Solution to Statistical Physics Exam

29th June 2015

Name Student Number

Problem 1 Problem 2 Problem 3 Problem 4 Total Percentage Mark

Useful constants
Gas constant R 8.31J/ (K · mol)
Boltzmann constant kB 1.38·10−23 J/K
Avogadro number NA 6.02·1023 mol−1
Speed of light c 300·106 m/s

1
(25P) Problem 1
1. (2P) Find the entropy of rolling an octahedral and a dodecahedral “die”, respectively.The entropy
S reads
n
X 1
S = kB pi ln ,
i=1
pi

where pi is the probability that the system is in the i-th microstate, and n is the total number
of the microstates. For the octahdral “die” one finds pi = 1/8 and n = 8, yielding S = kB ln 8 .
For the dodecahedral “die” one finds pi = 1/12 and n = 12, yielding S = kB ln 12 .

2. (1P) How does the phase space density of a microcanonical ensemble in equilibrium change with
time? The phase space density of the microcanonical ensemble is constant in equilibrium.
3. (2P) Which are the units of the micro-canonical and the grand-canonical partition function,
respectively? The partition functions are dimensionless, they are the number of states, in the
microcanonical case, those which are compliant with a certain energy, in the grand-canonical
case, weighted by the dimensionless Boltzmann factor.

4. (3P) Which quantities are constant in a

(a) micro-canonical ensemble ? N, V, E number of particles, volume, energy

(b) canonical ensemble ?N, V, T number of particles, volume, temperature
(c) grand-canonical ensemble?µ, V, T chemical potential, volume, temperature

5. (1P) How does the heat capacity of a classical ideal monoatomic gas depend on temperature?
Cv = 23 N kB no temperature dependence.
6. (1P) Which thermodynamic potential is constant during a Joule-Thomson process? Enthalpy
H

7. (2P) Express the heat capacity at constant volume CV and the heat capacity at constant
pressure CP as a second derivative of an
 appropriate
 thermodynamic
 2 potential
 (N =const)
∂Q dS ∂ F
since ∂F



heat capacity at constant volume CV = ∂T = T dT V = −T ∂T 2 ∂T V,N =
V,N V,N
−S (N, p, T )
   
∂Q dS ∂2G ∂G



heat capacity at constant pressure Cp = ∂T = T dT p = −T ∂T 2 since ∂T p,N =
p,N p,N
−S (N, p, T )
1. (2P) What is the qualitative difference between the phase diagram of water and that of a “normal
substance”, e.g. carbon dioxide? The volume of solid water is larger than that of liquid water
at most temperatures. Hence, the melting curve has a negative slope, whereas it has a positive
slope for CO2.
2. (3P) Write down the total differential for the Helmholtz Free energy ? The Helmholtz free
energy A dependes on number of particles N , volume V , and temperature T , hence
∂A ∂A ∂A
dA = dN + dV + dT
∂N ∂V ∂T
= µdN − P dV − SdT

2
3. (4P) Which quantities are intensive: volume, temperature, particle number, pressure, entropy?
Temperature, pressure.
4. (3P) Sketch the isotherms for the van der Waals equation in a P, V (pressure,Volume) dia-
gram for temperature T > Tc , T = Tc , and T < Tc where T C is the critical temperature.

5. (1P) What is the maximal occupation number of an energy level in an ideal Fermi gas? One for
each spin level, or two if spin is not considered.

3
 (25P) Problem 2
Consider a system of N ideal gas particles with mass m at a temperature T . The mean pressure is P .
Ng gas particles move freely in a volume V = L3g . Another Ns ideal gas particles are absorbed on a
surface with area L2s , forming a two-dimensional gas and N = Ng + Ns . The energy of an absorbed
2
particle is ε = |p|
2m − ε0 where p is the two-dimensional momentum and ε0 is the binding energy per
particle.
1. (6P) Calculate the classical partition functions of the free gas Zg and of the adsorbed gas Zs .
The particles are indistinguishable. In continuous space
ˆ
X V
≈ 3 d3 p (1)
p
h

Such that ˆ
V
Z (1) ≈ 3 d3 p exp [−βεp ] (2)
h
p2
Plugging in βεp = 2mkB T and integrating over the sphere
ˆ
p2
 
V 2
Z (1) = 3 4π dp · p exp − (3)
h 2mkB T
p2
We now substitute x2 = 2mkB T and obtain a standard integral
ˆ ∞
V 3
dx · x2 exp −x2
 
Z (1) = 3 4π (2mkB T ) 2 (4)
h −∞
| {z }
1√
4 π

such that
V 3
Zg (1, V, T ) = 3
(2πmkB T ) 2 (5)
h
V Ng
Zg (Ng , V, T ) = (6)
Ng !λ3Ng

L2s
 
Ns ε 0
Zs (1, V, T ) = (2πmkB T ) · exp (7)
h2 kB T
L2N
 
s N ε
s 0
Zs (Ns , V, T ) = · exp (8)
Ns !λ2N kB T

 3N2 g
V Ng 2πmkB T

Zg =
Ng ! h2
 2N
Ls 2Ng 2πmkB T 2
  
Ns ε 0
Zs = · exp
Ns ! h2 kB T

2. (4P) Find an expression for the Gibbs free energies Gg and Gs of the free and the absorbed gas,
respectively, using the partition functions from above.

G = A + P V = −kB T ln Z + N kB T

4
 where we have used the ideal gas law P V = N kB T. Hence
 
Ng
  3N2 g
V 2πmk B T
Gg = −kB T ln   + Ng kB T
Ng ! h2

V Ng
  
3Ng 2πmkB T
= −kB T ln + ln + Ng k B T
Ng ! 2 h2
  
3Ng 2πmkB T
= −kB T ln V N − ln Ng ! + ln + Ng kB T
2 h2
  
3Ng 2πmkB T
≈ −kB T ln V N − Ng ln Ng − Ng + ln + Ng kB T
2 h2
  
3 2πmkB T
= −Ng kB T ln V − ln Ng + ln
2 h2
  
Ng 3 2πmkB T
= Ng kB T ln − ln
V 2 h2
" N #
L2N
 
s
s
2πmkB T Ns ε 0
Gs = −kB T ln · exp + Ns kB T
Ns ! h2 kB T
Ls 2Ns
   
2πmkB T Ns ε 0
= −kB T ln + Ns ln + + Ns k B T
Ns ! h2 kB T
   
2Ns 2πmkB T Ns ε 0
= −kB T ln Ls − ln Ns ! + Ns ln + + Ns kB T
h2 kB T
   
2N 2πmkB T Ns ε0
≈ −kB T ln Ls − Ns ln Ns − Ns + Ns ln + + Ns k B T
h2 kB T
   
2πmkB T ε0
= −Ns kB T ln L2s − ln Ns + ln +
h2 kB T
   
Ns 2πmkB T ε0
= Ns kB T ln 2 − ln −
Ls h2 kB T

3. (4P) Find the chemical potentials µg and µs of the free and the absorbed gas, respectively. The
G ∂G


chemical potential is the Gibbs free energy per particle µ = N or also µ = ∂N p,T
. Hence,
  
V 3 2πmkB T
µg = −kB T ln + ln
Ng 2 h2
and
L2
   
2πmkB T ε0
µs = −kB T ln s + ln +
Ns h2 kB T

4. (4P) At temperature T , the free gas particles and the absorbed gas particles are in equilibrium.
Calculate the mean number of gas particles absorbed per unit area in terms of of the given
conditions (temperature pressure, ...) In equilibrium, the two chemical potentials are equal
µg = µs
L2s
      
V 3 2πmkB T 2πmkB T ε0
−kB T ln + ln = −kB T ln + ln +
Ng 2 h2 Ns h2 kB T
L2s
   
V 3 2πmkB T 2πmkB T ε0
ln + ln = ln + ln + (9)
Ng 2 h2 Ns h2 kB T

5
 We are interested in the number of particles per area, i.e. solving for LN2 :
s

L2s
   
2πmkB T ε0 V 3 2πmkB T
− ln = ln + − ln − ln
Ns h2 kB T Ng 2 h2
 
1 2πmkB T ε0 V
= − ln 2
+ − ln
2 h kB T Ng
1
h2
 
Ns Ng 2
ε0
ln 2 = ln + ln +
Ls V 2πmkB T kB T
 2
 12  
Ns Ng h ε0
= · exp (10)
L2s V 2πmkB T kB T

5. (2P) Keeping temperature T and total number of particles N the same, now the volume in
which the free gas particles can move is increased. What consequences does this have for the
equilibrium between the free gas particles and the gas particles absorbed on the surface, i.e. are
more or less particles absorbed compared to the case with smaller volume? Justify your answer.
According to eq. 10 the number of absorbed particles per area (L2s ) decreases. One can think of
there is more space for particles to move freely, and hence the probability to “find” space in the
larger volume is larger.
6. (3P) What do partition function, Gibbs free energy, and chemical potential for gas particles
absorbed on a one-dimensional “surface” look like?
 1N
Ls 2N 2πmkB T 2
  
Nl ε0
Zl = · exp
Nl ! h2 kB T
   
Nl 1 2πmkB T ε0
Gl = Nl kB T ln − ln −
Ll 2 h2 kB T
L1l
   
1 2πmkB T ε0
µl = −kB T ln + ln +
Nl 2 h2 kB T

7. (2P) The Nl particles absorbed on the one-dimensional surface are now in equilibrium with the
free gas particles. (There are no particles on a two-dimensional surface and N = Ng + Nl ).
How does the number of particles per unit length in this equilibrium compare to the number
of particles per unit area (from the equilibrium above) under the same (temperature, pressure,
mass...) conditions?
µg = µl

L1l
   
V 3 2πmkB T 1 2πmkB T ε0
ln + ln = ln + ln +
Ng 2 h2 Ng 2 h2 kB T
   
Nl 1 2πmkB T ε0 V 3 2πmkB T
ln 1 = ln + − ln − ln
Ll 2 h2 kB T Ng 2 h2
 
Ng 2πmkB T ε0
= ln − ln +
V h2 kB T
h2
   
Nl Ng ε0
= · exp
L1l V 2πmkB T kB T
Hence
Nl  21
h2 √

L1l
Ns
= = λ
L2s
2πmkB T

6
(25P) Problem 3
Consider an ideal Fermi gas of N particles in a volume V and in a magnetic field H at temperature
T = 0. The energy of a particle is
p2
ε= ± µB H (11)
2m
where µB is Bohr’s mageton.
1. (6P) Give an expression for the chemical potential µ0 in terms of the number of particles N for
vanishing magnetic field. For zero temperature, the Fermi-Dirac distribution is a step function.
To get the number of particles N , we integrate up to the Fermi momentum pF
ˆ pF
V
N = 2· 3 d3 p
h 0
ˆ pF
V
= 2 · 3 4π p2 dp
h 0
4πV 3
= 2· p
3h3 F
At T = 0 and for vanishing magnetic field, the chemical potential equals the Fermi energy
p2F
µ0 = εF = 2m . Hence,
  13
N 3
pF = 3h
2 · 4πV
and
N

2
3 3
8πV 3h
µ0 = (12)
2m
2
~3 3π 2 N

3
V
= (13)
2m

2. (4P) Write down the Fermi momenta of the spins oriented parallel and anti-parallel, respectively,
to the external magnetic field H 6= 0 For finite magnetic field
p2F
εF = ± µB H (14)
2m
and with µ0 = εF
1
pF + = [2m (µ0 − µB H)] 2
1
pF − = [2m (µ0 + µB H)] 2

3. (8P) Find the total energy of spins oriented parallel E− and anti-parallel E+ , respectively. The
total energy of the spins oriented anti-parallel to the external magnetic field is
ˆ pF +  2 
V p
E+ = + µB H dp
h3 0 2m
ˆ pF +  2 
4πV p
= + µB H p2 dp
h3 0 2m
ˆ pF +  4 
4πV p 2
= + µB Hp dp
h3 0 2m
4πV p5F + µB Hp3F +
 
= +
h3 10m 3

7
 and the total energy of the spins oriented parallel to the external magnetic field is

4πV p5F − µB Hp3F −

 
E− = 3 −
h 10m 3

4. Bonus Task (3P) Show that for weak external magnetic fields, µ0  µB H, the average energy
per particle "  2 #
E 3 5 µB H
≈ µ0 1 − (15)
N 5 2 µ0
The average energy per particle is
E E+ + E−
=
N N
4πV p5F + p3F + µB H 4πV p5F − p3F − µB H
    
1
= N + + −
h3 10m 3 h3 10m 3
5 5 3 3

!
4πV pF + + pF − µB H pF + − pF −
= 3
+
Nh 10m 3

Using eq.12
4πV 1
= 3
N h3 2 · (2mµ0 ) 2
we have
!
E 3 p5F + + p5F − µB H p3F + − p3F −
= 3 +
N 2 · (2mµ0 ) 2 10m 3
we can write
5 5
p5F + + p5F − [2m (µ0 − µB H)] 2 + [2m (µ0 + µB H)] 2
=
10m 10m
3 5 3 5
(2m) (µ0 − µB H) 2 + (2m) 2 (µ0 + µB H) 2
2
= 2m
h 10m i
3 5 5
(2m) 2 (µ0 − µB H) 2 + (µ0 + µB H) 2
=
5
and

µB H µB H  3 3

p3F + − p3F − = [2m (µ0 − µB H)] 2 − [2m (µ0 + µB H)] 2
3 3
For µ0  µB H we can expand
   
h 5 5
i 55 32 5 5 32
(µ0 − µB H) 2 + (µ0 + µB H) 2 ≈ µ0 − µ0 µB H... + µ0 + µ0 µB H...
2 2

2 2
5
≈ 2µ02 (16)

and
   
h 3 3
i 33 12 3 3 12
(µ0 − µB H) 2 − (µ0 + µB H) 2 = µ0 − µ0 µB H... − µ0 + µ0 µB H...
2 2

2 2
1
≈ −3µ02 µB H (17)

8
 Such that plugging this in
3 5 !
E 3 (2m) 2 2µ02 µB H 3
 1

≈ 3 + (2m) 2 −3µ02 µB H
N 2 · (2mµ0 ) 2 5 3
5 !
3 2µ02 µB H  1

= 3 + −3µ0 µB H
2

2 · (µ0 ) 2 5 3
 
3 2µ0 µB H
= − (µB H)
2 5 µ0
!
2
3 5 (µB H)
= µ0 −
5 2 (µ0 )
!
2
3 5 (µB H)
= µ0 1 − (18)
5 2 (µ0 )2

∂M
5. Bonus Task (4P) Calculate the susceptibility χ = ∂H for weak external magnetic fields, where
M is the average magnetisation per volume.
µB (N− − N+ )
M =
V
µB 4πV 4πV
3


3h3 pF − − 3h3 p3F +
=
 V 3 3

µB 4πV
3h3 [2m (µ0 + µB H)] 2 − [2m (µ0 − µB H)] 2
=
V
and again with eq. 12
4πV 1
= 3
3h3 N 2 · (2mµ0 ) 2
thus
 3 3

N
µB 3 [2m (µ0 + µB H)] 2 − [2m (µ0 − µB H)] 2
2·(2mµ0 ) 2
M =
V
N  3 3

= µB 3 [(µ0 + µB H)] 2 − [(µ0 − µB H)] 2
2V (µ0 ) 2

In the limit of low magnetic field and µB H  µ ≈ εF we expand again

   
h 3 3
i 3 3 12 3 3 12
(µ0 + µB H) − (µ0 − µB H)
2 2
= µ0 + µ0 µB H + ... − µ0 − µ0 µB H...
2 2

2 2
1
≈ 3µ02 µB H

N  1

M = µB 3 3µ02 µB H
2V (µ0 ) 2
3N
= µB (µB H) (19)
2V (µ0 )
3N 2
= µ H
2V µ0 B

9
Figure 1: Pressure-Volume (left) and temperature-entropy (right) diagram of a three-step cyclic pro-
cess.

(25P) Problem 4
A.) Figure 1 shows a three-steps cyclic process as Pressures-volume (P, V ) and temperature-entropy
(T, S) diagram, respectively. Consider 1 mol of an ideal gas which has initially volume V1 , pressure
P1 , and temperature T1 , undergoing such a cyclic process.
1. (3P) Describe the steps 1 → 2; 2 → 3; and 3 → 1 sketched in the two diagrams: Which
thermodynamic quantity changes, which is kept constant ? 1 → 2 isochoric heating to T2 ;2 → 3
isentropic (adiabatic) expansion to V3 ; and 3 → 1 isobaric cooling to T1

2. (12P) For each step calculate the performed work W and the heat transfer Q in terms of pressure
P1 and volumes V1 , V3 .
1 → 2: ˆ V2
W12 = P1 dV = 0
V1

since V2 = V1

Q12 = CV (T2 − T1 )
V1
= CV (P2 − P1 )
R  
V 1 P1 P2
= CV −1
R P1
 
V 1 P1 P2
= CV −1
R P3
 γ 
V 1 P1 V3
= CV −1
R V2
 γ 
V 1 P1 V3
= CV −1
R V1

where we have used the ideal gas law P V = RT and the adiabatic relation V1γ P1 = V2γ P2 and
V1 = V2 together with P1 = P3 .
2 → 3:
Q23 = 0

10
 since S3 = S2 (adiabatic process). Hence ∆U = W and

W23 = U3 − U2

From the definition of the heat capacity at constant volume we rewrite

U3 − U2 = CV (T3 − T2 )
CV
= (P1 V3 − P2 V1 )
R
where we have used the ideal gas law again as P 3 V 3 =RT 3 =P1 V 3 and P2 V2 = RT2 =P2 V1 . Using
the adiabatic relations we then arrive at
 
C V P1 P2
W23 = V3 − V1
R P1
   
CV P1 P2
= V3 − V1
R P3
  γ 
C V P1 V3
= V3 − V1
R V2
  γ 
C V P1 V3
= V3 − V1
R V1

3 → 1: ˆ V1
W31 = P1 dV = P1 (V1 − V3 )
V3

From the first law we know

Q31 = W31 + U1 − U3
plugging in the expression for work and rearranging

(U1 + P1 V1 ) − (U3 + P1 V3 ) = (H1 − H3 )

= CP (T1 − T3 )
= γCV (T1 − T3 )

where we have used the definition if the enthalpy, and the heat capacity at constant pressure
such that we finally obtain
γCV P1
Q31 = − (V3 − V1 ) < 0
R

B.) Bonus Task For a low-density gas the virial expansion can be terminated at first order in the
density and the equation of state is:
N kB T N
P = [1 + B2 (T )] (20)
V V
where P :pressure, V : volume, N : number of particles, T :temperature and kB : Boltzmann’s constant.
B2 (T ) is the second virial coefficient. The heat capacity will have corrections to its ideal gas value.
We can write it in the form
3 N 2 kB
CV,N = N kB − f (T ) (21)
2 V

11
1. (3P) Find the entropy and internal energy as partial derivatives of the Helmholtz free energy, A
(Hint: use the relation between pressure P and Helmholtz free energy at constant temperature).
Regarding to Helmholtz free energy for canonical ensemble we have
∂A
P = −( )T (22)
∂V
N kB T N 2 kB T
P = + B2 (T ) = Pideal + δP (23)
V V2
Using equ.22 the free energy is

N2
A = Aideal + kB T B2 (T ) (24)
V
Then
∂A N2
S = −( )V = Sideal + δS, δS = −kB [B2 (T ) + T B20 (T )] (25)
∂T V
Since we know both entropy and Helmholtz free energy, we find the internal energy as

N2 N2
U = A + T S = Uideal + kB T B2 (T ) − kB T [B2 (T ) + T B20 (T )] = Uideal + δU (26)
V V
N2
δU = −kB T 2 B20 (T ) (27)
V

2. (4P) Find the form that f (T ) must have in order for the two equations to be thermodynamically
consistent. As a result of subtask 1,

∂U N2
CV = ( )V = CVideal + δCV =⇒ δCV = −kB T [2B20 (T ) + T B200 (T )] (28)
∂T V
We find in this way,
f (T ) = 2T B20 (T ) + T 2 B200 (T ) (29)

3. (3P) Find the heat capacity at constant pressure and constant number of partivles CP,N . Let
us express the equation of state as V (P, T ). Since the density is assumed to be small in the
correction one can use equation for the ideal gas to find the the volume. We have,
N kB T
V = + N B2 (T ) (30)
P
Consequently, the entropy can be expressed as

S = Sideal + δS1 (31)

with replacing equ.30 in δS in equ.25 we have

B2 + T B20
δS1 = −kB N (32)
kB T
+ B2
P
now
∂δS1
CP = CPideal + T ( )P (33)
∂T

12
 (B2 + T B20 )0 ( kBPT + B2 ) − ( kPB + B20 )(B2 + T B20 )
= −kB N (34)
( kBPT + B2 )2
N2
= CPideal − kB [T (2B20 + T B200 ) − (B2 + T B20 )] (35)
V
N2
= CPideal + δCV + kB (B2 + T B20 ) (36)
V
As a result,
N2
CP − CV = (CP − CV )ideal + kB (B2 + T B20 ) (37)
V

13