Chapter 10: Superconductivity

Bardeen, Cooper, & Schrieer

May 9, 2001
Contents
1 Introduction 2
1.1 Evidence of a Phase Transition . . . . . . . . . . . . . . . . . . . . . 2
1.2 Meissner Eect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 The London Equations 7
3 Cooper Pairing 10
3.1 The Retarded Pairing Potential . . . . . . . . . . . . . . . . . . . . . 11
3.2 Scattering of Cooper Pairs . . . . . . . . . . . . . . . . . . . . . . . . 12
3.3 The Cooper Instability of the Fermi Sea . . . . . . . . . . . . . . . . 14
4 The BCS Ground State 17
4.1 The Energy of the BCS Ground State . . . . . . . . . . . . . . . . . . 18
4.2 The BCS Gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5 Consequences of BCS and Experiment 28
5.1 Specic Heat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5.2 Microwave Absorption and Reection . . . . . . . . . . . . . . . . . . 28
5.3 The Isotope Eect . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1
6 BCS Superconducting Phenomenology 32
7 Coherence of the Superconductor Meisner eects 37
8 Quantization of Magnetic Flux 41
9 Tunnel Junctions 43
2
1 Introduction
From what we have learned about transport, we know that there
is no such thing as an ideal ( = 0) conventional conductor.
All materials have defects and phonons (and to a lessor degree
of importance, electron-electron interactions). As a result, from
our basic understanding of metallic conduction must be nite,
even at T = 0. Nevertheless many superconductors, for which
= 0, exist. The rst one Hg was discovered by Onnes in
1911. It becomes superconducting for T < 4.2

K. Clearly this
superconducting state must be fundamentally dierent than the
normal metallic state. Ie., the superconducting state must
be a dierent phase, separated by a phase transition, from the
normal state.
1.1 Evidence of a Phase Transition
Evidence of the phase transition can be seen in the specic heat
(See Fig. 1). The jump in the superconducting specic heat C
s
indicates that there is a phase transition without a latent heat
3
C (J/molK)
T
T
c
C
S
C T
n
Figure 1: The specic heat of a superconductor C
S
and and normal metal C
n
. Below
the transition, the superconductor specic heat shows activated behavior, as if there is
a minimum energy for thermal excitations.
(i.e. the transition is continuous or second order). Furthermore,
the activated nature of C for T < T
c
C
s
e

(1)
gives us a clue to the nature of the superconducting state. It is
as if excitations require a minimum energy .
1.2 Meissner Eect
There is another, much more fundamental characteristic which
distinguishes the superconductor from a normal, but ideal, con-
4
ductor. The superconductor expels magnetic ux, ie., B = 0
within the bulk of a superconductor. This is fundamentally dif-
ferent than an ideal conductor, for which

B = 0 since for any
closed path
C
S
Superconductor
Figure 2: A closed path and the surface it contains within a superconductor.
0 = IR = V =
_
c dl =
_
S
c dS =
1
c
_
S
B
t
dS, (2)
or, since S and C are arbitrary
0 =
1
c

B S

B = 0 (3)
Thus, for an ideal conductor, it matters if it is eld cooled or
zero eld cooled. Where as for a superconductor, regardless
of the external eld and its history, if T < T
c
, then B = 0
inside the bulk. This eect, which uniquely distinguishes an
5
Zero-Field Cooled Field Cooled
B = 0
B = 0
B = 0
B 0
B 0
B 0
T < T
c
T < T
c
T < T
c
T < T
c
T > T
c
T > T
c
Ideal Conductor
Figure 3: For an ideal conductor, ux penetration in the ground state depends on
whether the sample was cooled in a eld through the transition.
ideal conductor from a superconductor, is called the Meissner
eect.
For this reason a superconductor is an ideal diamagnet. I.e.
B = H = 0 = 0 M = H =
1
4
H (4)
6

SC
=
1
4
(5)
Ie., the measured , Fig. 4, in a superconducting metal is very
large and negative (diamagnetic). This can also be interpreted

0
T

T
c
D(E )
Pauli
F
4
-1
H
M
j
s
Figure 4: LEFT: A sketch of the magnetic susceptibility versus temperature of a su-
perconductor. RIGHT: Surface currents on a superconductor are induced to expel the
external ux. The diamagnetic response of a superconductor is orders of magnitude
larger than the Pauli paramagnetic response of the normal metal at T > T
C
as the presence of persistent surface currents which maintain a
magnetization of
M =
1
4
H
ext
(6)
in the interior of the superconductor in a direction opposite
to the applied eld. The energy associated with this currents
7
increases with H
ext
. At some point it is then more favorable
(ie., a lower free energy is obtained) if the system returns to a
normal metallic state and these screening currents abate. Thus
there exists an upper critical eld H
c
Normal
S.C.
T
c
H
c
T
H
Figure 5: Superconductivity is destroyed by either raising the temperature or by ap-
plying a magnetic eld.
2 The London Equations
London and London derived a phenomenological theory of su-
perconductivity which correctly describes the Meissner eect.
They assumed that the electrons move in a frictionless state, so
that
8
m v = ec (7)
or, since
j
t
= en
s
v,
j
s
t
=
e
2
n
s
m
c (First London Eqn.) (8)
Then, using the Maxwell equation
c =
1
c
B
t

m
n
s
e
2

j
s
t
+
1
c
B
t
= 0 (9)
or

t
_
_
m
n
s
e
2
j
s
+
1
c
B
_
_
= 0 (10)
This described the behavior of an ideal conductor (for which
= 0), but not the Meissner eect. To describe this, the
constant of integration must be chosen to be zero. Then
j
s
=
n
s
e
2
mc
B (Second London Eqn.) (11)
or dening
L
=
m
n
s
e
2
, the London Equations become
B
c
=
L
j
s
c =
L
j
s
t
(12)
9
If we now apply the Maxwell equation H =
4
c
j B =
4
c
j then we get
(B) =
4
c
j =
4
c
2

L
B (13)
and
(j) =
1

L
c
B =
4
c
2

L
j (14)
or since B = 0, j =
1
c

t
= 0 and ( a) =
( a)
2
a we get

2
B
4
c
2

L
B = 0
2
j
4
c
2

L
j = 0 (15)
B
SC
j B z x
s

z
B
x
x
y
z

j
^ ^
Figure 6: A superconducting slab in an external eld. The eld penetrates into the
slab a distance
L
=
_
mc
2
4ne
2

.
10
Now consider a the superconductor in an external eld shown
in Fig. 6. The eld is only in the x-direction, and can vary in
space only in the z-direction, then since B =
4
c
j, the
current is in the y-direction, so

2
B
x
z
2

4
c
2

L
B
x
= 0

2
j
sy
z
2

4
c
2

L
j
sy
= 0 (16)
with the solutions
B
x
= B
0
x
e

L
j
sy
= j
sy
e

L
(17)

L
=

c
2

L
4
=

mc
2
4ne
2

is the penetration depth.

3 Cooper Pairing
The superconducting state is fundamentally dierent than any
possible normal metallic state (ie a perfect metal at T = 0).
Thus, the transition from the normal metal state to the super-
conducting state must be a phase transition. A phase transition
is accompanied by an instability of the normal state. Cooper
rst quantied this instability as due to a small attractive(!?)
interaction between two electrons above the Fermi surface.
11
3.1 The Retarded Pairing Potential
The attraction comes from the exchange of phonons. The lat-
v 10 cm/s
F
8
+
+ +
+ +
+
+
+
+
+
+
region of
positive charge
attracts a second
electron
ions
+ +
+
+
e
-
e
-
Figure 7: Origin of the retarded attractive potential. Electrons at the Fermi surface
travel with a high velocity v
F
. As they pass through the lattice (left), the positive ions
respond slowly. By the time they have reached their maximum excursion, the rst
electron is far away, leaving behind a region of positive charge which attracts a second
electron.
tice deforms slowly in the time scale of the electron. It reaches
its maximum deformation at a time
2

D
10
13
s after the
electron has passed. In this time the rst electron has traveled
v
F
10
8cm
s
10
13
s 1000

A
. The positive charge of
the lattice deformation can then attract another electron with-
out feeling the Coulomb repulsion of the rst electron. Due
to retardation, the electron-electron Coulomb repulsion may be
neglected!
12
The net eect of the phonons is then to create an attrac-
tive interaction which tends to pair time-reversed quasiparticle
states. They form an antisymmetric spin singlet so that the
e
e
1000

k
- k
Figure 8: To take full advantage of the attractive potential illustrated in Fig. 7, the
spatial part of the electronic pair wave function is symmetric and hence nodeless. To
obey the Pauli principle, the spin part must then be antisymmetric or a singlet.
spatial part of the wave function can be symmetric and nodeless
and so take advantage of the attractive interaction. Further-
more they tend to pair in a zero center of mass (cm) state so
that the two electrons can chase each other around the lattice.
3.2 Scattering of Cooper Pairs
This latter point may be quantied a bit better by considering
two electrons above a lled Fermi sphere. These two electrons
13
are attracted by the exchange of phonons. However, the max-
imum energy which may be exchanged in this way is h
D
.
Thus the scattering in phase space is restricted to a narrow
shell of energy width h
D
. Furthermore, the momentum in
k
2
k
1
k
2
k
1
k
1
k
1
k
2
k
2

D
E k
2
k
Figure 9: Pair states scattered by the exchange of phonons are restricted to a narrow
scattering shell of width h
D
around the Fermi surface.
this scattering process is also conserved
k
1
+ k
2
= k

1
+ k

2
= K (18)
Thus the scattering of k
1
and k
2
into k

1
and k

2
is restricted to
the overlap of the two scattering shells, Clearly this is negligible
unless K 0. Thus the interaction is strongest (most likely)
if k
1
= k
2
and
1
=
2
; ie., pairing is primarily between
14
time-reversed eigenstates.
k
1
-k
2
K
scattering shell
Figure 10: If the pair has a nite center of mass momentum, so that k
1
+ k
2
= K,
then there are few states which it can scatter into through the exchange of a phonon.
3.3 The Cooper Instability of the Fermi Sea
Now consider these two electrons above the Fermi surface. They
will obey the Schroedinger equation.

h
2
2m
(
2
1
+
2
2
)(r
1
r
2
) +V (r
1
r
2
)(r
1
r
2
) = ( +2E
F
)(r
1
r
2
)
(19)
If V = 0, then = 0, and

V =0
=
1
L
3/2
e
ik
1
r
1
1
L
3/2
e
ik
2
r
2
=
1
L
3
e
ik(r
1
r
2
)
, (20)
15
where we assume that k
1
= k
2
= k. For small V, we will
perturb around the V = 0 state, so that
(r
1
r
2
) =
1
L
3

k
g(k)e
ik(r
1
r
2
)
(21)
The sum must be restricted so that
E
F
<
h
2
k
2
2m
< E
F
+ h
D
(22)
this may be imposed by g(k), since [g(k)[
2
is the probability of
nding an electron in a state k and the other in k. Thus we
take
g(k) = 0 for
_

_
k < k
F
k >

2m(E
F
+ h
D
)
h
(23)
The Schroedinger equations may be converted to a k-space
equation by multiplying it by
1
L
3
_
d
3
r e
ik

r
S.E. (24)
so that
h
2
k
2
m
g(k) +
1
L
3

g(k

)V
kk
= ( + 2E
F
)g(k) (25)
where
V
kk
=
_
V (r)e
i(k k

)r
d
3
r (26)
16
now describes the scattering from (k, k) to (k

, k

). It is
usually approximated as a constant for all k and k

which obey
the Pauli-principle and scattering shell restrictions
V
kk
=
_

_
V
0
E
F
<
h
2
k
2
2m
,
h
2
k

2
2m
< E
F
+ h
D
0 otherwise
. (27)
so
_
_
_
h
2
k
2
m
+ + 2E
F
_
_
_ g(k) =
V
0
L
3

g(k

) A (28)
or
g(k) =
A

h
2
k
2
m
+ + 2E
F
(i.e. for E
F
<
h
2
k
2
2m
< E
F
+ h
D
)
(29)
Summing over k
V
0
L
3

k
A
h
2
k
2
m
2E
F
= +A (30)
or
1 =
V
0
L
3

k
1
h
2
k
2
m
2E
F
(31)
This may be converted to a density of states integral on E =
h
2
k
2
2m
17
1 = V
0
_
E
F
+ h
D
E
F
Z(E
F
)
dE
2E 2E
F
(32)
1 =
1
2
V
0
Z(E
F
) ln
_
_
2 h
D

_
_
(33)
=
2 h
D
1 e
2/(V
0
Z(E
F
))
2 h
D
e
2/(V
0
Z(E
F
))
< 0, as
V
0
E
F
0
(34)
4 The BCS Ground State
In the preceding section, we saw that the weak phonon-mediated
attractive interaction was sucient to destabilize the Fermi sea,
and promote the formation of a Cooper pair (k , k ). The
scattering
(k , k ) (k

, k

) (35)
yields an energy V
0
if k and k

are in the scattering shell E

F
<
E
k
, E
k
< E
F
+ h
D
. Many electrons can participate in this
process and many Cooper pairs are formed, yielding a new state
(phase) of the system. The energy of this new state is not just
18
N
2
less than that of the old state, since the Fermi surface is
renormalized by the formation of each Cooper pair.
4.1 The Energy of the BCS Ground State
Of course, to study the thermodynamics of this new phase, it is
necessary to determine its energy. It will have both kinetic and
potential contributions. Since pairing only occurs for electrons
above the Fermi surface, the kinetic energy actually increases:
if w
k
is the probability that a pair state (k , k ) is occupied
then
E
kin
= 2

k
w
k

k
,
k
=
h
2
k
2
2m
E
F
(36)
The potential energy requires a bit more thought. It may be
written in terms of annihilation and creation operators for the
pair states labeled by k
[1)
k
(k , k )occupied (37)
[0)
k
(k , k )unoccupied (38)
or
[
k
) = u
k
[0)
k
+ v
k
[1)
k
(39)
19
where v
2
k
= w
k
and u
2
k
= 1 w
k
. Then the BCS state, which
is a collection of these pairs, may be written as
[
BCS
)

k
u
k
[0)
k
+ v
k
[1)
k
. (40)
We will assume that u
k
, v
k
1. Physically this amounts to
taking the phase of the order parameter to be zero (or ), so
that it is real. However the validity of this assumption can only
be veried for a more microscopically based theory.
By the Pauli principle, the state (k , k ) can be, at most,
singly occupied, thus a (s =
1
2
) Pauli representation is possible
[1)
k
=
_
_
_
_
1
0
_
_
_
_
k
[0)
k
=
_
_
_
_
0
1
_
_
_
_
k
(41)
Where
+
k
and

k
, describe the creation and anhialation of the
state (k , k )

+
k
=
1
2
(
1
k
+ i
2
k
) =
_
_
_
_
0 1
0 0
_
_
_
_
(42)

k
=
1
2
(
1
k
i
2
k
) =
_
_
_
_
0 0
1 0
_
_
_
_
(43)
Of course
+
k
_
_
_
_
0
1
_
_
_
_
k
=
_
_
_
_
1
0
_
_
_
_

+
k
[1)
k
= 0
+
k
[0)
k
= [1)
k
(44)
20

k
[1)
k
= [0)
k

+
k
[0)
k
= 0 (45)
The process (k , k ) (k

, k

), if allowed, is
associated with an energy reduction V
0
. In our Pauli matrix
representation this process is represented by operators
+
k

k
,
so
V =
V
0
L
3

kk

+
k

k
(Note that this is Hermitian) (46)
Thus the reduction of the potential energy is given by
BCS
[V [
BCS
)

BCS
[V [
BCS
) =
V
0
L
3
_
_
_

p
(u
p
0[ + v
p
1[)

kk

+
k

_
u
p
[0)
p

+ v
p
[1)
p

_
_

_
(47)
Then as
k
1[1)
k

=
kk
,
k
0[0)
k

=
kk
and
k
0[1)
k

= 0

BCS
[V [
BCS
) =
V
0
L
3

kk

v
k
u
k
u
k
v
k
(48)
Thus, the total energy (kinetic plus potential) of the system of
Cooper pairs is
W
BCS
= 2

k
v
2
k

V
0
L
3

kk

v
k
u
k
u
k
v
k
(49)
As yet v
k
and u
k
are unknown. They may be treated as
variational parameters. Since w
k
= v
2
k
and 1 w
k
= u
2
k
, we
21
may impose this constraint by choosing
v
k
= cos
k
, u
k
= sin
k
(50)
At T = 0, we require W
BCS
to be a minimum.
W
BCS
=

k
2
k
cos
2

V
0
L
3

kk
cos
k
sin
k
cos
k
sin
k
=

k
2
k
cos
2

V
0
L
3

kk

1
4
sin 2
k
sin 2
k
(51)
W
BCS

k
= 0 = 4
k
cos
k
sin
k

V
0
L
3

cos 2
k
sin 2
k
(52)

k
tan 2
k
=
1
2
V
0
L
3

sin 2
k
(53)
Conventionally, one introduces the parameters E
k
=
_

2
k
+
2
, =
V
0
L
3

k
u
k
v
k
=
V
0
L
3

k
cos
k
sin
k
. Then we get

k
tan 2
k
= 2u
k
v
k
= sin 2
k
=

E
k
(54)
cos 2
k
=

k
E
k
= cos
2

k
sin
2

k
= v
2
k
u
2
k
= 2v
2
k
1 (55)
w
k
= v
2
k
=
1
2
_
_
1

k
E
k
_
_
=
1
2
_
_
_
_
1

k
_

2
k
+
2
_
_
_
_
(56)
If we now make these substitutions
_
2u
k
v
k
=

E
k
, v
2
k
=
1
2
_
1

k
E
k
__
into W
BCS
, then we get
W
BCS
=

k

k
_
_
1

k
E
k
_
_

L
3
V
0

2
. (57)
22
0
w = v
k k
2
T = 0
k F
= -E +
h k
2 2
2m
clearly kinetic
energy increases
1
Figure 11: Sketch of the ground state pair distribution function.
Compare this to the normal state energy, again measured
relative to E
F
W
n
=

k<k
F
2
k
(58)
or
W
BCS
W
n
L
3
=
1
L
3

k
_
_
1 +

k
E
k
_
_

2
V
0
(59)

1
2
Z(E
F
)
2
< 0. (60)
So the formation of superconductivity reduces the ground state
energy. This can also be interpreted as Z(E
F
) electrons pairs
per and volume condensed into a state below E
F
. The aver-
age energy gain per electron is

2
.
23
4.2 The BCS Gap
The gap parameter is fundamental to the BCS theory. It tells
us both the energy gain of the BCS state, and about its excita-
tions. Thus is usually what is measured by experiments. To
see this consider
W
BCS
=

k
2
k
1
2
_
_
1

k
E
k
_
_

L
3

2
V
0
(61)
Lots of algebra (See I&L)
W
BCS
=

2E
k
v
4
k
(62)
Now recall that the probability that the Cooper state (k , k )
was occupied, is given by w
k
= v
2
k
. Thus the rst pair breaking
excitation takes v
2
k
= 1 to v
2
k
= 0, for a change in energy
E =

k=k

2v
4
k
E
k
+

k
2v
4
k
E
k
= 2E
k
= 2
_

2
k

+
2
(63)
Then since
k
=
h
2
k
2
2m
E
F
, the smallest such excitation is just
E
min
= 2 (64)
This is the minimum energy required to break a pair, or create
an excitation in the BCS ground state. It is what is measured
by the specic heat C e
2
for T < T
c
.
24
k
-k
k k
w = v = 1
2
k
2
v = 0
e
e
Figure 12: Breaking a pair requires an energy 2
_

2
k
+
2
2
Now consider some experiment which adds a single electron,
or perhaps a few unpaired electrons, to a superconductor (ie
tunneling). This additional electron cannot nd a partner for
superconductor
normal
metal
Figure 13:
pairing. Thus it must enter one of the excited states discussed
25
above. Since it is a single electron, its energy will be
E
k
=
_

2
k
+
2
(65)
For
2
k
, E
k
=
k
=
h
2
k
2
2m
E
F
, which is just the energy of
a normal metal state. Thus for energies well above the gap, the
normal metal continuum is recovered for unpaired electrons.
To calculate the density of unpaired electron states, recall
that the density of states was determined by counting k-states.
These are unaected by any phase transition. Thus it must be
that the number of states in d
3
k is equal.
d k
3
k
x
k
y
k
z

L
3
Figure 14: The number of k-states within a volume d
3
k of k-space is unaected by
any phase transition.
D
s
(E
k
)dE
k
= D
n
(
k
)d
k
(66)
In the vicinity of
k
, D
n
(
k
) D
n
(E
F
) since [[ E
F
26
(we shall see that 2w
D
). Thus for
k

D
s
(E
k
)
D
n
(E
F
)
=
d
x
dE
k
=
d
dE
k
_
E
2
k

2
=
E
k
_
E
2
k

2
E
k
>
(67)

1
E
D
s D
n
Density of additional
electron states only!
Figure 15:
Given the experimental and theoretical importance of , it
should be calculated.
=
V
0
L
3

k
sin
k
cos
k
=
V
0
L
3

k
u
k
v
k
=
V
0
L
3

2E
k
(68)
=
1
2
V
0
L
3

2
k
+
2
(69)
Convert this to sum over energy states (at T = 0 all states with
27
< 0 are occupied since
k
=
h
2
k
2
2m
E
F
).
=
V
0
2

_
h
D
h
D
Z(E
F
+ )d

2
+
2
(70)
1
V
0
Z(E
F
)
=
_
h
D
0
d

2
+
2
(71)
1
V
0
Z(E
F
)
= sinh
1
_
_
h
D

_
_
(72)
For small ,
h
D

e
1
V
0
Z(E
F
)
(73)
h
D
e

1
V
0
Z(E
F
)
(74)
sinh x
x
e
x
Figure 16:
28
5 Consequences of BCS and Experiment
5.1 Specic Heat
As mentioned before, the gap is fundamental to experiment.
The simplest excitation which can be induced in a supercon-
ductor has energy 2. Thus
E 2e
2
T T
c
(75)
C
E

T


2
T
2
e
2
(76)
5.2 Microwave Absorption and Reection
Another direct measurement of the gap is reectivity/absorption.
A phonon impacting a superconductor can either be reected
or absorbed. Unless h > 2, the phonon cannot create an ex-
citation and is reected. Only if h > 2 is there absorption.
Consider a small cavity within a superconductor. The cavity
has a small hole which allows microwave radiation to enter the
cavity. If h < 2 and if B < B
c
, then the microwave in-
tensity is high I = I
s
. On the other hand, if h > 2 ,or
29
h
cavity
B
10
h
h = 2
I
n
I - I
s n
microwave
superconductor
B=0
Figure 17: If B > B
c
or h > 2, then absorption reduces the intensity to the
normal-state value I = I
n
. For B = 0 the microwave intensity within the cavity is
large so long as h < 2
B > B
c
, then the intensity falls in the cavity I = I
n
due to
absorbs ion by the walls.
Note that this also allows us to measure as a function of
T. At T = T
c
, = 0, since thermal excitations reduce the
number of Cooper pairs and increase the number of unpaired
electrons, which obey Fermi-statistics. The size of (Eqn. 71) is
only eected by the presence of a Cooper pair . The proba-
bility that an electron is unpaired is f
_

2
+
2
+ E
F
, T
_
=
1
exp

2
+
2
+1
so, the probability that a Cooper pair exists is
30
k
-k
e
e
kT 2
Figure 18:
1 2f
_

2
+
2
+ E
F
, T
_
. Thus for T ,= 0
1
V
0
Z(E
F
)
=
_
h
D
0
d

2
+
2
_
1 2f
_
_

2
+
2
+ E
F
, T
__
(77)
Note that as

2
+
2
0, when we recover the
T = 0 result.
This equation may be solved for (T) and for T
c
. To nd T
c
(T)
(0)
T/T
c
1
In
Sn
Pb
Real SC data (reflectivity)
Figure 19: The evolution of the gap (as measured by reectivity) as a function of tem-
perature. The BCS approximation is in reasonably good agreement with experiment.
31
consider this equation as
T
T
c
1, the rst solution to the gap
equation, with = 0
+
, occurs at T = T
c
. Here
1
V
0
Z(E
F
)
=
_
h
D
0
d

tanh
_
_

2k
B
T
c
_
_
(78)
which may be solved numerically to yield
1 = V
0
Z(E
F
) ln
1.14 h
D
k
B
T
c
(79)
k
B
T
c
= 1.14 h
D
e
1/{V
0
Z(E
F
)}
(80)
but recall that = 2 h
D
e
1/{V
0
Z(E
F
)}
, so
(0)
k
B
T
c
=
2
1.14
= 1.764 (81)
metal T
c

K Z(E
F
)V
0
(0)/k
B
T
c
Zn 0.9 0.18 1.6
Al 1.2 0.18 1.7
Pb 7.22 0.39 2.15
Table 1: Note that the value 2.15 for (0)/k
B
T
c
for Pb is higher than BCS predicts.
Such systems are labeled strong coupling superconductors and are better described by
the Eliashberg-Migdal theory.
32
5.3 The Isotope Eect
Finally, one should discuss the isotope eect. We know that
V
kk
, results from phonon exchange. If we change the mass of
one of the vibrating members but not its charge, then V
0
N(E
F
)
etc are unchanged but

_
k
M
M

1
2
. (82)
Thus T
c
M

1
2
. This has been conrmed for most normal
superconductors, and is considered a smoking gun for phonon
mediated superconductivity.
6 BCS Superconducting Phenomenology
Using Maxwells equations, we may establish a relation between
the critical current and the critical eld necessary to destroy the
superconducting state. Consider a long thick wire (with radius
r
0

L
) and integrate the equation
H =
4
c
j (83)
33
d l

L
S
r
0 j
0
j = j e
0
0
(r - r )/
L
H

H
Figure 20: Integration contour within a long thick superconducting wire perpendicular
to a circulating magnetic eld. The eld only penetrates into the wire a distance
L
.
along the contour shown in Fig. 20.
_
HdS =
_
H dl =
4
c
_
j ds (84)
2r
0
H =
4
c
2r
0

L
j
0
(85)
If j
0
= j
c
(j
c
is the critical current), then
H
c
=
4
c

L
j
c
(86)
Since both H
c
and j
c
, they will share the temperature-
dependence of .
At T = 0, we could also get an expression for H
c
by noting
34
that, since the superconducting state excludes all ux,
1
L
3
(W
n
W
BCS
) =
1
8
H
2
c
(87)
However, since we have earlier
1
L
3
(W
n
W
BCS
) =
1
2
N(0)
2
, (88)
we get
H
c
= 2
_
N(0) (89)
We can use this, and the relation derived above j
c
=
c
4
L
H
c
,
to get a (properly derived) relationship for j
c
.
j
c
=
c
4
L
2
_
N(0) (90)
However, for most metals
N(0)
n
E
F
(91)

L
=

_
mc
2
4ne
2

(92)
taking = 1
j
c
=
c
4

_
4ne
2
mc
2
2

_
n2m
h
2
k
2
F
=

2
ne
hk
F
(93)
35
This gives a similar result to what Ibach and L uth get, but
for a completely dierent reason. Their argument is similar to
one originally proposed by Landau. Imagine that you have a
uid which must ow around an obstacle of mass M. From the
perspective of the uid, this is the same as an obstacle moving
in it. Suppose the obstacle makes an excitation of energy and
M
v
M
v
P
E
Figure 21: A superconducting uid which must ow around an obstacle of mass M.
From the perspective of the uid, this is the same as an obstacle, with a velocity equal
and opposite the uids, moving in it.
momentum p in the uid, then
E

= E P

= Pp (94)
or from squaring the second equation and dividing by 2M
M
P
E
M
P
E
p
(a)
(b)

Figure 22: A large mass M moving with momentum P in a superuid (a), creates an
excitation (b) of the uid of energy and momentum p
36
P
2
2M

P
2
2M
=
P p
M
+
p
2
2M
= E

E = (95)
P
P
p
v = P/M

Figure 23:
=
pP cos
M

p
2
2M
(96)
= pv cos
p
2
2M
(97)
If M (a defect in the tube which carries the uid could
have essentially an innite mass) then

p
= v cos (98)
Then since cos 1
v

p
(99)
Thus, if there is some minimum ,then there is also a mini-
mum velocity below which such excitations of the uid cannot
37
happen. For the superconductor
v
c
=

min
p
=
2
2 hk
F
(100)
Or
j
c
= env
c
=
ne
hk
F
(101)
This is the same relation as we obtained with the previous
thermodynamic argument (within a factor

2). However, the

former argument is more proper, since it would apply even for
gapless superconductors, and it takes into account the fact that
the S.C. state is a collective phenomena ie., a minuet, not a
waltz of electric pairs.
7 Coherence of the Superconductor Meisner
eects
Superconductivity is the Meissner eect, but thus far, we have
not yet shown that the BCS theory leads to the second London
equation which describes ux exclusion. In this subsection, we
will see that this requires an additional assumption: the rigidity
of the BCS wave function.
38
In the BCS approximation, the superconducting wave func-
tion is taken to be composed of products of Cooper pairs. One
can estimate the size of the pairs from the uncertainty principle
2 =
_
_
_
p
2
2m
_
_
_
p
F
m
p p 2m

p
F
(102)

cp
x
h
p

hp
F
2m
=
h
2
k
F
2m
=
E
F
k
F

(103)

cp
10
3
10
4

A
size of Cooper pair wave function (104)
Thus in the radius of the Cooper pair, about
4n
3
_
_

cp
2
_
_
3
10
8
(105)
other pairs have their center of mass.
Figure 24: Many electron pairs fall within the volume of a Cooper wavefunction.
This leads to a degree of correlation between the pairs and to rigidity of the pair
wavefunction.
39
The pairs are thus not independent of each other (regardless
of the BCS wave function approximation). In fact they are
specically anchored to each other; ie., they maintain coherence
over a length scale of at least
cp
.
SC
Normal Metal

BCS
2
>
cp coh
Figure 25:
In light of this coherence, lets reconsider the supercurrent
j =
2e
4m
p

p (106)
where pair mass = 2m and pair charge = 2e.
p = i h
2e
c
A (107)
A current, or a CM momentumK, modies the single pair state
(r
1
, r
2
) =
1
L
3

k
g(k)e
iK (r
1
+r
2
)/2
e
ik (r
1
r
2
)
(108)
(K, r
1
, r
2
) = (K = 0, r
1
, r
2
)e
iKR
(109)
40
where R =
r
1
+r
2
2
is the cm coordinate and hK is the cm mo-
mentum. Thus

BCS
e
i

BCS
(K = 0) = e
i
(0) (110)
= K (R
1
+ R
2
+ ) (111)
(In principle, we should also antisymmetrize this wave function;
however, we will see soon that this eect is negligible). Due to
the rigidity of the BCS state it is valid to approximate
=
R
+
r

R
(112)
Thus
j
s

2e
4m

_
_
_

BCS
_
_
i h
R

+
2eA
c
_
_

BCS
+
BCS
_
_
i h
R

+
2eA
c
_
_

BCS
_

_
(113)
or
j
s
=
2e
2m
_
_
_
[(0)[
2
4eA
c
+ 2 h[(0)[
2

_
_
_
(114)
Then since for any , = 0
j
s
=
2e
2
mc
[(0)[
2
A (115)
41
or since [(0)[
2
=
n
s
2
j =
ne
2
mc
B (116)
which is the second London equation which as we saw in Sec.??
leads to the Meissner eect. Thus the second London equation
can only be derived from the BCS theory by assuming that the
BCS state is spatially homogeneous.
8 Quantization of Magnetic Flux
The rigidity of the wave function (superconducting coherence)
also guarantees that the ux penetrating a superconducting
loop is quantized. This may be seen by integrating Eq. 114
along a contour within the superconducting bulk (at least a
distance
L
from the surface).
j
s
=
e
2
n
s
mc
A
e hn
s
2m

(117)
_
j
s
dl =
e
2
n
s
ms
_
A dl
e hn
s
2m

dl (118)
Presumably the phase of the BCS state
BCS
= e
i
(0) is
42
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
B

L
C
superconducting loop
Figure 26: Magnetic ux penetrating a superconducting loop is quantized. This may
be seen by integrating Eq. 114 along a contour within the superconducting bulk (a
distance
L
from the surface).
single valued, so

_

R

dl = 2N N Z (119)
Also since the path l may be taken inside the superconductor
by a depth of more than
L
, where j
s
= 0, we have that
_
j
s
dl = 0 (120)
so

e
2
n
s
ms
_
A dl =
e
2
n
s
ms
_
B ds = 2N
e hn
s
2m
(121)
Ie., the ux in the loop is quantized.
43
9 Tunnel Junctions
Imagine that we have an insulating gap between two metals,
and that a plane wave (electronic Block State) is propagating
towards this barrier from the left
a b c
V
0
metal
insulator
metal
d
dx
2
2
2m
2
h
+ E = 0
dx
2
d
2
2m
2
h
+ (E - V )
0
d
dx
2
2
2m
2
h
+ E = 0
V
x
0
d
Figure 27:

a
= A
1
e
ikx
+ B
1
e
ikx

b
= A
2
e
ik

x
+ B
2
e
ik

c
= B
3
e
ikx
(122)
These are solutions to the S.E. if
k =

2mE
h
in a & c (123)
k

=
_
2m(E V
0
)
h
in b (124)
44
The coecients are determined by the BC of continuity of
and

at the barriers x = 0 and x = d. If we take B

3
= 1
and E < V
0
, so that
k

= i =
_
2m(E V
0
)
h
(125)
then, the probability of having a particle tunnel from left to
right is
P
lr

[B
3
[
2
[B
1
[
2
=
1
[B
1
[
2
=
_

_
1
2

1
8
_
_
k

k
_
_
2
+
1
8
_
_
k

+

k
_
_
2
cosh 2d
_

_
1
(126)
For large d
P
lr
8
_
_
k

+

k
_
_
2
e
2d
(127)
8
_
_
k

+

k
_
_
2
exp
_

2d
_
2m(V
0
E)
h
_

_
(128)
Ie, the tunneling probability falls exponentially with distance.
Of course, this explains the physics of a single electron tun-
neling across a barrier, assuming that an appropriate state is
45
lled on the left-hand side and available on the right-hand side.
This, as can be seen in Fig. 28, is not always the case, es-
pecially in a conductor. Here, we must take into account the
densities of states and their occupation probabilities f. We will
be interested in applied voltages V which will shift the chemical
potential eV . To study the gap we will apply
X
eV
S I
N
N(E)
E
Figure 28: Electrons cannot tunnel accross the barrier since no unoccupied states are
available on the left with correspond in energy to occupied states on the right (and
vice-versa). However, the application of an appropriate bias voltage will promote the
state on the right in energy, inducing a current.
eV (129)
We know that
2
k
B
T
c
4,
4k
B
T
c
2
10

K. However typical
metallic densities of states have features on the scale of electron-
46
volts 10
4
K. Thus, on this energy scale we may approximate
the metallic density of states as featureless.
N
r
() = N
metal
() N
metal
(E
F
) (130)
The tunneling current is then, roughly,
I P
_
df( eV )N
r
(E
F
)N
l
()(1 f())
P
_
df()N
l
()N
r
(E
F
)(1 f( eV )) (131)
For eV = 0, clearly I = 0 i.e. a balance is achieved. For
E
F
Figure 29: If eV= 0, but there is a small overlap of occupied and unoccupied states on
the left and right sides, then there still will be no current due to a balance of particle
hopping.
eV ,= 0 a current may occur. Lets assume that eV > 0
and k
B
T . Then the rightward motion of electrons is
47
suppressed. Then
I PN
r
(E
F
)
_
df( eV )N
l
() (132)
and
dI
dV
PN
r
(E
F
)
_
d
f( eV )
V
N
l
() (133)
f
V
e( eV E
F
) (T E
F
) (134)
dI
dV
PN
r
(E
F
)N
l
(eV + E
F
) (135)
Thus the low temperature dierential conductance
dI
dV
is a mea-
sure of the superconducting density of states.
/e
I
V
V
/e
dI
dV
Figure 30: At low temperatures, the dierential conductance in a normal metal
superconductor tunnel junction is a measure of the quasiparticle density of states.
48