Académique Documents
Professionnel Documents
Culture Documents
7)
We will consider the gas of fermions in the degenerate regime, where the density n
exceeds by far the quantum density nQ, or, in terms of energies, where the Fermi energy
exceeds by far the temperature. We have seen that for such a gas is positive, and well
confine our attention to the limit in which is close to its T=0 value, the Fermi energy EF.
~ kBT
/EF
occupancy
1 kBT/EF
T=0
EF
h2 3
n
2/3
6.6 10 110
34 2
29 2 / 3
J 110 18 J 6 eV
8m
31
8 9 10
- at room temperature, this
TF EF / k B few eV few 10 K 4
Fermi gas is strongly
degenerate (EF >> kBT).
J 4.3 10 12 J 27 MeV
8 1.6 10 27
EF >>> kBT the system is strongly degenerate. The nucleons are very cold they
are all in their ground state!
The average kinetic energy in a degenerate Fermi gas = 0.6 of the Fermi energy
EF
h 2 3N
2/3
6.6 10
34 2
3
8.5 10 28
2/3
1.110 18 J 6.7 eV
8m V 8 9.110 31
(b) The electrons participate in the current flow if their energies correspond to the
occupancy n() that is not too close to 1 (no empty states available for the
accelerated electrons) and not too small (no electrons to accelerate). At T=300K,
calculate the energy interval that is occupied by the electrons that participate in the
current flow, assuming that for these electrons the occupancy varies between 0.1
and 0.9.
EF
n
1 1
0.9 exp 1 9 1 EF k BT ln 9
EF EF k BT
exp 1 exp 1 1
k BT k BT
1 EF 1
0.1 exp 2 2 EF k BT ln 9
2kBT ln 9 0.11 eV 2 EF B k T 9
exp 1
k BT
Problem (Final 2005, cont.)
(b)
(c) Using the assumptions of (b), calculate the ratio N1/N0 where N1 is the number of
current-carrying electrons, N0 is the total number of electrons in the conduction
band. Assume that within the range where the occupancy varies between 0.1 and
0.9, the occupancy varies linearly with energy (see the Figure), and the density of
states is almost energy-independent. The density of states for the three-
dimensional Fermi gas:
occupancy
g
3N
3/ 2
2 EF
- EF
E F / 2
3
n g d
3N
N1 3/ 2
EF 0.5 N N 0.012
E F / 2 2 EF 4 EF
Thus, at T=300K, the ratio of the current-carrying electrons to all electrons in the
conduction is 0.012 or 1.2 %.
The Heat Capacity of a Cold Fermi Gas
One of the greatest successes of the free electron model and FD statistics is the
explanation of the T dependence of the heat capacity of a metal.
So far, we have been dealing with the Fermi gas at T =
0. To calculate the heat capacity, we need to know how
the internal energy of the Fermi gas, U(T), depends on
the temperature. Firstly, lets predict the result
qualitatively.
By heating a Fermi gas, we populate some states
above the Fermi energy EF and deplete some states
below EF. This modification is significant within a
narrow energy range ~ kBT around EF (we assume that
kBT the system is cold - strong degeneracy).
The fraction of electrons that we transfer to higher energies ~ kBT/EF, the energy
increase for these electrons ~ kBT. Thus, the increase of the internal energy with
temperature is proportional to N(kBT/EF) (kBT) ~ N (kBT)2 /EF.
U T k B2T k T 3
CV N is much smaller (by B 1) than CV Nk B for an ideal gas
T V EF EF 2
The Fermi gas heat capacity is much smaller than that of a classical ideal gas with the
same energy and pressure. This is because only a small fraction kBT/EF of the electrons
are excited out of the ground state. As required by the Third Law, the electronic heat
capacity in metals goes to 0 at T 0.
The Heat Capacity of a Cold Fermi Gas (cont.)
A bit more quantitative approach: the difference
in energy between the gas at a finite temperature U T g f , T d U0
0
and the gas in the ground state (T = 0):
occupancy
When we consider the energies very close to ~ EF,
we will ignore the variation in the density of states, and
evaluate the integral with g() g(EF). The density of
states at the Fermi level is one of the most important
properties of a metal, since it determines the number
of states available to the electrons which can change
their state under weak excitations. with respect to EF
So long as kBT << EF, the distribution function f() is symmetric about EF. By heating up a
metal, we take a group of electrons at the energy - (with respect to EF), and lift them
up to . The number of electrons in this group g(EF)f()d and each electron has
increased its energy by 2 :
U U 0 2 g EF f , T d
- the lower limit is 0,
0
we integrate from EF
2 2
U U 0 2 g EF d 2k BT g EF kBT 2 g EF
x
dx
2
0
exp 1 0
exp x 1 12 6
dU T 2 2
3/ 2
1 2m k BT
g E F 2 2
3 n
g EF k B2T Ce
1/ 2
Ce EF Nk B
dT 3 2 2 EF 2 EF
- much less than the equipartition C. The small heat capacity is a direct consequence
of the Pauli principle: most of the electrons cannot change their energy.
Experimental Results
At low temperatures, the heat capacity of metals can be represented as the sum of
two terms:
CV K1T K 2T3
At T < 10 K, this linear (electronic) term dominates in the total heat capacity of
metals: the other term due to lattice vibrations dies out at T 0 faster, as T 3.
Pressure of the Fermi Gas
The internal energy and pressure of an ideal F
U 0 N ,V g d N EF
3
gas goes to 0 as T0. This is not the case
for a degenerate Fermi gas ! 0
5
2/3
2 2 N 3 2 N 2 N
2/3
On the other hand, EF 3 and U0 3 A N 5 / 3V 2 / 3
2m V 10 m V
At T=0, there is no distinction between the free energy F and the internal energy U0:
U 0
2/3
P
2 5 / 3 3 2 N 2 N
V 3
1 2n
2
3 2 n n EF
2/3
V N 3 10 m V 5 m 5
2/3 - it looks just as for an ideal gas. This is a
1 h2 3 2
P n 5/3
or PV U 0 direct consequence of the quadratic
20 m 3 relation between energy and momentum.
However, this is a non-zero pressure at T = 0, which does not depend on T at T << EF.
2
Lets estimate this pressure for a typical metal: P nEF 10 29 m-3 5 10 19 J 5 1010 Pa
5
In metals, this enormous pressure is counteracted by the Coulomb attraction of the
electrons to the positive ions.
When an electron is confined in a very small space, it "flies about its tiny cell at high
speed, kicking with great force against adjacent electrons in their cells. This degenerate
motion ... cannot be stopped by cooling the matter. Nothing can stop it; it is forced on the
electron by the laws of quantum mechanics, even when the matter is at absolute zero
temperature" (Thorne 1994).
Fermi Pressure in the Universe
2/3 - scales with the density as n5/3 provided that the electrons
3
2
1 h
Pnon rel n5 / 3 remain non-relativistic (speeds v << c). In a white dwarf
20 m or a neutron star, this approximation breaks down.
For the relativistic degenerate matter, the equation of state is softer Prel n 4 / 3
k3
The number of electrons per unit volume with k<kF: G k
2
P n EF
5 3 2
At T=0, all the states up to kF are occupied: n Gk F
k F3
n 2
3
k F 3 2 n
1/ 3
E pc ck EF c 3 n 2 1/ 3
In the ultra-relativistic case
Nobel 1983 The Fermi pressure of a degenerate electron gas prevents the gravitational
collapse of the star if the star is not too massive (the white dwarf).
An estimate the upper limit of the white dwarfs mass (the Chandrasekhar mass):
M M 3
Total number of electrons in the star ~ the number of protons: Ne ne
mp m p 4R3
GM 2
Total potential (gravitational) energy of the star: U g C g Cg ~ 1 proton mass
R
Kin. energy of the Fermi gas of electrons (relativistic case): -Ug
energy
1/ 3
M 2 M 3
K
c M 4 / 3
K C F N e EF C F c 3 4/3
m p m p 4R 3 mp R
If the mass is not too high (<3 MSun), the further contraction is stopped by the
degeneracy pressure of neutrons (otherwise black hole).
By knowing the parameters of neutron stars, we can explain why these stars are made
of neutrons. A typical neutron star has a mass like the Sun (MSun = 2 1033 g), but a
much smaller radius R ~10 km. The average density is M
density 4.8 1017 kg/m 3
R 3
4
3
about three times nuclear density. In terms of the concentration of neutrons, this
corresponds to n = 3 1044 m-3. Let the star be made, instead of neutrons, of protons
and electrons, each with the concentration n ~ 3 1044 m-3, the # of protons = the # of
electrons (the star is electrically neutral). The mass difference between a proton and a
neutron is greater than the mass of an electron:
M nc 2 M p c 2 mec 2 (940 MeV - 938 MeV 0.5 MeV)
so this configuration would seem to be energetically favored over the neutron
configuration. However, lets look at the Fermi energies of the protons and electrons in
this hypothetical star. The proton EF would be very nearly the same as that for the
neutrons above (because the concentration would be the same and the mass is nearly
the same). On the other hand, the electron EF would be larger by the ratio Mn/me ~1838,
so the electron EF would be ~160,000 MeV, enough to make about 170 nucleons!