saxler, D. Walker, R Kung, X. Zhang, M. Razheghi, J. Solomon, W. C. Mitchel, and H. R. Vydyanath, Appl. Phys. Lett. 71, 3272 (1997). 71. D. A. Tumbull, X. Li, S. Q. Gu, E. E. Renter, J. J. Coleman, and S. G. Bishop, J. Appl. Phys. 80, 4609 (1996). 72. C. Y. Yeh, S. H. Wei, and A. Zunger, Phys. Rev. B 50, 2715 (1995). 73. K. Cho, Phys. Rev. B 14, 4463 (1976). 74. G. E. Pikus and L. G. Bir, "Symmetry and Strain-Induced Effects in Semiconductors." Wiley, New York, 1974. 75. D. W. Langer, R. N. Euwema, K. Era, and T. Koda, Phys. Rev B 2, 4005 (1970). 76. M. Grynberg, Phys. Status. Solidi 27, 255 (1968). 77. J. E. Rowe, M. Cardona, and F. H. Pollak, in "Proceedings of the 1967 International Conference on II-VI Compounds" (D. G. Thomas Ed.). Benjamin, New York, 1967. 78. J. E. Rowe, M. Cardona, and F. H. Pollak, Solid State Commun. 6, 239 (1968). 79. A. A. Yamaguchi, Y. Mochizuki, C. Sasaoka, A. Kimura, M. Nido, and A. Usui, Appl. Phys. Lett. 71, 374 (1997). 80. J. R. Haynes, Phys. Rev. Lett. 4, 351 (1960). 81. H. Frohlich, Adv. Phys. 3, 325 (1954). 82. J. Conradi and R. R Haering, Phys. Rev. Lett. 20, 1344 (1968). 83. E. Gross, S. Permogorov, V. Travinokov, and A. Selkin, J. Phys. Chem. Solids 31, 2595 (1970). 84. S. Permogorov, in "Modem Problems in Condensed Matter Sciences" (E. I. Rashba and M. D. Sturge, Eds.), Vol. 2. North Holland, Amsterdam, 1982. 85. Y. S Park and J. R. Schneider, Phys. Rev Lett. 21, 798 (1968). 86. A. A Klochikhin, S. A. Permogorov, and A. N. Reznitsky, Sov Phys. JETP 44, 1176 (1976). 87. G. Popovici, H. Morkoc and S. N. Mohammad, in "Group III Nitride Semiconductor Compounds" (B. Gil, Ed.). Clarendon, Oxford, 1998. 88. H. Morkoc, IEEE J. Select. Topics Quantum Electron. 4, 537 (1998). 89. S. Strite, B. Sariel, D. J. Smith, H. Chen, and H. Morkoc, J. Cryst. Growth 127, 204 (1993). 90. A. Botchkarev, A. Salvador, B. Sverdlov, J. Myoung, and H. Morkoc, J. Appl. Phys. 11, 4455 (1995). 91. M. Smith, G. D. Chen, J. Y. Lin, and H. X. Jiang, A. Salvador, B. N. Sverdlov, A. Botchkarev, and H. Morkoc, Appl. Phys. Lett. 66, 3474 (1995). 92. D. J. Smith, D. Chandrasekhar, B. Sverdlov, A. Botchkarev, A. Salvador, and H. Morkoc, Appl. Phys. Lett. 67, 1830 (1995). 93. M. Smith, G. D. Chen, J. Z. Li, J. Y Lin, H. X. Jiang, A. Salvador, W. K. Kim, O. Aktas, A. Botchkarev, and H. Morkoc, Appl. Phys. Lett. 67, 3387 (1995). 94. Q. Zhu, A. Botchkarev, W. Kim, O. Aktas, A. Salvador, B. Sverdlov, H. Morkoc, S. C. Y. Tsen, and D. J. Smith, Appl. Phys. Lett. 68, 1141 (1996). 95. D. C. Reynolds, D. C. Look, W. Kim, O. Aktas, A. Botchkarev, A. Salvador, H. Morkoc, and D. N. Talwar, J. Appl. Phys. 80, 594 (1996). 96. G. D. Chen, M. Smith, J. Y Lin, H. X. Jiang, A. Salvador, B. N. Sverdlov, A. Botchkarev, and H. Morkoc, / Appl. Phys. 79, 2675 (1996). 97. W. Kim, O. Aktas, A. E. Botchkarev, A. Salvador, S. N. Mohammad, and H. Morkoc, J. Appl. Phys. 79, 7657 (1996). 98. Z. X. Liu, S. Pau, K. Syassen, J. Kuhl, W. Kim, H. Morkoc, M. A. Khan, and C. J. Sun, Phys. Rev. B 58, 6696 (1998). 99. J. Petalas, S. Logothetidis, S. Boultadakis, M. Alouani, and J. M. Wills, Phys. Rev B 52, 8082 (1995).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
137
100. H. Teisseyre, P. Pertin, T. Suski, I. Grzegory, S. Porowski, J. Jun, A. Pietraszko, and T. D. Moustakas, J. Appl. Phys. 76, 2429 (1994). 101. C. Trager-Cowan, K. P O. Donnell, S. E. Hooper, and C. T. Foxon, Appl Phys. Lett. 68, 355 (1996). 102. Z. Yu, S. L. Buczkowski, N. C. Giles, T. H. Myers, and M. R. Richards-Babb Appl Phys. Lett. 69, 2731 (1996). 103. Y. Zhao, C. W. Tu, I. T. Bae, and T. Y Seong, Appl Phys. Lett. 74, 3182 (1999). 104. S. H. Cho, T. Maruyama, and K. Akimoto, Jpn. J. Appl Phys. 34, L1575 (1995). 105. K. Iwata, H. Asahi, S. J. Yu, K. Asami, H. Fujita, M. Fushida, and S. Gonda, Jpn. J. Appl Phys. 35, L289 (1996). 106. K. Komitzer, K. Thonke, R. Sauer, M. Mayer, M. Kamp, and K. J. Ebeling, J. Appl Phys. 83, 4397 (1998). 107. C. Heinlein, J. K. Grepstad, S. Einfeldt, D. Hommel, T. Berge, and A. P Grande, J. Appl Phys. 83, 6023 (1998). 108. Z. J. Yu. B. S. Sywe, A. U. Ahmed, and J. H. Edgar, J. Electron. Mater. 21, 383 (1992). 109. Y Okamoto, S. Hashiguchi, Y Okada, and M. Kawabe, Jpn. J. Appl Phys. 37, LI 109 (1998). 110. Y Shimizu, T. Tominari, S. Hokuto, Y Chiba, and Y Nanishi, Jpn. J. Appl Phys. 37, L700 (1998). 111. T. W. Kang, S. H. Park, H. D. Cho, M. Y Kwak, G. S. Eom, and T. W. Kim, Jpn. J. Appl Phys. 37, 4417 (1998). 112. N. Grandjean, M. Leroux, M. Langt, and J. Massies, Appl Phys. Lett. 71, 240 (1997). 113. M. E. Lin, B. N. Sverdlov, and H. Morkoc, J. Appl Phys. 74, 5038 (1993). 114. N. Grandjean, J. Massies, P. Vennegues, M. Leroux, F. Demangeot, M. Renucci, and J. Frandon, J. Appl Phys. 83, 1379 (1998). 115. N. Grandjean, J. Massies, F. Semond, S. Y Karpov and R. A. Talalaev, Appl Phys. Lett. 74, 1854 (1999). 116. A. Krtschil, H. Witte, M. Lisker, J. Christen, U. Birkle, S. Einfeldt, and D. Hommel, Appl Phys. Lett. 74, 2032 (1999). 117. H. Tang and J. B. Webb, Appl Phys. Lett. 14, 2373 (1999). 118. X. Q. Shen, P Ramvall, P Riblet, and Y Aoyagi, Jpn. J Appl Phys. 38, L411 (1999). 119. Y Okamoto, S. Hashiguchi, Y Okada, and M. Kawabe, Jpn. J Appl Phys. 38, L230 (1999). 120. S. Strite, J. Ruan. Z. Li, N. Manning, A. Salvador, H. Chen, D. J. Smith, W. J. Choyke, and H. Morkoc, J Vac. ScL Technol, B 9, 1924 (1991). 121. E. Kim, L Berishev, A. Bensaoula, L Rusakova, K. Waters, and J. A. Schultz, J. Appl Phys. 85, 1178(1999). 122. N. Grandjean, M. Leroux, J. Massies, M. Mesrine, and M. Laugt, Jpn. J. Appl Phys. 38, 618 (1999). 123. W. Widmann, G. Feuillet, B. Daudin, and J. L. Rouviere, J. Appl Phys. 85, 1550 (1999). 124. M, E. Lin, B. Sverdlov, G L. Zhou, and H. Morkoc, Appl Phys. Lett. 62, 3479 (1993). 125. S. Yoshida, S. Misawa, and S. Gonda, Appl Phys. Lett. 42, 427 (1983). 126. Z. Yang, L. K. Li, and W. L Wang, Appl Phys. Lett. 67, 1686 (1995). 127. W. S. Wong, N. Y Li, H. K. Dong, F Deng, S. S. Lau, C. W. Tu, J. Hays, S. Bidnyk, and J. J. Song, J. Cry St. Growth 164, 159 (1996). 128. N. Grandjean and J. Massies, Appl Phys. Lett. 71, 240, 1816 (1997). 129. W. Shan, A. J. Fischer, J. J. Song, G E. Bulman, H. S. Kong, M. T. Leonard, W. G Perry, M. D. Bremser, and, R. E Davis, Appl Phys. Lett. 69, 740 (1996). 130. L A. Buyanova, J. P. Bergman, B. Monemar, H. Amano, and, L Akasaki, Appl Phys. Lett. 69, 1255 (1996). 131. T. Sasaki and T. Matsuoka, J Appl Phys. 64, 4531 (1988). 132. S. Chichibu, T. Azuhata, T. Sota, H. Amano, and I. Akasaki, Appl Phys. Lett. 70, 2085 (1997). 133. T. W. Weeks, M. D. Bremser, K. S. Ailey, E. Carlson, W. G Perry, and R. F. Davis, Appl Phys. Lett. 67, 401 (1995).
138
ANNAMRAJU KASI VISWANATH
134. F. R. Chien, X. J. Ning, S. Stemmer, P. Pirouz, M. D. Bremsev, and R. R Davis, Appl. Phys. Lett. 68, 2678 (1996). 135. T. Nishida, T. Akasaka, and N. Kobayashi, Jpn. J. Appl. Phys. 31, L459 (1998). 136. X. Q. Shen, S. Tanaka, S. Iwai, and Y. Aoyagi, Jpn. J. Appl. Phys. 37, L637 (1998). 137. V. M. Torres, M. Stevens, J. L. Edwards, D. J. Smith, R. B. Doak, and I. S. T. Tsong, Appl. Phys. Lett. 71, 1365 (1997). 138. F. Boscherini, R. Lantier, A. Rizzi, F. D'Acapito, and S. Mobilio, Appl. Phys. Lett. 74, 3308 (1999). 139. M. A. L. Johnson, J. D. Brown, N. A. El Masry, J. W. Cook, Jr., J. F Schetzina, H. S. Kong, and J. A. Edmond, J. Vac. Sci. Technol, B 16, 1282 (1998). 140. H. Okumura, H. Hamaguchi, G. Feuillet, Y. Ishida, and S. Yoshida, Appl. Phys. Lett. 72, 3056 (1998). 141. F Hamdani, A. Botchkarev, W. Kim, H. Morkoc, M. Yeadon, J. M. Gibson, S. C. Y Tsen, D. J. Smith, D. C. Reynolds, D. C. Look, K. Evans, C. W. Litton, W. C. Mitchel, and R Hemenger, Appl. Phys. Lett. 70, 467 (1997). 142. F Hamdani, A. E. Botchkarev, H. Tang, W. Kim, and H. Morkoc, Appl. Phys. Lett. 71, 3111 (1997). 143. G. Popovici, W. Kim, A. Botchkarev, H. Tang, H. Morkoc, and J. Soloman, Appl. Phys. Lett. 71, 3385 (1997). 144. F. Hamdani, M. Yeadon, D. J. Smith, H. Tang, W. Kim, A. Salvador, A. E. Botchkarev, J. M. Gibson, A. Y. Polyakov, M. Skowronski, and H. Morkoc, / Appl. Phys. 83, 983 (1998). 145. X. W. Sun, R. F. Xiao, and H. S. Kwok, J. Appl. Phys. 84, 5776 (1998). 146. R Kung, A. Saxler, X. Zhang, D. Walker, R. Lavado, and M. Rezeghi, Appl. Phys. Lett. 69, 2116(1996). 147. T. Ishi, M. Mukaida, T Nishihara, S. Hayashi, and M. Shinohara, Jpn. J. Appl. Phys. 37, L672 (1998). 148. A. Kuramata, K. Horino, K. Domen, and K. Shinohera, Appl. Phys. Lett. 67, 2521 (1995). 149. T George, E. Jacobson, W. Pike, P. Chang-chein, M. Khan, J. Yang, and S. Mahajan, Appl. Phys. Lett. 68, 337 (1996). 150. C. Sun, J. Yang, Q. Chen, M. Khan, T. George, P Chang-chien, and S. Mahajan, Appl. Phys. Lett. 68, 1129(1996). 151. S. A. Nikishin, H. Terkin, V. G. Antipov, A. I. Guriev, A. S. Zubrilov, V. A. Elyukhin, N. N. Faleev, R. N. Kyutt, and A. K. Chin, Appl. Phys. Lett. 72, 2361 (1998). 152. G. Ramirez-Flores, H. Navarro-Contreras, A. Lastras-Martinez, R. C. Powell, and J. E. Greene, Phys. Rev. B 50, 8433 (1994). 153. R. C. Powell, N. E. Lee, Y. W. Kim, and J. E. Greene, J. Appl. Phys. 73, 189 (1993). 154. M. A. Vidal, G. Ramirez-Flores, H. Navarro-Contreras, A. Lastras-Martinez, R. C. Powell, and J. E. Greene, Appl. Phys. Lett. 68, 441 (1996). 155. P. Perlin, Y. Gorczyca, N. E. Christensen, Y Grzegory, H. Teiserye, and T. Suski, Phys. Rev. B 45, 13,307 (1992). 156. A. Strittmatter, A. Krost, M. Strabburg, V. Turck, D. Bimberg, J. Biasing, and J. Christen, Appl. Phys. Lett. 74, 1242 (1999). 157. X. Zhang, S. J. Chua, R Li, K. B. Chong, and Z. C. Feng, Appl. Phys. Lett. 74, 1984 (1999). 158. H. Ishikawa, G. Y Zhao, N. Nakada, T. Egawa, T Jimbo, and M. Umeno, Jpn. J. Appl. Phys. 38, L492 (1999). 159. N. R Kobayashi, J. T. Kobayashi, R D. Dapkus, W. J. Choi, A. E. Bond, X. Zhang, and D. H. Rich, Appl. Phys. Lett. 71, 3569 (1997). 160. A. J. Steckl, J. Devrajan, C. Tran, and R. A. Stall, Appl. Phys. Lett. 69, 2264 (1996). 161. J. Cao, D. Pavlidis, A. Eisenbach, A. Philippe, C. Bru-Chevallier, and G. Guillot, Appl. Phys. Lett. 71, 3880 (1997). 162. J. Cao, D. Pavlidis, Y Park, J Singh, and A. Eisenbach, J. Appl. Phys. 83, 3829 (1998).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
139
163. S. I. Molina, A. M. Sanchez, F. J. Pacheco, R. Garcia, M. A. Sanchez-Garcia, F. J. Sanchez, and E. Callja, AppL Phys. Lett. 74, 3362 (1999). 164. Y. Nakada, I. Aksenov, and H. Okumura, Appl. Phys. Lett. 73, 827 (1998). 165. I. Berishev, A. Bensaoula, I. Rusakova, A. Karabutov, M. Ugarov, and V. P. Ageev, Appl. Phys. Lett. 73, 1808 (1998). 166. N. P Kobayashi, J. T. Kobayashi, W. J. Choi, and P D. Dapkus, Appl. Phys. Lett. 73,1553 (1998). 167. Y. Hiroyama and M. Tamura, Jpn. J. Appl. Phys. 31, L630 (1998). 168. T. Lei, M. Fanciulli, R. J. Molnar, T. D. Moustakas, R. J. Graham, and J. Scanlon, Appl. Phys. Lett. 59, 944 (1991). 169. E. Calleja, M. A. Sanchez-Garcia, D. Basak, F J. Sanchez, F. Calle, P. Youinou, E. Munoz, J. J. Serrano, J. M. Blanco, C. Villar, T. Laine, J. Gila, K. Sarrinen, P Hautojarvi, C. H. Molloy, D. J. Somerford, and I. Harrison, Phys. Rev. B 58, 1550 (1998). 170. C. H. Hong, D. Pavilidis, S. W. Brown, and S. C. Rand, / Appl. Phys. 11, 1705 (1995). 171. A. A. Yamaguchi, T. Manako, A. Akai, H. Sunakawa, A. Kimura, M. Nido, and A. Usui, Jpn. J. Appl. Phys. 35, L873 (1996). 172. X. L. Sun, H. Yang, L. X. Zheng, D. P Xu, J. B. Li, Y. T. Yang, G. H. Li, and Z. G. Wang, Appl. Phys. Lett. 74, 2827 (1999). 173. J. Wu, H. Yaguchi, K. Onabe, R. Ito, and Y Shiraki, Appl. Phys. Lett. 71, 2067 (1997). 174. J. Wu, H. Yaguchi, K. Onabe, Y Shiraki, and R. Ito, Jpn. J Appl. Phys. 37, 1440 (1998). 175. H. Tsuchiya, K. Sunaba, M. Minami, T. Suemasu, and F Hasegawa, Jpn. J. AppL Phys. 37, L568 (1998). 176. M. Sato, J Appl. Phys. 78, 2123 (1995). 177. S. Strite, J. Ruan, Z. Li, A. Salvador, H. Chen, D. J. Smith, W. J. Choyke, and H. Morkoc, / Vac. Sci. Technol. 89, 1924 (1991). 178. X. W. Lin, M. Behar, R. Maltez, W. Swider, Z. Liliental-Weber, and J. Washburn, Appl. Phys. Lett. 67, 2699 (1995). 179. D. E. Lacklison, J. W. Orton, I. Harrison, T. S. Cheng, L. C. Jenkins, C. T. Foxon, and S. E. Hooper, / Appl. Phys. 78, 1838 (1995). 180. J. Menniger, U. Jahn, O. Brandt, H. Yang, and K. Ploog, Phys. Rev. B 53, 1881 (1996). 181. H. Yang, O. Brandt, M. Wassermeier, J. Behrend, H. P Schonherr, and K. H. Ploog, Appl. Phys. Lett. 68, 244 (1996). 182. S. A. Nikishin, V. G. Antipov, S. S. Ruvimov, G A. Seryogin, and H. Temkin, Appl. Phys. Lett. 69, 3227 (1996). 183. O. Brandt, H. Yang, A. Trampert, M. Wassermeier, and K. H. Ploog, Appl. Phys. Lett. 71, 473 (1997). 184. D. J. As, R Schmilgus, C. Wang, B. Schottker, D. Schikora, and K. Lischka, Appl. Phys. Lett. 70, 1311 (1997). 185. G. Lupke, O. Busch, C. Meyer, H. Kurz, O. Brandt, H. Yang, A. Trampert, K. H. Ploog, and G. Lucovsky, Phys. Rev. B 57, 3722 (1998). 186. G. Mirjalili, T. J. Parker, S. F Shayesteh, M. M. Bulbul, S. R. P Smith, T. S. Cheng, and C. T. Foxon, Phys. Rev. B 57, 4656 (1998). 187. S. Shokhovets, R. Goldhahn, V. Cimalla, T. S. Cheng, and C. T. Foxon, J. Appl. Phys. 84, 1561 (1998). 188. U. Kohler, D. J. As, B. Schottker, T. Frey, K. Lischka, J. Scheiner, S. Shokhovets, and R. Goldhahn, / Appl. Phys. 85, 404 (1999). 189. J. H. Li, H. Chen, L. C. Cai, S. F Cui, W. X. Yu, J. M. Zhou, Q. Huang, Z. H. Mai, W. L. Zheng, and Q. J. Jia, Appl. Phys. Lett. 74, 2981 (1999). 190. K. Das and D. K. Ferry, Solid State Electron. 19, 851 (1976). 191. H. Yang, O. Brandt, and K. Ploog, Phys. Status Solidi B 194, 109 (1996). 192. A. Gassmann, T. Suski, N. Newman, C. Kisielowski, E. Jones, E. R. Weber, Z. Liliental-Weber, M. D. Rubin, H. I. Helava, I. Grzegory, M. Bockowski, J. Jun, and S. Porowski, J Appl. Phys. 80, 2195 (1996).
140
ANNAMRAJU KASI VISWANATH
193. K. P. Korona, A. Wysmolek, K. Pakula, R. Stepniewski, J. M. Baranowski, I. Grzegory, B. Lucznik, M. Wroblewski, and S. Porowski, Appl. Phys. Lett. 69, 788 (1996). 194. K. Pakula, A. Wysmolek, K. P. Korona, J. M. Baranowski, R. Stepniewski, I. Grzegory, M. Bockowski, J. Jun, S. Krukowski, M. Wroblewski, and S. Porowski, Solid State Commun. 97, 919 (1996). 195. S. Kurai, T. Abe, Y. Naoi, and S. Sakai, Jpn. J. Appl. Phys. 35, 1637 (1996). 196. K. Naniwae, S. Itoh, H. Amano, K. Itoh, K. Hiramatsu, and I. Akasaki, / Cryst. Growth 99, 381 (1990). 197. B. D. Joyce and J. A. Bradley, Nature 195, 458 (1962). 198. A. Sakai, Appl. Phys. Lett. 71, 2259 (1997). 199. D. Kapolnek, S. Keller, R. Vetury, R. D. Underwood, P Kozodoy, S. R Den Baars, and U. K. Mishra, Appl. Phys. Lett. 71, 1204 (1997). 200. D. Kapolnek, X. H. Wu, B. Keying, S. Keller, B. P Keller, U. K. Mishra, S. P Den Baars, and J. S. Speck, Appl. Phys. Lett. 67, 1541 (1995). 201. T. S. Zheleva, O. H. Nam, M. D. Bremser, and R. F. Davis, Appl. Phys. Lett. 71, 2472 (1997). 202. O. H. Nam, M. D. Bremser, T. S. Zheleva, and R. F. Davis, Appl. Phys. Lett. 71, 2638 (1997). 203. A. Usui, H. Sunakawa, A. Sakai, and A. A. Yamaguchi, Jpn. J. Appl. Phys. 36, 899 (1997). 204. A. Sakai, H. Sunakawa, and A. Usui, Appl. Phys. Lett. 71, 2259 (1997). 205. Y Kawaguchi, Y Honda, H. Matsushima, M. Yamaguchi, K. Hiramatsu, and N. Sawaki, Jpn. J Appl. Phys. 37, L966 (1998). 206. J. A. Freitas, Jr., O. H. Nam, R. F. Davis, G. V. Saparin, and S. K. Obyden, Appl. Phys. Lett. 72, 2990 (1998). 207. Y Kawaguchi, S. Nambu, H. Sone, T. Shibata, H. Matsushima, M. Yamaguchi, H. Miyake, K. Hiramatsu, and N. Sawaki, Jpn. J Appl. Phys. 37, L845 (1998). 208. J. Wang, S. Tattori, H. Sato, M. S. Hao, Y Ishikawa, T. Sugahara, K. Yamashita, and S. Sakai, Jpn. J Appl. Phys. 37, 4475 (1998). 209. J. Park, P A. Grudowski, C. J. Eiting, and R. D. Dupuis, Appl. Phys. Lett. 73, 333 (1998). 210. X. Li, G. Bishop, and J. J. Coleman, Appl. Phys. Lett. 73, 1179 (1998). 211. F. Bertram, T. Riemann, J. Christen, A. Kaschner, A. Hoffmann, C. Thomsen, K. Hiramatsu, T. Shibata, and N. Sawaki, Appl. Phys. Lett. 74, 359 (1999). 212. P Kung, D. Walker, M. Hamilton, J. Diaz, and M. Razeghi, Appl. Phys. Lett. 74, 570 (1999). 213. K. C. Zeng, J. Y Lin, H. X. Jiang, and W. Yang, Appl. Phys. Lett. 74, 1227 (1999). 214. S. J. Rosner, G. Girolami, H. Marchand, P T. Fini, J. P Ibbeston, L. Zhao, S. Keller, U. K. Mishra, S. P Den Baars, and J. S. Speck, Appl. Phys. Lett. 74, 2035 (1999). 215. T. S. Zheleva, W. M. Ashmawi, O. H. Nam, and R. F. Davis, Appl. Phys. Lett. 74, 2492 (1999). 216. M. Hao, S. Mahanty, T. Sugahara, Y Morishina, H. Takenaka, J. Wang, S. Tottori, K. Nishino, Y Naoi, and S. Sakai, J Appl. Phys. 85, 6497 (1999). 217. N. P Kobayashi, J. T. Kobayashi, X. Zhang, P D. Dapkus, and D. H. Rich, Appl. Phys. Lett. 74, 2836 (1999). 218. Q. K. K. Liu, A. Hoffmann, H. Siegle, A. Kachner, C. Thomsen, J. Christen, and F. Bertram, Appl. Phys. Lett. 74, 3122 (1999). 219. A. Kaschner, A. Hoffmann, C. Thomsen, F. Bertram, T Riemann, J. Christen, K. Hiramatsu, T. Shibata, and N. Sawaki, Appl. Phys. Lett. 74, 3320 (1999). 220. A. J. Fischer, W. Shan, G. H. Park, J. J. Song, D. S. Kim, D. S. Ye, R. Homing, and B. Goldenberg, Phys. Rev. B 56, 1077 (1997). 221. S. Pau, J. Kuhl, F. Scholz, V. Haerle, M. A. Khan, and C. J. Sun, Appl. Phys. Lett. 72, 557 (1998). 222. S. Pau, J. Kuhl, F. Scholz, V. Haerle, M. A. Khan, and C. J. Sun, Phys. Rev. B 56, R12,718 (1997). 223. R. Zimmermann, A. Euteneuer, J. Mobius, D. Weber, M. R. Hofmann, W. W Ruble, E. O. Gobel, B. K. Meyer, H. Amano, and I. Akasaki, Phys. Rev B 56, R12,722 (1997).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
141
224. T. J. Schmidt, J. J. Song, J. C. Chang, R. Homing, and B. Goldenberg, Appl. Phys. Lett. 72, 1504 (1998). 225. H. Haag, P. Gilliot, R. Levy, B. Honerlage, O. Briot, S. Rufenach-Clur, and R. L. Aulombard, Appl Phys. Lett. 74, 1436 (1999). 226. H. Haag, P Gilliot, R. Levy, B. Honerlage, O. Briot, S. Rufenach-Clur, and R. L. Aulombard, Phys. Rev. B 59, 2254 (1999). 227. T. Aoki, G. Mohs, M. Kuwata-Gonokami, and A. A. Yamaguchi, Phys. Rev. Lett. 82, 3108 (1999). 228. H. Ye, G. W. Wicks, and P M. Fauchet, Appl. Phys. Lett. 74, 711 (1999). 229. D. N. Hahn, G. T. Kiehne, J. B. Ketterson, G. K. L. Wong, P Kung, A. Saxler, and M. Razeghi, J. Appl. Phys. 85, 2497 (1999). 230. Y Narukawa, Y. Kawakami, M. Funato, S. Fujita, S. Fujita, and S. Nakamura, Appl. Phys. Lett. 70, 981 (1997). 231. Y Toyozawa, Prog. Theor. Phys. 20, 53 (1958); / Phys. Chem. Solids 25, 59 (1964). 232. B. Segall and G. D. Mahan, Phys. Rev. 171, 935 (1968). 233. B. Segall, Phys. Rev 150, 734 (1966); 163, 769 (1967); G E. Hite, D. T. F Marple, M. Aven, and B. Segall, Phys. Rev 156, 850 (1967). 234. B. segall, in "Proceedings of the IX Conference on the Physics of Semiconductors" (S. M. Ryvkin, Ed.), p. 425. Nauka, Leningrad, 1968. 235. S. Rudin, T. L. Reinecke, and B. Segall, Phys. Rev B 42, 11,218 (1990). 236. M. Ilegems, R. Dingle, and R. A. Logan, J. Appl. Phys. 43, 3797 (1972). 237. T. Azuhata, T. Sota, K. Suziki, and S. Nakamura, J. Phys.: Condens. Matter 7, L129 (1995). 238. A. S. Barker, Jr. and M. Ilegems, Phys. Rev B 7, 743 (1973). 239. A. J. Fischer, D. S. Kim, J. Hays, W. Shan, J. J. Song, D. B. Eason, J. Ren, J. F. Schetzina, H. Luo, J. K. Furdyna, Z. Q. Zhu, T. Yao, J. F Klem, and W. Schafer, Phys. Rev. Lett. 73, 2368 (1994). 240. R. Hellman, M. Koch, J. Feldman, S. T. Cundiff, E. O. Gobel, D. R. Yakovlev, A. Waag, and G. Landwehr, Phys. Rev B 48, 2847 (1999). 241. T. Li, A. J. Lozykowski, and J. L. Reno, Phys. Rev B 46, 6961 (1992). 242. M. O. Neil, M. Oestreich, W W. Ruble, and D. E. Ashenford, Phys. Rev B 48, 8980 (1993). 243. D. Lee, A. M. Johnson, J. E. Zucker, R. D. Feldman, and R. F. Austin, J. Appl. Phys. 69, 6722 (1991). 244. N. T. Pelekanos, J. Ding, M. Hagerott, A. V. Nurmikko, H. Luo, N. Samarth, and J. K. Furdyna, Phys. Rev B 45, 6037 (1992). 245. D. S. Chemla, D. A. B. Miller, P W. Smith, A. C. Gossard, and W. Wiegmann, IEEE J. Quantum Electron. 20, 265 (1984). 246. D. A. B. Miller, D. S. Chemla, D. J. Eilenberger, P W. Smith, A. C. Gossard, and W. T. Tsang, Appl. Phys. Lett. 41, 679 (1982). 247. Y S. Huang, A. Qiang, R H. Pollak, G D. Pettit, P D. Kirchner, and L. B. Sorensen, /. Appl. Phys. 70, 7537 (1991). 248. J. Lee, E. S. Koteles, and M. O. Vassell, Phys. Rev B 33, 5512 (1986). 249. A. Tookey, G Brown, B. Vogele, D. Bain, I. J. Blewett, A. K. Kar, I. Galbraith, K. A. Prior, B. C. Cavenett, and B. S. Wherrett, "Proceedings of the International Quantum Electronics Conference," paper QWE 2. Optical Society of America, San Francisco, 1998. 250. P. Borri, W. Langbein, J. M. Hvam, and F. Martelli, "Proceedings of the International Quantum Electronics Conference," paper QWE 3. Optical Society of America, San Francisco, 1998. 251. W. Shan, X. C. Xie, J. J. Song, and B. Goldenberg, Appl. Phys. Lett. 67, 2512 (1995). 252. C. I. Harris, B. Monemar, H. Amano, and I. Akasaki, Appl. Phys. Lett. 67, 840 (1995). 253. Y Kawakami, Z. G Peng, Y Narukawa, S. Fujita, S. Fujita, and S. Nakamura, Appl. Phys. Lett. 69, 1414 (1996). 254. J. S. Im, A. Moritz, F Steuber, V. Harle, F Scholz, and A. Hangleiter, Appl. Phys. Lett. 70, 631 (1997).
142
ANNAMRAJU KASI VISWANATH
255. S. Pau, J. Kuhl, M. A. Khan, and C. J. Sun, Phys. Rev. B 58, 12,916 (1998). 256. S. Pau, Z. X. Liu, J Kuhl, J. Ringling, H. T. Grahn, M. A. Khan, C. J. Sun, O. Ambacher, and M. Stutzman, Phys. Rev. B 57, 7066 (1998). 257. M. Godlewski, J. P. Bergman, B. Monemar, U. Rossner, and A. Barski, Appl Phys. Lett. 69, 2089 (1996). 258. L. Eckey, J. Ch. Hoist, R Maxim, R. Heitz, A. Hoffman, I. Broser, B. K. Meyer, C. Wetzel, E. N. Mokhov, and P G. Baranov, Appl. Phys. Lett. 68, 415 (1996). 259. O. Brandt, B. Yang, H. J. Wunsche, U. Jahn, J. Ringhng, G. Paris, H. T. Grahn, and K. H. Ploog, Phys. Rev. B 58, R 13,407 (1998). 260. B. Monemar, J. P. Bergman, I. G. Ivanov, J. M. Baranowski, K. Pakula, I. Grzegory, and S. Porowski, Solid State Commun. 104, 205 (1997). 261. R. Klann, O. Brandt, H. Yang, H. T. Grahn, and K. H. Ploog, Appl. Phys. Lett. 70, 1808 (1997). 262. R. Klann, O. Brandt, H. Yang, H. T. Grahn, K. Ploog, and A. Trampert, Phys. Rev. B 52, Rl 1,615 (1995). 263. D. Yonder Linde, J. Kuhl, and E. Rosengart, J. Lumin. 24/25, 675 (1981). 264. D. Rosen, A. G. Doukas, Y. Budansky, A. Katz, and R. R. Alfano, Appl. Phys. Lett. 39, 935 (1981). 265. R. Strobel, R. Eccleston, J. Kuhl, and K. Kohler, Phys. Rev. B 43, 12,564 (1991). 266. V. Emiliani, S. Ceccherini, F. Bogani, M. Colocci, A. Frova, and S. S. Shi, Phys. Rev. B 56, 4807 (1997). 267. M. Jorgensen and J. M. Hvam, Appl. Phys. Lett. 43, 460 (1983). 268. M. B. Johnson, T. C. McGill, and A. T. Hunter, J. Appl. Phys. 63, 2077 (1988). 269. A. M. de Paula, J. F Ryan, H. J. W. Eakin, M. Tatham, R. A. Taylor, and A. J. Turberfield, / Lumin. 59, 303 (1994). 270. J. L. A. Chilla, O. Buccafusca, and J. J. Rocca, Phys. Rev. B 48, 14,347 (1993). 271. H. J. W. Eakin and J. F Ryan, J. Lumin. 40/41, 553 (1998). 272. A. von Lehmen and T. M. Ballantyne, Appl. Phys. Lett. 44, 87 (1983). 273. J. I. Pankove, E. A. Miller, and J. E. Berkeyheiser, RCA Rev. 32, 383 (1971). 274. J. I. Pankove, E. A. Miller, D. Richman, and J. E. Berkeyheiser, J. Lumin. 4, 63 (1971). 275. J. I. Pankove, E. A. Miller, and J. E. Berkeyheiser, J. Lumin. 5, 84 (1972). 276. J. I. Pankove, E. A. Miller, and J. E. Berkeyheiser, J. Lumin. 6, 54 (1973). 277. H. P Maruska, D. A. Stevenson, and J. I. Pankove, Appl. Phys. Lett. 22, 303 (1973). 278. H. P Maruska, W. C. Rhines, and D. A. Stevenson, Mater. Res. Bull. 7, 777 (1972). 279. A. Shintani and S. Minagawa, J. Electrochem. Soc. 123, 1725 (1976). 280. M. Ilegems, R. Dingle, and R. A. Logan, / Appl. Phys. 43, 3797 (1972). 281. M. Ilegems and R. Dingle, J. Appl. Phys. 44, 4234 (1973). 282. O. Lagerstedt and B. Monemar, J. Appl. Phys. 45, 2266 (1974). 283. B. Monemar, O. Lagerstedt, and H. P Gislason, / Appl. Phys. 51, 625 (1980). 284. B. Monemar, H. P Gislason, and O. Lagerstedt, / Appl. Phys. 51, 640 (1980). 285. P Bergman, G. Ying, B. Monemar, and P O. Holtz, J. Appl. Phys. 61, 4589 (1987). 286. J. L Pankove, J. E. Berkeyheiser, and E. A. Miller, / Appl. Phys. 45, 1280 (1974); J. I. Pankove and J. E. Berkeyheiser, J. Appl. Phys. 45, 3892 (1974). 287. J. I. Pankove and J. A. Hutchby, J. Appl. Phys. 47, 5387 (1976). 288. M. Boulou, M. Furtado, G. Jacob, and D. Bois, J. Lumin. 18/19, 767 (1979). 289. G. Jacob, M. Boulou, and M. Furtado, J. Cryst. Growth 42, 136 (1977). 290. J. L Pankove, M. T. Duffy, E. A. Miller, and J. E. Berkeyheiser, J. Lumin. 8, 89 (1973). 291. T. Ogino and M. Aoki, Jpn. J. Appl. Phys. 19, 2395 (1980). 292. E. Ejder and H. G. Grimmeiss, J. Appl. Phys. 5, 275 (1974). 293. L Akasaki, H. Amano, M. Kito, and K. Hiramatsu, J. Lumin. 48/49, 666 (1991). 294. S. Nakamura, N. Iwasa, M. Senoh, and T. Mukai, Jpn. J. Appl. Phys. 31, L1258 (1992). 295. H. Amano, M. Kitoh, K. Hiramatsu, and I. Akasaki, J. Electrochem. Soc. 137, 1639 (1990).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
143
296. B. Goldenberg, J. D. Zook, and R. J. Ulmer, Appl. Phys. Lett. 62, 381 (1993). 297. T. Tanaka, A. Watanabe, H. Amano, Y. Kobayashi, I. Akasaki, S. Yamazaki, and M. Koike, Appl. Phys. Lett. 65, 593 (1994). 298. W. Gotz, N. M. Johnson, J. Walker, D. P. Bour, H. Amano, and I. Akasaki, Appl. Phys. Lett. 67, 2666 (1995). 299. M. A. Khan, Q. Chen, R. A. Skogman, and J. N. Kuznia, Appl. Phys. Lett. 66, 2046 (1995). 300. C. Johnson, J. Y Lin, H. X. Jiang, M. A. Khan, and C. J. Sun, Appl Phys. Lett. 68, 1808 (1996). 301. G. Yi, and B. W. Wessels, Appl. Phys. Lett. 70, 357 (1997). 302. X. Li, S. Q. Gu, E. E. Reuter, J. T. Verdeyn, S. G. Bishop, and J. J. Coleman, J. Appl. Phys. 80, 2687 (1996). 303. W. Gotz, N. M. Johnson, D. P Bour, M. D. McCelukey, and E. E. Haller, Appl. Phys. Lett. 69, 3725 (1996). 304. W. Gotz, N. M. Johnson, and D. P Bour, Appl. Phys. Lett. 68, 3470 (1996). 305. J. W. Huang, T. F. Kuech, H. Lu, and L Bhat, Appl. Phys. Lett. 69, 2392 (1996). 306. W. Gotz, N. M. Johnson, J. Walker, D. P Bour, and R. A. Street, Appl. Phys. Lett. 68, 667 (1996). 307. J. C. Zolper, M. H. Crawford, A. J. Howard, J. Ramer, and S. D. Hersee, Appl. Phys. Lett. 68, 200 (1996). 308. P Hacke, H. Nakayama, T. Detchprohm, K. Hiramatsu, and N. Sawaki, Appl. Phys. Lett. 68, 1362 (1996). 309. S. J. Pearton, J. W. Lee, and C. Yuan, Appl. Phys. Lett. 68, 2690 (1996). 310. H. Nakayama, P. Hacke, M. R. H. Khan, T. Detchprohm, K. Hiramatsu, and N. Sawaki, Jpn. J. Appl. Phys. 35, L282 (1996). 311. G. Yi and B. W. Wessels, Appl. Phys. Lett. 68, 3769 (1996). 312. U. Kaufmann, M. Kunzer, M. Maier, H. Obloh, A. Ramakrishnan, B. Santic, and P. Schlotter, Appl. Phys. Lett. 72, 1326 (1998). 313. L. Sugiura, M. Suziki, and J. Nishio, Appl. Phys. Lett. 72, 1748 (1998). 314. D. A. Tumbull, X. Li, S. Q. Gu, E. E. Reuter, J. J. Coleman, and S. G. Bishop, J. Appl. Phys. 80, 4609 (1996). 315. E. R. Glaser, T. A. Kennedy, K. Doverspike, L. B. Rowland, D. K. Gaskill, J. A. Freitas, Jr., M. A. Khan, D. T. Olson, J. N. Kuznia, and D. K. Wickenden, Phys. Rev. B 51, 13,326 (1995). 316. C. Wang and R. F. Davis, Appl. Phys. Lett. 63, 990 (1993). 317. M. E. Lin, G Xue, G. L. Zhou, J. E. Greene, and H. Morkoc, Appl Phys. Lett. 63, 932 (1993). 318. M. S. Brandt, J. W. Ager, W. Gotz, N. M. Johnson, J. S. Harris, Jr., R. J. Molnar, and T. D. Moustakas, Phys. Rev. B 49, 14,758 (1994). 319. M. Rubin, N. Newman, J. S. Chan, T. C. Fu, and J. T. Ross, Appl Phys. Lett. 64, 64 (1994). 320. M. S. Brandt, N. M. Johnson, R. J. Molnar, R. Singh, and T. D. Moustakas, Appl Phys. Lett. 64, 2264 (1994). 321. K. S. Stevens, M. Kinniburgh, and R. Beresford, Appl Phys. Lett. 66, 3518 (1995). 322. W. C. Hughes, W. H. Rowland, Jr., M. A. L. Johnson, S. Fujita, J. W. Cook, Jr., J. F. Schetzina, J. Ren, and J. A. Edmond, J. Vac. ScL Technol, B 13, 1571 (1995). 323. A. Botchkarev, A. Salvador, B. Sverdlov, J. Myoung, and H. Morkoc, / Appl Phys. 11, 4455 (1995). 324. R. J. Molnar, R. Singh, and T. D. Moustakas, Appl Phys. Lett. 66, 268 (1995). 325. J. M. Myoung, K. H. Shim, C. Kim, O. Glinschenkov, K. Kim, S. Kim, D. A. Tumbull, and S. G Bishop, Appl Phys. Lett. 69, 2722 (1996). 326. W. Kim, A. Salvador, A. E. Botchkarev, O. Aktas, S. N. Mohammad, and H. Morkoc, Appl Phys. Lett. 69, 559 (1996). 327. R. P Vaudo, L D. Goepfert, T. D. Moustakas, D. M. Beyea, T. J. Frey, and K. Meehan, J. Appl Phys. 79, 2779 (1996). 328. S. Guha, N. A. Bokerczuk, and D. W. Kisker, Appl Phys. Lett. 69, 2879 (1996).
144
ANNAMRAJU KASI VISWANATH
329. J. Z. Li, J. Y. Lin, H. X. Jiang , S. Salvador, A. Botchkarev, and H. Morkoc, AppL Phys. Lett. 69, 1474 (1996). 330. S. C. Y Tsen, D. J. Smith, K. T. Tsen, W. Kim, and H. Morkoc, J. AppL Phys. 82, 6008 (1997). 331. G. Popovici, G. Y Xu, A. Botchkarev, W. Kim, H. Tang, A. Salvador, H. Morkoc, R. Strange, and J. O. White, J. Appl. Phys. 82, 4020 (1997). 332. S. Guha, N. A. Bojarczuk, and F. Cardone, Appl. Phys. Lett. 71, 1685 (1997). 333. C. R. Abemathy, J. D. Mackenzie, S. J. Pearton, and W. S. Hobson, Appl. Phys. Lett. 66, 1969 (1995). 334. A. Salvador, W. Kim, O. Aktas, A. Botchkarev, Z. Fan, and H. Morkoc, Appl. Phys. Lett. 69, 2692 (1996). 335. J. L Pankove and J. A. Hutchby, / Appl. Phys. 47, 5387 (1976). 336. R. G. Wilson, S. J. Pearton, C. R. Abemathy, and J. M. Zavada, Appl. Phys. Lett. 66, 2238 (1995). 337. S. J. Pearton, C. B. Vartuli, J. C. Zolper, C. Yuan, and R. A. Stall, Appl. Phys. Lett. 67, 1435 (1995). 338. J. C. Zolper, R. G. Wilson, S. J. Pearton, and R. A. Stall, Appl. Phys. Lett. 68, 1945 (1996). 339. J. W Lee, S. J. Pearton, J. C. Zolper, and R. A. Stall, Appl. Phys. Lett. 68, 2102 (1996). 340. J. C. Zolper, R. J. Shul, A. G. Baca, R. G. Wilson, S. J. Pearton, and R. A. Stall, Appl. Phys. Lett. 68, 2273 (1996). 341. J. S. Chan, N. W. Cheung, L. Schloss, E. Jones, W. S. Wong, N. Newman, X. Liu, E. R. Weber, A. Gassman, and M. D. Rubin, Appl. Phys. Lett. 68, 2702 (1996). 342. T. Suski, J. Jun, M. Leszczynski, H. Teisseyre, S. Strite, A. Rockett, A. Pelzmann, M. Kamp, and K. J. Ebeling, J. Appl. Phys. 84, 1155 (1998). 343. S. Nakamura, T. Mukai , and M. Senoh, Jpn. J. Appl. Phys. 30, L1998 (1991). 344. Y. Kuga, T. Shirai, M. Haruiyama, H. Kawanishi, and Y Suematsu, Jpn. J. Appl. Phys. 34, 4085 (1995). 345. C. J. Sun, J. W. Yang, B. W. Lim, Q. Chen, M. Z. Anwar, M. A. Khan, A. Osinsky, H. Temkin, and J. R Schetzina, Appl. Phys. Lett. 70, 1444 (1997). 346. M. A. Khan, S. Krishnankutty, R. A. Skogman, J. N. Kuznia, and D. T. Olson, Appl. Phys. Lett. 65, 520 (1994). 347. A. S. Zubrilov, V. I. Nikolaev, D. Tsvetkov, V. A. Dmitriev, K. G. Irvine, J. A. Edmond, and C. H. Carter, Appl. Phys. Lett. 67, 533 (1995). 348. M. A. Khan, J. Kuznia, D. T. Olson, M. Blasinghame, and A. R. Bhattarai, Appl. Phys. Lett. 63, 2455 (1993). 349. D. Walker, A. Saxler, P Kung, X. Zhang, M. Hamitton, J. Diza, and M. Razeghi, Appl. Phys. Lett. 72, 3303 (1998). 350. J. B. Fedison, T. R Chow, H. Lu, and I. B. Bhat, Appl. Phys. Lett. 72, 2841 (1998). 351. J. Z. Li, J. Y Lin, H. X. Jiang, and M. A. Khan, Appl. Phys. Lett. 72, 2868 (1998). 352. A. Y Polyakov, N. B. Smimov, A. V. Govorkov, M. Shin, M. Skowronski, and D. W. Greve, J. Appl. Phys. 84, 870 (1998). 353. E. Calleja, M. A. Sanchez-Garcia, D. Basak, F J. Sanchez, F Calle, P Youinou, E. Munoz, J. J. Serrano, J. M. Blanco, C. Villar, T. Laine, J. Oila, K. Saariner, P Hautojarvi, C. H. Molloy, D. J. Somerford, and I. Harrison, Phys. Rev. B 58, 1550 (1998). 354. M. W. Wang, J. O. McCaldin, J. F Swenberg, T. C. McGill, and R. J. Haunstein, Appl. Phys. Lett. 66, 1974 (1995). 355. E. C. Piqnette, Z. Z. Bandic, J. O. McCaldin, and T. C. McGill, /. Vac. Sci. Technol., B 15, 1148(1997). 356. A. K. Viswanath, E. J. Shin, J. I. Lee, S. Yu, D. Kim, B. Kim, Y Choi, and C. H. Hong, /. Appl. Phys. 83, 2272 (1998). 357. J. J. Hopfield, D. G. Thomas, and M. Gershenzon, Phys. Rev. Utt. 10, 162 (1963).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
145
358. M. Smith, G. D. Chen, J. Y. Lin, H. X. Jiang, A. Salvador, B. N. Sverdlov, A. Botchkarev, H. M. Morkoc, and B. Goldenberg, Appl. Phys. Lett. 68, 1883 (1996). 359. S. H. Wei and A. Zunger, Phys. Rev. B 37, 8958 (1988). 360. S. Strite, Jpn. J. Appl Phys. 33, L699 (1994). 361. J. Neugebauer and C. G. Van de Walle, Phys. Rev. B 50, 8067 (1994). 362. J. Neugebauer and C. G. Van de Walle, Appl. Phys. Lett. 68, 1829 (1996). 363. J. Neugebauer and C. G. Van de Walle, Appl. Phys. Lett. 69, 503 (1996). 364. J. Neugebauer and C. G. Van de Walle, Phys. Rev. Lett. 75, 4452 (1995). 365. C. G. Van de Walle, Phys. Rev B 56, R10,020 (1997). 366. J. Neugebauer and C. G. Van de Walle, J. Appl. Phys. 85, 3003 (1999). 367. P. Boguslawski, E. L. Briggs, and J. Bemholc, Phys. Rev B 51, 17,255 (1995). 368. P Boguslawski, E. L. Briggs, and J. Bemholc, Appl. Phys. Lett. 69, 233 (1996). 369. P Boguslawski and J. Bemholc, Phys. Rev. B 56, 9496 (1997). 370. F. Mireles and S. E. Ulloa, Phys. Rev. B 58, 3879 (1998). 371. J. A. Van Vechten, J. D. Zook, R. D. Homig, and B. Goldenberg, Jpn. J. Appl. Phys. 31, 3662 (1992). 372. J. I. Pankove and N. M. Johnson, Eds., "Hydrogen in Semiconductors." Academic Press, Boston, 1991. 373. D. J. Chadi, Appl. Phys. Lett. 71, 2970 (1997). 374. U. Kaufmann, M. Kunzer, M. Maier, H. Obloh, A. Ramakrishnan, B. Santic, and P Schlotter, Appl. Phys. Lett. 72, 1326 (1998). 375. J. K. Sheu, Y. K. Su, G. C. Chi, B. J. Pong, C. Y Chen, C. N. Huang, and W. C. Chen, J. Appl. Phys. 84, 4590 (1998). 376. X. Zhang, S. J. Chua, P Li, K. B. Chong, and W. Wang, Appl. Phys. Lett. 73, 1772 (1998). 377. L. Sugiura, M. Suziki, and J. Nishio, Appl. Phys. Lett. 72, 1748 (1998). 378. L. Eckey, U. Von Gfug, J. Hoist, A. Hoffmann, A. Kaschner, H. Siegle, C. Thomsen, B. Schineller, K. Heime, M. Heuken, O. Schon, and R. Beccard, J. Appl. Phys. 84, 5828 (1998). 379. T. Suski, J. Jun, M. Leszczynski, H. Teisseyre, S. Strite, A. Rockett, A. Pelzmann, M. Kamp, and K. J. Ebeling, J. Appl Phys. 84, 1155 (1998). 380. B. J. Pong, C. J. Pan, Y C. Teng, G. C. Chi, W. H. Li, K. C. Lee, and C. H. Lee, /. Appl Phys. 83, 5992 (1998). 381. C. Ronning, E. P Carlson, D. B. Thomson, and R. F. Davis, Appl Phys. Lett. 73, 1622 (1998). 382. E. Oh, H. Park, and Y Park, Appl Phys. Lett. 72, 70 (1998). 383. R. Zhang and T. F Kuech, Appl Phys. Lett. 72, 1611 (1998). 384. J. Jayapalan, B. J. Skromme, R. P Vaudo, and V. M. Phanse, Appl Phys. Lett. 13, 1188 (1998). 385. T. S. Cheng, S. V. Novikov, C. T. Foxon, and J. W. Orton, Solid State Commun. 109, 439 (1999). 386. J. L Pankove, J. T. Torvik, C. H. Qiu, I. Grzegory, S. Porowski, P Quigley, and B. Martin, Appl Phys. Lett. 74, 416 (1999). 387. D. H. Youn, M. Lachab, M. Hao, T. Sugahara, H. Takenaka, Y Naoi, and S. Sakai, Jpn. J Appl Phys. 38, 631 (1999). 388. M. H. Zaldivar, P Femandez, J. Piqueras, and J. Solis, J Appl Phys. 85, 1120 (1999). 389. M. A. Reshchikov, G C. Yi, and B. W. Wessels, Phys. Rev. B 59, 13,176 (1999). 390. D. J. As, T. Simonsmeier, B. Schottker, T. Frey, D. Schikora, W. Kriegseis, W. Burkhardt, and B. K. Meyer, Appl Phys. Lett. 73, 1835 (1998). 391. H. P Mamska and J. J. Tietjen, Appl Phys. Lett. 15, 327 (1969). 392. M. Ilegems and M. C. Montgomery, J Phys. Chem. Solids. 34, 885 (1973). 393. B. Monemar and O. Lagerstedt, J. Appl Phys. 50, 480 (1979). 394. T. L. Tansley and R. J. Egan, Phys. Rev. B 45, 10,942 (1992). 395. D. W. Jenkins, J. D. Dow, and M. H. Tsai, J. Appl Phys. 72, 4130 (1992). 396. P Boguslawski, E. L. Briggs, and J. Bemholc, Phys. Rev B 5\, 17,255 (1995).
146
ANNAMRAJU KASI VISWANATH
397. J. I. Pankove, in "Non-Stoichiometry in Semiconductors." Symposia Proceedings of the International Conference on Advanced Materials (K. J. Bachmann, H. L. Hwang, and C. Schwab, Eds.), p. 143. North-Holland, Amsterdam, 1992. 398. W. Siefert, R. Franzheld, E. Butter, H. Sobotta, and V. Riede, Cryst. Res. Technol 18, 383 (1983). 399. B. C. Chung and M. Gershenzon, / Appl Phys. 72, 651 (1992). 400. I. Gorczyca, A. Svane, and N. Christensen, Solid State Commun. 101, 747 (1997). 401. T. Mattila and R. M. Nieminen, Phys. Rev. B 55, 9571 (1997). 402. J. B. Xia, K. W. Cheah, X. L. Wang, D. Z. Sun, and M. Y. Kong Phys. Rev. B 59, 10,119 (1999). 403. F. A. Roboredo and S. T. Pantelides, Phys. Rev Lett. 82, 1887 (1999). 404. D. W. Jenkins and J. D. Dow, Phys. Rev B 39, 3317 (1989). 405. C. G. Van de Walle, Phys. Rev. B 57, R2033 (1998). 406. W. E. Carlos, J. A. Freitas, Jr., M. A. Khan, D. T. Olson, and J. N. Kuznia, Phys. Rev. B 48, 17,878 (1993). 407. M. A. Khan, D. T. Olson, J. N. Kuznia, W. E. Carlos, and J. A. Freitas, Jr., / Appl. Phys. 74, 5901 (1993). 408. W. E. Carlos, J. A. Freitas, Jr., M. A. Khan, D. T. Olson, and J. N. Kuznia, Mater. Res. Forum 143, 99 (1994). 409. W. E. Carlos, J. A. Freitas, Jr., E. R. Glaser, T. A. Kennedy, and M. A. Khan, Inst. Phys. Conf. Sen 5, 443 (1993). 410. E. R. Glaser, T. A. Kennedy, H. C. Crookham, J. A. Freitas, Jr., M. A. Khan, D. T. Olson, and J. N. Kuznia, Appl. Phys. Lett. 63, 2673 (1993). 411. M. Kunzer, U. Kaufmann, K. Maier, J. Schneider, N. Herres, I. Akasaki, and H. Amano, Mater Sci. Forum 143, 87 (1994). 412. E. R. Glaser, T. A. Kennedy, J. A. Freitas, Jr., M. A. Khan, D. T. Olson, and J. N. Kuznia, Inst. Phys. Conf. Ser 5, 453 (1993). 413. C. Blozdog, H. Przybylinska, G. D. Watkins, V. Harle, F Scholz, M. Mayer, M. Kamp, R. J. Molnar, A. E. Wickenden, D. D. Koleske, and R. L. Henry, Phys. Rev. B 59, 12,479 (1999). 414. G. Denninger, R. Beerhalter, D. Reiser, K. Maier, J. Schneider, T. Detchprohm, and K. Hiramatsu, Solid State Commun. 99, 347 (1996). 415. W. J. Moore, J. A. Freitas, Jr., and R. J. Molnar, Phys. Rev. B 56, 12,073 (1997). 416. C. Wetzel, W Walukiewicz, E. E. Haller, J. Ager, III, I. Grzegory, S. Porowski, and T. Suski, Phys. Rev. B 53, 1322 (1996). 417. P. Perlin, T. Suski, H. Teisseyre, M. Leszczynski, I. Grzegory, J. Jun, S. Porowski, P. Boguslawski, J. Bemholc, J. C. Chevrin, A. Polian, and T. D. Moustakas, Phys. Rev. Lett. 75, 296 (1995). 418. B. K. Meyer, D. Volm, A. Graber, H. C. Alt, T. Detchprohm, A. Amano, and I. Akasaki, Sold State Commun. 95, 597 (1995). 419. P. Perlin, E. Litwin-Staszewska, B. Suchanek, W. Knapp, J. Camassel, T. Suski, R. Piotrzkowski, I. Grzegory, S. Porowski, E. Kaminska, and J. C. Chevrin, Appl. Phys. Lett. 68, 1114(1996). 420. W. Gotz, N. M. Johnson, C. Chen, H. Lieu, C. Kuo, and W. Imler, App. Phys. Lett. 68, 3144 (1996). 421. C. F. Lin, G. C. Chi, M. S. Feng, J. D. Guo, J. S. Tsang, and J. M. Hong, Appl. Phys. Lett. 68, 3758 (1996). 422. W T. Gotz, L. T. Romano, B. S. Krusor, N. M. Johnson, and R. J. Molnar, Appl. Phys. Lett. 69, 242 (1996). 423. D. C. Look and R. J. Molnar, Appl. Phys. Lett. 70, 3377 (1997). 424. G. Y. Zhang, Y Z. long, Z. T. Yang, S. X. Jing, J. Li, and Z. Z. Gan, Appl. Phys. Lett. 71, 3376 (1997).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
147
425. N. Kaneda, T. Detchprohm, K. Hiramatsu, and N. Sawaki, Jpn. J. Appl. Phys. 35, L468 (1996). 426. P. Hacke, A. Maekawa, N. Koide, K. Hiramatsu, and N. Sawaki, J. Appl, Phys. 33, 6443 (1994). 427. W. Krystek, F. Pollak, Z. C. Feng, M. Schurman, and R. A. Stall, Appl. Phys. Lett. 72, 1353 (1998). 428. M. Hirsch, J. A. Wolk, W. Walukiewicz, and E. E. Haller, Appl Phys. Lett. 71, 1098 (1997). 429. C. H. Qiu and J. I. Pankove, Appl. Phys. Lett. 70, 1983 (1997). 430. P. Hacke, P. Ramvall, S. Tanaka, Y. Aoyagi, A. Kuramatta, K. Hoirino, and H. Munekata, Appl Phys. Lett. 74, 543 (1999). 431. M. Toth, K. Fleischer, and M. R. Phillips, Phys. Rev. B 59, 1575 (1999). 432. W. M. Chen, I. A. Buyanova, M. T. Wagner, B. Monemar, J. L. Lindstrom, H. Amano, and I. Akasaki, Phys. Rev. B 58, R13,351 (1998). 433. I. A. Buyanova, J. P. Bergman, B. Monemar, H. Amano, I. Akasaki, A. Wysmolek, P. Lomiak, J. M. Baranowski, K. Pakula, R. Stepniewski, K. P. Korona, I. Grzegory, M. Bockowski, and S. Porowski, Solid State Commun. 105, 497 (1998). 434. J. J. Hopfield, Phys. Rev 112, 1555 (1958); Y. Toyozawa, Prog. Theor. Phys. Suppl 12, 111 (1959). 435. S. Permogorov, in "Excitons" (E. I. Rashba and M. D. Sturge, Eds.), p. 177. North-Holland, Amsterdam, 1982. 436. E. Gross, S. Permogorov, and B. S. Razbirin, J. Phys. Chem. Solids 27, 1647 (1966). 437. A. A. Klochikhin, S. A. Permogorov, and A. N. Reznitsky, Sov Phys. Solid State 18, 1304 (1976). 438. T. Steiner, M. L. W. Thewalt, E. S. Koteles, and J. P Salaemo, Phys. Rev. B 34, 1006 (1986). 439. S. Ruvimov, Z. Liliental-Weber, T. Suski, J. W. Ager, III, J. Washburn, J. Krueger, C. Kisielowski, E. R. Weber, H. Amano, and I. Akasaki, Appl Phys. Lett. 69, 990 (1996). 440. D. C. Look, D. C. Reynolds, W Kim, O. Aktas, A. Botchkarev, A. Salvador, and H. Morkoc, J. Appl Phys. 80, 2960 (1996). 441. J. C. Zolper, H. H. Tan, J. S. WiUiams, J. Zou, D. J. H. Cockayne, S. J. Pearton, M. H. Crawford, and R. F Karlicek, Jr., Appl Phys. Lett. 70, 2729 (1997). 442. E. F Schubert, I. D. Goepfert, W. Grieshaber, and J. M. Redwing, Appl Phys. Lett. 71, 921 (1997). 443. G Beadie, W. S. Rabinovich, A. E. Wickenden, D. D. Koleske, S. C. Binari, and J. A. Freitas, Jr., Appl Phys. Lett. 71, 1092 (1997). 444. E. Iliopoulous, D. Doppalapudi, H. M. Ng, and T. D. Moustakas, Appl Phys. Lett. 73, 375 (1998). 445. Z. Q. Fang, J. W. Hemsky, D. C. Look, and M. P Mack, Appl Phys. Lett. 72, 448 (1998). 446. X. Zhang, S. J. Chua, W. Liu, and K. B. Chong, Appl Phys. Lett. 72, 1890 (1998). 447. I. H. Lee, J. J. Lee, P Kung, F J. Sanchez, and M. Razeghi, Appl Phys. Lett. 74, 102 (1999). 448. M. H. Zaldivar, P Fernandez, and J. Piqueras, J. Appl Phys. 83, 462 (1998). 449. E. Oh, H. Park, and Y Park, Appl Phys. Lett. 72, 1848 (1998). 450. V. A. Joshkin, C. A. Parker, S. M. Bedair, L. V. Krasnobaev, J. J. Cuomo, R. F Davis, and A. Suvkhanov, Appl Phys. Lett. 72, 2838 (1998). 451. R. Singh, R. J. Molnar, M. S. Unlu, and T. D. Moustakas, Appl Phys. Lett. 64, 336 (1994). 452. D. M. Hofmann, D. Kovalev, G Steude, B. K. Meyer, A. Hoffmann, L. Eckey, R. Heitz, T. Detchprohm, H. Amano, and I. Akasaki, Phys. Rev. B 52, 16,702 (1995). 453. T. Suski, P. Perlin, H. Teisseyre, M. Leszczynski, I. Grzegory, J. Jun, M. Bockowski, S. Porowski, and T. D. Moustakas, Appl Phys. Lett. 67, 2188 (1995). 454. F K. Koschnick, K. Michael, J. M. Spaeth, B. Beaumont, and P Gibart, Phys. Rev. B 54, Rl 1,042 (1996). 455. W. Grieshaber, E. F Schubert, I. D. Goepfert, R. F Karlicek, Jr., M. J. Schurman, and C. Iran, J. Appl Phys. 80, 4615 (1996).
148
ANNAMRAJU KASI VISWANATH
456. A. Billeb, W. Grieshaber, E. F. Schubert, and R. F. Karlicek, Jr., Appl Phys. Lett. 70, 2790 (1997). 457. P. Hacke and H. Okushi, Appl. Phys. Lett. 71, 524 (1997). 458. H. Liu, J. G. Kim, M. H. Ludwig, and R. M. Park, Appl. Phys. Lett. 71, 347 (1997). 459. M. Ramsteiner, A. J. Menniger, O. Brandt, H. Yang, and K. H. Ploog, Appl. Phys. Lett. 69, 1276 (1996). 460. H. Siegle, I. Loa, P. Thurian, L. Eckey, A. Hoffmann, I. Broser, and C. Thomsen, Appl. Phys. Lett. 70, 909 (1997). 461. E. F Schubert, I. D. Goepfert, and J. M. Redwing, Appl. Phys. Lett. 71, 3224 (1997). 462. D. C. Reynolds, D. C. Look, B. Jogai, and H. Morkoc, Solid State Commun. 101, 643 (1997). 463. H. M. Chen, Y. F. Chen, M. C. Lee, and M. S. Feng, Phys. Rev. B 56, 6942 (1997). 464. E. Calleja, F. J. Sanchez, D. Basak, M. A. Sanchez-Garcia, E. Munoz, L Izpura, F. Calle, J. M. G. Tijero, J. L. Sanchez- Rojas, B. Beaumont, P. Lorenzini, and P. Gibart, Phys. Rev. B 55, 4689 (1997). 465. J. Eisner, R. Jones, M. L Heggie, P. K. Sitch, M. Haugk, Th. Frauenheim, S. Oberg, and P R. Briddon, Phys. Rev. B 58, 12,571 (1998). 466. M. Godlewski, E. M. Goldys, M. R. Phillips, R. Langer, and A. Barski, Appl. Phys. Lett. 73, 3686 (1998). 467. L. W. Tu, Y. C. Lee, D. Stocker, and E. F Schubert, Phys. Rev. B 58, 10,696 (1998). 468. C. V. Reddy, K. Balakrishnan, H. Okumura, and S. Yoshida, Appl. Phys. Lett. 73, 244 (1998). 469. L. W. Tu, Y C. Lee, S. J. Chen, L Lo, D. Stocker, and E. F Schubert, Appl. Phys. Lett. 73, 2802 (1998). 470. S. J. Rhee, S. Kim, E. E. Reuter, S. G. Bishop, and R. J. Molnar, Appl. Phys. Lett. 73, 2636 (1998). 471. E. R. Glaser, T. A. Kennedy, W. E. Carlos, J. A. Freitas, Jr., A. E. Wickenden, and D. Koleske, Phys. Rev. B 57, 8957 (1998). 472. G. Li, S. J. Chua, S. J. Xu, W. Wang, P Li, B. Beaumont, and P Gibart, Appl. Phys. Lett. 14, 2821 (1999). 473. C. K. Sun, Y L. Huang, S. Keller, U. K. Mishra, and S. P Den Baars, Phys. Rev. B 59, 13,535 (1999). 474. R. A. Mair, J. Li, S. K. Duan, J. Y Lin, and H. X. Jiang, Appl. Phys. Lett. 74, 513 (1999). 475. R. Dingle, K. L. Shaklee, R. F Leheny, and R. B. Zetterstrom, Appl. Phys. Lett. 19, 5 (1971). 476. M. A. Khan, D. T. Olson, J. M. Van Hove, and J. N. Kuznia, Appl. Phys. Lett. 58, 1515 (1991). 477. H. Amano, T. Asahi, and L Akasaki, Jpn. J. Appl. Phys. 29, L205 (1990). 478. H. Amano, N. Watanabe, N. Koide, and L Akasaki, Jpn. J. Appl. Phys. 32, LIOOO (1993). 479. S. T. Kim, H. Amano, I. Akasaki, and K. Noide, Appl. Phys. Lett. 64, 1535 (1994). 480. X. Zhang, P Kung, A. Saxler, D. Walker, and M. Razeghi, J. Appl. Phys. 80, 6544 (1996). 481. J. M. Redwing, D. A. S. Loeber, N. G. Anderson, M. A. Tischler, and J. S. Flynn, Appl. Phys. Lett. 69, 1 (1996). 482. S. Kurai, Y. Naoi, T. Abe, S. Ohmi, and S. Sakai, Jpn. J Appl. Phys. 35, L77 (1996). 483. T. Tanaka, K. Uchida, A. Watanabe, and S. Minagawa, Appl. Phys. Lett. 68, 976 (1996). 484. X. H. Yang, T. Schmidt, W. Shan, J. J. Song, and B. Goldenberg, Appl. Phys. Lett. 66, 1 (1995). 485. T. J. Schmidt, X. H. Yang, W Shan, J. J. Song, A. Salvador, W. Kim, O. Aktas, A. Botchkarev, and H. Morkoc, Appl. Phys. Lett. 68, 1820 (1996). 486. A. Nakadaira and H. Tanaka, Appl. Phys. Lett. 71, 812 (1997). 487. O. Glushenkov, J. M. Myoung, K. H. Shim, K. Kim, Z. G. Figen, J. Gao, and J. G. Eden, Appl. Phys. Lett. 70, 811 (1997). 488. R. Klan, O. Brandt, H. Yang, H. T. Grahn, and K. H. Ploog, Appl. Phys. Lett. 70, 1076 (1997). 489. S. Bidnyk, T. J. Schmidt, G. H. Park, and J. J. Song, Appl. Phys. Lett. 71, 729 (1997). 490. K. Funato, F. Nakamura, S. Hashimoto, and M. Ikeda, Jpn. J Appl. Phys. 37, L1023 (1998). 491. Y Ohba and H. Yoshida, Jpn. J Appl. Phys. 37, L905 (1998).
3
GROWTH AND OPTICAL PROPERTIES OF GAN
149
492. H. Yang, L. X. Zheng, J. B. Li, X. J. Wang, D. P. Xu, Y. T. Wang, X. W. Hu, and P. D. Han, Appl. Phys. Lett. 74, 2498 (1999). 493. S. Hess, R. A. Taylor, J. F. Ryan, B. Beaumont, and R Gibart, Appl Phys. Lett. 73, 199 (1998). 494. N. Grandjean, J. Massies, M. Leroux, and R Lorenzini, Appl. Phys. Lett. 72, 82 (1998). 495. J. Wu, H. Yaguchi, K. Onabe, and Y. Shiraki, Appl. Phys. Lett. 73, 1931 (1998). 496. J. Han, M. H. Crawford, R. J. Shul, J. J. Figiel, M. Banas, L. Zhang, Y. K. Song, H. Zhou, and A. V. Nurmikko, Appl. Phys. Lett. 73, 1688 (1998). 497. S. Bidnyk, B. D. Little, Y. H. Cho, J. Krasinski, J. J. Song, W. Yang, and S. A. McPherson, Appl. Phys. Lett. 73, 2242 (1998). 498. S. Guha and N. A. Bojarczuk, Appl. Phys. Lett. 72, 415 (1998). 499. J. Hoist, A. Hoffmann, I. Broser, B. Schottker, D. J. As, D. Schikora, and K. Lischka, Appl. Phys. Lett. 74, 1966 (1999). 500. H. Cao, Y. G. Zhao, S. T. Ho, E. W. Seelig, Q. H. Wang, and R. R H. Chang, Phys. Rev. Lett. 82, 2278 (1999). 501. M. A. Khan, R. A. Skogman, J. M. van Hove, S. Krishnankutty, and R. M. Kolbas, Appl. Phys. Lett. 56, 1257 (1990). 502. A. Salvador, G Liu, W. Kim, O. Aktas, A. Botchkarev, and H. Morkoc, Appl. Phys. Lett. 67, 3322 (1995). 503. M. Smith, J. Y Lin, H. X. Jiang, A. Salvador, A. Botchkarev, W. Kim, and H. Morkoc, Appl. Phys. Lett. 69, 2453 (1996). 504. R. Cingolani, G Coli, R. Rinaldi, L. Calcagnile, H. Tang, A. Botchkarev, W. Kim, A. Salvador, and H. Morkoc, Phys. Rev. B 56, 1491 (1997). 505. R. A. Mair, K. C. Zeng, J. Y Lin, H. X. Jiang, B. Zhang, L. Dai, H. Tang, A. Botchkarev, W Kim, and H. Morkoc, Appl. Phys. Lett. 71, 2898 (1997). 506. A. Niwar, T. Ohtoshi, and T Kuroda, Appl. Phys. Lett. 70, 2159 (1997). 507. W. Shan, S. Xu, B. D. Little, X. C. Xie, J. J. Song, G E. Bulman, H. S. Kong, M. T. Leonard, and S. Krishnankutty, / Appl. Phys. 82, 3158 (1997). 508. K. C. Zeng, J. Y Lin, H. X. Jiang, A. Salvador, G Popovici, H. Tang, W. Kim, and H. Morkoc, Appl. Phys. Lett. 71, 1368 (1997). 509. D. Behr, R. Niebuhr, J. Wagner, K. H. Bachem, and U. Kaufmann, Appl. Phys. Lett. 70, 363 (1997). 510. R. A. Mair, K. C. Zeng, J. Y Lin, H. X. Jiang, and B. Zhang, Appl. Phys. Lett. 72, 1530 (1998). 511. N. Grandjean and J. Massies, Appl. Phys. Lett. 73, 1260 (1998). 512. N. A. Zakhleniuk, C. R. Bennet, B. K. Ridley, and M. Babiker, Appl. Phys. Lett. 73, 2485 (1998). 513. K. C. Zeng, R. Mair, J. Y Lin, H. X. Jiang, W. W. Chow, A. Botchkarev, and H. Morkoc, Appl. Phys. Lett. 73, 2476 (1998). 514. M. Leroux, N. Grandjean, M. Laugt, J. Massies, B. Gil, P. Lefebvre, and P. Bigenwald, Phys. Rev. B 58, R13,371 (1998). 515. N. Suziki and N. lizuka, Jpn. J. Appl. Phys. 37, L369 (1998). 516. J. S. Im, H. Kollmer, J. Off, A. Sohmer, F. Scholz, and A. Hangleiter, Phys. Rev B 57, R9435 (1998). 517. R Lefebvre, J. Allegre, B. Gil, A. Kavokine, H. Mathieu, W. Kim, A. Salvador, A. Botchkarev, and H. Morkoc, Phys. Rev B 57, R9447 (1998). 518. N. Suziki and N. lizuka, Jpn. J. Appl. Phys. 38, L363 (1999). 519. N. Grandjean, J. Massies, and M. Leroux, Appl. Phys. Lett. 74, 2361 (1999). 520. T. J. Ochalski, B. Gil, P. Lefebvre, N. Grandjean, J. Massies, and M. Leroux, Solid State Commun. 109, 567 (1999). 521. B. Gil, P. Lefebvre, J. Allegre, H. Mathieu, N. Grandjean, M. Leroux, J. Massies, P. Bigenwald, and R Christol, Phys. Rev B 59, 10,246 (1999).
150
ANNAMRAJU KASI VISWANATH
Conclusions Photoluminescence and time-resolved spectroscopy of GaN materials and devices have been reviewed in this article. Most of the work on material growth and device development was done using MOCVD techniques. However, other techniques such as MBE were also tried. To date, MOCVD methods have provided better materials and optoelectronic devices compared to MBE. Similarly, sapphire has been the substrate most commonly employed. In this review, we also quoted work done with all the other possible substrates. Development of large area single-crystal growth of GaN may provide the ultimate solution to the problem of lattice mismatch. The literature on Si and GaAs substratebased III-V nitrides has been quite small. There is a lot of room to improve these technologies for the obvious reason that GaN technology can be integrated with the most important and well established semiconductor technologies of Si and GaAs. Most of the long-standing problems in GaN have been solved by the highly commendable achievements of two Japanese groups led by Amano and Akasaki, and Nakamura. Amano and Akasaki made a major breakthrough in achieving p-type GaN, which can be made by a very easy process. It must be remembered that the development of blue lasers based on ZnSe could not take off, because very good quality p-type ZnSe was a major problem. Later, Nakamura was very successful in reaching a series of milestones that made possible the realization of blue lasers and what not! In conclusion, the future of nitrides looks extremely bright. We hope in the future we will have many more devices, such as UV lasers and nonlinear photonic switches, based on III-IV nitrides. Acknowledgments I thank the brain pool program of South Korea for allowing me the opportunity to work in South Korea for three years. My original work that is discussed in this article was done at the National Creative Research Initiative Center for Ultrafast Optics Control at the Korean Research Institute of Standards and Science (KRISS) in Taejon. I thank Dr. Dongho Kim for providing the finest laser laboratory to investigate optical phenomena. I also thank a number of collaborators, Dr. Joo In Lee, Dr. Sunkyu Yu, Dr. Eunjoo Shin, and Dr. S. C. Jeoung, of KRISS. I am very grateful to many scientists who have provided a number of samples of GaN. Thanks are owed to Dr. Chang Hee Hong, who was the group leader of the GaN project at L. G. Electronics, and is presently at the Department of Physics and National Semiconductor Research Centre, Jeoungbook National University, Korea, for giving me the first batch of samples that I worked with. I also thank Dr. Yoonho Choi, the present group leader of the GaN project, and Dr. Baeyoung Kim of L. G. Electronics. I thank C. R. Lee and J. Y. Leem of KRISS and S. T. Kim and A. G. Lee of Chungbuk National University, Korea, for their cooperation. I also thank my wife Mrs. Annapuma Viswanath and son Mr. Srinivas for their understanding and support during the course of experimental investigations as well as during the write up of this review.
SEMICONDUCTORS AND SEMIMETALS, VOL. 73
CHAPTER
4
SiGe/Si Processing D. Y, C, Lie COMMUNICATIONS RESEARCH AND DEVELOPMENT CENTER ( C R D C ) , IBM MICROELECTRONICS, ENCINITAS, CALIFORNIA, USA
K. L. Wang DEPARTMENT OF ELECTRICAL ENGINEERING, UNIVERSITY OF CALIFORNIA, LOS ANGELES, CALIFORNIA, USA
1. INTRODUCTION
151
2. SIGE/SI MATERIAL PROPERTIES AND PROCESSING CHALLENGES
2.1. 2.2. 2.3. 2.4.
Si/SiGe Hetewstructures: Lattice Mismatch and Bandgap Engineering Materials Growth Characterization Techniques for Si/SiGe Heterostructures General Processing Challenges to the Fabrication of Si/SiGe Devices
REFERENCES
153
. .
. . .
153 157 164 174 192
1.
Introduction
The worldwide integrated circuits (ICs) market has reached more than a hundred bilUon dollars per year, where the majority of the IC chips sold are fabricated using silicon as the substrate material [1, 2]. The technology of silicon ICs, however, is approaching fundamental limits set by the atomic nature of matter. One cannot count on doubling the chip capacity by shrinking the size of Si complementary metal-oxide-semiconductor (CMOS) devices forever, and many people expect that the Moore law will be invalid in 10-15 years. In the meantime, rapidly growing telecommunications industries are driving the development of reliable and economical high-frequency devices with lower-power dissipation. Si/Sij.^Ge^ heterostructures are, therefore, under extensive study because they can provide adjustable bandgaps and improved carrier mobilities compared with Si homostructures [3]. Heterojunction bipolar transistors (HBTs) that utilize Si/SiGe heterolayers promise a very impressive extension of the high-frequency limit of Si-based bipolar technology to cutoff frequencies well above 100 GHz, a frequency range that, so far, has been dominated by GaAsbased devices [4, 5]. Modulation-doped field-effect transistors (MODFETs) that 151 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved. ISBN 0-12-752182-8 ISSN 0080-8784/01 $35.00
152
D. Y. C. LIE AND K. L. WANG
employ SiGe as the channel layer have also shown considerable improvement in both speed and gain over their Si counterparts [6, 7]. High-sensitivity photodetectors made with Si/SiGe have also been fabricated [8]. Several quantum size effects in strained SiGe layers and their potential in device applications were reviewed by Karunasiri and Wang [9]. In the near future, Si/SiGe heterojunction and superlattice-based devices may even play an important role in the integration of complex electronic circuitry with optoelectronic functionality on a single IC chip. For example, growing GaAs on high-quality SiGe buffer layers on Si may combine laser diodes made of III-V materials with Si ICs [10]. Tensilely strained Si can also be grown on these SiGe buffer layers and improved surface-channel devices have been demonstrated [11]. Si/SiGe heterostructures have, therefore, opened up a new, exciting avenue of research opportunities. Fabrication processes for Si/SiGe devices are rather compatible with those routinely used for Si ICs, and this is a major advantage for Si/SiGe over III-V compounds. This compatibility ensures the continued use of the existing multibillion-dollar Si IC fabrication facilities for the manufacture of Si/SiGe devices, which makes SiGe more cost effective than GaAs for technological evolution. However, SiGe has its problems too. One of the most important problems in fabricating Si/SiGe devices is associated with the thermal stability of the heterostructures under various processing steps, such as ion implantation and postimplant annealing [10, 12, 13]. The intrinsic strain in a Si/SiGe heterostructure is, therefore, both a blessing and a curse. The presence of intrinsic pseudomorphic strain changes the band structure of Si/SiGe and can enhance the mobility of carriers. However, if strain relaxation takes place during annealing, unwanted defects are introduced and the performance of Si/SiGe devices will be degraded considerably [14, 15]. To make millions of Si/SiGe transistors on a planar IC chip, one has to be able to process the heterostructure with great precision in a repeatable fashion and with a very low defect density level. In general, defects generated by advanced deep-submicrometer wafer processing can act as nucleation sites for dislocation formation, which is shown to enhance the undesired strain relaxation of SiGe [16]. High-temperature thermal treatment of Si/SiGe >950 °C can also introduce very significant interdiffusion of Ge and Si [3]. Due to the problems of interdiffusion and strain relaxation, it is evident that processing of Si/SiGe devices requires a substantial decrease in the thermal budget over devices made of Si, which can make the manufacture of these devices difficult. It is also challenging to design and control the exact Ge profiles in the devices. The epitaxial growth, etching, isolation, and salicidation modules required for Si/SiGe devices may also need to be redeveloped and reoptimized. All of these materials properties and processing issues are covered in Section 2. There are many exciting and novel devices that can be built with Si/SiGe heterostructures, but, in our opinions, by far the most important device of all is the Si/SiGe HBT. Numerous new circuits and products just announced in the past few years use Si/SiGe HBTs. We discuss in detail the device physics and the
4
SIGE/SI PROCESSING
153
design optimization for this important device in Section 3, where other devices such as Si/SiGe metal-oxide-semiconductor field-effect transistors (MOSFETs) also are discussed. We provide some insight on what Si/SiGe heterostructures can do to benefit the successful scaling of devices for the IC industry. Whereas the device characteristics need to be accurately modeled for circuit designers to use, we include in Section 3 a discussion on the device modeling issues for Si/SiGe devices, with an emphasis on rf device modeling. Limitations on the SPICE-Gummel-Pooh model and some rf testing issues are addressed. The reliability issues for the Si/SiGe devices also are discussed briefly in Section 3. After the topics on Si/SiGe materials properties, processing, device physics, and compact modeling are discussed, in Section 4 we are able to show the readers the exact advantages of using Si/SiGe for advanced IC designs, particularly for rf communication circuits and high-speed ICs. A few circuit examples are given to illustrate the points more clearly. We then discuss the existing products that use the Si/SiGe technology and share some examples for potential Si/SiGe system-on-a-chip applications. In this way we hope the readers can better appreciate the real impact of the Si/SiGe technology. This review chapter is an up to date and comprehensive treatment on how and why the exciting Si/SiGe technology is so important to the advancement of the entire IC industry. From the basic discussions on the materials properties to real-life Si/SiGe circuits and products, we believe that no other single publication of this ambitious scope currently exists in the literature. Hence, the one major reason for the format and content of this review chapter is to cover and explain the most important characteristics of this Si/SiGe technology and its applications. It was extremely challenging to make this chapter comprehensive and up to date because many new publications on this subject are appearing at a very fast pace. However, it is certainly exciting and rewarding to see this technology, which roughly started about 25 years ago from work in research labs, finally come to life and make some real breakthroughs with viable industrial products. In a few years, the impact of Si/SiGe technology probably will be felt even more in our daily lives if the "Dick Tracy" type of wrist watch can be successfully realized in Si/SiGe technology.
2. 2.1.
SiGe/Si Material Properties and Processing Challenges
SI/SIGE HETEROSTRUCTURES: LATTICE MISMATCH AND BANDGAP ENGINEERING
The concept of bandgap engineering has been around for a long time. Realization of these heterostructures for electronic and photonic device applications requires both (1) a detailed understanding of the science of heteroepitaxy and
154
D. Y. C. LIE AND K. L. WANG
(2) precise atomic-layer control during thin-film growth. These are made possible by modem technological advances such as molecular beam epitaxy (MBE) and ultra-high-vaccum chemical-vapor deposition (UHV-CVD). Heteroepitaxy is defined as the epitaxial growth of dissimilar materials upon each other. A good example here is growing Ge on a Si substrate, where a very large (4.17%) lattice mismatch exists. As will be discussed more later, this large lattice mismatch between the two dissimilar materials means that the Ge atoms in epi-Ge film are under a large amount of strain. The SiGe epilayer can therefore be fully elastically strained, partially relaxed, or fully relaxed. The exact strain state of the epifilm depends on the film thickness, growth conditions, postgrowth processing temperatures, and so forth. Whereas elastically relaxed films are usually associated with relatively large amounts of defects, it is preferable to grow fully strained epi-SiGe on Si for most electronic applications. The lattice mismatch can be designed and tailored by controlling the Ge content in the alloy to satisfy the specific appHcations. If elasticity theory holds, the lattice constant of an ideal Si^.^Ge^ (x < 1) film is defined by linear interpolation as a(Sii_^GeJ = (1 - x) • a(Si) + x • a{Si) where a (Si) and fl(Ge) are the lattice constants of bulk Si and Ge, respectively. This equation is also known as Vegard's law. It is obvious that the higher the Ge content is in the Si^.^Ge^ film, the larger is its lattice constant a{Sii_ficJ. Note that to be exact, a(Si), a(Sii_^Ge^) and fl(Ge) are all functions of temperature, and at room temperature ^(Si) = 5.431 A, whereas ^(Ge) = 5.658 A. Let us conduct a thought experiment next: when the epi-SiGe film (of a larger lattice constant) is placed on top of a Si substrate (of a smaller lattice constant), what happens? Is the epi-SiGe film forced to "compress" to the same lattice constant as the Si substrate? If so, the epilayer is under compressive strain, where the lattice constant of the epilayer in the plane is the same as that of the substrate (see Fig. 1). Hence, according to linear elasticity theory, the lattice cell of the epi-SiGe is distorted and the lattice is stretched and elongated along the vertical (growth) direction. To facilitate subsequent discussions, we need to introduce definitions for some commonly used terms. The total lattice mismatch (/) for any lattice-mismatched heterostructure is defined as f=-^
(1)
where GQ is the lattice constant of the substrate and Gf is the lattice constant of the epifilm material in the unstrained state (i.e., in the bulk state). Note that the lattice mismatch between Ge and Si is 4.17% at room temperature and it increases only slightly with increasing temperature. The in-plane strain e" (also
4
SIGE/SI PROCESSING
155
io(P0o#o^' o o # 6 o^o#
OOOO'CDS' 000(s)00
FIG. 1. A schematic drawing that shows the concept of strained epitaxy. The epi-SiGe film is pseudomorphically grown on a Si substrate; its in-plane lattice constant « | is equal to the lattice constant «o of the unstrained Si bulk substrate. However, the lattice in epi-SiGe is stretched longer along the perpendicular direction (i.e., normal to the growth interface), resulting in a larger perpendicular lattice constant aj > GQ.
called parallel strain or coherency strain) is therefore defined as (2) where a^ is the in-plane lattice constant of the film. We can similarly define the perpendicular strain of the film e^ as
s- = f i ^
(3)
and likewise aj is the perpendicular lattice constant of the film (along the growth direction). If the SiGe epilayer deposited on Si is fully strained, then by definition the in-plane lattice constant of the film needs to be squeezed to match that of the Si substrate. The SiGe cells, therefore, are stretched along the perpendicular direction and it is obvious that aj is not be equal to a^ in this case. The growth of fully strained films like this is called pseudomorphic or commensurate growth, because « | = a^^, but aj ^ a^. In the real world, however, as we deposit a thin epifilm on a given substrate, the in-plane lattice constant of film may not conform to that of the substrate for various reasons. The film, therefore, can be entirely unstrained and still retain its bulk lattice constant (i.e., be fully relaxed). The film can also be partially strained (also called partially relaxed). Therefore, for these elastically relaxed films, there is no one-to-one atomic alignment at the interface between the film and the substrate, and this lattice mismatch has to be accommodated by defects that are called misfit dislocations. The average spacing of misfit dislocation is inversely proportional to the lattice mismatch and can be roughly calculated as 5 ^ a^/f. For the case of deposition of pure Ge films on Si, the lattice
156
D. Y. C. LIE AND K. L. WANG
mismatch / = 4.17% and GQ = 5.431 A, and, therefore, S is calculated to be roughly 130 A. This corresponds to a very high dislocation density that causes serious electron scattering, which reduces electron mobility, because the typical electron mean free path is on the order of 1000 A. These defects can also cause severe junction leakage via pipe diffusion of metallic impurities and so forth. For modem Si processing, the dislocation spacing in the initial Si substrate can be as large as '^ 1 mm. It is, therefore, clear that most practical device applications require the growth of pseudomorphic SiGe films (i.e., fully strained) and limit the dislocation density to a very low level. Kline et al. [17] reported the first electroreflectance measurement data on the electronic band structure for bulk SiosGcos alloys. They found that the bandgaps vary linearly with the Ge content in the alloys. In strained SiGe there are important effects, such as strain-induced sphtting (of the degenerate states) and changes on the effective masses of carriers, that also affect band structure significantly [18, 19]. Figure 2 shows the calculated bandgap of epi-SiGe as a function of Ge content in the film on a Si(lOO) substrate [18]. Because the deformation potentials for Si and Ge are reasonably well known, the deformation potential theory has been used to calculate the perturbed conduction band levels and the strain-induced band splitting [19]. The A conduction band minimum in Si has sixfold degeneracy, and the strain will split the band into a fourfold and a twofold state [19, 20]. The L conduction band minimum for Ge has fourfold degeneracy, and the strain does not split it. For practical applications
20
T—I—I—I—I—r—1—I—r Conduction Bonds
0 5 Si
Ge Froction.
x
Ge
FIG. 2. Calculated valence and conduction bands in strained Si^Ge,_^ films grown on a Si(lOO) substrate. The dashed lines are the weighted averages of the valence bands and the A conduction bands. Adapted from C. G. Van de Walle and R. M. Martin, Phys. Rev. B 34, 5621 (1986) and S. C. Jain and W. Hayes, Semicond. Sci. Technol. 6, 547 (1991).
4
SIGE/SI PROCESSING
157
of pseudomorphic SiGe on Si(lOO), the net result of bandgap reduction in the pseudomorphic SiGe film on Si(lOO) is roughly 7.4 meV for each 1% of Ge in the film, which means LEy = 0.74x
(4)
where x is the Ge content in the film (0 < jc < 1) and the valence bandgap difference is given in electronvolts [21]. Note that Eq. (4) is an approximation, because the conduction band offset is assumed to be zero and the density of states Ny in the epi-SiGe is also assumed to be constant and unchanging with the Ge content in the film. Also note that for a heavily doped emitter, the bandgap-narrowing effects due to heavy doping cannot be neglected; that can be expected to be
KT^)
A£,(Ar,)=. 0.0187 x l n ( ^ ^ - ^ j
(5)
where A^^ is the doping concentration in the emitter [22]. To summarize the points discussed better, one clearly wants to have pseudomorphic growth of the epi-SiGe because (1) significant bandgap reduction (compared with the bulk Si) can be achieved with a strained epi-SiGe due to the splitting of the degenerate conduction and valence bands (this change of band structures also enhances the carrier mobility), (2) the dislocations at the interface for relaxed films could become electrically active defects, causing leakage currents, interface states, and band discontinuity, and (3) the defects associated with the strain relaxation process can act as scattering centers to degrade carrier mobilities [23].
2.2.
MATERIALS GROWTH
Thanks to the considerable research advancements made in material growth (i.e., epitaxy) techniques in the past 25 years, the concept of bandgap engineering can finally be applied to realize Si-based strained heterostructures. Historically, the growth of strained-layer heterostructures was demonstrated very successfully in III-V compound semiconductor systems and compUcated devices (such as laser diodes) were made in the early 1960s. Epi-SiGe films on Si substrates are useful for electronic or photonic device applications only if they can be grown with very low defect density. Therefore, a low-temperature material growth method is attractive to prevent strain relaxation by dislocation nucleation and it also can minimize island formation during growth. In addition to manipulating the growth kinetics, surface cleaning and passivation methods are of paramount importance too. Hydrogen-passivated Si/SiGe interfaces can considerably reduce incorporated the oxygen concentration to minimize interfacial defects. Therefore, low-temperature epitaxial growth techniques
158
D. Y. C. LIE AND K. L. WANG
such as molecular beam epitaxy (MBE) or chemical-vapor deposition (CVD) are useful to realize these pseudomorphically strained films. Rapid-thermal chemical-vapor deposition techniques and ultra-high-vacuum chemical-vapor deposition (UHV-CVD) are particularly attractive, because they help either to limit dislocation nucleation during growth or to maintain an oxygen-deficient surface. 2.2.7. UHV-CVD Meyerson [24, 25] at IBM was the pioneer who successfully grew devicequality Si/SiGe heterostructures using the UHV-CVD technique. The UHV-CVD system uses a load-lock chamber to transfer wafers to a hot-walled isothermal furnace (growth temperature ^^400-650 °C for SiGe). High-purity gas sources commonly seen in semiconductor processing fabrications are also used inside UHV-CVD tools for the deposition and doping of Si/SiGe (i.e., silane, germane, diborane, phosphine, etc.). To prevent the presence of any contaminant on the surface before the growth of epi-SiGe, a conventional Si CVD epitaxy system typically uses a very high-temperature baking process (>1000 °C) to volatize any oxide or organic contaminants that may otherwise stay on the interface and act as defect nucleation sites. The deposition temperature for conventional Si epitaxy is, therefore, usually rather high (i.e., -^800-1000 °C). One drawback of this kind of conventional CVD Si epitaxy is that at such high temperatures, dopant diffusion can be very significant, which would increase junction depths and prevent the scaling of advanced devices. The main reason this UHV-CVD technique can be used to successfully grow SiGe epilayers at much lower temperatures is because surface preparation is done with a standard clean, followed by a dilute HF dip to create a H-passivated surface. In terms of Si surface preparation, UHV-CVD was the first commercial growth method to rely on hydrogen passivation [24, 25]. The surface is then kept in ultra-high vacuum of ^10"^ torr, with very low partial pressures for O2 and H2O to prevent oxidation and recontamination of carbon [26]. The high-purity gases are free of oxygen and water (10 MV/cm can be routinely achieved [146-148]. These almost perfect characteristics of the Si/Si02 system enabled complex planar ICs to be manufactured at low cost and has led to tremendous advances in Si technology over the past 30 years. As we start to process the Si/SiGe system to make heterojunction devices, we find out immediately that direct thermal oxidation of SiGe ends up with Ge piling up at the interface, messing up the interface properties of the oxide with high Djt and oxide charges [149, 150]. Germanium Oxide is also water soluble, which is not ideal for masking. Patton et al. [149], Nayak et al. [75], and LeGoues et al. [150, 151] all studied direct oxidation of strained SiGe and found that the initial oxidation rate of SiGe is better than in Si, and they proposed different mechanisms on the oxidation kinetics (such as catalytic interaction of Ge with H2O and the presence of Ge-reduced interstitial injection during oxidation). Therefore, for practical purposes, direct thermal oxidation of SiGe should always be avoided by placing a Si cap layer on top of the strained
190
D. Y. C. LIE AND K. L. WANG
SiGe layer [152]. An alternative is to prepare the SiOj film by deposition with the help of microwave electron cyclotron resonance plasma processing, rapidthermal oxidation, or high-pressure oxidation to alter the oxidation kinetics to realize excellent oxide interface properties [153-155]. 2.4.4.
Other Practical Process Integration Issues with Si/SiGe
Most of the published Si/SiGe BiCMOS processes adopt the "base = gate" integration scheme, where the base polycrystalline regions of the n-p-n bipolar transistors are patterned at the same time as the MOSFET polycrystalline gate [4, 156]. All devices undergo the same high-temperature rapid-thermal annealing step that activates all dopants [such as after the high-dose extrinsic base linkup implant for HBTs and the gate and Source-to-drain (S-D) implants for MOSFETs]. For this kind of base = gate process, a thin polycrystalline silicon (poly-Si) "protection" layer typically is deposited on the CMOS gate oxide area right after the MOS gate oxidation is done. This poly-Si layer is patterned and removed in the Si/SiGe HBT region and then followed by a low-temperature epitaxy to grow single crystal epibase in the HBT regions and the polycrystalline materials everywhere else [156]. The MOSFET gate stack, therefore, includes both the initial poly-Si protection layer and the polycrystalline material deposited by this epibase growth. This kind of integration approach may work fine with 0.5-fim CMOS processes, but can have problems integrating with the advanced sub-0.25-/im CMOS processes. Due to the significant polycrystalline depletion effects that adversely degrade the device current drive for short-channel MOSFETs, high-temperature rapid-thermal annealing (RTA) typically is required after the S-D implants to activate the dopants. This poly-si depletion issue in nMOS (n^ poly-si) and in pMOS (p^ polycrystals) typically cannot be optimized by a single annealing step. The n+ polycrystals (typically implanted with arsenic or with both arsenic and phosphorus atoms) suffer a more serious polycrystalUne depletion effect and higher annealing temperatures are required to activate the dopants than for the p"^ polycrystals. However, at high RTA temperatures, two issues arise for pMOS: (1) B penetration from the p+ polycrystals diffusing through the very thin gate oxide (^30 A) cause large threshold voltage variation, large mobility degradation, and possible local current leakage; (2) B from the S-D p+ implant can diffuse laterally into the channel and shorten the device channel length in a noncontrollable fashion (transientenhanced diffusion), which drastically increases the device off-state leakage current. There is obviously a trade-off in the thermal budget for the sub-O.lS-fim MOSFET design between the polycrystalline depletion issues associated with n"^ gate polycrystals for nMOS and the boron penetration-diffusion problem for pMOS [157]. To reduce the manufacturing and process development costs, a popular approach for IC fabrication is to develop special Si/SiGe BiCMOS processes in
4
SIGE/SI PROCESSING
191
house that can be built on a standard CMOS process developed and acquired from a typical low-cost wafer fabrication foundry. In this case, the required SD annealing temperature and hot time for the Si/SiGe HBTs are obviously not optimized, even though they may have been painstakingly optimized for the advanced sub-0.25-/xm CMOS processes. Note that because the digital CMOS process is mature and all design libraries, circuits, and products were already developed for the foundry's advanced (but low-cost) CMOS process, one certainly would not want to tamper with the hot time of the foundry's CMOS process. Therefore, the device performance for the Si/SiGe HBTs, which are supposed to provide the highest speed for the critical paths on BiCMOS chips, cannot be optimized without changing the performances of the MOSFETs. To resolve this dilemma for a successful integration of Si/SiGe HBT technology with advanced sub-0.25-)Ltm CMOS processes, a practical approach is to adopt the "base after gate" scheme. A recent article from IBM reported some details of this integration method [158]. The base after gate process typically performs the S-D implant and drive-in annealing for the nMOS first, and then proceeds with the HBT process module. After the HBTs are entirely formed, this base after gate process continues with the pMOS S-D implant and annealing, and then concludes with the salicidation and backend metallization processes. Note that many film stacks are actually deposited during the HBT module process, including deposition of the epibase and the emitter polycrystalline material and deposition of the oxide-nitride-polycrystalline layers during the formation of the extrinsic base. One of the most challenging process integration issue with this base after gate approach is complete removal of these films from the MOSFET areas so that the unwanted polycrystalline stringers and/or oxide-nitride spacer residuals do not remain on the MOSFET areas [158]. To enable a very robust Si/SiGe HBT process module, simulations must be performed to optimize strain conditions, process flow parameters, and overall device performance carefully. Initially, establishing a thermal budget range and evaluating its impact on overall process parameters as well as device performance is an important task. Of particular importance is the effect on Ge segregation, SiGe interdiffusion, and strain relaxation. It should be recognized that the decision of the thermal budget is most important during process development because it impacts many critical steps such as gate oxidation, source-drain activation, and device isolation, to name a few [159]. The performance improvement needs to be assessed and compared with bulk Si devices. Isolation and salicide modules may need to be specially developed to accommodate the SiGe layers also. Another important point is that the early work on Si/SiGe HBTs showed that both the collector and the base currents increased significantly when compared with those in Si bipolar junction transistors (BJTs). Undesirable higher base leakage currents of 5-100 times greater than those in Si BJTs were reported [14]. The causes of the high base leakage currents were partly identified as
192
D. Y. C. LIE AND K. L. WANG
being due to the defects at the Si/SiGe interface; later devices no longer exhibit these serious base leakage issues [160]. It is also interesting to note that typically when a pseudomorphically strained SiGe layer is deposited across an entire wafer, the film stack on the active areas (i.e., areas not covered by oxide) consists of (1) an intrinsic Si cap layer, (2) the pseudomorphic SiGe layer (with Ge content ramped or not ramped), and (3) another intrinsic Si layer (or an undoped SiGe strained layer with constant Ge content). This stack lies directly on top of the Si substrate. However, when the epi-SiGe film is deposited over the oxide area, the layer become polycrystalline and becomes the extrinsic base contact of the Si/SiGe HBT and/or the polycrystalline gate electrode of the CMOS device. In addition to the issues described in the preceding text (such as thermal budget, implantation and annealing, and film removal), many other practical process integration issues exist for the Si/SiGe BiCMOS processes. Because these issues are highly dependent on device architectures and tool sets, it is impossible to cover all of them here. Harame et al. [4] nicely reviewed several Si/SiGe device architectures in their article, where both non-self-aligned and self-aligned devices are mentioned (the more practical self-aligned device architectures covered in their article include mesalike, epibase before the epibase, epibase after the epibase, and disposable emitter mandrel). In our opinion, many of the processing integration issues are not too difficult to resolve as long as the device architecture is carefully selected and the process development and transfers to manufacturing are carefully done and planned. Other processing difficulties, such as the need to develop selective poly-SiGe etch with respect to Si or Si02, are not as fundamental as those listed, so they are not discussed further here. The interested reader is referred to [161].
References 1. T. Makimoto, "Symposium on VLSI," 1996, Technical Digest, pp. 6-9. 2. US News and World Report July 31, p. 41 (1995). 3. R. Hull and J. C. Bean, in "Strained-Layer Superlattices: Materials Science and Technology" (T. P. Pearsall, Ed.), Chap. 1, pp. 1-72. Academic Press, London, 1991. 4. D. L. Harame, J. H. Comfort, J. D. Cressler, E. E Crabbe, J. Y.-C. Sun, B. S. Meyerson, and T. Tice, IEEE Trans. Electron Devices 42, 469 (1995). 5. E Schaffler, Solid-State Electron. 37, 765 (1994). 6. Y. J. Mii, Y. H. Xie, E. A. Eitzgerald, D. Monroe, P J. Silverman, J. M. Kuo, A. R. Kortan, E A. Thiel, and B. E. Weir, J. Vac. Sci. TechnoL, B 10, 1807 (1992). 7. U. Konig, A. J. Boers, E Schaffler and E. Kasper, Electron. Lett. 28, 160 (1992). 8. T. P Pearsall, H. Temkin, J. C. Bean, and S. Luryi, IEEE Electron Device Lett. 7, 330 (1986). 9. R. P G. Karunasiri and K. L. Wang, J. Vac. Sci. TechnoL, B 9, 2064 (1991). 10. E. Kasper and E. Schaffler, "Strained-Layer Superlattices: Materials Science and Technology" (T. P Pearsall, Ed.), Vol. 33, Chap. 4. Academic Press, London, 1991. 11. Y H. Xie, E. A. Eitzgerald, D. Monroe, G. P. Watson, and P. J. Silverman, Jpn. J. Appl. Phys. 33, 2372 (1994).
4
SIGE/SI PROCESSING
193
12. M. D. Giles, in "VLSI Technology" (S. M. Sze, Ed.), Chap. 8, p. 370. McGraw-Hill, Singapore, 1988. M. I. Current, I. Yamada, N. W. Cheung, P. L. F. Hemment, and K. J. Reeson, in "Handbook of Ion Implantation Technology" (J. F. Ziegler, Ed.), pp. 363^33. Elsevier Science, Amsterdam, 1992. 13. J. F. Gibbons, Proc. IEEE 60, 1062 (1972). 14. C. A. King, J. L. Hoyt, and J. F. Gibbons, IEEE Trans. Electron Devices 36, 2093 (1989). 15. P S. Peercy, B. W. Dodson, J. Y. Tsao, E. D. Jones, D. R. Myers, T. E. Zipperian, L. R. Dawson, R. M. Biefeld, J. F Klem, and C. R. Hills, IEEE Electron Devices Lett. 9, 621 (1988). 16. R. Hull, J. C. Bean, J. M. Bonar, G S. Higashi, K. T. Short, H. Temkin, and A. E. White, Appl. Phys. Lett 56, 2445 (1990). 17. J. S. Kline, R H. Pollak, and M. J. Cardona, Helv Phys. Acta 41, 968 (1968). 18. M. Gell, Phys. Rev. B 38, 7535 (1988). 19. C. G. Van de Walle and R. M. Martin, Phys. Rev B 34, 5621 (1986). 20. S. C. Jain and W. Hayes, Semicond. Sci. Technol. 6, 547 (1991). 21. R. People and J. C. Bean, Appl. Phys. Lett. 48, 538 (1986). 22. J. Del Alamo, S. Swirhun, and R. M. Swanson, "International Electron Device Meeting," 1985, Technical Digest, p. 290. 23. J. Baslev, Phys. Rev. 143, 636 (1966). 24. B. S. Meyerson, Proc. IEEE 80, 1592 (1992). 25. B. S. Meyerson, Appl. Phys. Lett. 48, 797 (1986). 26. B. S. Meyerson, F J. Himpsel, and K. J. Uram, Appl Phys. Lett. 57, 1034 (1990). 27. S. S. Iyer, in "Epitaxial Silicon Technology" (B. J. Baliga, Ed.), pp. 97-175. Academic Press, New York, 1985. 28. A. Y Cho, Thin Solid Films 100, 291 (1983). 29. Y Ota, / Appl. Phys. 51, 1102 (1980). 30. J. C. Bean and E. A. Sadowski, /. Vac. Sci. Technol. 20, 137 (1982). 31. H. Kibbel, E. Kasper, P Narozny, and H.-U. Schreiber, Thin Solid Films 184, 163 (1990). 32. E. Kasper and J. C. Bean, Eds., "Silicon Molecular Beam Epitaxy." CRC Press, Boca Raton, FL, 1988. 33. K. N. Tu, J. W. Mayer, and L. C. Feldman, "Electronic Thin Film Science." Macmillian, New York, 1992. 34. S. S. Iyer, M. Arienzo, and E. de Fresart, Appl. Phys. 57, 893 (1990). 35. C. A. King, J. L. Hoyt, D. B. Noble, C. M. Gronet, J. F Gibbons, M. P Scott, T. I. Kamins, and S. T. Laderman, IEEE Electron Device Lett. 10, 159 (1989). 36. J. C. Sturm, E. J. Prinz, and C. W. Magee, IEEE Electron Device Lett. 12, 303 (1991). 37. A. Vollmer, Electronic Design, Jan. 23 (1998). 38. M. Hong, E. de Fresart, J. Steele, A. Zlotnicka, C. Stein, G Tam, M. Racanelli, L. Knoch, Y C. See, and K. Evans, IEEE Electron Device Lett. 14, 450 (1993). 39. T. F Meister, H. Schafer, M. Franosch, W. Molzer, K. Aufinger, U. Scheler, C. Walz, M. Stolz, S. Boguth, and J. Bock, "International Electron Device Meeting," 1995, Technical Digest, p. 739. IEEE, New York, 1995. 40. A. Monroy, M. Laurens, M. Marty, D. Dutartre, D. Gloria, J. L. Carbonero, A. Perrotin, M. Roche, and A. Chantre, BCTM, 1999, Technical Digest, p. 121. IEEE, New York, 1999. 41. J. W. Matthews and A. E. Blakeslee, J. Cryst. Growth 27, 118 (1974). 42. J. W. Matthews and A. E. Blakeslee, / Cryst. Growth 29, 273 (1975). 43. Van der Merwe, Surf. Sci. 31, 198 (1972). 44. E. Kasper, Surf. Sci. 174, 630 (1986). 45. J. P Hirth and J. Lothe, "Theory of Dislocations," Chap. 2. McGraw-Hill, New York, 1969. 46. E. Kasper and H. J. Herzog, Thin Solid Films 44, 357 (1977). 47. J. W. Matthews, in "Epitaxial Growth" (J. W. Matthews, Ed.), Chap. 8. Academic Press, New York, 1975.
194
D. Y. C. LIE AND K. L. WANG
48. J. Willis, S. C. Jain, and R. Bullough, Philos. Mag. A 62, 115 (1990). 49. R. People, IEEE J. Quantum Electron. QE-22, 1696 (1986). 50. D. N. Theodore, G. Tarn, J. Whitefield, J. Christiansen, and J. Steele, Mater. Res. Soc. Symp. Proc. 319, 159 (1994). 51. D. N. Theodore and G. Tarn, Microscopy Soc. Amer Proc. 1112 (1993). 52. J. M. Baribeau, T. E. Jackman, D. C. Houghton, R Maigne, and M. W. Denhoff, /. Appl Phys. 63, 5738 (1988). 53. E. A. Fitzgerald, Mater Sci. Rep. 7, 81 (1991). 54. V. S. Sperioso, J. Appl. Phys. 52, 6094 (1981). 55. S. T. Picraux, B. L. Doyle, and J. Y. Tsao, "Strained-Layer Superiattices: Materials Science and Technology" (T. R Pearsall, Ed.), Vol. 33, Chap. 3, pp. 139-222. Academic Press, London, 1991. 56. T. Vreeland, Jr., and B. M. Paine, J. Vac. Sci. Technol, A 4, 3153 (1986). 57. C. Kittel, "Introduction to Solid State Physics," 7th ed. Wiley, New York, 1996. 58. D. A. Neumann, H. Zabel, and H. Morkoc, Mater Res. Soc. Symp. Proc. 37, 47 (1985). 59. C. R. Wie, T. A. Tombrello, and T. Vreeland, Jr., J. Appl. Phys. 59, 3743 (1986). 60. C. J. Tsai, A. Dommann, M.-A. Nicolet, and T. Vreeland, Jr., J. Appl. Phys. 69, 2076 (1991). 61. G. Bai and M.-A. Nicolet, / Appl. Phys. 70, 649 (1991). 62. D. Y. C. Lie, J. H. Song, N. D. Theodore, A. Vantomme, M.-A. Nicolet, T. K. Cams, and K. L. Wang, / Appl. Phys. 11, 2329 (1995). 63. R R Fewster, J. Appl. Crystallogr 25, 714 (1992). 64. B. K. Tanner, "X-ray Diffraction Topography," Pergamon, Oxford, 1976. 65. W. K. Chu, J. W Mayer, and M.-A. Nicolet, "Backscattering Spectrometry," Chap. 8, pp. 223-275. Academic Press, New York, 1978. 66. L. C. Feldman, J. W. Mayer, and S. T. Picraux, "Materials Analysis by Ion Channeling." Academic Press, London, 1982. 67. F Eisen, in "Channeling: Theory, Observation and Applications" (D. V. Morgan, Ed.), Chap. 14, p. 426. Wiley, London, 1973. 68. F Cedeira, A. Pinczuk, J. C. Bean, B. Batlogg, and B. A. Wilson, Appl. Phys. Lett. 45, 1138 (1984). 69. G. Bai and M.-A. Nicolet, J. Appl. Phys. 70, 3551 (1991). 70. J. W Edington, "Practical Electron Microscopy in Materials Science." Philips, Eindhoven, 1974. 71. S. M. Sze, "Physics of Semiconductor Devices," 2nd ed. Wiley, New York, 1981. 72. M. Arienzo, S. S. Iyer, B. S. Meyerson, G. L. Patton, and J. M. C. Stork, Appl. Surf. Sci. 48/49, 377 (1991). 73. V. Venkartaraman, C. W. Liu, and J. C. Sturm, Appl. Phys. Lett. 63, 2795 (1993). 74. J. Welser, J. L. Hoyt, and J. F. Gibbons, "International Electron Device Meeting," 1992, Technical Digest, p. 1000. IEEE, New York, 1992. 75. D. K. Nayak, J. C. S. Woo, J. S. Park, K. L. Wang, and K. R MacWilliams, Appl. Phys. Lett. 62, 2853 (1993). 76. M. Glicksman, Phys. Rev. I l l , 125 (1958). 77. Y C. Chen, S. H. Li, R K. Bhattacharya, J. Singh, and J. M. Hinckley, Appl. Phys. Lett. 64, 3110(1994). 78. G. Busch and O. Vogt, Helv. Phys. Acta 33, 437 (1960). 79. J. M. McGregor, T. Manku, J. R Noel, D. J. Roulston, A. Nathan, and D. C. Houghton, J. Electron. Mater 22, 319 (1993). 80. S. K. Chun and K. L. Wang, IEEE Trans. Electron Devices 39, 2153 (1992). 81. T. Manku and A. Nathan, IEEE Electron Device Lett. 12, 704 (1991). 82. J. M. Hinckley and J. Singh, Phys. Rev. B4\, 2912 (1990). 83. T. K. Cams, S. K. Chun, M. O. Tanner, K. L. Wang, T. I. Kamins, J. E. Tumer, D. Y C. Lie, M.-A. Nicolet, and R. G. Wilson, IEEE Trans. Electron Devices 41, 1273 (1994).
4
SIGE/SI PROCESSING
195
84. K. Hess, "Advanced Theory of Semiconductor Devices," Chaps. 8 and 9. Prentice-Hall, Englewood Cliffs, NJ, 1988. 85. P. M. Asbeck, in "Modem Semiconductor Device Physics" (S. M. Sze, Ed.), Chap. 1, p. 47. Wiley, New York, 1998. 86. D. K. Nayak and S. K. Chun, Appl. Phys. Lett. 64, 2514 (1994). 87. D. Rosenfeld and S. A. Alterovitz, IEEE Trans. Electron Devices 41, 848 (1994). 88. M. M. Rieger and P Vogl, Phys. Rev. B 48, 14276 (1993). 89. J. Gyulai, in "Ion implantation: Science and Technology" (J. F. Ziegler, Ed.), pp. 139-210. Academic Press, Orlando, 1984. 90. L. J. van der Pauw, Philips Tech. Rev 20, 220 (1958). 91. N. G. E. Johansson, J. W. Mayer, and O. J. Marsh, Solid State Electron. 13, 317 (1970). 92. D. Y. C. Lie, J. Electron. Mater. 27, 377 (1998). 93. Fisful, "Heavily Doped Semiconductors," p. 133, Plenum Press, New York, 1969. 94. D. Y C. Lie, N. D. Theodore, and J. H. Song, Appl. Surf. Sci. 92, 557 (1996). 95. D. Y C. Lie, J. H. Song, M.-A. Nicolet, and N. D. Theodore, J. Electron. Mater 25, 87 (1996). 96. D. Y C. Lie, J. H. Song, M.-A. Nicolet, N. D. Theodore, M. O. Tanner, S. Thomas, and K. L. Wang, J. Appl. Phys. 79, 8341 (1996). 97. S. Im, F. Eisen, M.-A. Nicolet, M. O. Tanner, K. L. Wang, and N. D. Theodore, J. Appl. Phys. 81, 1695 (1997). 98. D. Y C. Lie, J. H. Song, M.-A. Nicolet, and N. D. Theodore, Appl. Phys. Lett. 66, 592 (1995). 99. S. Im, D. Y C. Lie, and M.-A. Nicolet, J. Appl. Phys. 79, 7389 (1996). 100. Z. Atzman, M. Eisenberg, P. Revesz, J. W. Mayer, S. Q. Hong, and F. Schaffer, Appl. Phys. Lett. 60, 2243 (1992). 101. L. Eriksson, J. A. Davies, N. G. E. Johassson, and J. W. Mayer, J. Appl. Phys. 40, 842 (1969). 102. J. W. Mayer, L. Eriksson, and J. A Davis, "Ion Implantation in Semiconductors." Academic Press, New York, 1970. 103. G. L. Olson and J. A. Roth, Mater Sci. Rep. 3, 1 (1988). 104. J. S. Williams in "Surface Modification and Alloying" (J. M. Poate and G. Foti, Eds.). Plenum, New York, 1983. 105. E. P. Donovan, F. Spaepen, D. Tumbull, J. M. Poate, and D. C. Jacobson, / Appl. Phys. 57, 1795 (1985). 106. D. C. Paine, J. Miner Metal. Mater Soc. 2, 55 (1993). 107. L. Csepregi, J. W. Mayer, and T W Sigmon, Phys. Lett. A 54, 157 (1975). 108. I. Suni, G. Goltz, M.-A. Nicolet, and S. S. Lau, Thin Solid Film 93, 171 (1982). 109. E Spaepen, Acta Metall. Mater 26, 1167 (1978). 110. T. Saito and I. Ohdomari, Philos. Mag. B 43, 673 (1981). 111. S. S. Lau and W. F. Van der Weg, "Thin Films: Interdiffusion and Reactions" (J. M. Poate, K. N. Tu, and J. W. Mayer, Eds.), Chap. 12. Wiley-Interscience, New York, 1978. 112. P. J. Germain, M. A. Paesler, D. E. Sayers, and K. Zellama, Mater Res. Soc. Symp. Proc. 13, 135 (1983). 113. J. Narayan, J. Appl. Phys. 53, 8607 (1982). 114. I. Suni, G. Goltz, M. G. Grimaldi, M.-A. Nicolet, and S. S. Lau, Appl Phys. Lett. 40, 269 (1982). 115. A. Lietolia, A. Wakita, T W. Sigmon, and J. R Gibbons, / Appl. Phys. 53, 4399 (1982). 116. J. M. Poate and J. S. Williams, "Ion Implantation and Beam Processing," pp. 13-57. Academic Press, New York, 1984. 117. M. J. Aziz, Mater Res. Soc. Symp. Proc. 321, 1, (1994). 118. G.-Q. Lu, E. Nygren and M. J. Aziz, J. Appl. Phys. 70, 5323 (1991). 119. P Kringh0j and R. G. Elliman, Phys. Rev Lett. 73, 858 (1994). 120. D. C. Paine, N. D. Evans, and N. G. Stoffel, J. Appl. Phys. 70, 4278 (1991). 121. S. Q. Hong, Q. Z. Hong, and J. W. Mayer, J. Appl. Phys. 72, 3821 (1993). 122. D. Y C. Lie, N. D. Theodore, J. H. Song, and M.-A. Nicolet, J. Appl. Phys. 11, 5160 (1995).
196
D. Y. C. LIE AND K. L. WANG
123. D. Y. C. Lie, T. K. Cams, N. D. Theodore, F. Eisen, M.-A. Nicolet, and K. L. Wang, Mater. Res. Soc. Symp. Proc. 321, 485 (1994). 124. Q. Z. Hong, J. G. Zhu, J. W. Mayer, W. Xia, and S. S. Lau, J. Appl. Phys. 71, 1768 (1992). 125. C. Lee, T. E. Haynes, and K. S. Jones, Appl. Phys. Lett. 62, 501 (1993). 126. Z. Atzman, M. Eisenberg, Y Shacham-Diamand, J. W. Mayer, and F. Schaffler, J. Appl. Phys. 75, 377 (1994). 127. T. E. Haynes, M. J. Antonell, C. A. Lee, and K. S. Jones, Phys. Rev. B 51, 7762 (1995). 128. B. T. Chilton, B. J. Robinson, D. A. Thompson, T. E. Jackman, and J.-M. Baribeau, Appl Phys. Lett. 54, 2 (1989). 129. M. J. Aziz, P. C. Sabin, and G.-Q. Lu, Phys. Rev. 5 44, 9812 (1991). 130. G.-Q. Lu, E. Nygren, M. J. Aziz, D. Tumbull, and C. W. White, Appl. Phys. Lett. 54, 2583 (1989). 131. J. Fratello, J. F. Hays, and D. Tumbull, J. Appl. Phys. 51, 4718 (1980). 132. D. Y C. Lie, A. Vantomme, R Eisen, T. Vreeland, Jr., M.-A. Nicolet, T. K. Cams, and K. L. Wang, J. Appl. Phys. 14, 6039 (1993). 133. P. Kringh0j, R. G. Elliman, and J. L. Hansen, Mater. Res. Soc. Symp. Proc. 321, 461 (1994). 134. R. G. Elliman, W.-C. Wong, and P Kringh0j, Mater. Res. Soc. Symp. Proc. 321, 375 (1994). 135. D. C. Paine, D. J. Howard, N. G. Stoffel, and J. H. Horton, J. Mater. Res. 5, 1023 (1990). 136. D. C. Paine, D. J. Howard, and N. G. Stoffel, J. Electron. Mater. 20, 735 (1991). 137. D. J. Howard, W. E. Bailey, and D. C. Paine, Appl. Phys. Lett. 63, 2893 (1993). 138. W.-C. Wong and R. G. Elliman, Mater Res. Soc. Symp. Proc. 321, 491 (1994). 139. M. G. Grimaldi, B. M. Paine, M.-A. Nicolet, and D. K. Sadana, J. Appl. Phys. 52, 4038 (1981). 140. D. Y C. Lie, S. Im, and J. H. Song, unpublished data. 141. D. Y C. Lie, N. D. Theodore, J. Steele, and T. C. Smith, U.S. Patent 5,565,690, 1996. 142. D. K. Nayak, K. Kamjoo, J. S. Park, J. C. S. Woo, and K. L. Wang, IEEE Trans. Electron Devices 39, 56 (1992). 143. D. J. Eaglesham, J. M. Poate, D. C. Jacobson, M. Cemllo, L. N. Pfeiffer, and K. West, Appl. Phys. Lett. 58, 523 (1991). 144. J. Klappe, Ph.D. Thesis, University of Twente, Netherlands, 1994. 145. T. O. Sedgwick, Mater Res. Soc. Symp. Proc. 92, 3 (1987). 146. B. E. Deal and A. S. Grove, J. Appl. Phys. 36, 3770 (1965). 147. E. H. Nicollian and J. R. Brews, "MOS Physics and Technology." Wiley, New York, 1982. 148. P Balk, Ed., "The Si-SiOj System." Elsevier, Amsterdam, 1998. 149. G. L. Patton, S. S. Iyer, S. L. Delage, E. Ganin, and R. C. Mcintosh, Mater Res. Soc. Symp. Proc. 102, 295 (1988). 150. F. K. LeGoues, R. Rosenberg, T. Nguyen, F. Himpsel, and B. S. Meyersen, J. Appl. Phys. 65, 1724 (1989). 151. F. K. LeGoues, R. Rosenberg, and B. S. Meyerson, / Appl. Phys. 65, 1724 (1989). 152. D. K. Nayak, K. Kamjoo, J. S. Park, J. C. S. Woo, K. L. Wang, and K. P MacWilUams, IEEE Electron Device Lett. 12, 154 (1991). 153. S. S. Iyer, R M. Solomon, V. P Kesan, A. A. Bright, J. L. Freeouf, T. N. Nguyen, and A. C. Warren, IEEE Electron Device Lett. 12, 246 (1991). 154. R M. Garone, V. Vekatarrman, and J. C. Sturm, IEEE Electron Device Lett. 12, 230 (1991). 155. R W. Li, E. Yang, Y Yang, J. Chu, and B. Meyerson, IEEE Electron Device Lett. 15, 402 (1994). 156. D. Nguyen-Ngoc, D. L. Harame, J. C. Malinowski, S. J. Jeng, K. T. Schonenberg, M. M. Gilbert, G. D. Bert, M. Soyuer, K. A. Tallman, K. J. Stein, R. A. Groves, S. Subbanna, D. B. Colavito, D. A. Sunderiand, and B. S. Meyerson, BCTM, Technical Digest, pp. 89-92. IEEE, New York, 1995. 157. A. Hon, H. Umimoto, H. Nakaoka, M. Sekiguchi, and M. Segawa, "Intemational Electron Device Meeting," Technical Digest, p. 575. IEEE, New York, 1996.
4
SIGE/SI PROCESSING
197
158. S. A. St. Ogne, D. L. Harame, J. S. Dunn, S. Subbanna, D. C. Ahlgren, G. Freeman, B. Jagannathan, S. J. Jeng, K. Schonenberg, K. Stein, R. Grives, D. Coolbaugh, N. Feilchenfeld, P. Geiss, M. Gordon, P. Gray, D. Hershberger, S. Kilpatrick, R. Johnson, A. Joseph, L. Lanzerotti, J. Mahnowski, B. Omer, and M. Zierak, BCTM, Technical Digest, p. 117. IEEE, New York, 1999. 159. M. Racanelli, Z. Zhang, J. Zheng, A. Kar-Roy, P Joshi, A. Kalburge, L. Nathawad, M. Todd, C. Ukah, C. Hu, C. Compton, K. Schuefraf, P Ye, R. Dowlatshahi, G. Jolly, and P Kempf, BCTM, Technical Digest, p. 125. IEEE, New York, 1999. 160. J. D. Cressler, BCTM Short Course. IEEE, New York, 1999. 161. R Scott Johnson, J. Electron. Mater. 21, 805 (1992).
This Page Intentionally Left Blank
SEMICONDUCTORS AND SEMIMETALS, VOL. 73
CHAPTER
5
Advances in Quantum Dot Structures S. Kim and M. Razeghi CENTER FOR QUANTUM DEVICES, ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT, NORTHWESTERN UNIVERSITY, EVANSTON, ILLINOIS, USA
1. INTRODUCTION
199
2. PHYSICAL PROPERTIES
201
2.1. 2.2. 2.3. 2.4.
Density of States Energy States Optical Absorption and Transition in Quantum Dots Devices Based on Zero-Dimensional Quantum Structure
3. STATE OF THE ART
201 202 204 207 209
REFERENCES
212
1.
Introduction
Semiconductor quantum dots (QDs) represent one of the rigorous ongoing research areas for next generation optoelectronic devices. The strong interest in low-dimensional semiconductor structures originates from their exciting electronic properties that have an important impact on the performance of highspeed electronic and photonic devices and, moreover, on the development of novel device concepts such as the single electron transistor. The quantum dots known as quantum boxes are nanometer-scale islands in which electrons and holes are confined in three-dimensional potential boxes. They are expected to exhibit a zero-dimensional, S-function density of states and are able to quantize electrons free motion by trapping it in a quasi-zero-dimensional potential confinement. Due to these peculiar characteristics, quantum dots are expected to have superior characteristics for device performance in semiconductor lasers, detectors, and modulators. The condition for new electronic properties to occur in such device structures is that the lateral size of their active region must be smaller than the coherence length and the elastic scattering length of the carriers. Additional quantum-size effects require the structural features to be reduced to below 50 nm, that is, the range of the de Broglie wavelength. Therefore, the reproducible fabrication of these nanometer-scale quantum structures requires methods with atomic 199 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved. ISBN 0-12-752182-8 ISSN 0080-8784/01 $35.00
200
S. KIM AND M. RAZEGHI
scale precision, which is a major challenge for today's microstructure materials science. As a result of the strong confinement imposed in all three spatial dimensions, quantum dots are similar to atoms. They are frequently referred to as artificial atoms, superatoms, or quantum dot atoms [1]. What makes quantum dots such unusual objects is, first of all, the possibility to control their shape, their dimensions, the structure of energy levels, and the number of confined electrons. It is possible, for instance, to create and investigate, as a rectangular or parabolic potential well binding, one or several particles, as well as the Landau quantization of motion of a single electron, the radiative recombination from a few-particle system, and so on. Quantum dots were first realized by scientists from Texas Instruments Incorporated. Reed et al. reported the creation of a square quantum dot with a side length of 250 nm, etched by means of lithography. Since then, quantum dot and quantum wire structures have been fabricated by means of subsequent lateral patterning of two-dimensional heterostructures with lithographic techniques followed by chemical etching or selective crystal growth on prepattemed and masked substrates. However, although many fundamental properties of lowdimensional semiconductors can be demonstrated in these structures, it turns out that lithographic patterning processes and chemical etching always introduce defects that degrade the crystal quality and cause irregularities in size and shape of the quantum structures that are detrimental for practical applications in semiconductor devices. Especially to reduce the defect density, several methods for the direct fabrication of quantum dots and quantum wires based on the epitaxial growth process itself have been exploited. Quantum dots and wires have been grown by using the periodic step structure on vicinal surfaces, the generation of supersteps, and the breakup of high index surfaces into arrays of nanometer-scale facets. Recently, first breakthrough for growth of high quality highly strained epitaxial layers since the Stranski-Krastanow growth mode [2] was rediscovered during highly strained layer epitaxy; it is called self-assembled quantum dots. This discovery has great meaning for a new era of optoelectronic devices, providing new techniques for low-dimensional quantum structure [3-5]. When the lattice constants of the substrate and the crystallized material differ considerably, only the first deposited monolayer crystallizes in the form of epitaxial strained layers, where the lattice constant is equal to that of the substrate. When the critical thickness is exceeded, a significant strain that occurs in the layer leads to the breakdown of this ordered structure and to the spontaneous creation of randomly distributed islets of regular shape and similar sizes [6]. The small sizes of the self-assembled quantum dots (diameters in the range of 30 nm or even smaller), the homogeneity of their shapes and sizes in a macroscopic sample, the perfect crystal structure (without edge defects), and the fairly convenient growth process, without the necessity to precisely deposit electrodes or etching, are among their greatest advantages. Thus there is great hope with regard to their future applications in electronics and optoelectronics.
5
ADVANCES IN QUANTUM DOT STRUCTURES
2. 2.1.
201
Physical Properties
DENSITY OF STATES
Quantum confinement of charge carriers in semiconductors takes place when the carriers are trapped within potential wells of sufficiendy small dimensions. This quantum confinement can give rise to significant modification of the energy band structure and density of states (DOS) distribution in these materials. In the discussion that follows, we treat the quantum-confined structure in a simple, single-band model. Although this picture is adequate for the conduction band case, more elaborate multiband models, including the effect of band mixing, have been developed for the valence band. In the effective mass approximation, the energy spectrum E of the carriers is obtained by solving Schrodinger's equation 2
.^y^V{x.y^z)
^{x.y.z)
=
E^{x,y,z)
where "^ is the carrier envelope wave function, nf is the carrier effective mass, and y(jc, y, z) is the potential distribution. For potential wells of rectangular shape, one-, two-, and three-dimensional quantum confinement can be achieved in film-, wire-, and boxlike geometries by successively reducing the well dimensions r^, ty, and t^. For infinitely deep potential wells, the energy of the confined carriers (with respect to the band edge) is given by
'
lm*tl
2m*
one-dimensional (ID) confinement
Ei.n. = ^r-\-^ n* \tl \tl 2m*
+ t] -r\ + tl I) 2m*
two-dimensional (2D) confinement s2^2 / /2
^2
^2^
three-dimensional (3D) confinement where /, m, n = 1, 2 , . . . are the level quantum numbers and ky, k^ are the wave vector components along the unconfined dimensions. Quantum confinement in such quantum well (QWL), quantum wire (QWR), or quantum dot (QD) structures thus results in charge carriers of a quasi-2D, ID, or zero-dimensional (OD) nature.
202
S. KIM AND M. RAZEGHI
The density of states functions, including spin degeneracy, are given by
P3D =
V£
^^-^ 27r2
P2D X
E0(^-^/) I
*V/2 (2m*)
PlD
X y
EiE-Ei,J
-1/2
I, m
J:HE-E,„J
POD X y z
l,m,n
for 3D, quasi-2D, ID, and OD carriers, respectively. S{x) is the Heaviside function: ^ = 0 for jc < 0; ^ = 1 for jc > 0. The DOS distributions (Fig. 1) acquire sharper features as the carrier dimensionality is reduced, particularly in the case of ID and OD structures. Note, however, that these sharp features can be significandy smoothed out by well size fluctuations, leading to inhomogeneous broadening of the energy spectrum.
2.2.
ENERGY STATES
To explain three-dimensional quantum confinement, the quantum mechanical problem of the motion of a particle in a box was recalled. This problem turned out to be very complicated, because all the possibilities of electronhole Coulomb interaction—valence band structure or nonparabolic bands—were considered.
>
3D
A
2D
0
Q
•L
El
E2
>
\ )D
ID
y
Ell
^ — >
^12 Ei3
FIG. 1. Density of States.
— > •
11
b ji ^21
^r,1 1
^21
5
ADVANCES IN QUANTUM DOT STRUCTURES
203
The total Hamiltonian H for an electron-hole pair in the QD is given by
^h,kinVh)
T V.—i—v. + v — i — V
— ^
-e^
1
where ^^jcin and ///j^in are the kinetic energy of electrons and holes, and V^ and Vyj are the potential energy of electrons and holes, respectively. The H^ is the Coulomb interaction term, which is the only term dependent on coordinates of both electrons and holes, and couples of their motion. In spherical quantum dots, taking into account the Coulomb potential, there is a break in symmetry, because the Coulomb interaction depends on the spatial distance between the electrons and the holes. The simplest approach takes into account that the Coulomb energy scales like the inverse of the electron-hole distance (~ 1//^), whereas the kinetic energy scales like the square of the inverse radius (~ l/R^)One possible description for small dot radii in the so-called strong confinement range (R 2kT) from the band edges. In that case, the Fermi distribution can be approximated by the following Boltzmann-type expressions for the densities of conduction band electrons and of holes:
n = N,cx^[-{E,-E,)/kT]
(2)
or Ep = E^ + kT\n{n/N,)
(3)
and p = N,tx^[{E,-Ep)/kT]
(4)
or Ep = E^,-kT\n{plN,)
(5)
N^ and A^^ represent the effective density of states at the bottom of the conduction band and at the top of the valence band, respectively. From Eqs. (1), (2), and (4), it follows that np = N^N^Qxp{-EJkT)
(6)
Equation (6) implies that at equilibrium, when n is relatively high (n-type semiconductor), p will be negligibly small and vice versa. However, nonequilibrium conditions also must be considered. For example, when illuminating the semiconductor with photons that have an energy hv greater than E^, electrons will be excited from the valence band to the conduction band, leading to an excess of both electrons and holes but affecting, on a relative scale, primarily the density of minority carriers (holes in an n-type and electrons in a p-type semiconductor). Under constant illumination, a steady-state density of minority carriers
220
WALTER P. GOMES
will build up, depending on the generation and recombination rates of electrons and holes.
2.2.
THE SEMICONDUCTOR-LIQUID SOLUTION INTERFACE
Wet etching of semiconductors implies reactions that occur at a semiconductor-liquid solution interface. In most cases, an (aqueous) electrolyte solution is involved. Therefore, we will first deal with the properties of the semiconductor-electrolyte interface, and more specifically with the charge and potential distribution at this interface. 2.2.7.
Charge and Potential Distribution
Consider an electrochemical cell containing a semiconductor electrode SC, and a large area counterelectrode consisting of a metal ME (e.g., Pt) and a reference electrode RF. For the sake of simplicity, assume that all three electrodes are connected to wires consisting of the metal ME. The potential difference (p(hghi)—(p(\efi) in the circuit ME/RF/EL/SC/ME is called the electrode potential V of the semiconductor. It is the sum of the potential differences at the individual interfaces of the circuit. In what follows, the potential differences across ME/RF/EL are considered to be constant. Also the contact potential SC/ME can be considered to be constant for a given combination of materials. The potential difference of interest here is ut to the establishment of an acid-base or an adsorption equilibrium at the surface (see further). This V-independent contribution is hence included in the constant term of Eq. (35). It should be kept in mind that this contribution can vary with the composition of the electrolyte (e.g., with pH). In addition to depletion, other situations may occur at a semiconductor electrode surface, depending on bias. In Figure 2, depletion, flat-band situation, and accumulation are depicted for an n- as well as for a p-type electrode. One more possible situation is not shown in the figure, that is, inversion. By inversion we mean the following. In the case of an n-type electrode, for example, when the positive bias is so large that the Fermi level at the surface is below the middle of the bandgap, the equilibrium surface concentration of holes should exceed that of electrons; that is, holes should be created at the surface. When the bandbending is not too large, this inversion phenomenon is usually not observed with wide-bandgap semiconductors.
226
WALTER P. GOMES
Depletion I:
Accumulation
'(JJA«B
(45)
The surface species A • B is then an oxidation intermediate that has an unpaired electron that is no longer in a valence band level, but in a surface state, that is, an energy level in the bandgap at the surface [19]. Therefore, A#B may, in principle, be further oxidized either by capture of a second hole, A • B + h+ -^ products
(46)
or by injection of an electron into the conduction band, A • B -^ e~ + products
(47)
[We do not specify here the nature of the surface products formed in reactions (46) and (47). In most cases, several electrochemical steps are needed before soluble products are formed.] Electron injection [reaction (47)] is expected to occur only if the corresponding surface state is sufficiently close to the conduction band edge, which in many cases seems to be rather improbable [19]. Several methods may be used to investigate the contribution of electron injection to (photo)anodic dissolution. An appropriate technique for the photoanodic dissolution of n-type semiconductors is intensity-modulated photocurrent spectroscopy (IMPS). At a constant potential in the photocurrent saturation region, a harmonically modulated light source is superimposed upon the base illumination, creating a modulated hole flux toward the n-type semiconductor surface. It has been shown that, due to the extremely fast separation of photogenerated electron-hole pairs, the harmonically varying concentration of holes at the surface is always in phase with the light source. In the absence of reaction step (47), the current in the external circuit also is in phase with the light intensity. If, however, the second step of the photoanodic dissolution reaction is reaction (47) (for simplicity, we assume here that the overall reaction is two-equivalent), the fact has to be taken into account that such an electron injection step is characterized by a time constant, so that the current may lag behind with respect to the modulated light intensity over a certain phase angle. Only at very low
6
WET ETCHING OF III-V SEMICONDUCTORS
233
frequencies of the modulated light is the modulated injection current density in phase with the light source. In contrast, at very high frequencies, the hole current [reaction (46)] is measured exclusively. Hence by measuring the photoresponse as a function of frequency, the contribution of electron injection steps to photoanodic dissolution can be detected. More quantitative considerations on this method can be found in the literature [20, 21] and in further sections of this chapter. In addition to IMPS, electron injection during anodic dissolution of semiconductors may be investigated by other methods, which are summarized, for example, in [13], some of which are discussed further in this text.
2.4.
ELECTROCHEMICAL REACTIONS AT SEMICONDUCTORS IN REDOX ELECTROLYTES
From the electronic point of view, an oxidizing agent in solution, ox\ can be considered as an empty energy level, and the corresponding reducing agent red^~^ can be considered as a filled energy level (for simplicity in what follows, one-equivalent redox couples are considered). Cathodic reduction of an oxidizing agent at a semiconductor electrode then amounts to electron transfer from a semiconductor level to an empty level in solution; anodic oxidation of a reducing agent amounts to electron transfer from a filled level in solution to an empty level in the semiconductor. The generally accepted theory for these processes was developed by Gerischer [14, 15]. The basic idea is that electron transfer occurs by tunneling through the Helmholtz layer, and thus occurs between electron levels at equal energetic height. To understand electrochemical reactivity, the energy level distribution of the redox couple in solution with respect to that in the semiconductor must be known. Due to the difference in solvation structure between the oxidizing agent and the corresponding reducing agent, the energy of the most stable empty and filled redox states, E"^ and E^^, respectively, differs considerably. More specifically, it can be shown that the empty level is higher in energy than the filled level over an amount 2A^, E:,-E:,^2K
(48)
in which A^, the rearrangement energy, is related to the rearrangement taking place in the solvation structure of the ion when this ion either takes up or gives off an electron. Reliable data on A^ are not currendy available; values reported in the literature often are between 0.4 and 1.2 eV, depending on the redox couple considered. Owing to fluctuations of the solvation shell of the ions, the redox energy states do not constitute discrete levels, but are distributed over a certain energy range. In Gerischer's theory, the densities of energy functions are described by
234
WALTER P. GOMES
Gaussian functions, D^^ and Dj.^^, the widths of which are again determined by the rearrangement energy of the redox couple
D„,~exp[-(£-£-) V4A,A:r]
(49)
£ > r e d ~ e x p [ - ( £ - £ - , ) V4A,^r]
(50)
and
(see Fig. 5). It can be shown that the Fermi level of the redox couple at equal activities of the oxidizing and the reducing agent, ^^ redox' i^ energetically midway between the maxima of the empty and filled level curves: ^F, redox
=EI,+K=E:,
(51)
Whereas £^^ redox [measured with respect to the Fermi level in the standard hydrogen electrode (SHE)] is related to the standard redox potential of the couple U° by £;,redox = - ^ t / °
(52)
the U° value can be used as a rough measure for the energetic position of the redox levels. When also expressing the position of the semiconductor band edges at the surface E^^ and E^^^ with respect to the SHE, mutual comparison between energy level positions at both sides of the semiconductor-redox electrolyte interface becomes possible, which is essential for discussing electrochemical reactivity.
D(E) FIG. 5. Gaussian curves that represent the energetic distribution of redox levels in solution.
6
WET ETCHING OF III-V SEMICONDUCTORS
235
Since in principle, electrons for cathodic charge transfer may originate either from the conduction band or from the valence band of the semiconductor electrode, two types of cathodic reactions are possible: ox^-he--^red'-^
(53)
ox^-^^red^-^+h+
(54)
and
The first type is the capture of a conduction band electron, a process that was discussed previously (Section 2.3). Reaction (54) describes the uptake by the oxidizing agent of an electron from the valence band of the semiconductor, leading to the creation of a hole in that band; therefore this process is denoted as hole injection. Similarly, two types of anodic oxidation reactions may occur, depending on whether the conduction or the valence band is involved, that is, electron injection and hole capture [reactions (55) and (56), respectively]: r e d ^ - ^ - ^ o x ^ + e-
(55)
red'-*+h+-^ox^
(56)
Whether or not cathodic or anodic current flow can occur depends on the relative positions of the Fermi levels in the semiconductor and in the redox electrolyte. If a reaction is thermodynamically possible and if the surface reaction is ratedetermining, the current densities corresponding to reactions (53)-(56) can be expressed by the following respective equations, taking into account the sign conventions for electrochemical currents: (57)
Jc, c —
^^c, c^ox^s
Jc, V =
e^c.
Ja, c =
^^a. c^red
(59)
Ja, V =
^^a. u^redA
(60)
c
(58)
(CQX and Cred represent concentrations in the outer Helmholtz plane.) The rate constants in these equations evidently depend on the relative positions of electron energy levels at both sides of the interface. For example, hole injection requires that the empty redox levels overlap with the valence band of the semiconductor and hence have a relatively low position in the energy diagram. Considering Eq. (52), such behavior may be, in general, expected for a (one-equivalent) redox couple that has a high standard redox potential; that is, a relatively strong oxidizing agent.
236
WALTER P. GOMES
It also follows from these equations that the rate of the injection processes (54) and (55) is expected to be independent of the electrode potential V (as long as the position of the band edges at the surface is independent of V; see subsequent sections). This means, for example, that hole injection, when it occurs cathodically, also is expected to occur anodically. Anodically, however, the consumption of the oxidizing agent cannot be measured as an external current, but has to be detected by other methods. An appropriate method is rotating ring-disk voltammetry (see, e.g., [22, 23]). In this method, the semiconductor, in the shape of a circular disk, is concentrically surrounded by a thin metal ring. Ring and disk are electrically insulated from each other and are incorporated in separate electrochemical circuits. This electrode setup can be rotated at a controlled speed around an axis, perpendicular to the plane of the ring and the disk and passing through the center of the disk. Due to the rotation, the solution flows from the bulk to the disk and from there to the ring, so that products formed electrochemically at the semiconductor disk may be detected quantitatively through electrochemical reaction at the ring, that is, by measuring the ring current. The rates of the electron and hole capture reactions are proportional to the surface concentration of electrons and holes, respectively [Eqs. (57) and (60)]. At p-type semiconductors, reaction (53) therefore does not occur in darkness, since virtually no conduction band electrons are available. This reaction becomes possible under bandgap illumination however (cathodic photocurrent). The shape of the photocurrent curve is analogous to that discussed in Section 2.3. At n-type electrodes, the rate of reaction (53) is expected to increase exponentially when V is lowered. [Also here, in many cases k^ ^ is sufficiently large so that the reaction may occur under depletion (V > V^,) at considerable rates.] Similarly, hole capture can occur in darkness at p-type electrodes only; at n-type electrodes, illumination is required (anodic photocurrent). In the case of two-equivalent electrochemical reactions, a particular class of reactions—so-called current-doubling reactions—deserves special attention. The current-doubling effect was first observed at the (n-type) ZnO electrode [24, 25]. At given light intensity, the limiting (anodic) photocurrent was found to be approximately doubled when certain two-equivalent reducing agents were added to the indifferent electrolyte solution. This effect was explained by a mechanism analogous to that described by Eqs. (45) and (47) for the photoanodic oxidation of n-type semiconductors (chronologically, the latter was observed experimentally much later). The assumption is that the two-equivalent reducing agent A captures a photogenerated hole from the electrode to form a radical-type intermediate A"^* that has highly reducing properties so that the corresponding filled redox levels are close to or above the conduction band edge so that in a second step, the intermediate injects an electron into the conduction band to form A^+: A + h+-> A+*
(61)
A+->A2+ + e-
(62)
6
WET ETCHING OF III-V SEMICONDUCTORS
237
In this way, for each photogenerated hole reacting at the surface, two electrons flow through the external circuit. If, in an indifferent electrolyte, the photoanodic reaction involves hole capture steps only, the saturation photocurrent is doubled by adding the reactant. Shortly after photoanodic current-doubling was detected, its cathodic counterpart was also reported [26] on p-type GaP with two-equivalent oxidizing agents (B) such as hydrogen peroxide and the persulfate ion. The reduction mechanism can be denoted here as B+ e - ^ B -
(63)
B-->B^-H-h+
(64)
where B~* is a strongly oxidizing radical such as OH* with empty redox levels located near or below the valence band edge. In view of the electrochemical approach for studying etching mechanisms, the latter type of current-doubling reaction is especially important in the framework of this chapter on etching of III-V compounds, because current-doubling oxidizing agents are often found to be etchants for III-V compounds. Two important remarks have to be made at this point. The first one is that complications may arise in semiconductor electrode reactivity due to surface states. As an alternative route to direct charge transfer as described previously, charge transfer may occur in two steps, that is, first from the semiconductor bands to surface states and then from surface states to redox species in solution or vice versa. Furthermore, the filling or emptying of surface states may lead to charge accumulation at the electrode surface and hence to the unpinning of the band edges, with evident consequences on the magnitude of the electrochemical rate constants. The second remark is that at high current densities, mass transport may become the rate-limiting factor of electrochemical reactions. Indeed, due to the high reaction rate, the surface concentration of the oxidizing or the reducing agent may become considerably less than the bulk concentration, so that the electrochemical reactant has to be transported to the surface by diffusion. This leads to a potential-independent current density j^-^ff that, for rotating electrodes in which the transport is by convective diffusion, obeys to the Levich equation [23] j , ^ = 0.155 nFD'/\p/r]y/'w'/'c
(65)
In this equation, y^iff is the limiting current density in amperes per centimeter squared, n is the number of charge carriers involved in the reaction, D is the diffusion coefficient (in meters squared per second), and c is the concentration (in moles per liter) of the rate-limiting species, p is the density (in kilograms per cubic meter) and r] is the viscosity (in kilograms per meter per second) of the electrolyte solution, and w is the rotation rate of the electrode (per second).
238
WALTER R GOMES
Hence, a limiting current can be identified as diffusion-limited if, at a rotating (ring-) disk electrode, it is found to be proportional to the square root of the rotation rate of the electrode. We recall that diffusion limitation may also occur in an indifferent electrolyte; for example, in the (photo)anodic dissolution of semiconductors in an alkaline medium, the current density may be limited by diffusion toward the electrode of the 0H~ ions necessary to form soluble anions of the lattice constituents (see Section 2.3). 3. 3.1.
Types of Etching Reactions
(PHOTO)ELECTROCHEMICAL ETCHING
(Photo)electrochemical etching amounts to the (photo)anodic oxidation of the semiconductor under conditions in which the oxidation products are soluble (see Section 2.3). This type of etching has the practical disadvantage that the semiconductor has to be incorporated into an electrochemical circuit; that is, that electrical contact has to be made to connecting wires. On the other hand, (photo)electrochemical etching also offers certain advantages. The amount of semiconductor removed by etching can be controlled coulometrically, that is, through the total anodic charge drawn, provided that the equivalence of the reaction involved is known [cf. Eq. (44)]. Also, as mentioned later in this chapter, well-determined morphologies (e.g., porous surfaces) may be induced by choosing the appropriate range of applied potentials. Because holes are needed for the anodic dissolution of semiconductors, this process occurs in darkness with p-type semiconductors, whereas for n-type semiconductors, illumination with photons of the appropriate energies is required; that is, the dissolution is photoelectrochemical. Special advantages are in principle inherent to the latter procedure. Indeed, local photoelectrochemical etching of well-defined areas on the semiconductor surface may be achieved either by using a thin light beam or by covering chosen parts of the surface from the light. Also, when multilayer structures consisting of semiconductor materials with different bandgaps are involved, material-selective etching of a semiconductor with a narrower bandgap with respect to one having a wider bandgap may be performed by choosing the wavelength range of the light such that only in the former case are electron-hole pairs produced. Next to (photo)electrochemical etching, three types of etching reactions can be distinguished that involve no net current flow, which is obviously more attractive from the technological point of view: the semiconductor sample is simply immersed in the etchant solution and, in certain cases, illuminated with light of appropriate wavelengths. These three types of etching are photoetching, electroless etching, and chemical etching. The first type occurs under illumination; the other two in darkness. The first two involve an electrochemical mechanism; the third one does not. We now briefly introduce the principles and characteristics of these three types of etching reactions.
6 3.2.
WET ETCHING OF III-V SEMICONDUCTORS
239
PHOTOETCHING
Consider again photoelectrochemical etching as depicted in Figure 4a, which shows an n-type semiconductor under anodic bias and illuminated by photons with energy larger than the bandgap. The photogenerated holes are assumed to oxidize and hence to etch the semiconductor. The electrons flow through the external circuit and reduce some oxidizing agent at the counterelectrode. In the case of photoetching, there is no external circuit. (It should be noted here that a certain confusion in terminology exists in the literature; therefore it is stressed that by photoelectrochemical etching and by photoetching, we mean etching under illumination with and without external current flow, respectively.) Hence in photoetching, the photogenerated electrons have to be consumed at the semiconductor-electrolyte interface; that is, they react with an oxidizing agent. The levels of the oxidizing agent required for photoetching should hence overlap with the conduction band of the semiconductor (Fig. 6). The net photoetching process thus proceeds by an electrochemical mechanism that consists of two partial currents (photoanodic dissolution of the semiconductor and cathodic reduction of the oxidizing agent; Fig. 7a) that compensate each other electrically but not chemically. The corresponding reactions can be written as /zz^->e"+h+ SC-\-n h"^ ^- dissolution products n (ox^ 4-e~ -^ red^~^) SC-\-n ox^—> dissolution products+ /2 red^~^
FIG. 6. Mechanism of photoetching.
(66)
240
WALTER P. GOMES
n-type
SC + iih+ -> diss, prod
iz-l
p-type
1 1 1
SC +
nh+ ->
/I 1
diss. prod —^ //
1 /
71 1
•'^
i
i
/i/
V
i /V
J ''''
/' / \ /'' \ //
J'
ox^ + e~ -> red^--1
(b) FIG. 7. Schematic representation of partial and total current density curves at an illuminated semiconductor electrode in an electrolyte solution that contains an electron-capturing oxidizing agent, (a) n-type; (b) p-type. Photoetching occurs at V,.
(for simplicity, one formula unit of the semiconductor SC is assumed to be oxidized by n holes, whereas the reduction of the oxidizing agent is assumed to be one-equivalent). Combination of etch rate and electrochemical measurements in principle allows us to check whether this simple photoetching model holds. Indeed, in this case, the open-circuit etch rate should correlate with the cathodic current density at open-circuit potential under illumination V^^ (see Fig. 7a), which can be measured separately (e.g., by working with chopped light). From Figure 7a, it is obvious that the open-circuit photoetch rate is determined by the relative positions of both partial current curves. A high photoetch rate, at given light intensity and oxidizing agent concentration, implies a steep onset of the photocurrent
6
WET ETCHING OF III-V SEMICONDUCTORS
241
curve and hence a relatively low surface electron-hole recombination rate and a high electron capture rate constant k^^ [see Eq. (57)]. Figure 6 equally applies to the photoetching mechanism at p-type semiconductors. The corresponding partial current density curves are schematically represented in Figure 7b. The comments are analogous to those applying to n-type samples. The particular advantages of light-assisted etching are essentially the same as those enumerated in the previous section.
3.3.
ELECTROLESS ETCHING
Electroless etching occurs in darkness but is otherwise rather analogous to photoetching as far as the mechanism is concerned. Indeed, here also the net etching reaction is the sum of two electrochemical steps that cancel each other electrically but not chemically. The anodic step is the same as in photoanodic etching or in photoetching; that is, hole capture leads to dissolution of the semiconductor. The cathodic step is the injection of holes by an oxidizing agent. The empty levels of the etching reactant must hence overlap with the valence band of the semiconductor in this case. Under the same assumptions as those accompanying reaction Eq. (66), the electroless mechanism can be symbolized by SC H- n h^ ^- dissolution products n (ox'^red'~^-hh+) SC + n ox^ -^ dissolution products+ « red^~^
(67)
that is, the holes injected by the oxidizing agent are consumed in the oxidation reaction of the semiconductor. This mechanism is illustrated in Figure 8a by a current density vs. potential diagram pertaining to a p-type semiconductor. The diagram is based on the simple assumption that the hole injection current is potential-independent (see Section 2.4). The actual etching conditions correspond to the rest potential V^, which is a mixed potential, and imply that the anodic dissolution current adjusts itself to the constant cathodic reduction current. At V >V^, all injected holes are consumed in the etching reaction. At potentials sufficiently negative with respect to V^, the injected holes lead to cathodic current flow. In the case of an n-type semiconductor, where the current is transported by conduction band electrons, the graphical representation of electroless etching requires more attention. Indeed, in that case, the injected holes constitute the minority carriers, so that the situation is similar to that under illumination, where the holes are created by light. Hence, one has to construct a fictitious anodic curve in the current-potential diagram (Fig. 8b) that has a shape analogous to the photocurrent curve. At potentials close to the flat-band
242
WALTER P. GOMES
p-type
1
1
(1
1 SC + nh+ -> diss, prod -^
/ li
(a)
n-type SC + nh+ -^ diss, prod /
: ' \
V1
ox^ + e~ ^ red^
(b) FIG. 8. Schematic representation of partial and total current density curves at a dark semiconductor electrode in an electrolyte solution that contains a hole-injecting oxidizing agent, (a) p-type; (b) n-type. Electroless etching occurs at V^.
potential the injected holes recombine with electrons, which are here the majority charge carriers. At more positive potentials, the injected holes remain at the surface, where they oxidize the semiconductor. We note that, in contrast to the case of photoanodic dissolution, in electroless etching, current flow takes place when the holes recombine with electrons (the current is transported through the n-type sample by conduction band electrons), whereas no net current flows when the holes do not recombine. The fictitious anodic current-potential curve in Figure 8b reaches a limiting value that is equal to the absolute value of the Hmiting cathodic hole injection current (it is again assumed that the hole injection rate is potential-independent). The resulting current-potential curve does not
6
WET ETCHING OF III-V SEMICONDUCTORS
243
tend to zero at positive potentials due to the contribution of electron injection to the anodic dissolution reaction (see Sections 2.3 and 5.1) that causes a small net anodic current. As a result, also with n-type semiconductors, a rest potential V^ is defined (see Fig. 8b), at which the actual electroless etching occurs. It can be seen from Figure 8b that the etch rate at V^ corresponds to the limiting current density of the fictitious anodic dissolution curve. This point will be important in the discussion of the etch morphologies (see Section 7). It is obvious for electroless etching also that combining etch rate and electrochemical measurements may lead to detailed insight into the reaction mechanism. This will be illustrated by selected examples further on in this chapter.
3.4.
CHEMICAL ETCHING
In a chemical etching mechanism, no free charge carriers from the semiconductor (i.e., neither conduction band electrons nor valence band holes) are involved. The etching reaction amounts to the breaking of surface bonds of the semiconductor by the etchant and the formation of new bonds between the semiconductor constituents and species from solution. For example, the first step of the etching of InP by HCl can be represented by CI — H
CI H
-In-P-^-lLp/
\
/
(^^)
\
As this example shows, etchants operating by a chemical mechanism are not necessarily oxidizing agents, although they mostly are. For instance, oxidizing agents such as H2O2 and dihalogens act as chemical etchants. Although chemical etching reactions are not composed of electrochemical steps, electrochemical measurements can yield useful mechanistic information. Thus, electrochemical data may allow us to decide whether the attack of a semiconductor by an oxidizing agent in darkness is electroless or chemical, especially when p-type samples are available. Indeed, if a given oxidizing agent is found to etch the p-type sample under open-circuit conditions but not to enhance the cathodic dark current and, hence, not to inject holes, then electroless etching can be excluded, so therefore the mechanism must be chemical. If, on the other hand, the etchant does inject holes, it is informative to measure the etch rate under limiting cathodic current flow, that is, under circumstances in which the injected holes are drawn into the semiconductor by the electric field so that they cannot participate in etching. If the p-type sample is etched under these circumstances, this must involve a chemical mechanism, so that at rest potential, the etching will occur by both mechanisms in parallel. If, on the other hand, the p-type sample is not etched cathodically, this does not necessarily mean that chemical etching at rest potential can be excluded. Indeed, if the cathodic
244
WALTER P. GOMES
hole injection current is diffusion-limited, the competition for oxidizing species between hole injection and chemical reaction may be such that the rate of the latter becomes negligible. These principles of investigation will be further illustrated by specific examples. As will be demonstrated later, identification of the etching mechanism(s) is important in tackling problems of material-selective etching and of etch morphology. In case it is somehow established that an etching mechanism is chemical, semiconductor electrochemistry may yield further details on this mechanism, in the sense that it may distinguish between a concerted reaction [such as in the example of Eq. (68)] and a sequential reaction. Indeed, if the etchant is an oxidizing agent, it may first extract one electron from a surface bond, leaving a radical-type surface intermediate. Anodically, in a second step, the remaining surface electron may either be removed by the etchant or, if the position of the corresponding surface level allows, be injected into the conduction band. In certain cases, an increase in the anodic dark limiting current has been observed with n-type semiconductors in contact with a chemical etchant (see further), and this effect has been attributed to electron injection by surface oxidation intermediates of the semiconductor and hence to a sequential etching mechanism.
4.
Some Solid-State and Electrochemical Data on III-V Semiconductors
The binary and mixed III-V semiconductors to which this chapter pertains are listed in Table I, together with their crystal structures and their bandgap values Eg at room temperature. All these materials can be made either n-type or p-type semiconducting by appropriate doping, although there are presently still problems in making good quality p-type GaN samples. In Figure 9, the unit cell of the zincblende structure is represented. This structure belongs to the cubic system (class 43m) and has three inversion tetrad TABLE I III-V SEMICONDUCTOR MATERIALS
Semiconductor GaAs GaP InP -^10.25^^0.75 A s H.53Gao.47As
GaN
Crystal structure Zincblende Zincblende Zincblende Zincblende Zincblende Wurtzite (when grown on sapphire)
Bandgap E^ (eV) 1.43 2.25 1.35 1.75 0.75 3.39
6
245
WET ETCHING OF III-V SEMICONDUCTORS (100) [111]
[001]
FIG. 9. The zincblende unit cell. axes ([100], [010], and [001]) and four triad axes such as the [111] axis shown in the figure. In view of the discussion on etch morphology, it is important to note that the triad axes are polar, so that a sample cut perpendicular to the [111] axis has two faces that have different properties. The (111) face consists of group III atoms and the ( i l l ) face consists of group V atoms exclusively. The (lOO)-type faces, on the other hand, are mixed; that is, they contain group III as well as group V atoms. The wurtzite structure belongs to the hexagonal system (class 6mm). The hexagonal axis is polar. A sample cut perpendicularly to this axis has a (0001) face consisting of atoms of one constituent (Ga in the case considered) and a (0001) face consisting of the other constituent (N in the given case). In both structures, each atom is surrounded by four atoms of the other kind. Figure 10 is an energy scheme that shows the energetic position of the band edges (referred to the SHE scale) of the six semiconductor materials considered. The data pertain to an indifferent aqueous medium at pH = 0 and were obtained from differential capacitance measurements, as explained in Section 2. The sources are as follows: for GaAs, [18, 27]; for GaP, [27]; for InP, [18, 28]; for Alo.25Gao.75As, [29]; for lUo.saGao^vAs, [30]; for GaN, [31]. In all these cases, the band edges were found to shift upward over about 0.06 eV per unit pH increase, indicating that acid-base equilibria are established at the semiconductor-electrolyte interface [27]. The positions shown should be considered as approximate only. Indeed, in certain cases the flat-band potential and, hence, the energy band positions were found to differ, depending on the crystal face exposed to the electrolyte. Also several cases have been observed in which the bands may shift by adding redox components to the electrolyte solution, due either to adsorption at the semiconductor surface or to more complex processes
246
WALTER R GOMES
EvsSHE/feVA GaP
GaAs
0.25 _ 0 7 5
T InP
In Ga As * "0.53
HVH2
GaN
0/7
FelCNg)^'^
Fe 3*/2* O2/H2O
#,
m -1 H
^
^
^
-2]
-3
^
FIG. 10. Energy band edge positions for different III-V semiconductors and standard Fermi levels for different redox couples referred to the SHE; aqueous medium; pH = 0.
such as holes being injected by an oxidizing agent and being held at the surface. Such cases will be discussed on the occasion of the specific examples given further in the text. The left-hand side of Figure 10 shows the standard Fermi level [see Eq. (52)] of some common redox couples. Mutual comparison of the energy levels at both sides of the interface is essential for discussing the possibility of charge-transfer reactions. As far as the kinetics are concerned, it should be kept in mind that for one-equivalent redox couples, the empty levels are mainly above and the filled levels are mainly underneath the levels drawn [see Eqs. (49) to (51)], and that for two-equivalent redox couples, the E^ ^^^^^ values shown are actually values averaged over the two steps of the reaction, each of which is characterized by a Fermi level and by two Gaussian level distribution curves of the type of Eqs. (49) and (50).
Kinetics and Mechanisms of Etching Reactions at III-V Semiconductors 5.1.
(PHOTO)ELECTROCHEMICAL ETCHING
Anodic current flow at p-type and photoanodic current flow at n-type III-V compound semiconductors at either low or high pH and not too high current
6
WET ETCHING OF III-V SEMICONDUCTORS
247
densities lead mostly (see subsequent text) to the anodic dissolution and hence to etching of the semiconductor. At high current densities or in the intermediate pH range, the electrochemical oxidation of the semiconductor may lead to precipitation of products upon the surface and hence to passivation of the electrode. Several studies have been devoted to the electrochemical equivalence of the anodic dissolution reaction, that is, to the number of elementary charges n flowing through the external circuit per formula unit of semiconductor dissolved. Values ofn = 6 have been reported for GaAs [8], GaP [32, 33], and InP [34, 35], as well as for the mixed semiconductors Alo25Gao75As [36] and Ino53 Gao47As [37]. It hence follows that in all these cases the elements involved go into solution in the +3 oxidation state. This result is somewhat surprising as far as phosphorus is concerned, considering the negative standard redox potential of the (H3PO4, H3PO3) couple (t/° = -0.276 V vs. SHE). Since P O ^ and PH3 have been detected in solution [32, 38] in the anodic dissolution of GaP, it has been suggested that the H3PO3 or HPO3" formed electrochemically reacts further homogeneously in solution by chemical disproportionation to H3PO4 and PH3. As far as GaN is concerned, recent results demonstrate that n = 3 [31, 39], indicating that the oxidation products are Ga in the trivalent state and N2. It should be mentioned that certain authors claim that the photoanodic reaction at n-GaN in aqueous indifferent (H2SO4) electrolyte is H2O oxidation [40]. This apparent contradiction in the data may be due to a difference in behavior between the two polar faces perpendicular to the [0001] axis [31] in the sense that, presumably, the N-terminated (0001) face is etched photoanodically, whereas the Ga-terminated (0001) face is not. Whether the free surface is the (0001) or the (0001) face seems to depend on the circumstances under which the GaN film is grown. The growth occurs by MOCVD from (CH3)3Ga and NH3 upon a sapphire substrate. Gallium-terminated films can apparently be obtained only when, at the initiation of growth, all traces of ammonia have been removed from the reactor. On the basis of these summarized results and taking into account the chemistry of the elements involved in strong acidic and alkaline aqueous media [41], the following overall reaction equations can be proposed for the anodic dissolution of the binary III-V semiconductors under consideration:
At low pH, GaAs + 2 H2O 4- (6 - jc)h+ -> Ga^+ + HASO2 + 3 H+ + jc e"
(69)
GaP -h 3 H2O + (6 - x)h+ -> Ga^+ -f H3PO3 + 3 H+ -h xe'
(70)
InP + 3H20 + ( 6 - x ) h + ^ I n ^ + + H3P03-h3H+-hJceGaN + (3 - x)h+ -> Ga^+ + ^N2 + xe"
(71) (72)
248
WALTER P. GOMES
At high pH, GaAs + 10 OH- + (6 - jc)h+ -^ GaO^" + ASO2" + 5 H2O + xe"
(73)
GaP + 11 OH" + (6 - jc)h+ -^ GaO^" + HPO^" + 5 H2O + jce"
(74)
InP + 9 OH" + (6 - jc)h+ -^ InO" + HPO^" + 4 H2O + Jce"
(75)
GaN + 6 OH" + (3 - jc)h+ -> GaO^" + ^N2 + 3 H2O + jce"
(76)
The symbol x is used to take into account the possible participation of electron injection into the anodic oxidation reaction, a problem that is discussed subsequendy. In parallel with the foregoing reactions, other reactions may contribute to the anodic current to a minor extent. Thus, an enrichment in As at the GaAs electrode surface after anodic current flow has been demonstrated by various experimental techniques [42,43]. Hence, it must be concluded that three-equivalent oxidation of GaAs occurs to a small extent in parallel to the six-equivalent oxidation. In the intermediate pH range, the formation of an anodic oxide film on the electrode surface is often observed; that is, on GaAs, film formation in a weakly alkaline medium (pH = 11.5) was reported and attributed to slow dissolution of the Ga203 and AS2O3 formed anodically under these circumstances [44]. Thicker passivating layers on GaAs have been produced under various conditions. At n-GaN, a photoanodic current decrease due to Ga203 formation has been observed even in strongly acidic or alkaline media, indicating slow dissolution kinetics [31]. In view of the scope of this chapter, the focus is primarily on circumstances in which layer formation does not occur. From various experiments (see subsequent text), it follows that electron injection into the conduction band contributes to the (photo)anodic dissolution reaction only to a minor fraction in the case of GaAs [10,45], GaP [33,38,46], and GaN [31]; that is, that JC « 1 in Eqs. (69), (70), (72), (73), (74), and (76). In the case of n-InP, IMPS studies (see Section 2.3) have demonstrated that this fraction depends on the total photocurrent density and may be very significant [47,48]. Indeed, the ratio of the total photocurrent density j vs. the part of it due to hole capture jf^ has been found, for n-InP in an acidic medium, to range from 1.2 at high light intensities [implying JC = 1 in Eq. (71)] to 2 at very low light intensities [implying jc = 3 in Eq. (71)]; see, for example. Figure 11. This means that at low light intensities, three of the six electrochemical steps are hole-capture steps and three are electron-injection steps. At first, this suggests a sequence consisting of a hole-capture step leading to the formation of a radicaltype oxidation intermediate, followed by the injection of an electron by that intermediate, again followed by hole capture in a new surface bond, etc. However, from a detailed analysis of the IMPS data [47], it follows that three subsequent electron-injection steps are involved, indicating a more complex anodic dissolution mechanism. The decrease in j/jf^ from 2 to 1.2 with increasing light
6
WET ETCHING OF III-V SEMICONDUCTORS
249
logioGh in Acm-2) FIG. 11. Ratio of the total photocurrent density vs. the hole current density, y'/j,,, as a function of the hole current density, 7,,, measured by IMPS at n-InP in 1.2-moI \~^ HCl. The solid and the open symbols refer to measurements on two different electrodes. The curve has been calculated based on the assumptions mentioned in the text. Reprinted from Electwchim. Acta 38, B. H. Erne, D. Vanmaekelbergh, and I. E. Vermeir, pp. 2559-2567, ©1993, with permission from Elsevier Science.
intensity can be understood by assuming that for two electrochemical oxidation steps, a competition exists between electron injection and hole capture, the latter step being increasingly favored when the hole concentration at the surface, and hence the total photocurrent density, increases. The fact that even at very high photocurrent densities one step still occurs exclusively by electron injection leads to the conclusion that for this step, competition with hole capture is absent, the reason presumably being that the corresponding electron level (surface state) is located above the conduction band edge. The IMPS data also allow us to estimate the magnitude of the electron-injection rate constants associated with the photoanodic dissolution of n-InP. It was found that in 1-mol \~^ HCl medium, the rate constants are significantly larger (over a factor of 20-200) than in 1.3 mol 1"^ H2SO4 [48]. Apparently, the presence of CI" ions gives rise to decomposition intermediates with positions more favorable for electron injection as compared to those formed in SO4' containing solutions. It can hence be concluded that the properties of the surface states associated with the anodic decomposition intermediates of In? are influenced by the surface chemistry. The same conclusion follows from the observed enhanced electron-injection rate during photoanodic dissolution of n-InP in the presence of H2O2 [48]. Experimental results obtained at the n-Ino53Gao47As electrode in aqueous 1.3 mol 1~^ H2SO4 solution indicate that here, also the contribution of conduction band electron injection to the photoanodic dissolution reaction may be important when the photocurrent density is low [37]. All this suggests that efficient
250
WALTER R GOMES
electron injection during photoanodic dissolution of III-V compound semiconductors may be connected to indium-related surface states. There are strong indications that, in certain cases, a mixed chemicalelectrochemical dissolution mechanism of III-V semiconductors occurs. As far back as 1969, Gerischer et al. [10,11] reported an accelerating effect of Br2 upon the anodic dissolution of p-type GaAs, in the sense that the onset of the anodic current density curve was shifted toward less positive potentials as compared to the curve in an indifferent electrolyte. If the position of the band edges is unaffected by adding the oxidizing agent, this implies that at the same band-bending (i.e., the same hole concentration at the surface pj the anodic current and hence the corresponding rate constant is larger (see Section 2). The authors interpreted this effect as follows. It is generally accepted that the first hole-capture step in the anodic oxidation reation is rate-determining as a consequence of the high activation energy for breaking an intact surface bond. Since Br2 is known to be a chemical etchant for GaAs, Gerischer et al. then assumed that, in the presence of Br2, the slow first electrochemical step is substituted by a faster chemical bond-breaking step (or more likely, that the first two steps are substituted). Afterward, the same effect was observed with several other systems, including p-GaP-aqueous Br2 [49], p-GaP-methanoHc Br2 [50], and p-InP-aqueous HIO3 [51]. Such a negative shift of the anodic dissolution curve may serve as an indication that the oxidizing agent involved acts as a chemical etchant under open-circuit conditions. It should be checked by performing impedance measurements, however, whether the observed shift is due to an artefact, that is, to a shift in flat-band potential caused by the interaction of the oxidizing agent with the semiconductor surface. A particularly interesting case is that of p-GaP-Br2 in methanol [50]. By adding Br2 to the methanolic solution, the anodic dissolution curve of ( i l l ) p-GaP was found to shift up to 0.6 V on the potential scale referred to the flat-band potential, indicating that the preceding interpretation holds. At the (111) p-GaP face, however, no such effect was observed, in agreement with the observation that the open-circuit rate of etching by Br2 at the (111) face is much lower than at the opposite polar face [i.e. the ( i l l ) face]. However, rotating ring-disk experiments showed an increased Br2 consumption at the (111) face under anodic current flow. Moreover, the electrochemical equivalence of the anodic dissolution reaction was found to be about 4 in the presence of Br2, in contrast to the value of 6 measured in an indifferent electrolyte. All these data strongly indicate that a mixed dissolution mechanism holds here also, but the first oxidation steps are hole-capture steps, and two further steps are chemical; that is, chemical attack of the (111) face by Br2 can occur only after bond cleavage has been initiated by holes. It is evident that formulas such as Eq. (69) or (73) merely describe the overall anodic etching reactions and that these reactions involve various decomposition intermediates as well as chemical steps in which (in an aqueous medium) H2O molecules or 0H~ ions participate. Several experimental approaches based on
6
WET ETCHING OF III-V SEMICONDUCTORS
251
electrode kinetics have been used to obtain more detailed information on the anodic or photoanodic dissolution mechanisms of III-V semiconductors. One of these approaches consists of studying the kinetics of the competition between the (photo)anodic dissolution reaction of the semiconductor and the capture of holes by a one-equivalent reducing agent such as the Fe^^ ion or a Fe(II)-based complex. Interest in this subject was originally connected with the field of electrochemical solar energy conversion. Indeed, in 1975, Gerischer [52] proposed a type of solar cell based on a semiconductor electrode, called a regenerative or photovoltaic electrochemical cell, in which a dissolved reducing agent was photoanodically oxidized by holes at an n-type semiconductor electrode, while the corresponding oxidizing agent was reduced back at a dark counterelectrode, the net result being the conversion of solar energy into electrical energy. In view of the matching of the bandgap of the semiconductor with the spectral distribution of sunlight, GaAs and InP seemed to be the most suitable candidates for use as photoactive electrode materials. However, it was soon realized that one of the main problems with such cells was the stability of the semiconductor material with respect to photocorrosion, that is, in addition to hole capture by the dissolved reducing agent, reactions such as (69) or (71) take place also. To optimize the competition between both reaction types in favor of the added reducing agent, thorough kinetic studies were made in which the competing rates were investigated as a function of reducing agent concentration, pH, solvent composition, and light intensity. These investigations were carried out mainly by rotating ring-disk measurements; see Section 2.2. For a review on the results, see, for example, [13]. The fact that in nearly all cases the competition was more in favor of the photoanodic dissolution reaction when the light intensity was higher turned out to be an especially important clue as far as the competition mechanism is concerned. Indeed, if both reactions were simply occurring in parallel, they both would be first order in photogenerated holes [see Eqs. (44) and (60)], so that the competition would be light-intensity independent. A detailed analysis revealed the central role of the decomposition intermediate X^, formed when an intact surface bond of the semiconductor captures a hole from the valence band. For example, for GaAs, GaAs + h + ^ X +
(77)
where Xf is a surface bond with one electron missing. Many of the kinetic data could be explained by assuming that the dissolved reducing agent is not oxidized by a hole, but by donating an electron to X^, hence restoring the surface bond. More importantly in the present context, in most cases involving GaAs the kinetics indicated that the intermediate X^ and not the free hole h"^ constitutes the mobile species participating in the consecutive steps of the dissolution reaction, so that it must be concluded that the intermediate X^ is mobile within a two-dimensional layer; in other words, that a bonding electron may jump from an unbroken surface bond to a neighboring electron-deficient bond.
252
WALTER P. GOMES
More insight into the surface chemistry involved in the anodic dissolution reaction was obtained by studying the competition kinetics under conditions of varying pH and water activity, the latter being varied by adding large concentrations of a highly hydrated salt such as LiCl or by using mixed solvents such as water-methanol or water-acetonitrile. The results, which were primarily obtained on GaAs and GaP, led to a comprehensive model, proposed in [53]. The model assumes that after reaction (77), a chemical reaction between Xf and water occurs: X+ + mH^O ^ Xi—OH + H+(H20)^_i
(78)
In this reaction, m water molecules react with the mobile and positively charged intermediate X^, giving rise to a neutral intermediate X^—OH and a solvated proton. The intermediate X^—OH is then supposed to be immobile since it contains an OH group attached to the surface. The equilibrium (78) is thought to play a key role in the dissolution mechanism. For GaAs in an acidic aqueous medium, this equilibrium is supposed to be positioned somewhere in the middle, so that the second step of the photoanodic dissolution reaction is between Xi—OH and X^. In contrast, for GaP, this equilibrium is assumed to be positioned far to the right (i.e., practically no mobile intermediates are available), so that here the second oxidation step is between X^—OH and a free hole. The observed changes in competition mechanism induced by decreasing the water activity can then be explained as consequences of equilibrium (78) being shifted toward the left. The conclusions just mentioned were largely confirmed from independent data obtained by quantitatively studying the enhancement of the anodic dark current density at n-type electrodes, caused by hole injection. Indeed, as mentioned in earlier sections, under anodic bias, holes injected into an n-type semiconductor by an oxidizing agent in darkness are consumed, just like at rest potential, in the anodic dissolution reaction of the semiconductor. If the latter reaction involves the valence band solely, the anodic blocking current will remain unchanged by adding the hole-injecting reactant. Any participation of conduction band electrons in the dissolution reaction will show up as an increase in the dark anodic limiting current density. By this method, it has been demonstrated that the conduction band contributes considerably to the anodic dissolution reaction in the case of InP and IUQ 53Gag 47 As (as confirmed later by IMPS measurements; see preceding text), but only to a minor fraction (on the order of per thousand to per hundred) to the anodic dissolution of GaAs and GaP. Because j ^ , the anodic current density increase, was found to increase with temperature (for GaAs and GaP) [45,46], this process is believed to be activated; that is, the electron must be thermally excited from the surface state associated with the decomposition intermediate to the conduction band before injection can occur. Measuring the relationship between the additional anodic current density j^ and the cathodic hole injection current density j^ constitutes a powerful probe
6
WET ETCHING OF III-V SEMICONDUCTORS
253
for investigating the dissolution mechanism. The basic assumption used in the interpretation of such relationships, obtained on GaAs and GaP, is that the anodic current increase is due to injection of an electron by the first decomposition intermediate, X^, into the conduction band. Analysis of the results by standard steady-state kinetic treatment then leads to conclusions that confirm those obtained by studying the competition kinetics [54]. Recall at this point that anodic electron injection by dissolution intermediates can be observed not only as a consequence of hole injection (i.e., of electroless etching), but alternatively as a consequence of chemical etching (see Sections 3.4 and 5.4). Analogously, as before, quantitative studies of the relationship between the injection current density j^ and the chemical etch rate may yield important information on the etching mechanism, as will be shown further.
5.2.
PHOTOETCHING
The simple model for electroless photoetching, as proposed in Section 3.2, appears to hold in the case of the InP-Fe^^ system in an acidic medium [55, 56]. The partial current densities at the rest potential were determined as follows. The illuminated n- or p-type InP electrode in an Fe^^-containing solution was held at the potential at which the net current was zero (rest potential). Then the light was abruptly shut off. The value of the current density immediately after interruption of the light yields the partial "dark" current density, corresponding to cathodic Fe^"^ reduction in the n-type case and to anodic dissolution of the InP in the p-type case (see Fig. 7). The reason the light has to be shut off abruptly is that the energy bands may shift under illumination, so that the "dark" partial current density under illumination may be different from that in darkness. An alternative procedure consists of working with chopped light (see subsequent text). The (photo)anodic partial electrical current density at rest potential j^^ was then compared to the photoetch rate, the latter being determined analytically. To enable the comparison between both quantities, the photoetch rate rgj^h was converted into an etch current density j^^^^ through the relationship y'etch = «^^etch
C79)
where n is the number of charge carriers involved in the anodic dissolution of one formula unit of InP, equal to 6 [see Eq. (71)]. Figure 12 demonstrates the good quantitative correlation between the electrical and the etch current densities and hence the validity of the simple photoetch model of two compensating partial currents. Also according to this model, the photoetch rate is expected to be higher as the onset of the anodic dissolution curve is shifted toward negative potentials and the onset of the cathodic reduction curve is shifted toward positive potentials. This prediction is fulfilled in the given case, because the photoetch rate was found to be considerably higher in a HCl medium than in H2SO4 or
254
WALTER R GOMES
. etch , . -2 / / M.A.cm
PI
-2
FIG. 12. Photoetch current density obtained form etch rate measurements, jf^, vs. partial electrical current density, jf measured at rest potential at n-InP in aqueous FeCl3 +HC1 (pH = 0). O, (111) face; A, (111) face. Reprinted from J. Electrochem. Soc. 142, I. E. Vermeir, W. P. Gomes, and P. Van Daele, pp. 3226-3232, 1995. Reproduced by permission of the Electrochemical Society, Inc.
in HCIO4, and corresponding voltammetric results show that in a HCl medium, both currents are shifted favorably as compared to H2SO4 or HCIO4 media. Considering the general reaction scheme for simple photoetching [Eq. (66)], the anodic reaction here is Eq. (71) and the cathodic reaction is Fe^++e-
Fe2+
(80)
In view of the discussion in Section 7, it is worth mentioning that at the same light intensity, the photoetch rate at the (111) face is much lower than that at the ( i l l ) face [56]. This difference between both polar faces is well known for dark etching, but rather surprisingly holds as well for photoetching in the given case. Results similar to those outlined previously were obtained for GaP in alkaline OBr~ solution [57]. The partial reactions are now Eq. (74) and the reduction of OBr~, which appears to involve the current-doubling mechanism
6
200
255
WET ETCHING OF III-V SEMICONDUCTORS
-
^
..A^dfiil
100
0
a •100
-200 Voc
yr.
.inn
-1.2
-1.1
-0.9
-0.8
-0.7
-0.6
-0.5 -0.4 Vvs.SCE l\/
-0.3
FIG. 13. Voltammograms, obtained at n-GaN in aqueous 1-mol 1"' KOH with chopped illumination. Curve a, light off; curve b, light on. V^^ is the open-circuit potential under illumination. Courtesy of I. M. Huygens, unpublished results.
(see Section 2.4): OBr- + e + H2O ^ Br* + 2 OH" Br* -> B r + h +
(81) (82)
Also in the case of (0001) n-GaN, photoetching can be achieved in aqueous alkaline-indifferent electrolyte [31]. This is visualized in Figure 13, which shows voltammetric results in 1-mol 1"^ KOH obtained with chopped illumination. Curves a and b connect the points with the light off and on, respectively. The potential indicated in the figure represents zero net current under illumination. This situation results from two mutually compensating currents, that is, the dark cathodic current and the photoanodic dissolution current, corresponding to reaction Eq. (76), the net result being the photoetching of GaN. The oxidizing agent in indifferent electrolyte is H2O and/or residual dissolved O2. This photoetching effect is essentially due to the fact that the onset of the dark cathodic j vs. V curve is situated in a potential range that is markedly positive with respect to the flat-band potential, implying a very high reactivity of conduction band electrons toward the oxidizing agent. Note that in acidic media, no such
256
WALTER P. GOMES
high reactivity of conduction band electrons occurs and hence no photoetching of GaN is observed. Presumably, a special reaction pathway exists in alkaline media through surface states, the presence and/or activity of which is somehow controlled by the surface chemistry. Due to the high chemical stability of GaN, etching recipes for this semiconductor material are scarce, so the proposed photoetching procedure constitutes an interesting possibility for wet processing of GaN in device technology. We recall, however, that it is very likely that this possibility may not exist as far as the (0001) face is concerned, because the photoanodic current does not involve dissolution of the GaN (see Section 5.1). Photoetching processes do not always consist of a simple superposition of an anodic and a cathodic partial process, and may exhibit various types of complications. First, even in the "simple" case of the photoetching of GaP single crystals in alkaline OBr~ solutions, the situation is actually more complex than depicted, because in n-type crystals, the photoetching process itself induces a hole injection reaction and hence an electroless etching effect [57]. Initially, OBr" is reduced at the GaP surface via the previously mentioned currentdoubUng mechanism. However, during the reduction of OBr~ ions, Br* radicals are formed as intermediates [cf. reaction (81)], which appear to initiate an autocatalytic reaction mechanism: surface states are formed, through which holes are injected into the valence band (at least at not too high OBr~ concentrations). These surface states, which are experimentally detected as a peak in the capacitance-potential plot [57, 58], are believed to be associated with adsorbed OBr". Furthermore, voltammetric experiments demonstrate that these surface states can be annihilated by a sufficiently large concentration of holes at the surface. The latter effect may explain why this induced electroless etching effect is not observed at p-GaP, since in this case the holes are the majority carriers. Another more complicated case is the photoetching of GaAs by several twoor multiequivalent oxidizing agents such as H2O2, Br2, OBr", OCl", S208~, and Cr03-HF [12,59-66]. The mechanisms of the etching reactions appear to be rather complex. The oxidizing agent was found to react at the semiconductor through various pathways, such as chemical etching, hole injection, electron capture followed by hole injection (current doubling), and hence photoetching. A model has been proposed in which these reactions were supposed to be mutually linked via a common precursor, that is, a chemisorbed surface complex formed by transfer of an electron from a surface bond to the oxidizing agent. A more detailed discussion on the competing reactions subsequent to the formation of the precursor will be given later in this chapter. Particularly interesting features have been observed in the photoetching of n-GaAs surfaces that are partially illuminated with laser Ught [65,66]. Photoetching occurs only at the illuminated spots with a quantum efficiency that is considerably higher than under uniform illumination (in the latter case, the quantum efficiency is usually low because the rest potential is generally situated in the onset region of the photocurrent curve; see Figure 7). The quantum
6
WET ETCHING OF III-V SEMICONDUCTORS
257
efficiency increases with the ratio of dark vs. illuminated surface area and may reach a value of 1 when this ratio is sufficiently high. It was shown experimentally that anodic dissolution proceeds in the illuminated area only, whereas electron capture by the oxidizing agent necessary to maintain charge neutrality takes place all over the surface, the reason being that the electrons are the majority carriers so they are available anywhere at the surface. For that reason, a high dark vs. illuminated area ratio leads to high quantum efficiencies. The photoetching mechanism thus amounts to the operation of a corroding shortcircuited photogalvanic element.
5.3.
ELECTROLESS ETCHING
When a given oxidizing agent etches a semiconductor at the rest potential, it can be concluded that the etching mechanism is purely electroless if, first, the reactant is observed to cause a limiting cathodic dark current at a p-type electrode (demonstrating that hole injection takes place) and if, second, no etching occurs when this limiting cathodic dark current flows [indicating that no (chemical) etching occurs when all injected holes are drawn into the bulk of the sample]. In fact, the latter criterion is unambiguous only if the hole-injection current is not limited by diffusion of the oxidizing agent, since in the opposite case, the possibility should, in principle, be considered that chemical etching is prevented because of the competition for the diffusing species by the holeinjection reaction. Because cases have been observed where the hole-injection rate is lower at the rest potential than under limiting cathodic current flow (see subsequent text), a parallel chemical etching reaction could then participate at the rest potential. Using the foregoing diagnostic criteria, it has been established that an electroless etching mechanism operates, for example, with Fe(CN)^~ at GaAs [67] and GaP [68,69] in an alkaline medium, with Fe(CN)6" at InP at pH = 14 (this high pH being necessary to avoid passivation) [70], with acidic Fe^^ solutions at GaAs [71,72] and GaP [38], and with acidic Ce"^^ solutions at GaAs [10,71,72] and GaP [38]. More examples are given in Section 5.5, pertaining to simultaneous electroless and chemical etching, and in Section 6, dealing with material-selective etching. The simple mechanism of electroless etching was presented in Section 3.3 and illustrated by Figure 8. Various factors may complicate the kinetics and mechanisms of electroless etching reactions, however, as will be explained by taking the GaP-Fe(CN)^~ system as an example [68,69]. The behavior of the GaAsFe(CN)^~ system is closely analogous. The complications are the following: First, the etch rate at pH = 13 was found to be dependent on the crystal face. More specifically, a difference was found between the two polar faces: whereas at the (111) face, the rate is kinetically controlled, at the (111) it is controlled by diffusion, either of Fe(CN)^~ (at concentrations below 0.3 mol 1"^) or of OH"
258
WALTER P. GOMES
(at Fe(CN)6~ concentrations > 0.3 mol \~^). The faster kinetics at the (111) face as compared to the ( H I ) face can be related to the observed difference in valence band position. Indeed, the flat-band potential was found to be 0.1-0.2 V more negative for the ( i l l ) face than for the ( H I ) face [68]. The same observation was made with GaAs and can be rationalized from the structural point of view: the ( H I ) face consists of Ga atoms that are triply bonded to the nearest atoms of the crystal, so that all valence electrons are used in these bonds; the ( i l l ) face consists of P atoms, again triply bonded to their neighboring atoms so that one valence electron pair remains free. This leads to a higher surface electron density at the ( i l l ) face as compared to the ( H I ) face and hence to a more negative potential, that is, to a higher position of the band edges. It is interesting to note that the difference in electron density around group III and group V atoms in the zincblende lattice has been directly visualized by recording scanning tunneling microscopy (STM) images of the (110) GaAs surface [73]. The higher position of the ( i l l ) valence band edge apparently leads to a better overlap with the empty redox levels of the (Fe(CN)^~, Fe(CN)^~) system and hence to a higher hole-injection rate. At Fe(CN)^~ concentrations below 03_mo\ ^ ^ the hole-injection rate and hence the open-circuit etch rate at the ( i l l ) face are limited by diffusion of Fe(CN)6". At higher Fe(CN)^~ concentrations, at which the hole-injection rate under cathodic polarization exceeds the diffusion-limited value of the anodic dissolution rate, the etch rate at rest potential is determined by the anodic partial current density and hence controlled by the diffusion rate of 0H~ ions. Second, from rotating ring-disk experiments, it was found that the holeinjection rate at the ( H I ) face is not potential-independent, but is lower at anodic than at cathodic potentials [in the latter case, it corresponds to the cathodic Fe(CN)^~ reduction current density]. An example is shown in Figure 14. This result implies that the plateau value of the cathodic current density cannot be used as a measure for the etch rate at the rest potential. The reason for this decrease in the hole-injection rate at higher potentials should be sought along the same lines as before. Indeed, when the potential is increased so that the cathodic current decreases, part of the injected holes are used in the oxidation of the semiconductor. In this reaction, positively charged decomposition intermediates such as Xf are formed (see Section 5.1). Thus, positive charges are accumulated at the crystal surface, which causes a downward shift of the band edges and, hence, a poorer overlap with the empty redox levels. Whereas at the (111) face, the reaction is kinetically controlled, a lower hole-injection rate results. A decrease of the hole-injection rate when the electrode potential is increased is commonly observed in III-V semiconductor electrochemistry. For example, in the case of GaAs-Fe^^, it was shown by means of rotating ring-disk experiments that the reduction rate of Fe^^ in the anodic potential region is drastically decreased in comparison to that in the cathodic potential region. In the case of GaAs-Ce"^"^, on the contrary, holes are injected at a diffusion-limited rate over
6
r2 /7 m A.cm' 0
n— '—1
u-^:???^- ^=T^
/^"[y^''
-0.4 r
259
WET ETCHING OF III-V SEMICONDUCTORS
\
1/
T3 - 0 . 8 o JC
-u
I
-1.6h J
!
-1.5
1
1
1
- L .— L -
L.
-1.0
1
1
—L-J
VvsSOEIM
-0.5
FIG. 14. Current density vs. potential curves at (111) n-GaP in aqueous 10~^-mol 1"^ K3Fe(CN)6 + 0.1-mol r^ KOH. —, net current density; - - -, partial current density due to reduction of Fe(CN)6". Reprinted from Electrochim. Acta 35, H. H. Goossens, I. E. Vermeir, F. Vanden Kerchove, and W. P. Gomes, pp. 1351-1358, ©1990, with permission from Elsevier Science.
the whole potential region [71, 72]. This difference in behavior is explained on the basis of the positions of the redox levels concerned. Whereas the (Ce^^^, Ce^"^) standard Fermi level lies well below the valence band edge, the (Fe^^, Fe^"^) standard Fermi level is located close to the valence band edge, so that the overlap of the unoccupied (Fe^^) levels with the valence band, and hence the Fe^"^ reduction rate, is very sensitive to the downward displacement of the valence band edge under anodic polarization caused by the accumulation of holes at the surface.
5.4.
CHEMICAL ETCHING
In a chemical etching mechanism, no free charge carriers (no valence band holes) are involved. Whereas most chemical etchants for III-V compound semiconductors are oxidizing agents, just as electroless etchants are, it is essential to distinguish between the two dark etching mechanisms. The obvious way to do this is by observing the cathodic behavior of the p-type semiconductor in darkness in the presence of the etchant: if no cathodic reduction of the etchant (hence no hole injection) occurs, the etching mechanism cannot be electroless and, hence, must be chemical. In addition to allowing a decision on whether the etching mechanism is electroless or chemical, electrochemical measurements also yield essential information on chemical etching reactions, that is, on the stoichiometry of the surface reaction, on the problem of whether the mechanism is concerted (synchronous) or sequential, etc. This information will be explained by taking a study on the
260
WALTER R GOMES
etching of GaP by methanolic bromine solution [50,74] as an example. This solution is commonly used for the etching various III-V and II-VI semiconductor materials. Considering that bromine-methanol mixtures may be hazardous at high Br2 concentrations [75] and that methanol is toxic, one of the main objectives of this study was to find the possible advantages of using methanol instead of water as the solvent for bromine-based etching solutions. Measurements were performed at the ( i l l ) and the (111) GaP face. At the rest potential, the etch rate at the (111) face appeared to be diffusion-limited, whereas that at the (111) face was low and independent of the rotation rate of the sample. The etch rate under given circumstances was the same for n- and p-type samples. The etch rate and the current density at the ( i l l ) face were studied at an n-type ( i l l ) GaP electrode as a function of electrode potential. The results are represented in Figure 15. A cathodic diffusion-controlled plateau is observed at sufficiently negative potentials. In this potential region, the etch rate is low. In Figure 15, the etch rates have been converted into etch current densities, based on Eq. (79), in which n has been put equal to 6. The underlying hypothesis here is that, similar to the anodic dissolution, the oxidation of GaP by Br2 is a 6-equivalent redox reaction. When the electrode potential is increased, the cathodic current decreases and, simultaneously, the etch rate increases. Finally, at sufficiently positive potentials, a low limiting anodic current is observed and
rofe/mA.cm"2
^^^
K
\
•
J
2.5i
L
-_^,—•
-5 -0.8
-0.6
-O.A
-0.2
0.2
0.A
Vvs.AglAgCilCr/si FIG. 15. Combined etch rate and current density vs. potential plots at ( i l l ) n-GaP in methanolic 4 x 10~^-mol r^ Br2+0.25-mol \~^ LiCl. x, etch rate (expressed as an etch current density); •, electrical current density. Reprinted from J. Electrochem. Soc. 140, K. Strubbe and W. P. Gomes, pp. 3294-3300, 1993. Reproduced by permission of The Electrochemical Society, Inc.
6
WET ETCHING OF III-V SEMICONDUCTORS
261
the etch rate reaches a Umiting value that is proportional to the square root of the rotation speed of the electrode. The sum of the etch rate and the absolute value of the electrical current density is found to be constant over the entire potential range. Rotating ring-disk experiments show that the Br2 consumption at the GaP disk is potential-independent and equal to the diffusion-controlled value. Measurements at the p-type ( i l l ) face show that Br2 is not reduced cathodically in darkness, which allows us to exclude electroless etching. Combined current-potential and impedance measurements reveal that the anodic dissolution current starts at a considerably lower surface hole concentration in the presence of Br2 than in an indifferent electrolyte solution. Also this phenomenon points to a chemical etching mechanism, since it can be explained by assuming that in the presence of Br2, the first (slow) anodic dissolution steps are replaced by faster chemical steps; the faster subsequent hole-capture steps can then occur at a higher rate for the same hole concentration than in a totally electrochemical mechanism (see Section 5.1). The fact that the cathodic limiting current density j^y^^^ in Figure 15 is found to be equal to the anodic limiting etch current density yetch,iim confirms that this chemical etching process is a 6-equivalent redox reaction, as assumed. Indeed, because the cathodic reduction of Br2 is 2-equivalent, j^ ,jn, can be written as l,,^
(83)
= 2FJ^,^
where J^^^ is the diffusion flux of Br2 toward the electrode surface. Putting Eq. (83) equal to yet,h,iim = ^Fr,,^hMm yields ^Br2 — ^ '"etch, lim
\^^)
that is, the number of moles of Br2 arriving at the surface by diffusion is 3 times the number of moles of GaP consumed. Taking into account chemical considerations [76,77], the reaction equation can then be proposed as O
t
GaP + 3 Br2 + 3 CH3OH -> GaBr3 + (CH30)2 P—H + CH3Br -h 2 HBr (85) Note that the experiments mentioned yield direct information on the stoichiometry of the surface reaction and that it cannot be excluded that (CH30)2PHO is further oxidized by Br2 homogeneously in solution. The fact that a small limiting current due to the presence of Br2 is observed anodically at ( i l l ) n-GaP allows us to conclude that the etching mechanism is not concerted, but sequential. Indeed, this anodic current can be attributed to electron injection by decomposition intermediates of GaP; see Section 5.1. If the chemical etching mechanism were synchronous, no such intermediates would be formed. More details on the reaction mechanism can be obtained by studying the relationship between this limiting anodic current and the consumption rate
262
WALTER P. GOMES
|;^l/mA.cm"2
FIG. 16. Anodic j^ vs. cathodic \j^\ current density at ( i l l ) n-GaP in methanolic LiCl +various Br2 concentrations; V^ and V^ equal 0.4 and -0.8 V vs. Ag-AgCl-Cl", respectively. The closed symbols refer to 0.25 mol \~^ LiCl; the open ones refer to 4-mol 1"' LiCl. Reproduced from J. Electrochem. Soc. 140, K. Strubbe and W. P Gomes, pp. 3294-3300, 1993. Reproduced by permission of The Electrochemical Society, Inc.
of Br2. As appears from the considerations mentioned previously for the latter, the cathodic Hmiting current density can be taken as a measure. Figure 16 shows the measured relationship between the anodic and the cathodic limiting current densities at two different concentrations of the indifferent electrolyte (LiCl). The relationship appears to be linear in both cases and the slope is lower at higher LiCl concentration. These data can be interpreted as follows [74]. Reaction of an unbroken surface bond with Br2 first leads to the formation of a mobile surface intermediate X^: GaP + Br2 -> X+ + Br" -f Br*
(86)
6
WET ETCHING OF III-V SEMICONDUCTORS
263
(see Section 5.1). In the next step, the positively charged X^ intermediate is converted into an immobile neutral intermediate Xj—A by the reaction
K X+ + A- ^ X|—A
(87)
In principle, A~ may be either CI", present in the indifferent electrolyte solution, or Br~, formed during the chemical etching process [Eq. (86)], or methanolate [in which case a proton is released in reaction (87)]. Whereas P is more electronegative than Ga, the A~ anion is localized on the gallium. We assume that the equilibrium (87) is positioned to the right, so that few mobile Xf intermediates are present. The neutral species Xj—A is then further oxidized by the Br* radical to the intermediate X^: Xi—A + Br*-^X2+ + Br-
(88)
The anodic current increase is attributed to electron injection from Xf into the conduction band of the semiconductor, occurring in parallel with reaction (87): X^-^X^'^+e-
(89)
Whereas the chemical structure of the intermediate X\^ is obviously different from that of X j , step (89) must be followed by other chemical reaction steps. However, since these steps do not influence the competition between reactions (87) and (89), they are not taken into account in the kinetic derivation. Further oxidation of the semiconductor is then assumed to proceed in subsequent steps, in which two more Br2 molecules per GaP formula unit are consumed. The small excess of Br* at the surface [formed during reaction (86), but not used up in reaction (89), since X^ is converted into X^^ in reaction (89) without consumption of Br*] is assumed to desorb from the surface and to react further in solution: Br* ^
(Br-),„„
(90)
By a standard steady-state kinetic derivation, the proposed mechanism is shown to lead to the following relationship between the anodic current density j^ and the cathodic current density \j^\\ i =
^
ek,k_XcA
.g^.
in which c^ represents the concentration of A. Equation (91) predicts a linear relationship between j^ and \j^\ for the ( i l l ) face with a positive intercept on the j^ axis, as is found experimentally (Fig. 16). Equation (91) also predicts that an increase of the A concentration leads to a
264
WALTER P. GOMES
decrease of the slope of the j^ vs. |7^| curve. In Figure 16, we see that the slope of the j^ vs. \j^\ plot in 4-mol 1"^ LiCl is lower than in 0.25-mol T^ LiCl, indicating that step (87) occurs between X^ and a CI" ion in that case. In practical etching conditions in which no LiCl is present, the anion reacting with Xf is presumably the Br~ ion formed in the first step. When comparing the results of this study on the etching of GaP by methanolic Br2 to those by aqueous Br2 [49], it appears that both processes occur quite similarly. Hence, there seems to be no clear advantage of using methanol as the solvent, except maybe for the fact that the solubility of bromine is higher in methanol than in water. However, especially at high concentrations, bromine solutions in methanol are hazardous. Moreover, it is possible to increase the solubility of bromine in aqueous solutions considerably by complexing it with bromide to form Br^ and Br^. Other oxidizing substances besides Br2 that usually act as chemical etchants for III-V compound semiconductors are CI2, I2, OCP, H2O2, and HIO3. Most often it is found that the etch rate at the (100) and ( i l l ) faces is comparable, whereas that at the (111) face is considerably lower. The difference in kinetics between both polar faces perpendicular to the [111] axis has been attributed to the fact that electrophilic reactants are concerned and that the electron density at the ( i l l ) face is higher than that at the (111) face, due to the difference in orientation of the group III atom-group V atom surface dipole [78,79]. By exception, the chemical etch rate of InP by HIO3 is diffusion-controlled even at the (111) face [51]; otherwise, the electrochemical and etching behavior of this system is quite similar to that of GaP-methanolic Br2 described previously. Kelly and co-workers performed extensive studies on the etching of GaAs in acidic solutions containing CI2, Br2, I2, or H2O2 [60,61] and in alkaUne solutions containing the hypohalites C10~ and BrO~ [62]. Their observations will be examplified by the case of GaAs-H202. An enhanced anodic dark current at n-type electrodes in the presence of the oxidizing agent was observed, as was an interdependence between the rate of chemical etching and that of cathodic reduction. Indeed, in darkness, the cathodic current at a p-type electrode is low, indicating weak hole injection by the oxidizing agent. Under the same circumstances, GaAs is etched at a potential-independent rate. When the p-type electrode is illuminated, a photocurrent is observed that involves the currentdoubling mechanism (see Section 2.3), and the etch rate decreases by a corresponding amount. At high light intensities, the etching is completely suppressed cathodically even though no significant depletion of the oxidizing agent at the surface occurs. The conclusion was reached that hole injection, electron capture, and chemical etching are coupled through a common precursor: the oxidizing agent is thought to adsorb on the semiconductor surface under injection of a hole in a surface bond, so that an electron-deficient species of the type as depicted in Figure 17 is formed. Electronically, this species acts as a filled surface state, the electron of which can be injected into the conduction band or can participate in competing surface reactions (see the scheme of Figure 17, which
6
WET ETCHING OF III-V SEMICONDUCTORS
H^Oo •
265
Ga:As'