superconductivity by kitterson & song
TRANSCRIPT
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 1/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 2/511
- J ,
This is a text on the subject of superconductivity, an area of intense research
activity worldwide. The book is in three parts: the first deals with
phenomenological aspects of superconductivity, the second with the microscopic
theory of uniform superconductors, and the third with the microscopic theory of
nonuniform superconductors.
The first part of the book covers the London, Pippard, and
Ginzburg-Landau theories, which are used to discuss a wide range of phenomena
involving surface energies, vorticity, the intermediate and mixed states,
boundaries and boundary conditions, the upper critical field in bulk, thin-film,
and anisotropic superconductors, and surface superconductivity. The second part
discusses the microscopic theory of Bardeen, Cooper, and Schrieffer. Finite
temperature effects are treated using the Bcgoliubov-Valatin transformation. The
theory is used to discuss quasiparticle tunneling and the Josephson effects from a
microscopic point of view. The final part of the book treats nonuniform
superconductors using the Bogoliubov-defiennes approach with which it is
possible to extract many important results without invoking Green's function
methods.
This text will be of great interest to graduate students taking courses in
superconductivity, superfluidity, many-body theory, and quantum liquids. Itwill
also be of value to research workers in the field of superconductivity.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 3/511
Superconductivity
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 4/511
Superconductivity1. B. KETTERSON and S. N. SONGNorthwestern University
CAMBRIDGEUNIVERSITY PRESS
...-----.----..--.-~R U H R -U N lV E R S :T i 'T , - , . . : ' . '
Fak. f. Phys':':~; "', ,p.-:-,.;' i'" ~.
In",v:r~~.;;'" 6 0 ~ , · , 5 : s .
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 5/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 6/511
PUBLISHED BY TIlE PRESS SYNDIC."TE OFTHE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge CB21RP, United Kingdom
CA:VIBRIDGE UNIVERSITY PRESS
The Edinburgh Building, Cambridge CB2 2RU, LK http://www.cup.cam.ac.uk
40 West 20th Street, New York. NY 10011-4211, USA httpr/www.cup.org
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
< £ ) Cambridge University Press 1999
This hook is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 1999
Printed in the United Kingdom at the University Press, Cambridge
Typeset in 9.5/13pt Times New Roman [VN]
A catalogue recordfor this hook is available/rom the British Library
Library a/Congress Casaloquinq in Publication data
Ketterson, J. B. (John Boyd)
Superconductivity - J. B. Ketterson and S. N. Song.p. ern.
ISBN 0--521-56295-3 (hardcover). - ISBN 0-521-56562-6 (pbk.)
1. Superconductivity. 2. Superconductors. 1. Song, S. N.
(Shengnian N.) II. Title.
QC611.95.K48 1998
621.3'5-dc21 97-3060 CTP
ISBN 0 521 562953 hardback
ISBN 0 521 565626 paper hack
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 7/511
Contents
Preface Xlii
Part I Phenomenological theories of superconductivity
1 Introduction
2 The London-London equation 5
3 Pippard's equation 8
4 Thermodynamics of a Type I superconductor 1 2
5 The intermediate state 1 5
6 Surface energy between a normal and a supcrconducting metal 1 7
7 Quantized vorticity 1 9
8 Type II superconductivity 2 4
8 . 1 Magnetic fields slightly greater than H c1 2 4
8 . 2 The region Hc1 < < H < < Hc2 2 7
8 . 3 Microscopic magnetic probes of the mixed state 2 9
9 The Ginzburg-Landau theory 3 1
9 . 1 Basic equations 3 1
9 . 2 Gauge invariance 3 5
9 . 3 Boundaries and boundary conditions 3 6
10 The upper critical field of a Type II superconductor 4 3
11 The anisotropic superconductor 4 5
1 2 Thin superconducting slabs 4 7
13 Surface superconductivity 5 0
1 4 The Type II superconductor for H just below H c2 5 2
1 5 The Josephson effects 5 9
1 5 . 1 The Josephson equations 5 9
1 5 . 2 Magnetic field effects: the two-junction SQUID 6 1
1 5 . 3 The extended Josephsonjunction 6 4
1 5 . 4 Effect of an applied rf field 6 5
1 5 . 5 The resistively shunted junction (RSJ) model 6 6
1 5 . 6 The rf biased SQUID 73
1 6 The Josephson lattice in ID 77
1 7 Vortex structure in layered superconductors 8 2
1 7 . 1 3D anisotropic London model 8 2
vii
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 8/511
VIII ~onrenrs
17.2 Lawrence-Doniach model 85
17.3 Vortex struct ure in a 2D film 87
18 Granular superconductors: the Josephson lattice in 2D and 3D 91
19 Wave propagation in Josephson junctions, superlattices, and arrays 95
19.1 Wave propagation in a junction 95
19.2 Wave propagation in a superlattice 104
19.3 The Josephson transmission line 108
20 Flux pinning and flux motion 110
20.1 Nonideal Type II superconductors 110
20.2 Microscopic description 111
20.3 The Lorentz force 113
20.4 Pinning centers and pinning forces 115
20.4.1 The core interaction 11620.4.2 Surface magnetic interaction 117
20.4.3 Summation of the pinning forces 118
20.5 The equation of motion 118
20.6 The critical state 120
20.7 The elastic constants of a flux-line lattice 123
20.8 Collective flux pinning 127
20.9 Mechanisms of flux motion 131
20.10 Relaxation of the magnetization with time 134
20.11 Phase diagram of high To oxide superconductors 137
21 Time-dependent G~L theory 140
22 Fluctuation effects 144
22.1 The Ginzburg criterion 144
22.2 The diamagnetic susceptibility for T> t; 147
22.3 Paraconductivity for T> t; 149
23 G~L theory of an unconventional superfluid 152
23.1 The order parameter of an unconventional superfluid 154
23.1.1 Superfluid 3He: isotropic p-wave pairing 154
23.1.2 Isotropic d-wave pairing 157
23.2 Crystal-field and spin-orbit effects 158
23.3 TheG L theory of an unconventional superfluid 163
23.3.1 G L theory for 3He 163
23.3.2 G L theory for an isotropic d-paired superfluid 168
23.3.3 Unconventional G~L theory in metals 168
23.4 Inhomogeneities in the order parameter 171
23.5 Collective modes in an unconventional superfluid 172
23.5.1 Collective modes of 3He B 173
23.5.2 E state collective modes 177
24 Landau Fermi liquid theory 180
24.1 Basic equations 180
24.2 Collisionless collective modes 187
24.2.1 The kinetic equation 187
24.2.2 Collisionless longitudinal zero sound 189
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 9/511
24.2.3 Collisionless transverse zero sound
24.2.4 Collisionless spin waves
190
191
Part" The microscopic theory of a uniform superconductor
25 The Cooper problem: pairing of two electrons above a filled
Fermi sea 195
26 The BCS theory of the superconducting ground state 199
27 Elementary excitations: the Bogoliubov Valatin transformation 208
28 Calculation of the thermodynamic properties using the
Bogoliubov-Valatin method 212
29 Quasiparticle tunneling 216
30 Pair tunneling: the microscopic theory of the Josephson effects 222
31 Simplified discussion of pairing mechanisms31.1 The electron-phonon interaction
31.2 The spin fluctuation mechanism
32 The effect of Coulomb repulsion on To
33 The two-band superconductor
34 Time-dependent perturbations
34.1 Ultrasonic attenuation
34.2 Nuclear spin relaxation
230230
234
240
243
245
248
248
35 Nonequilibrium superconductivity 251
35.1 Elastic and inelastic scattering processes 251
35.2 Quasiparticle and phonon populations in a noncquilibrium
superconductor 254
Part '" Nonuniform superconductivity
36 Bogoliubov's self-consistent potential equations
37 Self-consistency conditions and the free energy
257
262
38 Linearized self-consistency condition and the correlation function 265
38.1 Treating the gap function as a perturbation 265
38.2 Relation to a correlation function 267
39 Behavior of the correlation function in the clean and dirty limits 272
39.1 A simple model for the clean limit 272
39.2 The dirty limit 273
39.3 The general case
40 The self-consistency condition
40.1 The dirty limit at zero magnetic field
40.2 The dirty limit at finite magnetic field
40.3 The clean limit at zero magnetic field
41 Effects involving electron spin
41.1 Spin generalized Bogoliubov equations
4l.2 The density matrix
4l.3 The linearized gap equation
41.4 Spin-dependent potentials
275
277
277
279
282
284
284
286
288
289
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 10/511
x Contents
41.5 Paramagnetic impurities, electron paramagnetism, and
spin orbit coupling 290
41.6 The Fulde-Ferrcll state 293
41.7 Gapless superconductivity 295
42 Boundary conditions 298
43 The proximity effect at zero field 304
43.1 Governing equations 304
43.2 The thin-film (Cooper-deGennes) limit 306
43.3 The general1D case 307
43.4 A microscopic theory of the IDJosephson superlattice 314
44 The proximity effect in a magnetic field 316
44.1 Governing equations in the presence of magnetic fields,
spin susceptibility, paramagnetic impurities, and spin -orbit
coupling 316
44.2 Representative numerical solutions 321
45 Derivation of the G-L theory 326
45.1 The first G-L equation 326
45.2 The gradient term in the clean limit 327
45.3 The gradient term in the dirty limit 328
45.4 The gradient term in the general case 329
45.5 The second G-L equation 33046 Gauge invariance: diamagnetism in the low field limit 332
46.1 Gauge invariance 332
46.2 The magnetic field as a perturbation 332
46.3 The diamagnetic current 335
46.4 Diamagnetism of the superconducting Fermi gas 338
46.5 Magnetic field behavior near a vacuum-superconductor
interface 341
46.6 Relation between normal-state conductivity and the
superconducting diamagnetic response 343
46.7 Calculations of the diamagnetic response using Chambers'
method 347
47 The quasiclassical case 350
47.1 Quasiclassicallimit of the Schrodinger equation 350
47.2 Quasiclassical limit of the Bogoliubov equations 353
47.3 Andreev scattering 355
48 The isolated vortex line 360
48.1 Bogoliubov's equations for the isolated vortex line 36048.2 The quasiclassical equations for a vortex line 363
48.3 A model calculation for the bound core states 365
49 Time-dependent Bogoliubov equations 369
49.1 Basic equations 369
49.2 The time-dependent, linearized, self-consistency condition 370
49.3 The linearized, time-dependent, G-L equation 372
50 The response of a superconductor to an electromagnetic field 374
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 11/511
Contents xi
50.1 The vector potential as a time-dependent perturbation 374
50.2 Relation between the current density and the vector potential 376
51 The Bogoliubov equations for an unconventional superfluid 380
52 Mean potentials, density matrices, and distribution functions in
a uniform superfluid
52.1 The mean potentials
52.2 Singlet-triplet separation
52.3 Density matrices and distribution functions
53 Superfluid 3He
53.1 Some experimental properties of super fluid 3He
53.2 Structure of the gap in an t=1, S =1 superfluid
53.3 The Balian-Werthamcr (BW) phase
53.4 The Anderson-Brinkman-Morel (ABM) or A phase
54 Collective modes in normal and supcrfluid Fermi systems
54.1 General formalism
54.2 Zero sound in a normal Fermi liquid
54.3 Collective modes in superfluid 3He B
55 Green's functions
55.1 Green's functions for the Bogoliubov equations: the
Nambu-Gorkov equations 421
55.2 Quasiclassical Green's functions in a normal system 42655.3 Quasiclassical Green's functions in a superfluid system 430
55.3.1 Keldysh formalism 437
55.3.2 Quasiclassical equations of motion
55.3.3 Eilenberger equations
55.4 Temperature Green's functions
55.5 Dirty superconductors
55.5.1 The dirty-normal-metal Green's function
55.5.2 The dirty superconductor Green's function
55.6 G-L theory revisited
55.7 The Usadcl equations
384
384
389
393
397
397
397
401
404
406
406
412
414
421
443
443
437
443
443
447
451
452
Appendix A Identical particles and spin: the occupation number
representation 458
A.l The symmetry of many-particle wavefunctions 458
A.2 Occupation number representation: Bose statistics 461
A.3 Occupation number representation: Fermi statistics 465
Appendix B Some calculations involving the BCS wavefunction 467Appendix C The gap as a perturbation through third order 470
Superconductinq transition temperature, thermodynamic criticaljield,
Debye temperature and specific heat coefficientfor the elements 475
References 476
Additional readinq 481
List of mathematical and physical symbols 483
Index 493
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 12/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 13/511
Preface
A preface is supposed to alert potential readers to the contents and style of what lies within so
they can decide whether to proceed further. Superconductivity is now a vast subject extending
from the esoteric to the very practical; the people who study or work on it have different
preparations, goals, and talents. No treatment can or should address all these dimensions.
Part I is devoted to the phenomenological aspects of superconductivity, e.g., London's and
Pippard's electrodynamics, the Ginzburg-Landau theory and the Landau Fermi liquid theory.
These theories allow a discussion of the effects of magnetic fields, interfaces and boundaries.
fluctuations, and collective response (which may all be thought of as different manifestations of
inhomogeneities).
Since there is currently much interest in unconventional (non-s-wave) superconductivity,
we have included a discussion of the associated Ginzburg-Landau theory (which then has a
multidimensional, complex order parameter). 3He is the only established example of an uncon-ventional superftuid (triplet p-wave) and therefore our discussions of this substance are somewhat
longer.
Part II is devoted to the microscopic theory of uniform superconductors: the theory of
Bardeen, Cooper, and Schrieffer (BCS) and the Bogoliubov-Valatin canonical transformation,
where the latter so greatly simplifies the discussion of excited states and finite temperature effects.
Although not strictly a uniform-superconductor phenomenon, the theory of tunneling and the
accompanying Josephson effects are also discussed in Part II .
Part III deals with the microscopic theory of nonuniform superconductors exclusively
through the apparatus of the self-consistent Bogoliubov equations as developed extensively by
deGennesand coworkers (and hence sometimes referred to as the Bogoliubov-decIennes theory).
Inhomogeneities associated with a magnetic field, impurities, and boundaries are discussed;
temporal 'inhomogeneities' (e.g., relaxation phenomena and collective modes) are also discussed
via the time-dependent Bogoliubov equations. Bogoliubov theory is extended to include uncon-
ventional BCS superl1uids, such as 3Hc and (possibly) high temperature superconductors and
heavy fermion superconductors (UPt3). Part II I ends with a simplified discussion of Green's
functions starting from the Bogoliubov theory. This formalism serves as an appropriate depar-
ture point for those wishing to go deeper into the theory of superconductivity. Real time Green's
functions, thermal or Ma tsu bara Green's functions, as well as q uasiclassical Green's functions are
all discussed.
Our goal in this book is to focus primarily on the physics of superconductivity. The
materials aspect is generally ignored; material properties enter the discussions only via idealized
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 14/511
xiv Preface
parameters. For this reason the book may not contain enough material on high temperature
superconductors to suit some readers' taste, although idealized layered systems are treated in
some detail, However, it is our view that much of the high T; literature does, in fact, involve the
systematics of the superconducting properties with regard to their chemical and physical make-
up. On the other hand, the underlying physical pairing mechanism and the order parameter
symmetry are still controversial at this time (1997).
The level of the treatment varies between sections. Much of Part I is accessible to fourth
year undergraduate students or first year graduate students in physics who have had exposure to
mechanics, electricity and magnetism, and elementary quantum mechanics. Parts II and III
involve some use of second quantization at a level usually arrived at by the end ofthe first year of
graduate school. With some omissions the book could form the basis of a one semester graduate
course on superconductivity; if all topics are discussed it would likely extend to a year.
The microscopic origin of the attractive electron-electron interaction involving the ex-
change of phonons is discussed (in Part II) only in terms of the jellium model which is sufficient to
bring out the basic physics of the electron-phonon interaction. A proper discussion, however,
requires solving the coupled Green's function equations of motion for the electrons and phonons,
which is beyond the scope of this book. A simplified discussion of the spin fluctuation pairing
mechanism is included since pairing in superfluid 3He likely arises from this eJTect.Other pairing
mechanisms, as have been proposed for high T; materials, are ignored entirely.
The character of many of our discussions was strongly influenced by deGennes' 1966
classic, Superconduct ivity of AIetals and Alloys; any similari ty of our discussions and his is likelyintentional. Both of the present authors are primarily experimentalists; their involvement in
writing a largely theoretical book about superconductivity was primarily a self-education excr-
cise. We have tried to use the simplest mathematical methods we can while minimizing the
(pedagogically useless) 'it-can-be-shown' approach. Our decision to use a uniform theoretical
approach to the topics in Part IIImeant that in many cases we had to devise our own methods; in
so doing we have strived to get things right but inevitably errors will creep in (herc and elsewhere),
for which we apologize in advance. Incidentally we would appreciate hearing about any errors
detected.
This book is not meant to be a treatise on superconductivity (for this kind of treatment we
recommend the well-known text edited by R. Parks); i.e., it is not our purpose to collect most of
what is known and useful, but rather to present the nuts and bolts used to obtain some of the
important results.
The literature on superconductivity is enormous; it is therefore not possible to be aware of
anything but a small fraction of it (this situation was 'worsened' by the discovery of high
temperature superconductivity). For this reason we decided to minimize the number of original
citations to those which are particularly important historically or pedagogically (a difficult and
dangerous decision) or to those where the reader may seek further details. We hope the manyworkers not cited will forgive us.
We have benefited from discussions with (or encouragement by) P. Auvil, K. Bennemann,
S. Doniach, A. Patashinskii, P. Wolfle, and S.K. Yip. The text was typed (again and again) by Ms
A. Jackson who deserves the most thanks.
Northwestern University J. B. Ketterson
S.N. Song
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 15/511
P A R T I PHENOMENOLOGICAL THEORIES
OF SUPERCONDUCTIVITY
Introduction
Many metals, alloys, and intermetallic compounds 1 undergo a phase transition at some (gen-
erally low) temperature, Tc' to a state having zero electrical resistance (see Fig. 1.1), a phenom-
enon which is call cd superconductivity. Superconductivity was discovered by Kamerlingh Onnes
in 1911; it has been sensitively probed by observing the magnetic field produced by a circulating
supercurrent using a sensitive technique such as nuclear magnetic resonance. Favorable ma-
terials exhibit no detectable decay of this current for periods limited by the patience of the
observer ( .~ years). However, superconductivity is better defined by the nature of the associated
phase transition and it then includes materials or measurement conditions where the resistance
may not vanish; here we postpone the discussion of these exceptions and initially restrict
ourselves to perfect superconductors.
0.001
I.>V j
1,
!,,
: ,'Hg
,,
I,,
< 10-6 ,I'
0.0015
~:::: 0.0010~
0.0005
0.0000-w o 4.10 4.20 4.30
T(K)
4.40 450
Figure 1.1 The original R vs T curve of Kamcrlingh Onnes showing the
superconducting transition in mercury. (After Kamerlingh-Onnes (1911).)
A second fundamental characteristic associated with a superconductor is the exclusion of
magnetic flux, discovered by W. Meissner and R. Ochsenfeld in 1933 (which we will often refer to
1. An alloy is a solid solution of two or more clements at least one of which is a metal. An intermetal1ic
compound involves a metal and one or more other elements which form a chemical compound
having nearly precise ratios of its constituents.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 16/511
2 Part I Phenomenological theories
Perfect conductor Supercond uctor
I ) rf./r\\.
1 ,1
'\\, ~!. \''__j I
I . \ . ..1,I I : I
, illI ' .i / \ I
fr~
I\VJi\ \ , (! \ i I
Experiment 2:
Perfect conductor Supercondncto
: ' I
/ ; 1 t
l · r D . ii I I
i I !
i . :
II \,1
,( "I ! ~
\~Jr \ /
Figure 1.2 Hypothetical experiments showing the difference between a
superconductor and a perfect conductor. Experiment 1:sample cooled in zero magnetic
field after which a field is applied; experiment 2: sample cooled in applied magnetic field.
simply as the Meissner effect in what follows). The degree of flux exclusion can depend on the
material or measurement conditions; we will also assume this property is nearly perfectlydisplayed for our initial discussion.
The combination of zero resistance and perfect diamagnetism results in a clear distinction
between a superconductor (which as we will see is in a thermodynamic state) and a hypothetical
'perfect conductor' (which has the unique transport property of zero resistance); this difference is
illustrated in Fig. 1.2 and involves the differing response each would have for different histories of
cooling below the transition temperature and applying a magnetic field. If we start by cooling
through T, and then apply a magnetic field, both the superconductor and the perfect conductor
would exclude the field, For the perfect conductor, induced currents arising from Faraday's law
would screen the flux, Flux exclusion in the superconductor could be assigned to the same
mechanism or (more fundamentally) to the Meissner effect itself. If we reverse the order by first
applying the field and then cooling through Te, the superconductor and the perfect conductor
beha ve differently: the superconductor excludes the flux (the Meissner effect); the perfect conduc-
tor would remain fully permeated by the field.
These experimental observations argue that the transition associated with superconductiv-
ity is indeed a phase transition since an equilibrium thermodynamic state is defined by its
independent thermodynamic variables (in this case T and H), and is independent of its history
(which as we see is not true for the perfect conductor).Superconductivity, and with it the Meissner effect, does not persist to arbitrarily high
magnetic fields, For each temperature there is a well-defined critical field, He(T) , at which
superconductivity disappears." Fig. 1.3shows a universal curve of the temperature dependence of
He vs T.
2. This statement is restricted to so-called Type I materials having a shape for which there is no
demagnetization effect, as will be discussed in later sections.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 17/511
Introduction 3
H
Normal
Superconducting
Figure 1.3 The temperature dependence of the critical field H J T).
,8x 10-J I"
ca V g K I Ti n
6
<;Q)
si:
~U
'" '0en
2
I/
////
/
".~.:;': .,,'- : : ..'.'.'.'...
o3 4
Tempera ture
o
Figure 1.4 Specific heat of tin as a function of temperature: open circles without an
external magnetic field; filled circles with an external field H > He' Shown also are the
individual contributions from the electrons and the lattice for H > He: chain line
lattice contribution; dashed line - electron contribution. (After Keesorn and van Laer
(1938).)
A superconductor also exhibits a discontinuous increase in its specific heat at T; (there is no
latent heat at zero field) below which it drops rapidly (approaching an exponential dependence at
low temperatures), A discontinuity in the specific heat is a signature of a second order phasetransition providing added evidence that superconductivity is associated with a distinct ther-
modynamic phase. Fig. 1.4 shows the heat capacity of a typical superconductor.
There is a related low temperature phenomenon known as superfluidity which occurs in
liquid helium. There are actually two such liquids involving the isotopes 3He and 4He having
superfluid transitions near 2 x 10- 3 K and 2 K, respectively, depending on the pressure. For
some purposes a superconductor may be regarded as a superfluid having an electric charge. 4He
superfluidity involves a Bose condensation, a phenomenon which is related to superconductivity
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 18/511
4 Part I Phenomenological theories
in a somewhat subtle or indirect way and will not be discussed in this book. The superfluidity of
3He is intimately related to superconductivity and will be addressed in later sections. Another
related superfluid is the neutron liquid in the interior of neutron stars.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 19/511
The London-London
equation
Wenow present two derivations of equations which are useful in describing many of the magnetic
properties of superconductors. The older approach, used by F. London and H. London (1935)(see also London (1950)), starts with the Drude-Lorentz equation of motion for electrons in a
metal, which is Newton's law for the velocity, v, of an electron with mass, m, and charge, e, in an
electric field, E, with a phenomenological viscous drag proportional to v/r:
(2.1)
For a perfect conductor r - - - + cc, Introducing the current density j = nev, where n is the conduc-
tion electron density, Eq. (2.1) can be written
dj ne1-=-Edt In '
(2.2)
which is referred to as the first London equation. The time derivative of Maxwell's fourth
equation (in cgs units) is
DH 4rr oj r. o2EV x -- =- - + - -.
a t C o t c D t2·(2.3)
where E is the dielectric constant. Taking the curl of (2.3) and using (2.2) we have
( C H ) (4rrne2
r. (2
)V x V x -- =_- + - - , V x E;a t m (' c D l-
(2.4)
using V x E =- ( l jc)DH/Dt we have
( a ) ( 1 I: a 1) ax V x -. H + - + - -.- - H = .a t ; . ~ ('2 D t2 a t .
(2.5)
where we introduced the London depth, i.L, defined by
--,-.me:
(2.6)
Eq. (2.5) has been obtained for a perfect conductor model. In order to conform with the
experimentally observed Meissner effect, we must exclude time-independent field solutions
5
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 20/511
6 Part I Phenomenological theories
arising from integrating (2.5) once with respect to time and we therefore write
(1 8 (
2)
V X (V x H ) + -:;-,----::;:;-. H =0;J'L C~ o r
(2.7)
this is referred to as the second London equation. In what follows we will refer to Eq, (2.7) simply
as the London equation.
An alternate derivation of (2.7) is motivated by the idea that some of the moving electrons
behave collectively as a superfiuid, a liquid possessing no viscosity. This concept is borrowed
from the physics of liquid 4He; below 2.l9 K this system behaves as if it were composed of a
mixture of two liquids: a superfluid, having no viscosity, and a normal liquid, having a finite
viscosity. We assume that the free energy of a superfiuid consists of three parts
(2.8)
where F N is the free energy associated with the normal liquid, Ekin is the kinetic energy of the
moving superfluid, and Emag is the magnetic field energy. We may write these latter two terms as
(2.9)
and
(2.10)
where p ( r ) is the mass density associated with the supcrfluid. Writing p = 1 m and v=l /ne)j and
using the fourth Maxwell equation V x H = (4rr/c)j, Eq, (2.l0) becomes
(2.l1)
n IS now interpreted as the density of supcrconducting electrons, We will assume that the
superconducting electrons adjust their motion so as to minimize the total free energy; thisrequires c 5 ( E m a g + E kin) =0 or
I fme? }
H(r)' bH(r) + --2 [V x H(r)] . [V x bH(r)] d3=0,4r rne
(2.12)
where bH(r) is a variation of the (initially unknown) function H(r). Integrating the second term by
parts (and placing the resulting surface outside the superconductor) we obtain
I[H(r) + A~Vx (V x H)]' c)H(r)d
3
r = . (2.13)
Since the variation c'iH(r) is arbitrary, the term in the square brackets must vanish; therefore
V x (V x H ) + ic~H =0, (2.14)
which is eq uivalent to (2.7) (including the displacement term in Maxwell's equation yields the last
term in (2.7), which is negligible for most applications).
As a simple application of Eq. (2.l4) we now discuss the behavior of a superconductor in a
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 21/511
The London-London equation 7
magnetic field near a plane boundary. Consider first the case of a field perpendicular to a
superconductor surface lying in the x y plane with no current flowing in the z direction. From the
second Maxwell equation, V' H = 0, we obtain DHz /u z = 0 or J J = const. From the fourth
Maxwell equation, V x H = (4n/c)j, the first term in (2.14) vanishes and hence H = 0 is the onlysolution. Thus a superconductor exhibiting the Meissner effect cannot have a field component
perpendicular to its surface.
As the second example consider a field lying parallel to the superconductor surface, e.g.,
H I I X,which we may write as H =H(z)x (which satisfies V' H =0). Using the vector identity
V x (V x H) = V (V 'H ) - V 2H , (2.15)
Eq. (2.14)becomes
(j2H 1__ x __ H =0(jZ2 A l x
(2.16)
or (for a superconductor occupying the region z > 0)
(2.17)
A field parallel to the surface is therefore allowed; however, it decays exponentially, with a
characteristic length, Au in the interior; ).L (T =0) ranges from 500 to 10000 A , depend ing on the
material. Accompanying this parallel field is a surface current density, which, from Maxwell's
fourth equation, is
c .j(z) =- -.-HAO)c-Z!'.LY.
4nl'L(2.18)
This current density shields or screens the magnetic field from the interior of the superconductor.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 22/511
Gf----p-;-p-p-a-r-d-'-S-e-q-u-a-t-;-o-n---
At temperatures well below the superconducting transition temperature the heat capacity of a
superconductor displays an exponential behavior C ~ e - MkB T (see Fig. 1.4).This suggests that theconduction electron spectrum develops an energy gap, L '\ (not to be confused with the gap in a
semiconductor); recall that electrons in normal metals have a continuous (gapless) distribution o f
energy levels near the Fermi energy, fl. On dimensional grounds one can construct a quantity
having the units of length from L '\ and the Fermi velocity, 1!F; we define the so-called coherence
length by
(3.1)
This length bears no resemblance to the London depth, Au and hence represents a different length
scale affecting the behavior of a superconductor; it can be interpreted as a characteristic length
which measures the spatial response of the superconductor to some perturbation (e.g. the
distance over which the superconducting state develops at a normal metal-superconductor
boundary). Length scales of this kind were introduced independently by Ginzburg and Landau
(1950) and by Pippard (1953).1 We first discuss Pippard's phenomenological theory (which
scmiquantitatively captures the main features of the microscopic theory to be discussed later). We
begin by writing London's equation in an alternative form. Substituting V x H =4n/c)j (from
the fourth Maxwell equation) in the London equation yields
. cVXJ=---H
4n}~
ne2
=--Hme
(3.2)
We next write H = V x A . where A is the magnetic vector potential, and restrict the gauge to
satisfy
V·A = 0 (London gauge) (3.3)
and the boundary condition
I.These length scales are not identical, however; the Pippard length is temperature-independent while
the Ginzburg-Landau length depends on temperature. The Pippard coherence length is related to
the ReS coherence length.
8
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 23/511
Pippard's equation 9
(3.4)
where Anis the component of A perpendicular to the superconductor surface. London's equation
may then be written
ne2
i= --Ame
(3.5)
Note that the condition (3.4) builds in the reasonable boundary condition that the normal
component ofthc supercurrent.j., vanishes at a boundary (this is a good boundary condition at a
superconductor-insulator boundary but will require modification for metal-superconductor or
superconductor-superconductor boundaries).
To generalize (3.5), Pippard reasoned that the relation between j and A should be nonlocal,
meaning that the current j(r) at a point r involves contributions from A(r') at neighboring points r'located in a volume with a radius of order ~o surrounding r. The mathematical form he selected
was guided by the non local relation between the electric field, Etr'], and the current, j(r), which
had been developed earlier by Chambers (1952). The expression employed by Pippard was
f [ A(r')' R]R ..( ) C -R'<odJ r
J r =- c ,. rR4 '
(3.6)
where R= - r'. The constant C is fixed by requiring (3.6) reduce to (3.5) in the quasiuniform
limit where we may take A from under the integral sign; we then have
C fCO s2 (!dQ fe - R!~odR =ne2
,
me(3.7)
fromwhich we obtain C = 3ne2 /4n~omc. Pippard's generalization of London's equation is then
,. 3ne2.f[A(r')'R]R -R~ J,
J(r) =---- e ! ,Od r.4n;ome R4
(3.8)
Since Eq. (3.8) involves two functions. A(r) and j(r), a complete description requires a second
equation which is obtained by substituting H =V x A in the fourth Maxwell equation to obtain
4nVx(VxA)=~j
e(3.9)
(resulting in an overall integrodifferential equation for A).
Eq. (3.8)applies only to a bulk superconductor. An important question we would like to
examine is the behavior of a magnetic field near a surface, which will require a modification (or
reinterpretation) of (3.8). To model the effect of the surface the integration over points r' is
restricted to the interior of the superconductor. If the surface is highly contorted, then it can
happen that two points near the surface and separated by about a coherence length cannot be
connected by a straight electron trajectory, without passing through the vacuum; one then has to
account for this shadowing effect. We restrict ourselves here to plane boundaries which we take to
be normal to the z direction.
In the limit A L >> ~o Eq. (3.8) reduces to the London equation, as discussed above. (By
expandingA(r') in a power series in R, we may obtain corrections to the London equation due to
nonlocality.) In the opposite limit, A L < < ;0' A(r') varies rapidly. Let us assume that A(r) falls off
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 24/511
10 Part I Phenomenological theories
over a characteristic distance ;_ (which we will determine shortly through a self-consistency
argument). When )~< ~O, the value of the integral (3.8) will be reduced roughly by a factor ( A . / ( o ) ;
I.e.,
;. ne2j(r) =-~- A(r).
C ;o mc(3.10)
We may also write (3.10) in the 'London-like' form
(3.11)
This equation has solutions which decay in a characteristic length (I.~~0!;_)1/2; to achieve self-
consistency we set this length equal to A :
(3.12)
(A more rigorous derivation from the microscopic theory carried out in Sec. 46 yields
2=0.62i. t~0.) We conclude that in the Pippard limit the effective penetration depth A is larger
than the London depth, )'L: ;'!;'L =~oI)L)l/.l > 1. At the same time ), remains smaller than the
coherence length: }'/(o =(I'LgO)2/3 < l.
If our metal has impurities it is natural to assume the relation between the current and
vector potential will be altered. To account for the effects of electron scattering, Pippard modified
the coherence length factor in the exponent of(3.8) as 1!~0 --> (1/(0) + (lit) where tis the electronmean free path;' the coefficient in front of the integral was not altered. Eq. (3.8) then becomes
. 3ne2
f[A(r')'R]R -R(., '-+~) 1,J(r)=--- ..- e '" Id'r.
47 [~omc R4(3.l3)
In the limit ). >> (t, ~o) we may again take A(r) from under the integral sign; carrying out
the integration we obtain
ne2 1
j(r) =-- ..·--A(r).mcc;o 1 1-+-~o t
(3.14)
For the case of a very dirty metal, where l< < ~o,
ne? tj(r) =-- ~A(r).
me C ;o
(3.15)
The effective penetration depth is then obtained from the expression
(3.16)
If ;.< < (t, ~o) we continue to use Eq. (3.12).
The magnetic response of a superconductor depends on whether A ; ; : ; ~ or A . : S ~ , as will be
developed later." These regimes are designated in Table 3.l. A Type I superconductor displays a
2. Other expressions, such as ~=({;0)1/2, are sometimes used to estimate the effective coherence length.
Such ambiguities are removed by the microscopic theory to be discussed in Part III.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 25/511
Pippard's equation 11
Table 3.1 Regimes defininq Type J and
Type II superconductors?
Type I (or Pippard)
Type II (or London)
complete Meissner effect (flux exclusion) up to some critical field H e' above which it becomes
normal. The magnetic response of a Type II superconductor is more complex and will be
developed later.
3. The precise criterion separating the regimes is A =ufi.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 26/511
.I--~-h-e-r-m-O-d-y-n-a-m-ic-s-o-f-a--
Type I superconductor
In this section we consider the thermodynamics of a Type 1superconductor (Gorter and Casimir
1934a,b); our discussion follows that of London (1950). We restrict the geometry of the supercon-ductor to a form for which the external field, H, is not distorted by the presence of the
superconductor (examples being an infinitely long cylinder with H parallel to the axis or a plane
slab of infinite extent with H parallel to its surface). Far inside the superconductor (i.e., several
London depths from the surface) the magnetic field essentially vanishes in the superconducting
state and is equal to H in the normal state. In the thermodynamic identities that follow we identify
this interior field as B, the flux density. The H field will be taken as the applied external field. The
relation between Band H is shown in Fig. 4.l.
We recall the thermodynamic identity for the response of a system in a magnetic field
1dIS=TdY + -H·dB,
4n(4.1)
where C f · is the energy density and Y is the entropy density. When T and B arc the independent
variables we use the Helmholtz free energy density, Y =f f - TCf', and when T and H are the
independent variables we usc the Gibbs free energy density, '§ = . ' 1 7 - (1/4n)B· H; taking the
differential of these two quantities and using (4.1) yields
1dY = - 9'dT+ -H·dB
4n(4.2)
and
1d~1j= - .'f'dT - - B· dH.
4n(4.3)
A Type I superconductor displays the Mcissner-Ochsenfeld properties:
B =0,
B=H, H> H e.
(4.4)
(4.5)
H e is called the thermodynamic critical field. Since H and Twil l be our independent variables, we
integrate (4.3) at constant T to obtain
< fJ(T , H) =§(T ,O ) - L f B·dH;12
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 27/511
Thermodynamics of a Type I superconductor 13
B
/
/
/
o H
Figure 4.1 B vs H curve for a Type I superconductor.
therefore
1~fj (T,H) = 0'(T ,O ) - _(H 2 - H D
Sn( 4 . 6 )
or
~fj (T,H) =0' (T,O) ( 4 . 7 )
Note that qj is continuous at the transition. We define a function
1 0
'f Jo = = ' fJ(T,O) +-II~.Sn
(4 .8 )
Wemay then write
(H>HJ
(4 .9)
and
1' f J (T ,H) =fJ o - -H ~
8n
(4.10)
where qjN and qjs denote the normal and superconducting states, respectively. We may interpret
qjo as the Gibbs free energy of the normal metal at zero field (were it stable); hence the Gibbs
free energy density of the superconducting state is lower than that of the normal state by
( 1 /8n)H~(T) ; this quantity is referred to as the c on den satio n e ne rgy . Since ~ fj= =.'F - (1/4n)H' B,
§ is(T ,O) = jo - ( I /Sn)H~. The normal state Helmholtz free energy density is then
(4.ll)
From Eq. (4.3),
.r;'=_~ 'fJ ) ;6T H
(4.12)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 28/511
14 Part I Phenomenological theories
thus
aqjoY' ---
N - D T(4.l3)
and
~,(f 1 ( ' H ):~ O,'lo () cJs= --+-Hc - .
a T 4n a T H
(4.14)
Note the entropy is discontinuous across the transition and hence we have a first order transition
(when H = F 0):
(4.l5
The heat of the transition is
(4.16
this equation corresponds to the Clausius-Clapeyron equation of a (P, V,T) system. The specifn
heat (at constant H) is defined as
(4.17
or
_ T [ 32Hc (OHc)2]---H--+-4n c cT2 a T .
(4.l8
At 11=0, where the transition is second order,
t;( 3 1 1 c)2
/I,.«jHIT=T'=-4n a T T~T';
this is sometimes called Rutgers' formula (Rutgers 1933).
(4.15
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 29/511
.I---~-'h-e-in-t-e-r-m-e-d-ia-t-e-s-t-a-t-e--
I f a superconducting body of arbitrary shape is placed in a magnetic field, the flux exclusion
associated with the Meissner effect will in general distort the magnetic field. Exceptions are aninfinite cylinder with the field parallel to the axis, or a sheet or half space with H 0 parallel to the
plane of symmetry. For situations involving lower symmetry the local magnetic field can vary
over the surface, being both higher and lower than the applied field, Ho. As a simple example
consider the case of a spherical superconductor shown in Fig. 5.1. From magnetostatics the field
willbe highest at the equator (on the cirele C in Fig. 5.1)where it is H =~Ho- Hence flux enters the
sample, not at the thermodynamic critical field, H c ' but at a value flo =~Hc . For magnetic fields
He> Ho > iH c the sample consists of alternating domains of normal metal and superconductor.
A superconductor in such a regime is said to be in the intermediate state (Landau 1937).
Figure 5.1 The magnetic field distribution about a superconducting sphere of radius a.
For an applied field H o < ~Hc ' there is a complete Meissner efTectand the field at the
equator (at any point on circle C) is ~H 0 ; the field at the poles (Q, Q') is zero. For
~Hc < Ho < He' the sphere is in the intermediate state.
15
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 30/511
16 Part I Phenomenological theories
Figure 5.2 Schematic of the magnetic field distribution in a slab in a perpendicular
field. For all f ields J J 0 < ll; the sample is in an intermediate state having a laminar
structure of normal and superconducting domains.
We will limit our discussion to the simple case of a plane superconducting sheet with H o
parallel to the surface normal. From our earlier discussion we know that a superconductor
cannot sustain a field component perpendicular to its surface. The field behavior is shown
qualitatively in Fig. 5.2. It has the following features.
(i) For a magnetic field 0 < H o < He the sample consists of adjacent domains which are
wholly superconducting (with no internal flux) or normal (with Hloeal =HcJ,
(ii) In the interior of the superconductor and far from the surface, the domain walls are parallel
to the applied field direction; the fraction of the cross section that is supercond ucting is fixed
by Ho and H, such that the total flux through the sample is conserved
normal cross section H 0
total cross section He(5.1)
Hence the superconducting fraction is 1 - HolHe (which vanishes, as it should, for
H o =H c l .
(iii) Near the surfaces the flux sheets 'flare out' (which reduces the field curvature, which would
otherwise raise the local fields at the interface). Were the local field to remain fixed at He' to
sustain the average flux, needle-like superconducting domains would have to encroach on
the normal domains in the vicinity of the surface. Such needles do not occur since the field
profile is controlled by a minimization of the surface energy (between supcrconducting and
free surface or the normal regions, see next section), the superfluid kinetic energy, and the
total magnetic field energy.
A detailed treatment of the domain structure is mathematically complex and will not be dealt
with here (see Landau, Lifshitz, and Pitaevskii (1984), deGcnncs (1966), London (1950)).
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 31/511
8 f - - - - s-u - r - , . - a - c - e - e - n - e - r - g - Y - b - e - t w - e - e - n -anormal and a
superconducting metal
Consider a slab parallel to the x-y plane in a perpendicular magnetic field parallel to z. Assume
we have a phase boundary perpendicular to the x axis with the superconductor occupying theregion x> O . The total free energy, F, in the London model is:
IYJ l H2(r),i2 ]
F =sf dx .'1's + -- + - - - - ' = : [ V x H(r)]2 ,o Sn Sn
(6.1)
where s1 is the interface area,:Fs is the condensation energy density, H~(T)!8n, and the second
and third terms are the magnetic field energy density and the superfluid electron-kinetic energy
density, respectively. At our phase boundary in the intermediate state, where H =He for x < 0,
we have (see Eqs. (4.6) and (4.7))
(6.2)
By definition
e W I ,< ,
G =F - - H e' B(x)dx.4n 0
(6.3)
According to the discussion at the beginning of Sec. 4 w e set B(x) =H(x); from Eq. (2.17) we have
H(x) =Hcz e -X
/AL
• Inserting this form in (6.3) and using (6.1) we obtain
IYc
= "I' d3" + " _4- .7" S I ,yY,
o
(6.4)
where y is the surface energy per unit area, i.e., the 'surface tension', given by
(6.5)
Note this surface energy is negative. This suggests the system can lower its energy by maximizing
17
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 32/511
18 Part I Phenomenological theories
the interfacial area (i.e., the system is unstable to the formation of multiple domains with
associated interfaces),
Type I superconductors (for geometries not possessing an intermediate state) display a
single domain for H :::;;H e; hence they must have a positive domain wall energy. Physically this
positive surface energy arises because superconductivity is destroyed over a region of order ~
perpendicular to the interface; i.e., we lose the condensation energy over a volume of order .#~,
where ( is a coherence length. This is equivalent to a positive contribution to the surface energy of
order
H2},=~_c.
Sn(6.6)
In a Type I material ( > )'L and hence the positive contribution (6.6) outweighs the negative
contribution (6.5) and the interface is stable. In Sec, 9.3 we will continue this discussion and derive
an expression for the surface tension.
For a Type II material the system does, in some sense, try to maximize the amount of
internal interfacial area above some field (referred to as the lower critical field); however, it is
subject to a constraint imposed by quantum mechanics, as we discuss in the next two sections.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 33/511
8I---o-u-a-n-t-;z-e-d-v-o-r-t-ic-;-t-y---
The previous discussion of the surface energy of a normal metal/superconductor interface
suggests that Type II materials, where ¢ : : S A are unstable to the formation of domain structureswhichin some way maximize the amount of interface area. Two possible domain geometries are:
(i) an array of nested sheets (closed or open, depending on the geometry 1) and (ii) a two-
dimensional lattice of flux filaments. Calculations show the latter domain structure to be more
stable.
Since the filaments (are presumed to) admit flux into the interior of the superconductor, we
envisionthem as having a normal core with a diameter of order ¢, outside of which supercurrents
flowin a diameter of order }, which produce the internal field via Ampere's law.
Asa primitive model of a single flux filament we consider the extreme limit ¢ -- > 0 for which
the London approach should provide a good description. We recall Eq. (2.5) associated with our
first derivation of the London equation
a l 4nne2
]- v x (V x H) + -2- H = 0;D t m e (7.1)
we next integrate (7.1) over an area s~ intersecting the filament (for convenience we choose a
plane perpendicular to its axis) and use Ampere's law (Maxwell's fourth equation) in the form
a f l m e ];;- -2 V X + H . d,xi = O.o t ne
(7 .2)
In integrating (7.2) with respect to time we now allow the possibility of a nonzero constant of
integration (since the flux filament phenomen violates the Meissner behavior); thus
(7.3)
Applying Stokes' law to the first term in (7.3) yields
m e 1 fne2jj'dt + H·d.xi = 4 > . (7 .4)
Ifwe choose the contour to enclose a large area, we may expect the first term to be exponentially
1. For a superconducting slab we envision an array of interfaces parallel to the surface and for a
cylindrical sample an array of coaxial cylinders. Other shapes would have more complex structures.
19
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 34/511
20 Part I Phenomenological theories
small (since the currents fall off exponentially with a characteristic length ),);we can then identif
the constant of integration, ¢, as the total flux contained within the filament (most of which alsr
falls inside a radius of order A ) .
To gain further insight we substitute H = V x A into (7.4) and again apply Stokes' law tc
obtain
f[:~jA 1 M = ¢; (7.Sa
writing j=nev =ne(p/m), (7.Sa) becomes
l [ p + : A ] . dt = : ¢.J C C
(7.Sb
Wc identify the integrand as the canonical momentum associated with the motion of a charged
particle in the Hamiltonian formulation of mechanics. F. London correctly concluded that
superconductivity was a macroscopic quantum phenomenon, and guided by this insight he
suggested that Eqs. (7.S) must conform with the Bohr-Sommerfeld quantization rule for the
(quasiclassical) motion of an electron, i.e. (I e I/c)¢ = nh or ¢ = n(he/I e I),where n is an integer.
However, this assumes the orbiting entities are single electrons; Ginzburg and Landau allowed
for a marc general case where e -> e* ; we then have
(7.6a)
where
he
¢o=le*l·(7.6b)
From the Bardeen-Cooper-Schrieffer (ECS) theory it is now known that e* = 2e; i.e.,
he
¢o = 21 e 1
= 2.07 X 10-7 G crrr' (7.6c)
which is referred to as the flux quantum. Hence flux enters a Type II superconductor as an array
of quantized flux filaments; the lowest energy situation corresponds to singly quantizied (1 1 = )
filaments eaeh carrying a flux quantum ¢o.
In what immediately follows we adopt cylindrical coordinates (r, 0, z) and write the in-plane
radius vector as r. Let us next examine Eqs. (7.S) for a contour of radius A >> r >> ¢; the
amount of flux contained is then vanishingly small and the first term in the integrand of (7.Sb)
dominates yielding the condition 2npr = h or p =h/r; BCS theory also dictates that the mass ofthe orbiting entity is m* = 2m; hence
h ~v(r)=-{}
2mr(7.7)
where (j is an azimuthal unit vector. This velocity profile corresponds to the large r behavior of a
vortex in a fluid, although with the vorticity quantized. One then refers to the filaments as
qu antiz ed vortex lines or vortex lines for short.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 35/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 36/511
22 Part I Phenomenological theories
Figure 7.1 Structure of an isolated Abrikosov vortex line in a type IIsuperconductor
or
(7.14
where we choose a radius r =(for the (inner) line integral; this corresponds to the physicallj
reasonable assumption that the field divergence is removed at the coherence length scale (as,
more complete theory confirms). We expect that the l/r divergence of the superfluid velocitj
ultimately destroys superconductivity in the vortex core; we may model this effect by assumim
the density of supcrconducting electrons, n(r) , approaches zero (sufficiently rapidly) as r -- > O.Fig
7.1 shows the qualitative behavior of H(r) and Il/I(r) I near the center (core) of a vortex filamen
where I l/I(r) I = = [n(r)]t .
In the next section, where we discuss the mixed state of a Type II superconductor, we wil
require an expression for the interaction energy of two vortices. Returning to our London-lik:
model, we generalize Eq. (7.8) to the case of two (parallel) vortices
(7.15
Since this equation is linear the resulting magnetic field will be the sum of two terms having th
same form as (7.10):
(7.H
The total energy (which is quadratic in H) will involve three terms: two of these correspond to th
'self-energies' of the individual vortices (as given by (7.13)) and the third results from thei
interaction, The interaction energy in the extreme London limit can be evaluated from the secor»term in (7.12) along with (7.10) (we assuoe the path of the first integral involves a single circle c
very large radius encircling both vortices); the interaction energy is then given by
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 37/511
Ouantized vorticity 23
= 2 ( ~ ) 2 K o ( l r l ~ r 2 1 ) ,4n)·L AL
where,due to our assumptions, we require 1 r 1 - r 2 1 ;< ; ~ . The sign of (7.17) is positive and hence
the force (per unit length) between two vortices is repuls ive . We may rewrite Eq. (7.17) in the form
(7.17)
E(l2) '"
__ =~H(l2)
L 4n '(7 .18 )
whereH(12) =0 ll~l)(r = r2) = H~2)(r = r 1) with the latter given by (7.10); i.e., H(12) is the contribu-
tionto the field at one vortex resulting from the presence of another. Regarding ( - ¢oL/4n)z as a
magneticmoment, p, we could write E(12) = - p"H.where H = H(12)Z.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 38/511
v Type II superconductivity
8.1 Magnetic fields slightly greater than He1
Superconductors with ~ < )"termed Type II materials, behave differently in a magnetic field fron
those with ~ > I " (Type I materials). Fig. S.1 shows the field dependence of the magnetization of a:
ideal Type II superconductor (the sample geometry is assumed to be a plane slab or a cylinder tl
avoid geometrically induced field inhomogeneities as noted earlier in our discussion of th
intermediate state). For low fields the magnetization is - H/4n; i.e., the sample displays,
Meissner-like behavior. However, at a field H =He1, called the lower criticalfield, flux abrupt!
enters the sample (the susceptibility X =dM/dH) --+ +xfor an ideal sample at this field). Th
magnetization increases (becomes less negative) continuously above H el and reaches zero at ;
field H = H c2 called the upper critical field. The regime H el < H < H c2 represents a nev
thermodynamic superconducting state called the mixed state or Shubnikov phase (Shubnikov eal. 1937).
We now show that the Meissner state of a Type II superconductor becomes unstable to tb
entry of vortex filaments at a field which we identify with H c1• For magnetic fields which are onl
slightly above H e! we may write the Gibbs free energy per unit volume as
1 .. B·HI# (H) = I#(H = 0) + nLEIL + - IE(I,JI - --.
Li<i 4n(S.1
The second term represents the self-energy of the individual lines where nL is the number of lineper em? with energy E/ L per em, the third term is the (repulsive) interaction energy between th:
lines (and is summed over a unit of area), and the last term is the usual field term relating tl u
Gibbs and Helmholtz free energies. In the presence of a uniform array of flux lines the magnetn
induction is
(S.2
Near He! (above which flux first enters), we will initially neglect the interaction term and, usiru
(S.2),write
(S.3
For H < 4nE /¢oL , (S.3) is minimized by setting nL=O. However, for H > 4nE/4>oL Eq. (S.3
24
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 39/511
I ype " euperconaucttvtty
-41TM
(a) 0 H
-41TM
MeissnerPhase
AbrikoSOlJMixed Phase
Normal
Phase
Hb) 0
Figure 8.1 The reversible magnetization curves for a Type I (a) and a Type II (b)cylindrical superconductor. The magnetic field is applied parallel to the cylinder axis.
would suggest that q}ex - nL; i.e., the system can lower its free energy indefinitely simply by
creating more flux lines. It is clear that we can identify He Ias
(8Aa)
or
(8Ab)
where we used (7.14) for the vortex self-energy in (SAb).1 The negative divergence of C I J for
H> Hel is eliminated if we include vortex-vortex interaction (repulsion) effects; i.c., we must seek
the minimum with respect to B of the quantity
B l 1 4 > 0 ( d ) ]J.q}=- H -H+-z--K -4n c I 2 2n}~ O}L .
( 8 .5 )
The first two terms are the same as in (S.3). The last term is the result of the vortex-vortex
1. We can now make contact with the comment made at the end of Sec. 6. Taking the classical limit as
h - - -> 0, we see that the lower critical field would approach zero. At any finite field we would then
have a divergent number of flux lines; i.e., the system would have a divergent internal interface area.
Thus quantum mechanics imposes the constraint on the maximal amount of internal interfacial area.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 40/511
26 Part I Phenomenological theories
0.10
0.05
0.00
!J.'§ -0.05
-0.10
-0.15
-0.20
o
-50
-100
!J.,§-150
-200
-250
-3000.0
H/H =cl
(a)
1.1
1.2
1.3,_
2 3 4 5 6 7 8
dI", l
(b)
I
\ H/H z:
\cl
\ 8\ ->
\//-
,,/-
9 /-:
/
\ .> //
\ -~
///\\
\
\
\////10
\_-
0.2 0.4 0.6 0.8
diAL
Figure 8.2 Normalized Gibbs free energy density calculated from Eq. (8.5)as a function
of the reduced vortex lattice spacing in the low applied field regime (a) and high applied
field regime (b).
repulsion; the factor 1/2 assigns half of the interaction energy to each vortex (in Eq. (7.18)), z is the
number of nearest neighbors in the vortex lattice (the exponential fall-off of Ko justifies including
only nearest neighbor interactions at low fields where nL is small), and d (appearing in the
argument of Ko) is the vortex-vortex spacing, which depends on the symmetry of the vortex
lattice. Calculations show that a triangular (centered hexagonal) arrangement of lines has the
lowest energy. For this lattice nL = 2/J3d2 =BN o or d2 = 24>o/J3B, and z = 6. On substitu-
ting these values of d and z in Eq. (8.5) we obtain !J.~tJ=J.'§(B); this function is shown qualitatively
in Fig. 8.2 and the minimum value of M tJ yields the magnetic induction, B , for a given external
field, H. Carrying out the minimization process for each field H we can develop the function
B = B(H ) (or M = M(H ) through B = H + 4n:M) . (Further analysis of this model shows that
X = (3M/aH)T does diverge as H --> He! from above.)
The above analysis breaks down as nL increases (i.e., as B increases), and more powerful
techniques are required. The simplest method involves the Ginzburg-Landau (G-L) theory, a
discussion of which we begin in the next section.
In discussing the thermodynamics of a Type II superconductor the concept of the ther-
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 41/511
Type" superconductivity 27
modynamic critical field, H e' is retained in terms of the condensation energy, H~ /Sn ; however,
He! < He < Hc2' From the discussion in Sec. 4 (where we showed (l/4n)Sg,B'dH =0) we
identified H;/Sn as the condensation energy. The fact that M has a different field dependence in a
TypeIIsuperconductor does not alter this identity.
Therefore
fH 1
c M(H )dH = __ H ~.o Sn
(S .6 )
Hc2 greatly exceeds He in certain alloys and intermetallic compounds. This fact, coupled with the
experimental observation that a zero resistance state often persists up to H e2, makes Type II
materials of great importance in the manufacture of magnets and related technologies.?
8 . 2 The region H C 1 < < H < < H C 2
In the region He! < < H < < He2 there is a densely packed array of vortices; however, the
spacing between vortices ( ~ n C 1/2), d , satisfies the inequality ~ < < d < < A (here we assume that
( < < A so as to guarantee the existence of such a regime). When d >>~,H(r) is accurately given
by the solution of the inhomogeneous London equation
( S . 7 )
where iruns over all vortices in the lattice. Assuming the vortices lie on a periodic lattice:' H(r)
may be expanded in a two-dimensional Fourier series as
H(r) =HGe-iG,
c
(S.Sa)
where
(S.Sb)
here G denotes all vectors of the two-dimensional reciprocal lattice associated with the real space
lattice of the vortex array and nL is the reciprocal of the area of the unit cell. Inserting (S.Sa) into
(8.7),using the fact that V x [V x H(r)] =-V2H(r) (since V' H(r) =0), multiplying by eiG"', and
integrating over d2r. and noting G'·ri is a multiple of 2n we obtain
, s ¥ 'I1+ A 2G
2
)H(A;G' = ¢oNLzC
2. In the presence of a transport current (as one has in a magnet) the nux lines arc subjected to a
Lorentz force which would tend to make the flux lines move, which is a dissipative process.
However, flux lines are usually pinned (immobilized) by various inhomogeneities present in the
material (c.g., defects, grain boundaries, ctc.),
3 . In an inhomogeneous superconductor, vortices may be attached to so-called pinning sites, resulting
in deviations from periodicity. On the other hand, thermal agitation may cause the lattice to melt in
a high temperature superconductor.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 42/511
28 Part I Phenomenological theories
or
(S.9
where sf is the total cross sectional area of the vortex lattice, IVL is the total numbers of lines ant
n, = = NLied; henceforth we assume that H(r) and Hc are directed along the z axis and denote onl-
their amplitudes, H( r ) and H c '
The total energy of the vortex lattice follows from Eq. (7.11) as
Defining the energy density I f f ' =E/sf L we obtain
(S.10
Now the minimum nonzero value of I G I ~ 2 n /d( ~ 2 nn~ !2 ) and with our assumption d < < ;.w r
have ; . 2 G ; ; : ' i n >> 1; therefore we may approximate 1/(1 + }.2G 2) by 1/).2G 2 . The sum in (S.10
depends on the particular lattice adopted by the vortices. However, following deGennes (1966)
we will limit ourselves to a semiquantitative estimate and replace the sum by an integral;
(S.11
From the above argument Gmill ~ 2n /d ; on the other hand we expect Gmax to be cut off at the scale
of an inverse coherence length and write Gmax =2n /¢ . We then obtain the total energy density 0
the lattice as
(S.12
here we used Eq. (S.4b) for He! and included a numerical factor f 3 to offset partially the various
approximations used to obtain (S.12). Matricon (see deGennes (1996)) finds f 3 =0.3S1 for a
triangular lattice where d 2=2¢o /J3B . We note in passing that by considering the changes in the
free energy arising from small distortions of the lattice from its equilibrium form, we rna)
calculate the magnetic contribution to the elastic constants of the lattice. We will return to this
topic in Sec. 20.7. Ignoring the entropy contribution we may replace If f by IF ; using the definition
1'! J =ff - -BH
4n(8.13;
we may then calculate B =B(H) from the condition aG/3B = 0 which is appropriate for a
constant-H environment. Carrying out the calculation (including the contribution from the
implicit dependence of d on H ) we obtain
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 43/511
Type" superconductivity 29
(S.14)
where{ J ' = = {Je-I!2. From the definition B =H + 4nM we have
tn ( { J ' ~ )He! c ,
M=-4;- ...( l ) .tn -
(
(i\.15)
Sinced X. B-I!2 we see that M increases logarithmically (becomes less negative) with increasing
magneticfield. This behavior agrees well with experimental data for materials with )_>>~.
8 . 3 Microscopic magnetic probes of
the mixed state
Wewill now briefly discuss two experimental probes of the inhomogeneous magnetization in the
mixedstate: neutron diffraction and nuclear magnetic resonance. Other probes which have been
applied or proposed include: decorating the surface with magnetic atoms (the Bitter technique)
and magnetic force microscopy. (For more coarse grained magnetic structures magneto-optic
and scanning Hall probe microscopy are useful.)
We begin with neutron diffraction. The neutron has a magnetic moment fln = (gn /2)(eh/
2M pc) where gn is the anomalous 9 factor of the neutron (gn/2 = - 1.91354) and eh/2M pc is the
nuclearmagnet on with Mp the proton mass. The neutron interacts with a magnetic field via the
usual spin Hamiltonian
f1 = - f i n · H(r), (S.16)
where f i n = fln!! where g is the vector Pauli matrix. The coherent scattering cross section, (J c -followsfrom the standard Born approximation expression
wheref(q) is the scattering amplitude given by
A 1 n f ~ · ·3f(q) = -. H(r)e'Q"rdr ;
2nh1(S.17)
here Mn is the neutron mass and q = k - k' with k and k' the wave vectors of the incident and
scattered neutrons ( I k I = I k' I for a diffraction experiment). Inserting Eq. (S.Sa), with H G given by
(8.9),into (S.17)we have
(S.lS)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 44/511
30 Part I Phenomenological theories
where V is the sample volume and the Kronecker delta function reflects the usual Ewald
diffraction condition, q=G. Writing B =1L< :P o , < :P o =2nnc/2e and Jin ~ (gn/2)(elJ/2Mnc) we have
(g .19)
Note that as discussed above ; c 2G2 >> 1; hence 6(q) = If(q) 12 o: (AG)-4. For this reason it is
very difficult to observe more than the lowest order peak. Experiments were first performed on
Nb by Cribier et al. (1964). Since high T; and heavy fermion materials have anomalously large
London depths neutron diffraction has not been a useful probe of the vortex lattice in these
materials.
Wc now discuss the expected behavior of the line width, ~w, measured in an nmr
experiment. For many purposes the line width may be inferred from the root mean square
deviation of the magnetic field from its average value"; i.e.,
(S.20)
and we take ~w =y~H , where y is the gyromagnetic ratio of the nuclei which are in resonance.
Here the bar implies an average over the sample volume. For simplicity we assume the external
(applied) magnetic field is strictly uniform and that the sample has the shape of a long cylinder
parallel to the applied field so that demagnetization effects may be ignored. Now H ( r ) in (S.20) is
simply B =1r< : P o ; H 2 ( r ) is calculated as follows using (S.Sa) and (S.9):
where "I is the area of the sample perpendicular to the magnetic field. Separating ofTthe G =0--2
term (which cancels the [ H ( r ) ] term in (S.20)), restricting to the limit A 2G 2 >> 1, and again
replacing the sum by an integral we obtain
The integral is convergent at the upper limit so we extend Gmax to cc and we again take
Gillin ~ 2n/d (valid for a square lattice) yielding
(S.23)
Note that our result is independent of magnetic field and within a factor of order unity is equal to
H c L' Therefore nmr is readily observable in extreme Type II materials containing nuclei with
sufficiently large moments. We emphasize that this line width is not to be interpreted as thereciprocal of a magnetic relaxation time; it is referred to as an inhomogeneous broadening. The
relaxation of the magnetization will be discussed briefly in Subscc. 34.2.
4, For a discussion ofmoments in nmr seeAbragam (1961),p. 106.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 45/511
. f ---~- 'h-e-G-in-Z-b-U-r-g---L-a-n-d-a-u--
theory
9.1 Basic equations
In 1937Landau developed a model to describe second order phase transitions (those involving
nolatent heat). Central to this theory was the introduction of the concept of an order parameter.
The order parameter is an appropriate quantity which vanishes in the high temperature phase
(T> Te) but which acquires a nonzero value below the transition (T < Te) ' The identification of
theorder parameter is often obvious from the nature of the second order transition. Thus for the
ferromagnetic transition it is natural to identify the spontaneous magnetization, M, as the order
parameter.
The ferromagnet brings out a fundamental property of second order phase transitions: the
development of a spontaneous magnetization is accompanied by a reduction in the symmetry of
the system. Thus if the material is iron, where the high temperature (nonmagnetic) phase has a
body centered cubic (bee) structure, on passing through the transition (Curie) temperature
(T c =1043 K) the material chooses one of several symmetry-equivalent crystal axes along which
themagnetization develops; the choosing of one among several directions (for iron the directions
ofeasy magnetization are the cube axes and there are six such directions, ± x , ± y , ± z ) lowers
thesymmetry of the crystal. The system is then said to have 'spontaneously broken symmetry'. It
isa general property of second order phase transitions that the low temperature phase always has
a lower symmetry.
Since the order parameter evolves continuously from zero below Te, it is natural to expand
thefreeenergy as a power series in the order parameter. The free energy is a scalar but the order
parameter may be a higher-dimensional object (e.g., a vector, tensor, complex numher). For our
exampleof the ferromagnet, the order parameter is the magnetization, M, a vector. Thus in
making the expansion of F (M ) we can admit only symmetry-invariant combinations of the
c om pon ents, M i; since the magnetization may develop along any ofthe (easy) crystal axes the free
energy expansion must preserve the full symmetry of the high temperature phase. For a cubic
ferromagnet the expansion of the free energy satisfying the above requirements has the form
F (M , T) =F (O , T) + r x (M ; + M ; + M ;) + ~f:il(M ; + M ; + M;)2
1 ?
+ 2f:i2(M xM y + M yM z + MzMx l - · (9.1)
Expression (9.1) is invariant under all the symmetry operations of a cube. However, to simplify
31
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 46/511
32 Part I Phenomenological theories
F-Fo
F-Fo
0: < 0 j
(a)M
~~o\~ LLM
···~~a2/2~
(b)
Figure 9.1 G L freeenergyfunctions for T> T , (ex> 0)and [or T < T , (e t.< 0).
the discussion consider a (hypothetical) isotropic ferromagnetic liquid. The only rotationally
invariant scalar that can be formed is then M· M and (9.1) simplifies to
(9.2)
Limiting the expansion to fourth order will lead to an adequate description of ferromagnetism.
Thermodynamic stability requires / J >0 (otherwise the system would seek a divergingly large
magnetization). However, (f.may have either sign. If o . > 0 the minimum of(9.2) occurs at M =0;
if : x < 0 the minimum is for M #. O . These two situations are shown in Fig. 9.1. Hence we can
model a second order phase transition simply by arranging for the sign of (f. to change at T;which
is easily done by writing
(9 .3 )
it is sufficient to regard f J as a constant. Setting i J F / a M = 0 yields
(9.4)
This equation has the two solutions discussed above:
e x > 0, M =O (T> Tel;
(T < Te).
(9.Sa)
(9.Sb)x < 0,
In terms of Eq. (9.3) we have
1 \ . , f 2 =° (T> TJ; (9.6a)
(T< TJ. (9.6b)
It is important to recognize that the ferromagnetic state is degenerate. For the case of the
crystal it is degenerate with respect to the number of independent crystal axes (or directions) the
magnetization might orient along (six for our case of bee iron with easy (100) axes). For our
ferromagnetic liquid the ordered state is continuously degenerate: M can point in any direction.
There is an important aspect of the second order phase transition which is neglected in the
above model: fluctuations. All thermodynamic systems undergo fluctuations. As an example the
density of a liquid may fluctuate by an amount o p. However, this is at the expense of an increase
in the local pressure, ()P, governed by the bulk modules. This is a general feature: the 'restoring
force' (pressure in our example) is linear in the 'displacement' (here the density). For a system
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 47/511
The Ginzburg-Landau theory 33
whichundergoes a second order phase transition the order parameter becomes a thermodynamic
variable below the transition. For our magnetic system the generalized restoring force accom-
panying a change in the magnetization 3M follows from expanding the free energy about its
equilibrium value, M , However, as the transition is approached the linear restoring force goes to
zero:just at the transition the restoring force is proportional to (()M)3. We see that fluctuations
can have a profound effect on the system near its transition temperature. The Landau model
therefore assumes that the temperature is far enough from the transition temperature that
fluctuation effects may be ignored. This is referred to as the mean field model.
We now take up the case of the supercond uctor. Our earlier discussion has brough t out the
idea that superconductivity is some kind of macroscopic quantum state. Ginzburg and Landau
built this idea into the Landau second order phase transition theory by assuming the existence of
a macroscopic 'wave function', 1 / 1 , which they took as the order parameter associated with
superconductivity. Since wavcfunctions can be complex, only the form 1 / 1 1 / 1 * may enter the free
energyexpansion; we therefore write
(9.7)
Theminimization proceeds exactly as above, i.e.,
1 1 / 1 1 =0,
[a( T c - T)]1!2I I / f 1 = { 1 ,
(9.8a)
(9.8b)
To describe situations where the superconducting state is inhomogenous we must generalize (9.7).
F is then interpreted as a free energy density, .?(r), and we write
F = F o + l?(r)d3r
=Fo+ fd3r[xll/I(rW+~fill/l(r)14J(9.9)
F is now the total free energy. Eq. (9.9) in its present form does not model the increase in energy
associated with a spatial distortion of the order parameter, i.e., effects associated with a coherence
Iength.Z. To account for such effects Ginzburg and Landau added a 'gradient energy' term to (9.9)
ofthe form
(9.10)
with m* as a parameter; the choice of the coefficient f j2 /2m* makes (9.9) mimic the quantum
mechanical kinetic energy (introduced earlier in Eq. (2.10)). Ginzburg and Landau assumed that
if(9.10) was to be regarded as the kinetic energy contribution to the Hamiltonian density of the
superconducting electrons, then (as in Hamiltonian mechanics) the interaction of the electrons
with an electromagnetic field would be accomplished by the Hamiltonian prescription
e*p-->p--A
c
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 48/511
34 Part I Phenomenological theories
or, since p --->(~) V in quantum mechanics,
ie*
V--->V--k he ' (9.11J
the use of e* aIlows the superconducting entities to carry a different charge (e * = 2e in BCS
theory). Combining the above we have
f 1 i 2 / [ ie* ] / 2G = _ V - - A(r) ljI(r ) d3r.2m * he
(9.12)
Finally we must add the contribution of the magnetic field to the energy density
(2.9)
Combining the above we have
F = F 0 + f d 3r { x IIjI(r) 12+!i 1~J(r) 14 + 112 / [V _ ie* A(r)] ~J(r) / 22 2rn* he
+ ~H 2 (r)l. (9 .13)
gn 5The minimization of (9.13) must be carried out using the methods of the calculus of variations
since F is a functional involving the free energy density ff(l jI(r), I/J*(r), VIjI(r), V~J*(r), H(r)) which in
turn involves the unknown functions l jI( r) , 1 jI* (r ), and H(r)( = V x A(r)).
Minimizing (9.13) with respect to 1jI*(r) yields
f { 1 z 2 [ ie * ] 2 })F = d 31' - - V - - A(r) l jI( r) +x lj l( r) + fJll/J(r) 121j1(r) bl j l*(r)
2m * lie
f lil [ ie * ]+ d2r'- V - --:-A(r) l jI(r)bl j l*(r)2rn* he
(9.14)
(variation with respect to 1jI, which is an independent variable, yields the complex conjugate of
(9.14)). To minimize iF we set the integrand of the first part of (9.14) to zero; this yields the first
G-L equation
1 1 2 [ ie*]2- -* V - --:- A(r) l jI(r) + cxl j l (r) + Ii IIjI(r) 121j1(r)= O .
2m he
(9.15)
The surface term (which was generated by an integration by parts) can be used (with caution) to
establish certain boundary conditions and will be discussed later.
Variation of(9.13) with respect to A (with H = V x A(r)) yields Ampere's law
v006b;0114n
V x H(r) = -j(r)c
(9.16)
prov ided we identify j as
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 49/511
TheGinzburg-Landau theory 35
e* {[ e*] [ e* J }(r)=* 1jJ*(r) - ihV - - A(r) l jJ(r) + l jJ(r) + ihV - - A(r) 1jJ*(r) ,~n c c
(9.17a)
or equivalently,
- ie*h e*2j(r)=-- [~J*(r)V IjJ(r) - ljJ(r)VIjJ*(r)]- --IIjJ(r) 1 2 A(r).
2m* m*c(9.17b)
Eq.(9.17)is the second G-L equation; we note that (9.17) is the same as the expression for the
current density in quantum mechanics. Note the current density satisfies the equationj(r) = e(I)F /
bA(r));i.e., it is the variable conjugate to A(r).
9 . 2 Gauge invariance
Thesimplest solution of (9.15) is for the case of a uniform superconductor, IjJ #. l jJ(r), with A = 0,
as givenearlier in Eqs. (9.8). However, (9.15) possesses a continuum of other solutions having the
same free energy, which we now show. As we can with any complex function, we write
~(r)= a(r)ei<l>(r)hereatr) and cJJ(r)arc the position-dependent amplitude and phase. respectively.
Letus examine a class of solutions which satisfy the complex equation
(ie* )
V - -A(r) l jJ(r) = 0,he
(9.18)
whichis equivalent to the two real equations
Va(r) = 0 (9.19a)
and
[VcJJ- e* A] = O.
he(9.19b)
From (9.19a)we see that the only allowed solutions of(9.18) involve a constant amplitude, ,,; Eq.
(9.19b),on the other hand, has infinitely many solutions involving a vector potential and a
position-dependent phase (which does not affect the free energy) related by
heA = -VcJJ.
e*(9.20)
Anyvector potential satisfying (9.20) results in a uniform free energy and (on substituting (9.18)
into (9.17))a vanishing current density (note that H = V x A = 0 for all A of the form (9.20)).
The above exercise shows that the symmetry broken in superconductivity is qauqe symme-
try, or equivalently, phase symmetry. Superconductors having different phase functions, (D(r),are
in a real sense physically distinct/ this arbitrariness of the phase is the analogue for a supercon-
1. Strictly speaking we cannot determine the absolute phase of a superconductor, but in our later
discussion of the Josephson effects we will show that phase differences can be measured.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 50/511
36 Part I Phenomenological theories
ductor of the property that the magnetization may point in any direction in an isotropic (liquid
ferromagnet.
9.3 Boundaries and boundary conditions
We first examine a simple case involving an inhomogeneous order parameter generated by the
presence of a boundary, in the absence of a magnetic field. Assume we have a superconducting
half space occupying the region x > O . We further assume that the order parameter is driven tc
zero at this interface. Experimentally this can be accomplished by coating the surface of th e
superconductor with a film of ferromagnetic material.' We then seek a solution to the one-
dimensional G-L equation
(9.21;
Noting o : is negative in the superconducting state (r x = - I xl), defining the G-L coherence length
as
l i 2[2= __ -
- - 2 rn* I o : I'
and writing ( f J / I o : l ) t j J 2 =? we may rewrite (9.21) as
- ~2r -f + f3 = O .
(9.22)
(9.23)
Multiplying by f' we may rewrite (9.23) as
(9.24)
hence the quantity in square brackets must be a constant. Far from the boundaryJ' =0 and
[2 = 1 (equivalent to t j J 2 = I x l I f J ) ; then (9.24) becomes
(9.25)
which has the solution( = tanh(x/J2~) or
(9.26)
From (9.26) we see that ~ is a measure of distance over which the order parameter responds to a
perturbation. Since x = a(T - T e l we have
" ( l i 2 ) 1 , 2 ( T ) - 1 / 2dT)= -- 1--
2 rn*aT c t;
2. Paramagnetic impurities (those bearing a spin in a host material) or interfaceswith a ferromagnetic
metal strongly depress superconductivity. A normal metal interface has a much smaller effectand an
insulator or vacuum has a negligibleeffect formost purposes.
(9.27)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 51/511
Ine l2mzburg-LanClau theory 37
We see that the G L coherence length diverges as 1/(1 - T/TjiZ; this divergence is a general
property of the coherence length at all second order phase transitions (although the exponent
differs in general from this 'mean field' value of 1/2 close to TJ
We now examine Eq. (9.17b). In a limit where the first two terms are negligible, we
recognize the remainder as the London equation, as written in the form (3.5), provided we identify
n = I IjI IZ ; this supports the identification ofl j l as a (condcnsate) wave function associated with th e
superconducting electrons. We can immediately caleulate the London penetration depth as
(9.2Sa)
or
(9.28b)
Comparing Eqs. (9.27) and (9.2S) we see that in the G-L theory ),1. and ~ both diverge as
(1 - T/TJ-1i2. Their ratio, called the G L parameter, is therefore a constant which we write as
(9.29)
Let us now return to the discussion of the surface tension of a normal-superconductor
phase boundary begun in Sec. 6. From Eq. (6.3) the total Gibbs free energy, which is a constant, is
given by
G =W f + dx [§(X) - _ _ 1 _ _ H e .B(X)J;-"0 4n
(9.30)
here .9 1 is the interface area, : 1 F is the G-L free energy density and the integral must now be
extended into the region x > 0 since, due to the gradient term in the G-L model, the supercon-
ducting properties do not turn on abruptly at the interface (which we still locate, nominally, at
x =0) . Inserting (9.13) into (9.30), employing a gauge where A = A(x)z, and using the definition of
igiven by (6.4), we obtain the one-dimensional equation
f+ [ r x 2 1 7
z1 dljl(x) 1 2 e*2
}'= dx -+- ~- +--.zA(x)lljJ(xW_ < C D 2 f3 2m* dx 2m*c
1 BZ(x) H B(X)J+ r x I I j I ( xW + - f 3 1 1 j 1 ( x ) 1 4 + ~. __ ~e _._ •
2 Sn 4n(9.31)
The vanishing of the cross term iA' V in both (9.31) and the first G-L equation (9.15) allows us tochoose I j I real; it then follows from (9.17) that
(9.32a)
and
i,=,= . (9.32b)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 52/511
38 Part I Phenomenological theories
To compute (we must simultaneously solve the first and second G-L equations, (9.15) and (9.16),
for ~J(x) and B(x) ( = - (dA(x) /dx)) , subject to boundary conditions which will be specified
shortly, and insert the results in (9.31). To eliminate various parameters in the subsequent
calculations we rewrite all equations in terms of the scaled variables:
(9.33)
In terms of these variables the first and second G-L equations become (where we now drop the
bars over the variables)
(9.34)
and
(9.35)
The solutions of (9.34) and (9.35) for arbitrary I(must be obtained numerically. The appropriate
boundary conditions are
~J = 0 , B =A' = 1 at x =- e x ) (9.36a)
and
t / J =1, A' =0 at x = + ex; (9.36b
(where x > 0 is nominally the superconducting side). 3 The behavior of B(x) and t jJ(x) in the small k
regime where the field varies more rapidly than the order parameter is shown in Fig. 9.2. Tc
obtain a first integral ofthcse equations we multiply (9.34) by tjJ',which leads immediately to
1 d ' 2 d [(1 2 ) 2 1 4 ] 2 d (A2)-(~J )=- -A -1 tjJ +-tjJ -t jJ - -,1(2 dx dx 2 2 dx 2
(9.37a
and multiply (9.34) by A', which yields
~(A '2 ) _ tjJ2~ (A2)dx ,2 - dx, 2 .
Combining (9.37a) and (9.37b) we have for our first integral
(9.37b
(9.38
the value of the constant was fixed by the boundary conditions at either + 00 or - o : » . Usin:
(X2 !2~ =H:/8n and the scaled variables of Eq. (9.33) we may rewrite Eq. (9.31) as
) H Z f +c(=--
Sn _
3. From the structure of Eqs, (9.34) and (9.35) it follows from the boundary conditions (9.36a) and
(9,36b) that t i t ' =0 at x =± 00. Our restriction to real t / I requires that the constant A (which is
equivalent to a phase < 1 » vanishes at x=+ Xl.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 53/511
The Ginzburg-Landau theory 39
r-----s--~
K ll> I
Figure 9.2 Schematic diagram of the variation of Band 1 / / in a domain wall. The case
/{< < 1refers to a Type I superconductor with a positive surface energy; the case
/{>> Iefers to a Type II superconductor with negative surface energy.
Combining this expression with Eq, (9.38) we have
) H 2 f " % [ 2 ]=~ _1jJ'2 + A'(A' -1) dx;
4n K2(9.39)
givenA'(x)( = - B(x) ) and tj/(x) we can compute (.
Returning to Eq. (9.38) we verify that in the limit A ' =0 we recover the dimensionless form
ofEq. (9.25);
(9.40)
Thesolution of this equation is
K(X - x o)tj/(x) = tanh-~,------'--
)2(9,41)
(whichis the analogue of Eq. (9.26)); here Xo is the nominal position of the boundary. The
characteristic length scale of the order parameter variation associated with this normal/super-
conductor phase boundary is ~ (in un scaled units), as discussed earlier in connection with Eq.
(9.26).This form docs not satisfy the boundary conditions (9.36), as expected, since a stable phase
boundary in an unbounded superconductor can exist only in the presence of a field, but rather
satisfiesthe boundary conditions I j J =+ 1 or - 1 as x - - - > + cc , However, assuming the presence
ofan order parameter quenching mechanism at Xo =0 (e.g., a thin ferromagnetic plane embed-
ded in an otherwise homogeneous superconductor) we may evaluate the associated surface
tensionby substituting (9,41) into (9.39); carrying out the integration (with A = 0) we obtain
(9,42)
From the definition of K given in (9.29) we see that :' ~ ~(H; /8n) as anticipated earlier in Eq. (6.6).
WhenA is nonzero the characteristic length scale of a field variation may be estimated from Eq.
(9.35).In a Type I material at a vacuum/superconductor interface I j J may be regarded as constant
overdistances where the field varies. Since the length scale of Eq, (9.33) is unity this corresponds
to the field varying over a distance A L (the London depth) in unsealed units. However, near a
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 54/511
40 Part I Phenomenological theories
normal-metal-superconductor phase boundary the field variation occurs in a region where I / ! is
small. Ifwe seek an approximate solution to (9.35) as an exponentially decaying form, A .~ e~X!,j
where () is a characteristic length, then l/(F ;:::: I / ! 12. From an expansion of Eq, (9,41) about
x =xo, I / ! ~ K (X - x
o). The average value of 1 / ! 2 in a region of width b would be of order (K3f
Combining these forms we see that the field decay would be governed by (K6)2 ;:::: l/ b 2 or
() .~ K~ 1/2. This would result in a (negative) surface tension contribution of order - Kl/2y which,
although less than the expression (9,42) in a Type Imaterial, decreases slowly with K and hence
limits the accuracy of this expression.
With increasing K the surface tension continues to decrease and further analysis (see
Lifshitz and Pitaevskii (1980), Sec. 46) shows that it passes through zero for K =1//2.
We end this section with a discussion of the boundary conditions at an interface between
two dissimilar materials at least one of which is superconducting. A boundary condition which is
appropriate for the case when no current flows parallel to the surface normal, i i , is (see Fig.
9.3(a))4
[e* J 1- ihV - ~A . i i I / ! = - -I/f,C b
(9,43)
as is easily verified by substituting in Eq. (9.17a); here b is a parameter having the units of length.
The superconductor insulator case corresponds to the limit b - co,
Lastly we consider contact between two superconductors in the presence of a current
normal to the boundary. In the most general case the boundary, which we will locate at x =0,
may have properties different from both superconductors, an example being a thin layer of a third
(normal) metal, which is referred to as an SNS (supcrconductor-normal-superconductor)junc-
tion. The most general (Cauchy) boundary conditions connecting the two sides of the junction,
denoted 1 (x < 0) and 2 (x> 0), have the form (deGennes 1964)5
(9,44a)
and
(9,44b)
Ifwe choose a single gauge for regions 1 and 2 the vector potential will also be continuous, The
coefficients Mij are not independent but arc constrained by the requirement of current conserva-
tion through the boundary:
(9.45)
as a result we may choose the M ij to be real and they must satisfy the condition
M llM 22 - M 12M 21 = 1.
The resulting expression for the current through the boundary.j., is
4. We emphasize that Eq. (9.43) is a macroscopic boundary condition. The behavior of the microscopic
order parameter may differ substantially near the interface.
S . A boundary condition of this type wil l be discussed from the microscopic point of view in Part III,
Sec. 42.
(9,46)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 55/511
The Ginzburg-Landau theory 41
t \ 1 I f ( x ) \
I
hll}jx o
(b)
x
(a)
Figure 9.3 Schematic of the order parameter behavior: (a) near an NS boundary; (b) in
an SNS junction.
e*h= -* - 1 1 / 1 1(0) 1 1 1 / 12(0) I sin [(f)2(0) - (f)1 (0)].m J W l2
(9.47)
The maximum value of this current is
e*/iJ m " x =~ 1 1 / 1 1(0) 1 1 1 / 12(0) I ·
m 1Vl12
(9.48)
If jmax is much less than the bulk critical currents in the two superconductors then we may
approximate the I / I j ( x ) by the A = 0 forms analogous to (9.26)
(9.49a)
and
(9.49b)
here Xl and X2 are free parameters and ~1.2 are the G-L coherence lengths in the two media.
From Eqs. (9.49) we have immediately
(9.50a)
and
(9.50b)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 56/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 57/511
Of--~-'h-e-u-p-p-e-r-c-rl-·t-ic-a-'-~-ie-'-d-o-f-
a Type II superconductor
Asthe magnetic field is raised above He1, and the density of flux lines nL increases, a point is
eventually reached where the distance between flux lines becomes of the order of the vortex corediameter;i.e. n e 1/2 ~ ~. One would then expect a transition to the normal state, and the field at
whichthis occurs, called the upper critical field, is designated He2• In the region just below H c2 the
superconducting order parameter must be small and this is the regime where the G-L approach
shouldprovide a good description.
We may calculate Hc2 by linearizing the first G-L equation, since the order parameter is
vanishingly small just at Hc2' Eq. (9.15) then becomes
h2
( ie*)2_ - V _ - A ,I, + (X'/' = ·2m* he 'I' 'I' ,
(10.1)
thisequation is the same as the Schrodinger equation for a particle with energy, _ a, mass, m",
andcharge, e* , in a magnetic field, H,with an associated vector potential, A. We will assume a
uniformfield Ho I I i.The solution of(10.1) is easiest in the so-called Landau gauge where we write
thevector poten tial as
A =HoY x ; (10.2)
theresulting G-L equation is then
(10.3)
We may separate the variables in (10.3) by writing I j J n . k z . k J X , y , Z ) =; k z z + ; k y Y u n ( x ) (we do not
normalize I j J at this point as this property follows only from solving the nonlinear G-L equations,
whichwe address in Sec. 14). Inserting this wavefunction in (10.3) we obtain:
[1 1 2 d
21 ]
_ -_ --2 + - m*w~(x - XO)2 u n i x ) = n U n ( X ) ,
2m* dx 2
( lOA)
whereXo = ( l1e/e*H o)k y = (¢0 /2nHo)ky , and f,n = - rx- (h 2k; /2m*) . Eq. ( lOA) is the Schrodinger
equation of a harmonic oscillator with frequency we = I e* I H o/m *e (the cyclotron frequency of
theparticle) with energies
(10.5)
43
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 58/511
44 Part I Phenomenological theories
having its origin at the point xo' The full (unnormalized) wavefunctions are
'(k +k) ( )'2 2 ( x - x o )/, (x V z) = e' ) " y . ,z e " X-Xo~! " H H - - -V ' n.k;:,kv· ,~, n'
. all
(10.6)
where af r = = h /m*wc = hc /e*H o =¢o /2nHo·
Only the smallest eigenvalue, n =0, k; =0 solution (corresponding to the highest field at
which superconductivity can nucleate in the interior of a large sample) is meaningful since our
linearized theory is valid only as a description of the onset of superconductivity. Hence
(10.7a)
or
2m*caH ----c2 - he"
2m*c=--a(T -T)
ne* c
(10.7b)
or, as it is more commonly written,
H -~.c Z - 2n~2'
(10.7c)
in terms of the length aH, Eq. (1O.7c)may also be rewritten as aHel=~From (1O.7b)we see that
lld goes to zero at T ; and increases linearly below that temperature. Recalling nL= / 4 > o , we see
that (10.7c) is in accord with the order-of-magnitude estimate of H cz made at the beginning of this
section. However, we are then struck with the fact that in our particle analogy, we have
considered only a single quantum state (n =0, the 'ground state'), and yet the material contains
the largest possible number of flux lines. The resolution to this apparent contradiction is that our
n = 0 quantum state is highly degenerate (as evidenced by the fact that the quantum number k;
does not affect the energy).
The relation between this high degeneracy and the high density of flux lines will be dealt
with in Sec. 14, where we discuss the region of fields just below H cz .
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 59/511
.I--~-'h-e-a-n-i-s-o-t-ro-p-ic-----
superconductor
Anisotropy may be incorporated in the phenomenological G-L theory simply by introducing an
'effective-mass tensor' into the 'kinetic energy' term in Eq. (9.15) in the form
[ - in ( v - i:* A ) J ~[~* J [ - in ( v - i;: A ) ] ~ + a~ + P I ~ 12~ = 0,
(11.1)
wherel/m* is an effective-mass tensor.
For a system which may be regarded as uniaxial, the reciprocal effective-mass tensor may
bewritten as
0 0
[ ~ * l jmx
0 0 (11.2)mx
0 0mz
wherethe z axis is the symmetry direction. The inverse of Eq. (11.2) yields the mass tensor m*.
We now obtain an expression for the upper critical field Hc2(T). Near the field-dependent
transition temperature the order parameter is vanishingly small and we may again neglect the
nonlinear term. Our equation is then formally identical with the Schrodinger equation of a
particlewith charge e* and an anisotropic mass tensor m* in a uniform magnetic field Ho and the'energylevels' have the harmonic oscillator form
- a = (n + 1/2)nwc(O), (11.3)
wherewJ8) is the angle-dependent cyclotron frequency which is encountered in the effective-mass
theory of cyclotron resonance in semiconductors. The latter can be worked out classically simply
by solving Newton's equations of motion for our 'particle' moving under the influence of the
Lorentz force:
e*m* 'v = -v x Ho;c
(11.4)
here v is the velocity. The solution of this equation involves elliptical orbits which are traversed at
a frequency
(11.5)
45
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 60/511
46 Part I Phenomenological theories
here 0 measures the angle of the magnetic field from the z axis. The solution with the lowest free
energy corresponds to n = 0 in Eq. (11.3) and the upper critical field is thus given by
2ea(Tc - T )
Hd8) = . 2 2 2 12'ne*[slll 8jrnxrnz + cos 8/mJ ! (11.6)
Ifwe define two coherence lengths,
and
and recall the 'flux quantum' 1 > 0 = he/ I e* I, then Eq. (11.6) may be written as
(l1.7a)
in particular, for fields parallel and perpendicular to the symmetry plane of the material we have,
respectively,
(11.8a)
and
1 > 0HCH =~.
2n<,x(11.8b)
The angular dependence of the upper critical field is thus characterized by the two parameters Cand ((Lawrence and Doniach 1971). Note that (11.8a) and (11.8b) both lead to a linear behavior
of the upper critical field in T; - T. In terms of the definitions (11.8a) and (11.8b) we may rewrite
Eq. (l1.7a) as
H~2 (O) 2 LJ H~2(8) . 2 LJ _ 12 cos ()+ 2 Sill () - ,
HcH He211
(l1.7b)
which is the equation of an ellipse.
Eqs. (9.17) can be generalized to the case of an anisotropic superconductor by replacing
rn* -1by Ijm* from Eq. (11.2); an anisotropic London penetration depth then results.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 61/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 62/511
48 Part I Phenomenological theories
The behavior ofEq. (12.4) is radically different from that of Eq. (11.8aJ.The reduced dimensional-
ity has the effect of replacing ~z by d/(12)l!2. Since only one factor of'; enters Eq. (12.4), the
temperature dependence of He21i is altered from H c2 'I «: T; - T to He21i C£ iT; - T)112 . We can
refer to these two behaviors as 3D and 20, respectively.
Let us now consider the case where the field is at an arbitrary angle with respect to the thin
slab. We take the vector potential in the form
A = y(xcos8 - zsinO)Ho . (12.5)
Writing ljJ(r) =u(x,z)ei\y, the linearized G-L equation becomes
.itu + au=0, (12.6)
where
(12.7)
with
(12.8a)
and
- 112 e 2 1
.it' =-- ~ + - m*(j)~(z - Zo)2 s in ? 0 + m*(!)~x(z - zo)sin Ocos 8; (12.8b)2m* oz 2
here we have defined
Continuing our analogy to charged particle motion in a magnetic field, we note that there is no
constraint on the in-plane motion and thus the term in x2 would take on arbitrarily large values,
were we to treat it as a perturbation on an x-independent trial wavefunction; we have therefore
regarded this term as part of a zero order Hamiltonian, .ie' o - The eigenvalues of yt'o are simply
harmonic oscillator levels with a frequency We cos 8. The requirement that eu/ez vanishes at
z = ± d/2 again requires that (in lowest order) the wavefunction has no z dependence; i.e.,
u(x, z) =u(x). Since particle motion in the z direction is restricted, we can treat .ft' as a
perturbation. The eigenvalue is again minimized by setting k; =0 (the perturbation produced by
the last term in Eq. (12.8b) then vanishes by symmetry). The resulting perturbation produced by
x' is of the same form as Eq. (12.3). The total eigenvalue is then
(12.9)
Defining H e211 = (12)1/2cPO/2n(d and HcH =cP o!2n(2 we may rewrite Eq. (12.9) as
1 - H e2( O ) I III + H ~2 (8) " 2 O '- cos (J 2 Sill ,
».; He211
(12.10)
note that the angular dependence of this expression, first suggested by Tinkham (1963), is
radically different from that of Eq. (11.7b); in particular, it has a cusp at 8 = n/2.
We next treat the problem where the boundary condition is altered such that u(z) I z s= +dl,
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 63/511
I tun superconaucnng slaos
=O.T n practice, this condition is approximated by a superconducting film which is sandwiched
between two films containing a large concentration of paramagnetic impurities (such as fer-
romagnetic films) where 'pair breaking' effectively drives the order parameter to zero at the
boundary. We return to Eq. (12.1). For Ho = 0 (w e = 0) the eigenfunction satisfying our bound-
ary condition is
u = uocos (k z ) (12.11 )
with kdl2 =± n l2 for the ground state. This leads to a suppression of the transition temperature
governed by
h 1n2-rx=--
2m*d 2
or
(12.12)
here TeO is the bulk transition temperature and T, is the film transition temperature in the
presence of the new boundary condition. Application of a parallel magnetic field induces an
additional reduction in T; which we calculate from perturbation theory (Wong and Ketterson
1986):
<u( z ) 1 1 m *w~ z2 1 u (z )- brx=-'----__:_:-=----_.::_---'----'--
<u( z ) 1 u ( z )
1 2,(1 1)=2 m*w c d' 12 - 2n2
= 0.0l63m*w;d 2.
Combining Eqs. (12.12) and (12.13) gives
(12.13)
(12.14)
Eq. (12.14) may be written in terms of T', as
2 (m*C )2 2a(Tc - T)
Helll = 1 e* 1 0.0327m*d2·(12.15)
Ifwe define a coherence length involving Tc ' rather than Tco, by
(
h2 ) 1 / 2~= 2ma(Tc - T) ,
(12.16)
then we may write
5 . 53CPoH ---
e 2 1 1 - 2n,;d '(12.17)
whichis to be compared with Eq. (12.4). The angular dependence follows from inserting (12.17)
into (12.10).
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 64/511
~~s-u-r-t-a-c-e----------------
superconductivity
A superconducting half space in a parallel field may exhibit the phenomena of surface supercon-
ductivity (Saint-James and deGennes 1963). We take the interface to be the plane z =O.Thecritical field is still determined by the eigenvalue of Eq. (12.1) (k x =0 with the requirement that
a u / u z =0 at z=0). For values of Zo (i.e., k y ) such that the 'orbit center' is deep in the bulk of the
superconductor (z o -> - 00) the corresponding eigenvalue yields the bulk critical field; the
boundary condition is adequately satisfied since the ground-state wavefunction of our harmonic
oscillator is a Gaussian (centered on zo ) having negligible amplitude at the boundary. For the
opposite extreme, where we choose Zo = 0, the same Gaussian wavefunction also satisfies the
boundary condition, since the derivative at the peak of the Gaussian (now centered on Zo =0)
vanishes. The question arises whether there is a smaller eigenvalue (than the bulk eigenvalue) for
some value 0 > Zo > - 00. To examine this situation, we reflect the harmonic oscillator potential
V= tm*w;(z - zo)2 through the plane z = 0 such that
(13.1)
Note that this is not a harmonic oscillator potential but rather has two minima (located at
z = ± I Zo I ) and the slope a V/cz is discontinuous at z = O.Ifwe seek a nodeless solution for the
motion in this potential (which is symmetric under a sign change of z), the boundary condition at
z=0 will automatically be satisfied, and it should correspond to a possible ground state.
We can see immediately from Eq. (13.1) that an eigenvalue smaller than that of the bulk
must exist, since the potential described by Eq. (l3.1) is always less than or equal to that obtained
by extrapolating the potential from either half space into the other. The above-defined eigenvalue
problem admits an exact solution in terms of Weber functions (Saint-James, Thomas, and Sarma
1969). However, an adequate solution is obtained using a trial wavefunction u ~ e-rz2 (deGennes
1966). Minimization of o : with respect to rand Zo yields
(
2 ) 1 ! 2 I i W e I e* I I i H c 3- C i =1-- -=060---IT 2 . 2m*c' (13.2)
where we have defined the upper critical field associated with this surface superconducting state
as H e3. The exact solution yields 0.59 rather than 0.60 for the numerical factor. The upper critical
fields for bulk and surface superconductivity are then related by
(13.3)
50
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 65/511
Surface superconductivity 51
Surface superconductivity persists to higher magnetic fields (in a Type II material) and thus
may (in a conductivity measurement) mask the bulk transition. The state may be suppressed by
the following:
(1) rendering the surface rough (by sand blasting);
(2) using sufficiently high measuring currents to destroy surface superconductivity;
(3) depositing a normal-metal overlayer.
We now briefly discuss the case in which we have a thick film, of thickness d (with the origin
in the film center), rather than a superconducting half space. For films with d ; : ; ; ( , the amplitude
oftheG-L wavefunction at nucleation would be highest in the immediate vicinity of the surfaces
at z=± d/2 for He211 < H < H e3• As d is reduced, the eigenvalues (satisfying the boundary
conditions at z =± d/2 associated with bulk and surface superconductivity) approach each
other and, below some distance d =dc, only a state which has its wavefunction centered on z =0
isstable (Saint-James et al. 1969).
As we saw above, when the G-L equation for the magnetic field at an arbitrary angle is
written in terms of coordinates for which thc boundary conditions are conveniently applied, it is
nonseparable. Thus, for thick films where u ~ u(x, z) (rather than thin films where u ~ u(x)), we
must resort to a numerical, variational or perturbational technique. This complicated, and to
some extent not fully explored, problem has been discussed by Minenko (1983), Saint-James
(1965),Yamafuji, Kawashima, and Irie (1966) and Saint-James et al. (1969). However, we note
that, at some angle 8 = 8c' we expect the eigenvalues associated with the surface and bulksuperconducting states to merge.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 66/511
4 I > ~ ~ - h - e - T - ~ - p - e - ' - ' - - - - - - - - - - -superconductor for Hjust
belowHc2
The discussion of the upper critical field given in Sec. 10 involved the linearized G L theory and is
strictly valid only for the transition line in the H-Tplane where 11= HdT). For fields below thisvalue we must include the nonlinear terms in Eq. (9.15). T n this section we obtain an approximate,
but analytic, solution for the G-L equations which is useful for a small range of fields just below
He2. This problem was first examined by Abrikosov (1957).
We recall the solution of the linearized G-L equation obtained in Sec. 10 for a 'nucleation
wavefunction' at 11<2(for H I I z) which follows from Eq. (10.6) with n = 0 and kz = 0:
(14.1)
where we dropped the subscript from k; and Xo = = (hc/e*Hc2)k = 4Jok/2rrHc2 = (2k; here C is a'normalization constant', which vanishes for H =He2. As noted in Sec. 10, this wavefunction is
highly degenerate in that I < can have any value (so long as Xo lies within the superconductor).
The wavefunction which is the solution to the full G-L equations for H just below 1102 is
expected to resemble some particular solution ofthe linearized Eq. (10.3). At the same time, based
on our earlier discussion, we also expect some sort of vortex lattice to be present. We therefore
seek an approximate solution for t/Jas a superposition of terms of the form (14.1) chosen so as to
be appropriately periodic in both the x and y directions (thus forming the lattice). Eq. (14.1) is
periodic in the y direction as it stands with a period b = 2rr/l<. We may retain this periodicity in
the y direction while simultaneously forming a function which is quasiperiodic in the x direction
by substituting nk for k in (14.1) and summing over all n to yield
n= + o:
t/J(x,y) =C I einkY e-(X-XY/2I",
n~
(14.2)
where Xn =n~2k = 2r rn~2 / b and C is a (common) normalization constant. We define the period in
the x direction as a = = x, + 1 - Xn =2rr~2 lb. and the area of the unit cell is constrained by
ab = 2rr~2. Eq. (14.2) is not, however, strictly periodic in x: if we substitute x + a for x this is
equivalent to replacing n by It + 1 in the summation, which chanqes the phase of the waiefunction b y
eiky
; i.c.,
t/J(x+ a,y) =e-ikyt/J(x,y). (14.3)
This behavior is inherent in our choice of the gauge Ay =11 x ; under a coordinate displacement
x -> x + a, A y -> Ay + H a leading to a change in the canonical momentum nhk --+ nhk - (e* / c ) H a
52
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 67/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 68/511
54 Part I Phenomenological theories
We must also ineludc the effect of a change in the internal field, the magnetic induction Hir]. W
write this as a sum of two contributions arising from: (i) an explicit shift in the external fiel.
relative to He2 (since we have an exact solution for the latter), and (ii) a (screening) shift H(S)(I
arising from circulating supercurrents associated with the vortices (this latter effect is vanishingl
small for H = Hc2 and was therefore neglected in Sec. 10). We denote the sum of these contribu
tions by
(14.11
with
Associated with the internal field H(l) is a vector potential A( 1). Expanding (14.9) to first order i
A(l) yields (in a London gauge)
he" [( i e * ) ( i e * ) lI I j J 1 4+-. __ A(1)(f)· 1 j J * V--, Ao 1 j J - 1 j J V+-Ao 1 jJ * = 0 , ( 1 4 . 1 :21m*c he he
where the remaining terms cancel (since they satisfy the linear Schrodinger-like equation a
H =He2) '
The last two terms may be written as (l/c)j(S)(f)' A(ll(f) where we identify j(S)(f) as th
screening current density associated with the vortex lattice. Introducing V x H(s) s: (4rr!c)jls)(r
integrating by parts, and writing H<l)(f) =V x A(l)(f), Eq. (14.13) becomes
(14.14
where H(s) is parallel to H(l) by symmetry. To the accuracy we are carrying the calculation, H(s) i
the diamagnetic response field (H(s) = rrM(s)).
To obtain the field H(s) we first calculate the associated magnetization current, j(sl(f). Th
calculation of jIg) is facilitated by noting that the contours of constant 1 ~/(f) 1 and the (two
dimensional) streamlines of j(s) coincide, which we now demonstrate. We introduce the canonica
momentum operator n(=m*V where V is the velocity operator)
~ h e*II =-;-V--A.
Ie'(14.15
which has the commutation relations [IIx, I I y ] =(e*h/c)H. In analogy with the standard har
monic oscillator problem we introduce the raising (creation), n + =Ix + illy, and lowerin
(destruction), II_ = IIx - illy, operators." The ground state has the property I I _ 1 j J = 0 frorr
which we obtain
[(h a e* ) ( I i a e*
) l;----Ax -i -;----A, 1jJ=0.1 O X C l O Y C }
(14.16
Ifwe substitute th e wa vefunction in the form I j J =I j J 1 e;<!>nto (14.16), and identify the current ( in
lowest order) as
( 1 4 . 1 7 !
2. The linearized Schrodinger-like G-L Eq. (10.4) may be written Il + Il _ if ; =O.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 69/511
Type" superconductor tor Hjust below HC2 55
weobtain (on equating the real and imaginary parts separately to zero)
e* /1 Gj~ ) = I 1 / 1 I 2
2m* o y (14.1Sa)
and
(14.1Sb)
Fromthe structure of (14.1Sa) and (14.1Sb) we see that V I!/II.lj(s), The magnetization field, H(s),
followsfrom Maxwell's fourth equation, V x H(s)(r)= (4rr/c-jj<S\r);omparing this equation with
Eqs.(14,IS)yields
(14.19)
where the constant of integration was fixed such that H(s) vanishes when II/II= . We may
interpret H(s) as a magnetization through the usual definition B=H + 4rrM(s) where
M(S)(r)=- ( e*h /2m*c) I!/I1 2 z ; note it is negative as expected (since superconductors are diamag-
netic).
We now combine (14,12), (14.14), and (14.19) to obtain
---- e* /1 [ e*h
lJ I!/I1 4 + --I !/I1 2 Ho - H c2 - 4rr--I!/I 1 2 = .2m*c 2m*c (14.20)
Writing!/I(r)=~I of(r), where !/I6=Ix ] /fJ, and using the definitions (9.29) and (10.7) for I(and H c2
wemay write (14.20) as
4 ( 1 ) 2 ( Ho)1 - - - f 1 - - = .21(2 Hc2
(14.21)
The quantities p and r may be calculated provided we choose a symmetry (!/In or !/ID
from(14.6)or (14.S)) and a lattice spacing. It turns out that near H c2 the ratio
(14.22)
(introduced by Abrikosov (1957)) is independent of H 0' Introducing this quantity into (14.21) we
have
{2
. ( 1 )'r i A 1 - 2/,;2
(14.23a)
(14.23b)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 70/511
56 Part I Phenomenological theories
-- - -- ,--,2
From the above I V I I 2 =I e x l / f J ) f 2 and I if ; 1 4 =J A ( I if ; I 2 ) .
We now proceed to a consideration of the thermodynamics of a Type II superconductc
We begin with the magnetic induction which is defined as the position average of the intern
field, H(r): B = = Ho + I I ( s ) ; using (14.19) and the definitions of K and H c2 we have
(14.2
where we used the relation He2 = 2 K H c in the second step. Substituting (14.23a) for f2 v
obtain the thermodynamic magnetic induction B=B(H 0) as
(14.25
We may invert this relation to obtain Ho =Ha(B) as
(14.25
From (14.25a) we obtain the magnetization, (B - Ho)/4n, as
(14.25
note this is linear near H e2 as discussed earlier.
We now calculate the free energy. Comparing the variational expression (14.9) with E
(9.13) for the full free energy we have
Rewriting the first integral and substituting the unaveraged form of (14.24) this equation bccom
(14.2
Substituting Eqs. (14.23) we obtain F = F(Ho) as
( 1 4 . 2
Alternatively, using Eqs. (14.25), we may calculate F = F(B) for which we find
V[ 2 (B-Hc2)2 JF =Fo + gn B - - 1 - + - r - ' J
A
- ( 2 -K - - ; : . 2 - - - 1 ) . ( 1 4 . 2
For the case of a long cylinder with an external field, Hex!' parallel to the axis or a thin film w i
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 71/511
Type 1 / superconductor for Hjust below Hc2 57
000oo
(a)
o o(b)
Figure 14.1 Schematic diagram of square and triangular vortex lattices. The dashed
lines show the basic unit cells.
Figure 14.2 The spatial configuration of I if ; I 2 near H 02 for a triangular vortex lattice.
The numbers labeling the contours specify the square of the reduced order parameter.
(After Kleiner, Roth, and Autler (1964).)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 72/511
58 Part I Phenomenological theories
Hex ' lying in the plane (where Hex ' = Ho) wc would use (14,27): for the case where the external fiek
is perpendicular to the film (where Hex ' =B) we would use (14,28),
The only remaining problem is to calculate the Abrikosov parameter f 3 A' This calculation i
somewhat tedious and we refer the reader to Saint-James et al. (1969) for the details. The result
are
f 3 ~ = 1.18 (square lattice) (14.29a
and
r J ~ =1.16 (triangular lattice). (14.29b
The triangular lattice is thus slightly more stable. A schematic diagram of the square anc
triangular vortex lattices is shown in Fig. 14.1. The spatial configuration of 1 if ; 1 2 near Hc2 for,
triangular vortex lattice is shown in Fig. 14.2. The above treatment has neglected the elTectsof an]
in-plane anisotropy which, given that the structure is sensitive to f 3 A' can (and in some cases does
alter the observed symmetry.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 73/511
.,.1-------~ TheJosephson effects
15.1 The Josephson equations
Letus recall our expression for the G-L current density
ie*h e*2j = - -(lp*Vif; - if;Vif;*) - -I if;12A.
2m* m*e(9.17b)
Wewrite if; in the form if; = I if; I ei<I>s before, obtaining
e* fJ [ e* J=-1if;12 V<l>--A ;m* he
(ISJ)
i.e.,the current in a superconductor involves the gradient of the (gauge-invariant) phase (recall
that real G- L wavefunctions carry no current).
The G-L boundary condition at an insulator-superconductor interface was given earlier as
(1 1 e* )
fi· iV - ~ A if; = 0, (12.2)
whichis equivalent to no current flow through the boundary. Ifwe have a 'junction' which weakly
couplestwo superconductors (formed, for example, by a thin insulating layer between the two
superconductors through which electrons can tunnel), we must modify this condition. Taking fi
II, we replace (12.2) by more general (yet linear) boundary conditions 1
oif;! ie* I P 2-..,- - -h A,if;1= ---;-ox e A
(lS.2a)
an d
(1S.2b)
theparameter i is the same in both equations since it is a property of the boundary and not the
superconductor designations, 1 and 2. We insert (IS.2a) into Eq. (9.17b) yielding
. ie*h [ * oif;1 Oif;iJ e*2 2
t.= - -2 * if;! --;-. - if;! --;-. - -*-1 if;! I A,m uX ox m c
1 . This form is identical to Eq. (9.44) with Mll =M22 = 0, - MI2 = M';-/ = A .
59
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 74/511
60 Part I Phenomenological theories
ie*h [ (if; ie* ) (if; * ie* ) J e*2= _._ VI* __2. + -A if; - if; ~ - - A 1/1* - --11/1 12A .2m* 1 i " he xli J.* he x 1 m*e 1 x
(1 5
In the absence of magnetic a toms, the supercond ucting properties are in variant under til]
reversal, which results in if ; --> 1 / 1 * , j --> - j, and A --> - A. Under these operations both sides
Eqs. (15.2) turn into their complex conjugates and hence J . must be reaL We then obtain fro
(15.3)
(15.
Writing if;; = if;; I ei<I\ and assuming both sides arc prepared from the same kind of superconduc
ing material, I if ;1
I = I VI 2 I , we have
j=irnsin< l > 2 1 (15.:
where
(15.6,
and
(15.6t
Note j.; is the maximum current density that may be carried by the junction.
In deriving Eq. (15.5) we have assumed that no electric field and magnetic flux density an
present in the junction. When no electric field is present in the junction, the phase is time
independent. We generalize to the case in which a field is present by using a gauge invariano
argument. Under a gauge transformation (recall H =V x A and E = - V V - (l/c)(c'A/at))
A --> A + Vx (15.7
and
1 a XV-->V-----'-
c D t'(15.8
where x ( r , t) is an arbitrary single-valued function and Vis the potential. Since the G-L equation
contains the form V - (ie* / l7e)A, we must change the phase of the G-L wavefunction by
e*< l > --> < l> + ~ x( t).
he(15.9)
Comparing Eqs. (15.8) and (15.9) we see that the form
a < l > e*---V=Oat n
(15.10)
is qauqe-incariant. If V is (initially) assumed to be independent of time and denoted as V21> then
integration of (15.1 0) yields
(15.11)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 75/511
The Josephson effects 61
or
[e* J.' (0) .
J = J rn Slll 11>21 - Ii V21t . (15.12)
Introducing the frequency w] =ol1>21!ot we see that (15.11) leads to
(15.13)
Theoscillating currentj(t) given by (15.12) will be associated with an oscillating voltage which will
be superimposed on the static voltage, V2 I'
We next examine the Josephson effects in the presence of a magnetic field. We will restrict
ourselvesto the case of a relatively weak field where a quasielassical description is adequate; i.e.,
the dominant effect of a field, which is described by a vector potential A, is to make the phase
position-dependent. From the discussion surrounding Eqs. (9.19) we know that for a 'pure' gauge
field (one not involving a field H(r)) the only effect of the vector potential is to produce a
position-dependent phase; this suggests that in the presence of a vector potential associated with
aweakfield, H(r), the effect may be approximately incorporated in the phase of the wa vefunction.
Comparing Eqs. (15.7) and (15.9) we have the gauge-invariant form analogous to (15.10) as
e*VI1>--A=O.
he(15.14)
Therefore, the gauge-invariant phase difference is given by
_ (0) 2n f 2 .
11>21 -11>21 +- A dt.1 > 0 1
(15.15)
where 1 > 0 = = hel2e is the flux quantum.The fundamental equations governing the behavior of Josephson junctions are the cur-
rent-phase relation (15.5) , the voltage-phase relation (15.11) , and the gauge-invariant phase
relation (15.15) . They are believed to be exact. In the subsequent sections these equations are
applied to some simple junction structures and circuits. For example, electromagnetic radiation
isemitted from a Josephson junction in the presence of a potential (sec Sec. 17). Eq. (15.13) now
forms the basis for defining the standard volt in terms of a measured frequency and the
fundamental constants, e and h (Taylor, Parker, and Langenberg 1969).
15.2 Magnetic field effects:
the two-junction SQUID
Consider the superconducting circuit shown in Fig. 15.1 involving two Josephson junctions
connected by superconducting leads. The loop formed by this circuit is assumed to contain a
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 76/511
62 Part I Phenomenological theories
' l l = 'musin~<l>ll C
-- --a b
@H
d c-- ---
't = 'mtsin~<t>t
Figure 15.1 Schematic of a de SQUID consisting of two Josephson junctions
connected in parallel by supcrconducting links. The path of integration C is shown by
the dashed line.
magnetic flux, <D,arising from a position-dependent field, H(r), with some associated vector
potential A(r).
On entering the loop on the left in Fig. 15.1 (point 1), the current I splits into two
components III and It, where the subscripts refer to the upper and lower paths, respectively. From
Eq. (15.5) the total current arriving at point 2 is
I=Iu+1t
=ImusinL1l1>u+ Im t sin L1<D t, (15.16)
where Irnu and Imt correspond to the maximum currents associated with the upper and lower
junctions, and L1l1>llnd L111>{re the corresponding gauge-invariant phase shifts. Assuming a
matched pair of junctions for simplicity (lmu= Imz" = 1m), we may rewrite (15.16) as
(15.17)
The gauge invariant phase shifts may be obtained by integrating VII>around the elosed path
C shown in Fig. 15.1. Noting that <Dis a multivalued function that can change by 2nn upon
completing the path we have
where n is an integer. The phase differences across the upper and lower Josephson junctions are
given by Eq. (15.15) as
? n f bII>b- (D=L1(D+ . _ _ A .dt
a u c P o a
and
The second and fourth terms in Eq. (15.18) are phase differences in the superconducting leads
themselves and are found by using the supercurrent equation (15.1) and the expression for the
London penetration depth Eq. (2.6):
i c 2n i c ( 4n).2
)<Dc- <Db= VII>·dt =- A + _ _!j 'dtb ¢o h C
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 77/511
The Josephson effects 63
and
f
a 2n f a ( 4n ..[ 2 )<Pa-<Pd= V<D·dl=- A+ __ l.j ·dt.
d c P o d C
Substituting the above four equations in Eq. (15.18) gives
2n~ . 2n 4nill rl<Du-L1<Dt=2nn+- A'dt+--- .j·dt.
c P o c c P o C • C'
(IS.19a)
Theintegration of A is around a complete closed path C and is equal to the total flux c P inside the
area enclosed by the contour. The integration ofj follows a path C' which excludes the integration
over the insulators. If the superconducting leads are thicker than the London penetration depth.
the integration path can be taken deep inside the superconductors where the integral involvingthe supercurrent density is negligible. The phase difference is then simply related to the total flux
by
(IS.19b)
Using this equation to eliminate L1cI\,rom Eq. (15.17), the total current is
(15.20)
When the inductance L of the loop is taken into account, the total flux in Eq. (IS.19b)
consists of the externally applied flux c P C X L and the flux generated by the screening circulating
current Idr; i.e..
(15.21)
For the identical junction case,
(15.22)
In general, Eqs. (1 5.2 0), (1 5.2 1), and (15.22) must be solved self-consistently to describe the
behavior of the two-josephson-junction loop. For simplicity, we assume the loop inductance is
negligible and consider only the effect of the externally applied flux on the characteristics of the
loop.The maximum supercurrent density which can be carried by the loop is found by maximiz-
ing Eq. (15.20) with respect to L1<D{;.e.,
L1<Dt+ n ~ : X L =n + 1/2)n.
Hencethe maximum supercurrent density, I m a x , is given by
I(n c P C X L ) II m a x = 1 m cos To ' (15.23)
whichis periodic in the external flux. Since c P o ~ 2.07 X 10-7 Gcm ', it is clear that the device
pictured in Fig. 15.1 can be used to measure very small changes in magnetic field. It is sometimes
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 78/511
64 Part I Phenomenological theories
z
H
@ax
Figure 15.2 Schematic diagram of an SIS Josephson tunnel junction.
referred to as a two-junction SQUID where the latter is an acronym constructed from the words
superconducting quantum interference device.
15.3 The extended Josephson junction
We next discuss the behavior of a single planar Josephsonjunction in a magnetic field. We refer tothe junction cross section depicted in Fig. 1S.2. The magnetic field is directed into the page along
y . The middle of the junction is taken as the origin and the vector potential in the three regions is
_ { _ Hxe-(o-a,2)'AL
Az - - Hx
_ Hxe(o+1I'2)/'L
(z > a/2)
(a/2 > z > - a/2)
(z < - a/2)
(lS.24a)
(1S.24b)
(lS.24c)
where the exponential dependencies arise from the Meissner effect in the two superconductors.
The phase difference encountered in crossing the junction from z =-::I:) to z =+ C fJ at a given
horizontal coordinate x is
(1S.25)
Assuming a rectangular junction of dimensions L, and Lv, the total current, I , through the
junction is
f1.,12
1= jmLv . sin L1<D(x)dx-Lx/2
or
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 79/511
The Josephson effects 65
1.2 ,---,---,--r--,-,--.,.--,--,-----,,---,
1.0
0.80'
: = - E 0.6: : : E
0.4
0.2
Figure 15.3 J m vs r P characteristics for a short Josephson junction when the self-field
effect is negligible.
sin(rr¢l¢o). (0)
1 =1m sin < 1 >2 1 ,rr¢l¢o
where¢ = = H(a + 2),dLx , and 1m =jmLJ~y . The maximum value of the junction current, 1m""
occursfor sin < I > ~ O != ± I(depending on the sign of sin (rr¢l¢o)); thus
I'llax =1 I sin(rr¢l¢o) Im rr¢l¢o '
(15.26)
W e note we obtain a 'diffraction-like' (sin x / x ) pattern (see Fig. 15.3) involving the variable ¢ I ¢a
where¢ is the flux contained within the junction.
The effective junction area entering the definition of ¢ involves an effective thickness (made
u p of the insulator thickness plus a contribution of one London depth in each superconductor)
timesthe junction width, Lx '
15.4 Effect of an applied rf field
An interesting behavior results if an external rf voltage is applied to a de voltage-biased
Josephson junction, as first observed by Shapiro (1963). We take the applied voltage to have the
form
v =V o + V I cos tot. (15.27)
FromEq. (15.12) the resulting current in the junction will be
1=Imsin [;~ I ( V a + V I cos wt)dtj
= 1m sin [ < 1 > ( 0 ) + OlJt + ()sinwt], (15.28)
where ( ! J ( O ) is an arbitrary phase, WJ is given by Eq. (15.13) and b = = e* Vdhm is called the
modulation (or deviation) index. Using expressions which are well known in the theory of
frequencymodulatiorr' we may rewrite (15.28) as
2. SeeAbramowitz and Stegun (1970), Eqs, (9.1.42) and (9.1.43), p. 361.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 80/511
66 Part I Phenomenological theories
I(t) =i:{sin[<D(O) + wjtJ [ J 0(3) + 2 J I J 211(6)cos(2nwt) J
+ cos[(D(O) + wjtJ2 n~o .!2n+1
((»)sin[(2n + l)wtJ },(15.29)
where the Jn are Bessel functions of first kind of integer order. Using sin a cos b =
t[sin(a + b ) + sin(a - b)J we may rewrite (15.29) as
I(t) =t;(.! o((»)sin[w/ + <D(O)J
+ J l .!11(()){sin[(wj + nCU) l + <D(O)]+ ( - 1)"sin[(wj - nw)t + <D(O)J}).
(15.30)
I f we vary Wj (by sweeping Va) such that the condition OJ j = ± nco is met, then the time
dependence associated with this term in (15.30), having the amplitude I(()) and phase <D(O),would
be 'transformed to zero frequency' and would appear as a 'spike' in the current-voltage character-
istics.
The impedance of a Josephson junction is usually much smaller than the resistance of the
leads extending into the cryostat and hence a constant current source is a more accurate
representation than the constant voltage source assumed above. Further analysis shows (and
experiment confirms) that this results in a step-like (rather than a spike-like) current-voltage
behavior.
15.5 The resistively shunted junction
(RSJ) model
In analyzing the behavior of circuits involving small Josephson junctions, one may model the
effects of various dissipative processes and the distributed junction capacity with so-called
'lumped' circuit parameters. Fig. 15.4 shows such a model; here C and R represent the capacity
and effective resistance of the junction, 3 where the latter is represented by a cross. V j is the voltage
across the junction while I is the total current flowing through all three circuit elements which is
given by"
(15.31a)
hC .. I z .=Imsin<D + -<D + -<D,
2e 2eR(15.31b)
where we have used the Josephson equations
3. The capacitance is largely determined by the geometry of the junction and may he regarded as
constant. The resistance on the other hand may depend strongly on the junction voltage; for this
reason an additional shunt resistance, which is much smaller than the junction resistance, is
sometimes incorporated to bypass this effect.
4. We use MKS units in this subsection.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 81/511
The Josephson effects 67
I
+--- •
•
Figure 15.4 Equivalent circuit of a resistively shunted Josephson junction.
I~<P I
I
m ..
Figure 15.5 The simple pendulum with an applied torque.
(15.32)
and
I J =1m sin c D (15.33)
in obtaining (15.31b).
Let us examine the behavior of an RSJ to which a constant current is applied. If I is slowly
increased the voltage across the junction will remain zero (VJ = 0), implying <!>= sin -1 (1 1 1m ),
until I=m' Above this point a voltage develops and the junction becomes resistive with a
time-dependent phase. To examine the beha vior in this regime it is useful to note that the phase,
C l l ( t ) , ofajunction described by Eq. (15.31b) is in one-to-one correspondence with the angle of
rotation (which we will also call c I » ) of a damped pendulum, driven by a constant torque, in a
constant gravitational field (see Fig. 15.5): the applied torque, .r=mwyr , gravitationally induced
torque, mgl sin <!>,moment of inertia, .f=mr", and damping coefficient, k, are identified with I,
lmsin<!>,zeI2e, and hl2eR, respectively. The regime I < 1m corresponds to a situation in which
theapplied torque .r is less than a critical torque :Y m necessary to raise the pendulum to an angle
Cll=n/2 (where the opposing gravitational torque is maximal). for .'Y > :Ym the pendulum
rotates in a manner such that the average energy dissipated per rotation is equal to the average
work per rotation (were there no damping the average angular velocity c b would increase without
limit in this regime). We now go to a rotating reference frame by writing
< ! > ( t ) =cot + < ! > ' ( t ) , (15.34)
where
(15.35)
isthe average angular velocity and c D ' is periodic (with an average value of zero); Eq. (15.31b)then
becomes
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 82/511
68 Part I Phenomenological theories
he .. f z .-$+-$ +msin[wt +D'(t)J = .2e 2eR
(15.36)
In the limit where the departure of c I ) from a steady state rotation, oit, is small, i.e. <D'( t) < < 1, we
may neglect the $contribution to the last term in (15.36). We then have the equation for a driven
harmonic oscillator.
Writing $'(t) = < D ~ sin(wt + fi) l e a d s to the pair of equations
hC hI - - w2<D~ cos fi - - w$~ sin e =0m 2e 2eR
(15.37)
and
hC Ii
- - U )2$~ sin fi + - w$~ cos (J = .2e 2eR (15.38)
From the second condition we obtain
1tan f=-_
wRC(15.39a)
or
1
sin (J = 7 2 12(1+orr) /
where, = = RC. Substituting (15.39) into (15.37) yields
(15.39b)
, 2e ImR 1
< D o = Ii -:;;1 + (J)2,2)1/Z'(15.40)
Now
-s- 2e _ 2e< l > = - V =- IR.
h h
(15.41)
In the limit, (liT < < 1, $~=ImlI). In the opposite limit, OJ, >> 1, (15.40) becomes < D ~ =1/
(liT)/m/l. The requirement $;) < < I is therefore best satisfied in the limit of large I and large WI.
The limit OJ, ~ 0 may be obtained by eliminating the shunting capacitor in Fig. 15.4; the
differential equation (15.31b) then reduces to
. h.I =1 m Sill < D + - < D
2eR(15.42)
or
I z f < l > « ) d$
t= i R l m < 1 > 1 0 ) III", - sin < D '
which upon integration yields
[(
!X 2 - 1 ) + ( n t ) ](t) = 2tan-
1~ tan T -'X , (15.43)
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 83/511
The Josephson effects 69
a
~~
'/
4.0 B "'/
"/3.0
"/
/
/
/
/
2.0 "/
"/
"/
A "/
"/
1.0 /
/
/
/
1.0 2.0 3.0 4.0 5.0 T)
Figure 15.6 Time-averaged nonhysteretic I-V characteristic computed in the RS.J
model in the zero capacitance limit; a = 111m, '1 = VIRlm. Points A and B give the lime-
averaged voltages corresponding to o : = 1.2 and rJ . =4, respectively.
where we have defined
(15.44)
whereIX
is a parameter and Tis the period (which is the minimum time required for < D ( t ) to returnto its initial value <D(O)).The voltage across the junction from Eq. (15.5) is (h<D/2e). Ifwe assume the
voltage is sensed with a device which records the time average of V(t) (over many periods T) then
- Iz.. h 1 f T d<D h 2n 2 1/2V=-c))=--- -dt=--=Rlm(1X -1)·.
2e 2e T 0 dt 2e T(15.45)
Note that for I >> 1m ( I X >> 1) (15.45) approaches (15.41). Fig. 15.6 shows the J- V characteris-
tics f rom Eq. (15.45); note that the J- V characteristic is non hysteretic.
Wenow give a qualitative discussion of the general case when C * ' O.We begin by rewriting
Eq. (15.31b) in dimensionless form. We measure time in units of Wo and introduce a parameter f Jwhere
(
7 I ) 1 / 2__ C n . ' .
Wo =he ' (15.46)
• woisthe natural frequency for small oscillations (referred to as the Josephson plasma frequency).
Then Eq. (15.31 b) may be rewritten as
< i > + { 3 < D + sin C [) ='1, (15.4 7)
where the derivatives are now with respect to the dimensionless time variable. We define an
angular velocity O(t) = = < 1 > ; the angular acceleration is then < i> = n = (dO/dcD)ci>= (dO/d<D)O = (dj
dCll)(Q2/2)and Eq. (15.47) becomes
d ( 0 2
)- - + fin + sin <D=X .d<D 2
( 15.48)
The 'orbits' associated with the solutions of (15.48) for various initial conditions and parameter
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 84/511
70 Part I Phenomenological theories
t"J7
------~----~--~~--~$
Figure 15.7 The Q-1tJ plane trajectories (lower panel) associated with the solution of
Eq. (15.49). There are two kinds of orbits: an open orbit for (1/2)Q2(<ll =0) > 2 and a
closed orbit for (1/2)Q2(<ll =0) < 2. The upper panel shows the potential as a function of
<ll.
values (z and m are trajectories in the (HI> plane; we may think of this plane as the t w o -
dimensional 'phase space' associated with the motion. We now discuss the nature of these orbits.
We begin by discussing the orbits when : x = { J = O . Our equation is then
d ( 0 2
)d < 1 > 2+ sin c D = o . {15.49)
This equation may be integrated analytically in terms of Euler elliptic functions, as discussed in
mechanics textbooks. Wc will restrict ourselves to some qualitative statements. Since (15.49)is
invariant under the transformation c D ~ - < 1 > , we expect extrema at < 1 > = 0 and (when they exist]
± tt. For the rotating case, < t > advances continuously in time with 0 having a maximum at < 1 > =0
(where the kinetic energy of the pendulum is largest) and a minimum at c D =± tt (where thepotential energy is maximal); the phase space trajectory is shown in Fig. 15.7. Such 'open' orbits
req uire that the kinetic energy always be larger than the potential energy; i.e.,
(15.50)
When this condition is not satisfied the motion is oscillatory and the orbit in phase space is closed
as shown in Fig. 15.7.
We now turn on a small damping (Ii> 0), but still keeping the drive current (torque) zero
(« = 0); we will then have 0 ( < 1 » > 0 ( < 1 > + 2n) > 0 ( < 1 > + 4n), etc., i.e., the rate of rotation slows
down. Eventually a point is reached where 0passes through zero before the system (pendulum)
reaches the 'top' after which it reverses its motion. This point marks the transition between rotary
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 85/511
- I T it -it it -it it
A 8 c
Figure 15.8 The qualitative phase-plane trajectories, corresponding to the three
different situations (A, 8, C) in the fi e x plane in Fig. 15.10. (After Belykh, Pedersen, and
Soerensen (1977).)
andoscillatory motion. After the transition the phase point proceeds along a spiral, asymptoti-
callyapproaching the origin.
The above described sequence of events is shown as a phase space plot in Fig. 15.8. (Note
that sincedissipation is present the arrow of time is relevant.)
We now consider the general case;« > 0 , 1 3 >O.When a >> 1 (l>> 1m), Q is practically
constant and given by Q =a/f3. As a is lowered through unity from above the behavior of the
systemdepends on the magnitude of 1 3 . Since '- x is starting at a value greater than 1, the pendulum
motionmust evolve from a rotary state. I f we now decrease a to a value less than 1 (J < 1m) the
pendulumcan continue to rotate (i.e., we can maintain a steady state motion) provided the work
performedper cycle by the external torque, 2na, can compensate for the energy dissipated per
cycle;the latter is given by
f
T f 2 1 r dr{ 3 Q2(t)dt =3 Q2(<D)-d<D
o 0 d<D
f2 1 r
=3 0 Q(clJ)dclJ. (15.51 )
Thebehavior ofQ(<1»is not known without actually integrating the equation of the orbit (15.48).
However,we can estimate the critical torque, ac, by assuming Q(n) =0 and Q(O)is the angular
velocityresulting from 'free-fall' rotation under the influence of gra vity, which from Eq. (15.50) is
Q ( C l l=) = ; we linearly interpolate between 0 and tt by writing Q(<D)=Q(O)<D/n . We then have
f1 r <D
2n a c ~ 2jiQ (0) - d<Do n
or
Amore accurate expression can be obtained from the exact treatment of the free pendulumwhichyields
(15.52)
whichis valid for ji : : s 0.2 (Stewart 1968, 1974). Wc see that as fl ~ O,«, ~ 0; i.e., the junction is
hysteretic, switching from a constant-phase, zero-voltage state to a phase-precessing, finite-
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 86/511
72 Part I Phenomenological theories
/
r.t d~o_ _/(
[3 = 10
/
/
/
2.0
/
//
~ / 0.1
1.0 2.0 T)
Figure 15.9 Analog computer results for the time average of:x( = I / I m ) vs 1 1 ( =V/Rlm)
characteristics for different values of the parameter f J = l/woRC. (After Johnson (1968).)
1.0
.\
I
IIII
II
Bel---':"-1
II
IIIIII
I
0.5
0.5 1.0
Figure 15.10 Analog computer results for the hysteresis parameter !X c as a function of
{i. Tnthe region above the c x J f J ) curve only one stable solution exists. Tnthe region below
the curve two stable solutions exist. (After Johnson (1968).)
voltage state as a is increased above 1, but returning to the original (constant-phase) state ata
value a < 1. The J- V behavior of a hysteretic junction is shown in Fig. 15 .9 . The junctioninstability line in the a-Ii plane obtained by computer is shown in Fig. 15.10 (Johnson 1 9 6 8 ) .
For a discussion of the case in which the junction resistance is voltage-dependent s e e
Barone and Paterno (1982) .
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 87/511
The Josephson effects 73
Figure 15.11 A single junction SQUID coupled to a tank circuit via a mutual
inductance M.
15.6 The rf biased SQUID
Rather than measuring a magnetic flux with a two-junction interferometer (a de SQUID), as
discussed in Sec. 15.2, an alternative method involves the use of only a single junction (Silver and
Zimmerman 1967).
The device involves three circuits (see Fig. 15.11): (i) the primary SQUID, which is a
superconducting loop connected to a resistively shunted Josephson junction; (ii) a second loop(not shown in the figure), which is inductively coupled to the first loop and introduces an external
flux<Pex!=11;[ex!Iex! as a result of a mutual inductance, M ex!' coupling the two loops and a current,
lex!' applied to the second loop; and (iii) a third loop which is driven by an external oscillator
(typically operating at a frequency of order 107 Hz) which applies a periodically changing flux,
¢rf' through a mutual inductance, M.
To explain the operation of the device, which is somewhat subtle, we must examine the
action of CPex! and C P r f on the response of the primary SQUID loop. We begin by discussing the
equation of motion when CPex! = P r f = . The equivalent circuit consists of an RSJ connected to
an ind uctor, L(represented by the loop in Fig. 15.11). The equa tions of motion are"
(15.53a)
where
(15.53b)
Combining these equations with the forms in Eq. (15.31b) we have (where we multiply through by
L)
h<D hLC .. «i.:- + LIm sin < D + -- < D + - < D =O.2e 2e 2eR
(15.54)
The phase variable, (D , in Eq. (15.54) may be replaced by the internally generated flux, c P ,
according to the relation c P =cI)/2n)cpo' Incorporating the contribution of the external flux, CPex!,
Eq. (15.54) then becomes
5 . We use MKS units in this subsection.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 88/511
74 Part I Phenomenological theories
3(a)
2
3
3 4
2
(c) q ,
Figure 15.12 The static behavior of a single-junction superconducting loop in an
applied external flux ¢ex, : (a) c j ) m N > o =0.25; (b) ¢m/¢O = 1.25; (c) a 'linearized' version of
Fig. lS.12(b) showing the path traced out by ( p and ¢ext for the case ¢ =n c j ) o involving
three branches.
(IS.S5)
where ¢m = = LJ me w~=1/LC, and T =L/R. We examine the static solutions of this equation for
the responses ¢ =¢(¢ex,), which are shown in Fig. IS.12(a) and (b). Fig. IS.12(a) shows the
behavior when ¢m/¢O=0.2S, where a continuous, single-valued evolution of ¢ with ¢ext is
obtained; a line with a slope of 1 is shown for reference. Fig. IS.12(b) on the other hand shows the
behavior when ¢rn!c/JO= 1.2S. We now have a multi-valued behavior. Between successive points
4 ) c at which dcp/d¢ext diverges (indicated by the dashed vertical lines in Fig. IS.12(b)) there are
alternating regions with positive and negative slopes; the former are stable while the latter are
unstable and the system spontaneously oscillates. However, the frequency involved is typically in
the microwave regime and if T is sufficiently short the oscillations damp quickly (relative to the
period associated with ¢rr)' The system then, effectively, switches from one stable (positive slope)
region to an adjacent one. (By Taylor expanding in the region c/J/¢o ~ n (an integer) one sees that
the slope of the plateau at this point is (2n¢m/¢O + 1)-1.) The upward and downward directed i
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 89/511
, ne Josepnson eTTeCrS /'0
arrows in Fig. IS.12(c) indicate how the system switches for increasing and decreasing ¢ex ! '
respectively.
To use these characteristics to make measurements we need a means of locating a stable
branchand the position alongit. Todo this we introduce the third loop, which is driven sinusoidally
byan rfgenerator, and induces an additionalflux ¢rf( t ) = ¢ rf sin wrft into the primary loop. We will
assumethat a capacitor is associated with this additional loop and that the pair resonate at the rf
frequency.The quality factor, Q, of this resonant circuit along with the resonant frequency fixes its
bandwidth, ~OJ =wrrlQ, which in turn fixes the response time Trf ~ (~w) -1. Suppose that at t =0
therfgenerator is connected to the resonant circuit. The amplitude, ¢rf, of the oscillatory flux will
then increase with a characteristic time T rf . When the total externally induced flux, ¢ e x, + ¢ rf( t),
exceedsone of the thresholds, ¢e' (associated with the primary SQUID flux, ¢, depicted in Fig.
15.l2(b)),the SQUID loop becomes resistive, since it is then in a finite voltage state (oscillating at
somecharacteristic frequency much higher than Wrf' which we ignore). This results in a rapiddissipationleading to a partial quenching of the amplitude, ¢ r6 however, the process then repeats
itself.The peak amplitude of ¢rf, denoted rPrf (which can be measured with the appropriate
electroniccircuitry), is then a measure of ¢ex! as we will now discuss.
Referring to Fig. IS.l2(b), we identify two 'special' points: (i) the midpoint of a stable
branch,where ¢ e x , =n¢o (with n an integer), and (ii] the midpoint of an unstable branch, where
fex!=n + ! )¢o ' We first examine case (i).Assuming sufficient rf drive, as the oscillations build up
intime, the thresholds, ¢u, connecting the 11, n + 1and 11,11 - 1 branches are encountered and the
SQUIDswitches symmetrically between the three branches.Ifthe rf drive is further increased the
thresholds connecting the 11 + 1,11 + 2 and the 11 - 1 ,11 - 2 branches also are encountered and
theSQUID then oscillates symmetrically between five branches and so on. An example of a path
traversed by the system for a circuit involving three stable regions is shown in Fig. IS.12(c).
For case (ii), where ¢ex, =n + ~)¢o, the path traversed by the SQUID is nominally
centeredat the midpoint of an unstable region. As the rf drive is increased from zero, a level is
reachedwhere a complete circuit involves traversing the portion of the two stable branches lying
directlyabove, n + 1, and below, 11 , the unstable 11 + ~section. The threshold amplitude corre-
sponds to the extent of the unstable branch. Since this extent is smaller than that of the stable
region,the threshold for first completing circuits in which the SQUID switches irreversibly is
lowerfor case (ii) than for case (i).
Fig. IS.13 shows the peak amplitude of the rf tank circuit, VT, as a function of the rf drive
current, Irf• The slope is initially high but drops abruptly (nominally to zero for a peak reading
detector) when the thresholds for SQUID transitions are encountered, involving 3, S , ... stable
branches for case (i) and 2,4,6, ... for case (ii). From the symmetry it follows that cases (i) and (ii)
representthe maximum and minimum values of VT (or rPrf) for a given drive level, Irf'
Itis clear from the above discussion that if ¢ex, is continuously varied (at fixed Irf), the peak
amplitude (fjrf will move back and forth (in a triangular fashion) between maxima, at «;=l1¢o,andminima, at ¢ e x t =n +})¢o ' By observing the change in the number of flux quanta, one can
accurately measure a change in a current, Iext, associated with the loop generating ¢ o x t .
Fig. IS.14(a) shows the characteristic 'staircase' dependence on Trf for an rf SQUID
operated at 27 MHz that was shown schematically in Fig. IS.13. The two curves correspond to
thelimiting cases for the de external flux. Fig. IS.14(b) shows the triangular dependence of the
detectedrf voltage Vr vs Iex tcharacteristics. The different curves refer to different values of the rf
drivecurrent, T rf'
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 90/511
76 Part I Phenomenological theories
< il =(n+ 1/2)$0
t;--/
/
Figure 15.13 Relation between the tank circuit voltage Vr and the rf current amplitude
I"f for the cases of integral and half integral numbers of nux quanta. For intermediatevalues of flux, the voltage steps occur at intermediate values of VT•
f - <>. , -Cl
s'0>
- g. . . .o
2l.,o
(a) rf drive, Irf (b) de field, c/Jdc
Figure 15.14 Experimental responses of an rf SQUID in the dissipative mode: (a)
detected tank circuit voltage Vr vs rf current amplitude J rf (the two curves are the two
limiting cases for external dc flux ( I Jde = 11(/>0 and (/>dc = (n + })¢o); (b) VT vs appliedmagnetic flux curves at different values of I,.f. (After Zimmerman (1972).)
If the loop gencratingcjJext is coupled to another loop which, in turn, is coupled to a
magnetic sample in a fixed field (via an all superconducting circuit), changes in the magnetization
with temperature or some other parameter may be observed. (Note that current changes induced
through magnetization changes arc persistent.) The self-inductance of the two loops, together
with the mutual inductances coupling them to the SQUID and the sample, make up a 'flux
transformer' the characteristics of which can be optimized for maximal sensitivity.
For further discussion of the rf SQUID see Zimmerman (1972) or Barone and Paterno
(1982), where many additional references may be found.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 91/511
. f-----~-'h-e-J-o-S-e-p-h-s-o-n-,a-t-t-ic-e--
in 1D
Certain intermetallic compounds have a layered structure in which the in-plane conductivity
greatlyexceeds the conductivity normal to the layers. This applies for most of the high tempera-ture superconductors. Such structures may also be made artificially by depositing alternating
layersof a superconductor and a low conductivity material,
The simplest model is to assume infinitesimally thin superconducting layers which arc
coupledvia order parameter tunnelinq (Josephson coupling) through insulating layers of thickness
s.Following Lawrence and Doniach (1971), we introduce a modified free-energy functional of the
form
(16.1)
where
(16.2)
herewe have defined ,4 = = (lIs) J~~1),Azdz. The structure of this equation is similar to Eq. (9.13)
withrespect to the in-plane components while the intcrplane coupling is seen (on expanding theexponent) to be a generalization of the operator
a ie*---A_D z he ~
to the case of finite differences. Our order parameter J j J "(r)has a discrete dependence on the index
n and a continuous dependence on r =x x + y y : the total vector potential A I + Azz is, however,
definedat all points. Variation with respect to J j J * yields
1 1 2 ( ie* ) 2-- V--A ! If2m* he - "
(16.3)
77
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 92/511
78 Part I Phenomenological theories
Variation with respect to A _Land Az yields Eq , (9 .16 ) with
(16 .4)
and
. i 1 1 . , , + 1 . n =
Eq. (16.5) for the tunneling current flowing between planes nand n + 1 is equivalent to Eq. (15.4)
(written in a gauge-invariant form).
Solutions of E q. (16 .3) were first examined in detail by Klemm, Luther, and Beasley (1975) .
As before, we confine our interest to the evaluation of H c2, in which case we may neglect the last
(nonlinear) term. The external magnetic field will be assumed to lie in the x z plane. In order to
take advantage of the layer periodicity, we choose a gauge for which the vector potential has no z
dependence:
(16.6)
We seek a solution of the form
noting that I f z = - Hx sin 0, we find that
{
1 i2 d2 1- -- __ + -m*w2(x - x )2 cos? e2m* dx? 2 c 0
1 i
2
[ ( e* H s )]}-;;2cos kz s + --' x sin 8 - 1 + e x u,,(x) =0,mz s lie
(16.7)
where Xo = l1 eky /e * H cos e is an orbit center.For the field perpendicular to the layers, we again have a harmonic oscillator problem and
we recover our earlier result - c : = ~ 1 1 ( J ) e ( O ) . For H parallel to the planes Eq. (16.7) becomes
{1 1 2 d 2 / 1
2
[( e* H S ) ] }- -- __ - -- cos kz s + -, ,-x-I + c : un = O.2m* dx ' m:s2 he
(16 .8 )
Changing k z has the effect of shifting the origin on the x axis. We shall initially fix the origin such
that the argument of the cosine vanishes when x =O.Ncar the zero-field transition temperature,
where e x is small, we expect the critical magnetic field, H e 2 1 1 ' also to be small and further that thelowest eigenfunction u,which by Floquet's theorem is periodic (since the potential is periodic), is
concentrated ncar the minimum of the 'potential '
li2
[ (e*H S )](x ) =--2 1 - cos -, _. x .m:s he
(16.9)
Expanding Eq. (16 .9 ) to second order, we obtain a harmonic oscillator potential which, o n
substitution into Eq. (16.8), yields
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 93/511
The Josephson lattice in 1D 79
_! i__ u l! + [~m*w2 (~)X 2 + r x J u = 02m* 2 c 2 '
(16.10)
where
Thelowest eigenvalue is - o : = ±hm e(n/2), yielding an upper critical field H e 2 1 1 = cjJo!2n((z where
~ = = (h 2 /2m* I o : I )112 and (z = = (h 2!2m;I , :x I )112. For temperatures further from T e, where the eigen-
value- c : and the associated parallel critical field are both large, u becomes less concentrated at
thepotential minimum and our expansion of Vex) breaks down. To examine this limit, we write
Eq . (16.8)in its 'canonical' form (Abramowitz and Stegun 1970) by introducing new variables
e*H s 2cjJ~ m*2v = = - . - x + tt., 2q = = --0 - -
hc n2H-~m:
and
whichyields
d2u-2 + (a ' - 2qcos2v)u = .dv
(16.11a)
For large H, both at and q are small and arc related by the expansion (Abramowitz and Stegun
1970,p. 722)
(16.11b)
Inserting our expressions for a' and q into Eq. (16.11b) we obtain
(16.12)
Sincethis is a high field expansion, Eq, (16.l2) is valid only for temperatures such that (z ~ s/)2;
however, H c2 '! is real only for (z > s/j2. No (real) solution exists for (z < s/j2 and hence thecritical field becomes formally infinite (in this model) at a temperature T* such that (z(T*) = s /
)2. In a real system, this divergence would be removed by the effect of paramagnetic limiting (see
Part III, Eq. (41.45)).
Deutcher and Entin-Wohlman (1978) generalized the above model to the case of a
superlattice of thin slabs of thickness d separated by Josephson-coupled insulating layers of
thickness s. We shall continue to take the z axis normal to the layers, and we place z = 0 at the
interface with the lower surface of a metallic layer. The free energy is given by
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 94/511
80 Part I Phenomenological theories
f { f IlD t d [ h 2 1 ( ie*) 1 2F = I dxdy dz ~ V----:-A !11,,(X,y,Z)11 11f) _In I I (
, ( J
+ C t . 1 0 / l x , y, z ) 1 - + 2 1 o /,,(X , y , Z ) 1 4
+ ___ 2 _ , 1 o/l l+l(x,y,(n + I)D)exp( - ie*. All_ I S )2m;s" he
(16.13)
here , 4 , , + I = = (I/s) J ~ n D \ I . l D AJx, y, z )dz and D = d + s. We seek a solution to E q, (1 6.1 3) in the limit
d < < I C L and d < < ¢. The first of these assures us that we may neglect the diamagnetic screeningcurrents and hence we shall take the magnetic field as uniform and given by the external field, H.
We consider only the case where H is parallel to y and work in the gauge A = - Hxz. The limit
d < < ¢ implies that the magnitude of the order parameter is constant along z [or a given x ;
however, we do allow the phase to vary. We thus choose a solution of the form
The z component of the gradient energy in Ell. (16.13) is
1(
C < D e* ) 1 2--+-Hx 0 /jjz he
To avoid an arbitrarily large growth in this term for large x, we choose the phase in the form
e*( D = ---:-l1x(z - Zll);
he
21 1 is chosen to minimize the overall free energy. For an isolated slab it has the value Z/1 =D +
d/2 , i.c., the origin must be taken to be in the center of the nth slab. For the ground state, we assert
that, by symmetry, Zll will have the same value in the coupled superlattice. We include no
variation in the phase along y since this results in an increase in the free energy. Varying Eq .
(16.13) with respect to ! I I * yields the linearized equation
172
[ (e*llDx) l }--- cos ~-~ - 1 u + C t . 1 I dz = O.m;s2 he
(16.14)
Integration over z (noting that the second term in the first parentheses vanishes b y
symmetry) yields
(16.15)
where we h.ave introduced the notation we =e*H/m*c. Using Eqs. (12.3) and (12.4) we m a y I'
rewrite Eq. (16.15) as
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 95/511
The Josephson lattice in 1D 81
1 72
d2
11 112 [ (e*HD ) J ( H2)-- --0 - -- cos ----- x-I u + a 1- - u = 0,2m* d.x" m;s2 he H~
(16.16)
where H II is the parallel critical field of the isolated slab. The equation is identical with Eq. (16.X)
except for the additional term multiplying a.
To obtain the low-field behavior, we again expand the cosine term; this time we shall
include the H 4 term, the effect of which we calculate by perturbation theory. We require the
matrix element of X4 with respect to the unperturbed Gaussian ground-state wavefunction [/0
which is given by (Landau and Lifshitz 1977, p. 136)
(16.17)
Our eigenvalue equation is then
1 . D (m"'.') 112 m*(!)~ d2 ( 3 D 2)- ':i =-hwe- - + 1 - - - - - - - 0 + ...
2 s m~ 24 4 d:(16.18)
or
( 16.19)
For H -> 0 (T---7
Tel , we have the limiting behavior
H _ ¢o .e211 - 2nr;~z (D!s) '
(16.20)
thiswe refer to as 3D behavior with He211 o: T; - T.
To obtain the high-field behavior, we must write Eq. (16.15) in the canonical form of Eq.
(16.11).The constants now become
and
4¢2 ( H2 1'2 m*)
a' = (2n)2H 2oD 2r;2 1 - 1 1 1 i - 2 : 2 m; .
Using Eq. (16.12), we have
(16.21)
(16.22)
From Eq. (16.22), as H ---7 cc; we obtain He211 = HII; i.e., we obtain the upper critical field of an
isolated slab, where IIelll o: tT; - T)112, which we refer to as 2D behavior. Thus as the tempera-
ture is lowered a 3D to 2D 'crossover' occurs. Note that the divergence encountered with the
Lawrence-Doniach superlattice model involving infinitesimally thin superconducting layers is
avoided in the Deutcher-Entin-Wohlman model where d # 0.
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 96/511
Vortex structure in
layered superconductors
Dichalcogenides ' of transition metals such as NbSe2' superconductor/insulator supcrlattices,
and high T; oxide superconductors all have a layered structure. Most of these systems arc Type II
superconductors with relatively large I( values. One of the fascinating characteristics of the
layered superconductors is their strongly anisotropic magnetic properties. Usually, the coherence
length perpendicular to the layer plane ((J is much smaller than that parallel to the layer (SI) '
The anisotropy can be characterized by an anisotropy ratio, y, defined as )' = ~II/¢ L'Other length
scales of interest are the penetration length, A, and the scale of intrinsic inhomogeneities.
Depending on the relative size of ~j_ and the layer spacing, s, we may identify three different
regimes for the vortex structure in the layered, high-x superconductors: (1) If ¢j_(T) >> s, then
the layered structure is largely irrelevant and the superconductor may be regarded as three-
dimensional: anisotropic, but uniform. Since the coherence length diverges as T, is approached,
this regime will always occur sufficiently close to T" The vortex structure can be described using
the London or G-L theories by introducing an anisotropic mass tensor as discussed in Sec. 11.( 2 )
With decreasing temperature, we may have a regime where ¢J(T) < < s (especially in high T,
oxides). If both regimes can be entered by sweeping the temperature a 3D-2D crossover will
occur at some temperature. Below this temperature we have a quasi-2D regime; the layered
structure is then relevant and the Lawrence-Doniach-Iike model discussed in Sec. 16 is adequate
to describe the vortex structure. (3) For the case of extreme anisotropy, as in Bi- and Tl-based
oxide superconductors, with }' < : 10, the interlayer Josephson coupling is very weak. The flux
'lines' are then better viewed as stacks of 2D vortices residing in the superconducting layers
(so-called pancake vortices).
In this section we will give a brief account of the vortex structure in each of these three
regimes. We will focus on the weak field region and derive expressions for Hel, which contain
information on the anisotropy in addition to that contained in H c2'
17.1 3D anisotropic London model
As discussed in Sec. 11, the upper critical field, Hd in layered Type IIsuperconductors can be
described by the G-L equations with a phenomenological anisotropic mass tensor. Due to thelinearity of the equations near Helo a relatively simple solution can be found in this region (see
1. The chalcogenides are the group VI elements S, Te, and Se.
82
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 97/511
Vortex structure in layered superconductors 83
Sec.11). However, close to the lower critical field, H c1, it is difficult to solve the nonlinear G-L
equations to obtain H c l in the anisotropic case. Furthermore, the G-L expansion itself is
problematic in this regime since we are far from T'; For this reason, we will use the less accurate
London model, which provides a reasonable description, at least for a large G-L parameter K
(see,for example, Kogan (1981)). For K ( = = AR ) > > 1, the amplitude of the order parameter, 1 V I I,
variesconsiderably only for distances smaller than ~; one may then assume 1 1 / 1 1 to be constant
everywhereexcept in a narrow core of radius ~.
As in Sec. 11, we introduce a phenomenological effective mass tensor. In the reference frame
alignedwith the principal axes of a crystal with orthorhombic symmetry or higher, this mass
tensor is diagonal with elements which we now write as M j • For convenience we define a
geometricmean mass M=M 1M2M 3)1/3 and normalize the mass tensor such that mj =M J M.
Interms ofMwe define a mean penetration depth, A , and mean coherence Iength.z, by using Eqs.
( 2 . 6 ) and (9.22), respectively. The penetration depths, )'j =Aml/2 , describe the decay of the
components of the screening supercurrent along the principal directions, i. The coherence
lengths,~j =~/mj1/2, characterize the spatial variation of the order parameter along these direc-
tions.For a uniaxial crystal, m3 = 111j_,1 = m2 = m il (in high T, oxides, the anisotropy within the
Cu-O layers is very weak). The anisotropy ratio is defined as :! ~ ~ II/~j_= ; ' 1 / A ll = (M ljMII)1!2.
The free energy, given in Eq. (2.8) for the isotropic case, may be generalized to the
anisotropic case as
(17.1)
whereB is the local field. Straightforward minimization of the energy (17.1) yields the anisotropic
London equation
B + A 2V x [m- (V x B)] =O . (17.2)
Takinginto account the boundary condition, the local magnetic field around a single flux line
directedalong z and carrying a flux quantum C P o is determined by
B+ A 2V x (m V x B)= C P o z b ( x ) b ( y ) . (17.3)
In the isotropic case where mij = bij, Eq. (17.3) coincides with the usual London equation (2.14).
Todealwith an arbitrarily directed flux line, we define two reference frames as shown in Fig. 17.1.
Thevortex frame (x , y , z) is obtained from the crystal frame (x o, Y o, 1 .0) by rotating by an angle e
fromZoabout Y o axis. Restricting ourselves to the case of uniaxial symmetry, the components mij'
transformedto the vortex frame, are
(
ml COS2 e + m3 sin? () 0
m(8) = 0 1111
(m1 - 1113)sincos e 0
(17.4)
The algebra is greatly simplified by working in the vortex frame, since both Band j are
independentof z. Substituting (17.4) into Eq. (17.3) results in a set of equations for B x, B y , and B ;
Theseequations lead to a more complicated vortex structure in the anisotropic superconductor
thanthe isotropic Abrikosov vortex. For example, transverse fields, B x•y , are present even when
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 98/511
84 Part I Phenomenological theories
Figure 17.1 In the crystal lrame (xo , Y o , 7 .0) the plane xoYo coincides with the basal
crystal plane. The vortex frame (x, y , z) is obtained lrorn the crystal frame by rotating by
an angle 0 from 1 . 0 about the Y o axis. The axis 1 . is parallel to the vortex axis.
the applied field is along the z direction. For simplicity, we consider two special cases in which the
magnetic field is applied perpendicular to or parallel to the layer plane.
F or the perpendicular field case, ()=O. The mass tensor is diagonal;
mxx = myy = m1,H1zz = H13• Eq. (17.3) then becomes
2 ( a 2
B z a 2
B z ) .B , -}. ml- ..~2 + ml- ..~2 =<Po()(x)6(y).
A X o y(17.5)
This equation is equivalent to Eq. (7.S) and the lower critical field is given by (S.4b) as
< P o . ( A )clj_ ::::::-, tn -;:-.
4nJ~ C ;
When the applied magnetic field is parallel to the layer plane, e =n12 . The mass tensor is
(l7.6)
also diagonal: mxx = 1113, myy = mz z = mi' Eq. (17.3) now reads
(17.7a)
LC.;
(17.7b)
By interchanging Zo <-> x , this equation can be transformed back to the crystal frame with B along
the X o axis:
( 1 7 . 8 )
Eq. (17.S) can be solved by a Fourier transform method. Alternatively, we may map Eq, (17.8) to
the isotropic case by defining
(17.9)
From Sec. 7 we know that the solution of Eq. (17.S) at large distances is given by
(17.10)
where Kn(f j ) is a modified Bessel function of the second kind of order n . The free energy per unit
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 99/511
Vortex structure inlayered superconductors 85
length of the flux line, Ep can be calculated from Eq. (17.1). To logarithmic accuracy it is given by
(17.11)
Note for this case the normal core radius ~ j _ serves as the cut-off scale. From the usual
thermodynamic relation E1 =CPoHcl/4n, the parallel lower critical field is
(17.12)
Comparing Eqs. (17.6) and (17.12), we see that important information about the anisotropy
can be obtained by measuring H c 1 in both parallel and perpendicular fields.
17.2 Lawrence-Doniach model
InSec. 16 we discussed the upper critical fields for layered superconductors which can be modeled
asJosephson lattices in 10.We saw that whenthe temperature is lowered to a value T*such that
(,{'P) = s/J2, a 30-20 crossover occurs. Above T*, the layered superconductor behaves as an
anisotropic 30 superconductor. At temperatures lower than 1'*, the barriers of the layered
structure dominate, resulting in a 20 behavior of the temperature dependence of He211 (T). In this
subsection we will show that below 1'* the vortex structure in layered superconductors is also
drastically modified.
Wc will start with Eqs. (16.4) and (16.5) which were derived from the Lawrence-Doniach
freeenergy functional (16.2). For high tc supercond uctors in a weak magnetic field, the influence of
thefield on the value of 1 1 / 1 n I can be neglected, and we can regard 1 1 / 1 " I as a cons tan t; i.e., we use the
London approximation. We confine ourselves to the parallel field case. We take the field along
t h e x direction and choose a gauge such that A z = O . Writing 1 / 1 " = I V I " I ei<D",rom Eqs. (16.4) and
(16.5)we obtain
4nA~. CP oA j _ = - --h + - V < P n
c 2n(17.13)
and
(17.14)
where)'L is the bulk London penetration depth of the superconducting layers, < P n is the phase in
thenth superconducting layer and i;~e*h 1 1 / 1 1 2 /m;s is the maximum Josephson supercurrent
density.Note that in this model it is assumed, for simplicity, that each superconducting layer is
isotropic with an intrinsic bulk penetration depth AL• We use the rectangular contour C shown in
Fig.1 7 . 2 to compute the phase difference across one unit cell: < P = = < D " + 1 - ( D n . From E q . (17.13)
wehave
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 100/511
86 Part I Phenomenological theories
z
Figure 17.2 Schematic of a Josephson-coupled superconductor insulator superlattice.
The insulating layers of thickness d , alternate with superconducting layers
(crosshatched) of thickness d.. The modulation wavelength s=d, + d.. The rectangular
contour C is used to compute the magnetic flux in Eq. (17.15).
(17.I~
fy + "y f (n + 1), ~
dy' dz'B(y',z ')= "Aj_·dly ns (
C P o A
+ 2n [<D (y ) - ( D ( y + tly)].
The left-hand side in the above equation can be approximated as sLiyB(y,z) , and we obtain
41TA£ C P osLiyB (y , z ) ~ -- (j y n+ 1- i.n)Liy + - [<D (y ) - ( D ( y + Liy)] .
c' jO 2n
Hence
where we have used the following approximations: o jjoz ~ Uy ,n+ 1- .iy,n)!s a n d
c<D jD y ~ [<D (y + L 1 y ) - cD(y)] /L iy, For small < D the Josephson current relation (17.14) may b e
approximated as j, ~j m c D and substituting this form into Eq. (17.15) we obtain
4nA£ ojv 4n).; ojz- ~ - - ~ = B(y,z) ,
C o z C oy(17.1~
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 101/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 102/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 103/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 104/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 105/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 106/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 107/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 108/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 109/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 110/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 111/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 112/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 113/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 114/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 115/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 116/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 117/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 118/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 119/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 120/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 121/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 122/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 123/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 124/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 125/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 126/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 127/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 128/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 129/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 130/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 131/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 132/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 133/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 134/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 135/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 136/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 137/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 138/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 139/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 140/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 141/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 142/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 143/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 144/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 145/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 146/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 147/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 148/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 149/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 150/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 151/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 152/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 153/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 154/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 155/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 156/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 157/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 158/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 159/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 160/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 161/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 162/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 163/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 164/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 165/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 166/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 167/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 168/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 169/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 170/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 171/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 172/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 173/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 174/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 175/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 176/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 177/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 178/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 179/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 180/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 181/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 182/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 183/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 184/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 185/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 186/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 187/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 188/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 189/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 190/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 191/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 192/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 193/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 194/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 195/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 196/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 197/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 198/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 199/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 200/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 201/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 202/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 203/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 204/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 205/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 206/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 207/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 208/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 209/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 210/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 211/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 212/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 213/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 214/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 215/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 216/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 217/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 218/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 219/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 220/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 221/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 222/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 223/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 224/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 225/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 226/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 227/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 228/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 229/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 230/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 231/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 232/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 233/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 234/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 235/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 236/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 237/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 238/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 239/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 240/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 241/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 242/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 243/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 244/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 245/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 246/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 247/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 248/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 249/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 250/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 251/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 252/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 253/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 254/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 255/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 256/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 257/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 258/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 259/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 260/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 261/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 262/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 263/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 264/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 265/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 266/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 267/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 268/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 269/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 270/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 271/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 272/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 273/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 274/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 275/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 276/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 277/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 278/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 279/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 280/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 281/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 282/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 283/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 284/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 285/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 286/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 287/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 288/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 289/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 290/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 291/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 292/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 293/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 294/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 295/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 296/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 297/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 298/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 299/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 300/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 301/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 302/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 303/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 304/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 305/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 306/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 307/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 308/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 309/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 310/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 311/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 312/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 313/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 314/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 315/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 316/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 317/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 318/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 319/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 320/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 321/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 322/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 323/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 324/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 325/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 326/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 327/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 328/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 329/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 330/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 331/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 332/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 333/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 334/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 335/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 336/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 337/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 338/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 339/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 340/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 341/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 342/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 343/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 344/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 345/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 346/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 347/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 348/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 349/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 350/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 351/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 352/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 353/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 354/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 355/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 356/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 357/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 358/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 359/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 360/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 361/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 362/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 363/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 364/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 365/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 366/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 367/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 368/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 369/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 370/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 371/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 372/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 373/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 374/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 375/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 376/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 377/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 378/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 379/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 380/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 381/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 382/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 383/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 384/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 385/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 386/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 387/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 388/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 389/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 390/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 391/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 392/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 393/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 394/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 395/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 396/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 397/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 398/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 399/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 400/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 401/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 402/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 403/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 404/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 405/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 406/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 407/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 408/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 409/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 410/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 411/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 412/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 413/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 414/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 415/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 416/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 417/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 418/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 419/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 420/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 421/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 422/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 423/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 424/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 425/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 426/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 427/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 428/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 429/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 430/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 431/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 432/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 433/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 434/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 435/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 436/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 437/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 438/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 439/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 440/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 441/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 442/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 443/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 444/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 445/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 446/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 447/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 448/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 449/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 450/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 451/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 452/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 453/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 454/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 455/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 456/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 457/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 458/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 459/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 460/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 461/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 462/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 463/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 464/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 465/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 466/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 467/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 468/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 469/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 470/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 471/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 472/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 473/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 474/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 475/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 476/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 477/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 478/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 479/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 480/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 481/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 482/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 483/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 484/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 485/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 486/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 487/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 488/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 489/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 490/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 491/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 492/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 493/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 494/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 495/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 496/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 497/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 498/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 499/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 500/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 501/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 502/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 503/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 504/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 505/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 506/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 507/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 508/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 509/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 510/511
8/7/2019 superconductivity by kitterson & song
http://slidepdf.com/reader/full/superconductivity-by-kitterson-song 511/511