ADVANCED SOLID STATE PHYSICS Philip Phillips University of Illinois, Urbana-Cham;The Advanced Book Program All rights reserved. Printed in the United States of America. 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. Copyright © 2003 by Westview Press, A Member of the Perseus Books Group , Westview Press books are available at special discounts for bulk purchases in the United States by corporations, institutions, and other organizations. For more information, please contact the Special Markets Department at The Perseus Books Group, 11 Cambridge Center, Cambridge, MA 02142, or call (617) 252-5298. Published in 2003 in the United States of America by Westview Press, 5500 Central Avenue, Boulder, CO. 80301-2877, and in the United Kingdom by Westview Press, 12 Hid’s Copse Road, Cumnor Hill, Oxford OX2 9JJ. Find us on the World Wide Web at www.westviewpress.com A cataloging-in-publication data record for this book is available from the Library of Congress. ISBN 0-8133-4015-2 (HC) 0-8133-4014-4 (Pbk.) The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials, Z39.48-1984. 10987654321For AngelikiContents Foreword ix Preface xi Chapter 1 Noninteracting Electron Gas 1 Problems 8 Chapter 2 Born-Oppenheimer Approximation 9 2.1 Basic Hamiltonian 9 2.2 Adiabatic Approximation 10 2.3, Reduced-Electron Problem 15 References 15 Chapter 3 Second Quantization 17 3.1 Bosons 17 3.2 Fermions 19 3.3. Fermion Operators 20 Problems 24 References 25 Chapter 4 Hartree-Fock Approximation 27 4.1 Noninteracting Limit 27 4.2 Hartree-Fock Approximation 29 4.3 Diagrams 32 Problems 33 References 33 Chapter 5 Interacting Electron Gas 35 5.1 Uniform Electron Gas 36 5.2 Hartree-Fock Excitation Spectrum 39 5.3. Cohesive Energy of Metals 41 5.4 Summary 49 Problems 49 References 50vi Contents Chapter 6 Chapter 7 Chapter 8 Chapter 9 Local Magnetic Moments in Metals 53 6.1 Local Moments: Phenomenology 53 6.2 Mean-Field Solution 57 6.3 Summary 69 Problems 69 References 70 Quenching of Local Moments: The Kondo Problem 71 7.1 Kondo Hamiltonian 74 7.2 WhyIsJ Negative? 75 7.3 Scattering and the Resistivity Minimum 80 7.4 Electron-Impurity Scattering Amplitudes 88 7.5 | Kondo Temperature 94 7.6 Poor Man’s Scaling 97 7.7 Summary 106 7.8 Appendix: The Schrieffer-Wolff Transformation 106 Problems 112 References 112 Screening and Plasmons 115 8.1 Thomas-Fermi Screening 115 8.2 Plasma Oscillations 118 8.3. Linear-Response Theory 121 8.4 Dielectric-Response Function 127 8.5 Stopping Power of a Plasma 141 8.6 Summary 146 Problems 146 References 147 Bosonization 149 9.1 Luttinger Liquid 150 9.2. Pair Binding: Can Electrons Do It Alone? 165 9.3. Excitation Spectrum 167Chapter 10 Chapter 11 Contents vii 9.4 Summary 172 Problems 173 References 173 Electron-Lattice Interactions 175 10.1. Harmonic Chain 175 10.2 Acoustic Phonons 178 10.3. Electron-Phonon Interaction 179 10.4 Ultrasonic Attenuation 184 10.5 Electrical Conduction 186 10.6 Hydrodynamic Limit: Phonon Drag 197 10.7 Sound Propagation 199 10.8 Summary 202 Problems 203 References 203 Superconductivity 205 11.1 Superconductivity: Phenomenology 206 11.2 Electron-Phonon Effective Interaction 214 11.3. Model Interaction 217 11.4 Cooper Pairs 219 11.5 Fermi-Liquid Theory 224 11.6 Pair Amplitude 228 11.7. BCSGround State 234 11.8 Pair Fluctuations 237 11.9 Ground State Energy 239 11.10 Critical Magnetic Field 243 11.11 Energy Gap 244 11.12 Quasi-Particle Excitations 247 11.13. Thermodynamics 251 11.14 Experimental Applications 256 11.15 Josephson Tunneling 271 11.16 Summary 273 Problems 274 References 275viii Contents Chapter 12 Chapter 13 Chapter 14 Index 375 Localization: The Strong, the Weak, and the Defiant 277 12.1. Primer on Localization 278 12.2 Strong Localization: The Anderson Transition 281 12.3. Scaling Theory 285 12.4 Weak Localization 291 12.5 The Defiant: Exceptions to Localization 299 12.6 Summary 313 Problems 313 References 315 Quantum Phase Transitions 319 13.1 Quantum Rotor Model 322 13.2 Scaling 324 13.3. Mean-Field Solution 328 13.4 Landau-Ginsburg Theory 335 13.5 Transport Properties 340 13.6 Experiments 342 13.7. Summary 345 Problems 346 References 347 Quantum Hall Effect 349 14.1. What Is Quantum about the Hall Effect? 349 14.2. Landau Levels 354 14.3. The Role of Disorder 357 14.4 Currents at the Edge 359 14.5 Laughlin Liquid 364 14.6 Summary 372 References 372Foreword The problem of communicating, in a coherent fashion, recent develop- ments in the most exciting and active fields of physics continues to be with us. The enormous growth in the number of physicists has tended to make the familiar channels of communication considerably less effec- tive. It has become increasingly difficult for experts in a given field to keep up with current literature; the novice can only be confused. What is needed is both a consistent account of a field and the presentation of a definite point of view concerning it. Formal monographs cannot meet such a need in a rapidly developing field, while review articles seem to have fallen into disfavor. Indeed, it would seem that the people who are most actively engaged in developing a given field are the people least likely to write at length about it. The Frontiers in Physics series was conceived in 1961 in an effort to improve the situation in several ways. Leading physicists frequently give lectures, graduate seminars, or graduate courses in their special fields of interest. Such lectures serve to summarize the present status of a rapidly developing field and may well constitute the only coher- ent account available at the time. One of the principal purposes of the Frontiers in Physics series is to make notes on such lectures available to the wider physics community. As Frontiers in Physics has evolved, a second category of book, the informal text or monograph—an intermediate step between lecture notes and formal text or monographs—has played an increasingly im- portant role in the series. In an informal text or monograph, an author has reworked his or her lecture notes into a coherent summation of a newly developed field, complete with references and problems, suitable for either classroom teaching or individual study. Philip Phillips’ Advanced Solid State Physics is just such a book. The author provides a much-needed lucid introduction to the signifi- cant topics at the frontiers of condensed matter physics, while at thex Foreword same time providing, in clear pedagogical fashion, background mate- rial on the well-established fundamentals of the subject. His success in integrating introductory and advanced topics makes this volume partic- ularly appealing to graduate students and advanced researchers alike and should make it of considerable interest, as well, to mature scholars in other subfields of science who wish to obtain an overview of the very considerable intellectual challenge of contemporary solid state physics. It gives me great pleasure to welcome Philip Phillips to the Frontiers in Physics series. DAVID PINESPreface Solid state physics continues to be the most rapidly growing subdis- cipline in physics. As a result, entering graduate students wishing to pursue research in this field face the daunting task of not only mas- tering the old topics but also gaining competence in the problems of current interest, such as the fractional quantum Hall effect, strongly correlated electron systems, and quantum phase transitions. This book is written to serve the needs of such students. I have attempted in this book to present some of the standard topics in a way that makes it possible to move smoothly to current material. Hence, all the inter- esting topics are not presented at the end of the book. For example, immediately after the first 50 pages, Anderson’s analysis of local mag- netic moments is presented as an application of Hartree-Fock theory; this affords a discussion of the relationship with the Kondo model and how scaling ideas can be used to uncloak low-energy physics. As the key problems of current interest in solid state involve some aspects of electron-electron interactions or disorder or both, I have focused on the archetypal problems in which such physics is central. However, only those problems in which there is a consensus view are discussed extensively. In addition, I have placed the emphasis on physics rather than on techniques. Consequently, I focus on a clear presentation of the phenomenology along with a pedagogical derivation of the relevant equations. A key goal of the detailed derivations is to make it possible for the students who have read this book to immediately comprehend research papers on related topics. A key omission in this book is mag- netism beyond the Stoner criterion and local magnetic moments. This omission has arisen primarily because the topic is adequately treated in the book by Assa Auerbach. Most of this book grew out of lectures I have given in the one- semester advanced solid state physics class (Physics 490) taught herexii Preface at the University of Illinois. In teaching this course, I have relied on the advice of those who taught the course before me, in partic- ular my colleague Gordon Baym. Gordon had a significant impact in shaping this book: we have had numerous stimulating conversa- tions, he collaborated in extensively revising the first seven chap- ters of this book, and he shared with me his lecture notes, which influenced the presentation of the material for Chapters 1, 2, 4, 5, and parts of 8 and 11, in particular the derivative of the polarization function (Chapter 8), the pair amplitude formulation of T, (Chapter 11), and sound propagation in Chapter 10. Numerous others have been instrumental in the writing of this book. I start by acknowledging all the students in Physics 490 during the semesters Spring ’94, 95, 97, and ’98, who labored through various unfinished versions of the chapters that constitute this book. It is for these students that this book is written. While all of these students have pointed out typos or logical inconsistencies over the years, four stand out in particular for their copious and careful suggestions and revisions: Johannes Walcher, Julian Velev, Adrian Gozar, and Ke Dong. I also thank Harry Westfahl, who has been a constant sounding board and has been most willing to read and correct vast sections of this book, in par- ticular Chapters 8, 9, 11, and 13. Harry also assisted in the drawing of several figures. In a similar vein, I thank Revaz Ramazashvili, who care- fully read Chapters 6 and 12 and convinced me that a much-expanded version of the latter was necessary. I would also like to thank David Pines for providing the final encouragement I needed to write the chap- ter on quantum phase transitions. This chapter was written last and would truly not be included here had it not been for a conversation I had with David on what might be lacking from the book. In this context, I thank my student Denis Dalidovich for his numerous insights, which have shaped my approach to this field. Ali Yazdani and his students also offered useful comments on the Kondo and localization chapters. In addition, George Paroanou, Michael Stone, and Bob Laughlin pro- vided a sound critique of the quantum Hall chapter. I wish to thank Greg Whitlock for his masterful copyediting of the book as well as the staff at Publication Services for their willingness to accommodate my numerous revisions, especially Jason Brown, Ben Coblentz, and Susie Yates. I would also like to thank Michael Baym for the caring effort he took in drawing the figures in Chapters 4 through 9.4, 10 through 11, Figures 12.3, 12.6, and those in Chapter 14. Regarding the typeset- ting and layout of the manuscript, I have depended heavily on Craig Copi (Physics Department, Case Western Reserve University), who wrote all of the LaTeX macros that are responsible for the design ofPreface xiii each chapter. Craig also prepared the Table of Contents and the Index. I also thank Damian Menscher for typesetting assistance and Sandy Chancy and April Orwick for learning LaTeX and typing the initial versions of Chapters 1-7 and 9-11. Finally, I thank Nathan Hearne for scanning in and tweaking the experimental figures into the final format they now have. Above all, my enduring thanks go to Angeliki Tzanetou, who has encouraged me to stay focused on this manuscript, when its completion and ultimate utility seemed doubtful, at best. I have inflicted many sen- tences of this book on her and she has always told me when they were unclear. It is to her that this book is dedicated.—-l1- Noninteracting Electron Gas To a surprising extent, one can understand the elementary prop- erties of metals in terms of noninteracting electrons and phonons- quantized lattice vibrations. For example, the low-temperature specific heat of a metal is the sum of a term linear in the temperature, 7, from the electrons and a term proportional to T> from the phonons. This result follows from a noninteracting particle picture. The electrical conductivity limited by nonmagnetic impurity scattering is also well described by a noninteracting electron gas. In addition, from a knowl- edge of single-electron band theory, one can qualitatively discern the differences between metals, insulators, and semiconductors. The re- markable success of the noninteracting model is paradoxical because electrons and ions strongly interact both with themselves and with one another. Along with its successes, the noninteracting picture has colos- sal shortcomings, most notably its inability to describe old problems, such as cohesive energies, superconductivity, magnetism, and newer phenomena, such as the Kondo and fractional quantum Hall effects. We first review the physics of the noninteracting electron gas. It is only after we develop methodology for dealing with electron interac- tions that we can lay plain the reasons why the noninteracting model works so well. Electrons in metals are quantum mechanical particles with spin h/2, obeying Fermi-Dirac statistics. The Hamiltonian of a single elec- tron is #2/2m where pis the electron momentum (operator) and m the electron mass. Its eigenstates are plane waves of the form efpr/h/) (7 times a spinor that specifies the electron spin projection on a con- venient axis (usually 2), ha/2 where o = +1; here V is the sys- tem volume. The Hamiltonian (operator) for N such noninteracting2 Chapter1 Noninteracting Electron Gas electrons, gIR (1.1) is simply the sum of the kinetic energies of the individual particles. In this case, the eigenstates are products of the occupied single-particle plane-wave states. Each plane-wave state can be occupied at most by one electron of a given spin. We label these eigenstates by the distri- bution function fpo, which is 1 if the single-particle momentum-spin state is occupied and 0 otherwise. In the ground state, the lowest N /2 single-particle states are doubly occupied with electrons of opposite spin. Consequently, in the ground state (temperature T = 0), the dis- tribution function is foo = (uo — p?/2m), (1.2) where @(x) is the Heaviside function, @(x > 0) = 1, and 0 otherwise. Here, fg is the zero-temperature electron chemical potential, which in this case is simply the Fermi energy, the energy of the highest occupied state, p2/2m, where pr is the electron Fermi momentum. The Fermi temperature Tp equals yo/kp. In terms of fp, the total number of electrons is given by N => foo. (13) Bo In the ground state, we can replace the sum by an integral and find the electron density at T = 0: ae te) = 0) 2{" dp Dp =2> = P= . (14) p 3723 The average interparticle spacing is essentially the radius, re, of a sphere containing a single electron. 4 are 3 ne = 1. (1.5) Thus, from Eq. (1.4), the scale for the interparticle separation is Om\> fh h = (— — = 1.92—, 1.6 te ( 7) A = 190% (1.6)Chapter 1 Noninteracting ElectronGas 3 which is on the order of the lattice spacing. For re ~ 1A, we find that the Fermi velocity vp = pp/m ~ h/mre ~ 108 cm/s ~ c/300, where c is the speed of light. (Relativistic effects are generally not important for the motion of electrons in the ground state.) It is conventional to work with the dimensionless ratio r; = re/ a0, where ay = #2 /me? is the Bohr radius; this quantity provides a measure of the electron density. The dense limit corresponds to rs; < 1 and the dilute regime to rs >> 1. In metals, rs varies between 2 and 6. Listed below are values of r; for the alkali metals. Li Na K Rb Cs rs 3.25 3.93 486 5.2 5.62 Cesium is, in fact, the most dilute of all metals. It is this large value of rs that is responsible for the inhomogeneities in the density of Cs. In Chapter 5, we will discuss further physics associated with large rs, such as eventual formation of a Wigner crystal. We can also use Eq. (1.4) to solve for the zero-temperature chemical potential jg 9/ by = BP PPT nel”. (1.7) For Na, ao = 3.1 eV; typically in a metal, up ranges between 1 and 5 eV. The total energy of the system is the sum over the occupied states weighted by the single-particle energies, ep = p /2m: E => ehe. (8) Po At T = 0, the energy is given by 5 PrV 3 Eq = 5 = = Neo 19 °* Toman 5 MO (19) From the thermodynamic relation P = —(dE/dV)7y, we find the pressure in the ground state, Py = 2yigne /S. This quantity is of order 10° atm and arises entirely from the exclusion principle between the particles. At finite temperature, we define the distribution function that ranges between 0 and 1 and measures the average occupation of the single-particle states. For a system in equilibrium at chemical