Statistical Mechanics

Homework 1
Shiquan Ma 2021102020002

October 27, 2023

1. Problem 1.2
Demonstrate that gravitational energy is not extensive: show that the gravitational
energy of a sphere of radius r and uniform density varies with volume as V n and
find the exponent n.

Solution: An extensive property is a physical quantity whose value is proportional to

the size of the system it describes. This means that if the gravitational energy is
extensive, then we shall have
Egra (V ) ∝ V,
with the exponent n = 1.
To calculate gravitational energy we reiew the Newton’s law of gravity first
GM m
F =− .
r2
As r → ∞, the gravity force F → 0. It’s convenient to set the gravitational energy to
be 0 at infinity. Therefore, the gravitational energy of an object with mass m under
a gravitational field with a mass source M is equal to the work done to overcome
the gravity force when bringing it from infinity to R (distance to the origin),
Z R
GM m
Egra = (−F )dl = − .
∞ R

Thus, We get dE = − GM
R dm.
Now we start constructing the sphere (ball) of uniform density (ρ) from the infinity.
The mass of the sphere with radius R is
4
m = πR3 ρ.
3
The mass of sphere shell (the pieces we would add to the ball) is

dm = 4πR2 ρdR.

1
 Homework 1 - Statistical Mechanics

Thus the sphere shell gains gravitational energy:

GM
dE = − 4πR2 ρdR
R
16G 2 2 4
=− π ρ R dR,
3
where we substitute the source mass M to the semi-finished sphere mass 43 πR3 ρ.
The gravitational energy of the finished sphere is
Z r
Egra (V ) = dE
∞
Z r
16G 2 2 4
= − π ρ R dR
∞ 3
16
= − Gπ 2 ρ2 r5
15
16 3 5 13 5
= − ( ) 3 Gπ 5 ρ2 V 3 (r).
15 4
5
As the exponent n = 3 ̸= 1, gravitational energy is not a extensive (ofc it’s not
intensive).

2. Problem 1.3
In investigating the conditions for the establishment of equilibrium through the
transfer of thermal energy the fundamental requirement is that the entropy of the
equilibrium state should be a maximum. Equality of temperature was established
from the vanishing of the first derivative of S. What follows from a consideration of
the second derivative?

Solution: Reviewing the discussion on equilibrium conditions in Section 1.3.1 of the

book. Only thermal energy exchange is allowed between the two subsystems. The
microstates of the system equals to

Ωt (E) = Ω1 (E)Ω2 (E0 − E),

where E = E1 , E0 = E1 + E2 . Thus the entropy of the system is

St = S1 (E) + S2 (E0 − E).

The statistical temperature is defined as

1 ∂S
≡ .
T ∂E

2
 Homework 1 - Statistical Mechanics

We take the second derivative to S:

∂2S
   
∂ ∂S1 ∂ ∂S2
= +
∂2E ∂E ∂E ∂E ∂E
   
∂ ∂1 ∂ ∂1
= + .
∂E ∂T1 ∂E ∂T2
∂E
Aplying the defination of the isochoric thermal capacity CV = ∂T to the above
equation,
∂2S 1 1 1 1
=− 2 − .
∂2E T1 C1 T22 C2
We require that the entropy of a system in equilibrium takes a maximum value. This
∂S 2S
means ∂E = 0, ∂∂2 E < 0. When in equilibrium, the temperatures of the subsystems
are the same T1 = T2 . So that we get:
1 1
+ >0
C1 C2
This means that the thermal capacity is positive.

3. Problem 1.12
Starting from the expression for the Gibbs factor for a many-particle system, write
down the grand partition function Ξ and show how it may be expressed as the
product of Ξk , the grand partition function for the subsystem comprising particles
in the k th single-particle state.

Solution: Reviewing the discussions on the Gibbs factor in Section 1.3.5 and the grand
partition function in Section 1.4.6 of the book. The system can exchange both
energy and particles with the rest of the world. The probability of find the system
has N particles and is in jth microstate is given by Gibbs factor

e−[EN,j (N,V )−µN ]/kT

PN,j (V, T, µ) = ,
Ξ(V, T, µ)

where Ξ is the grand partition function given by

X
Ξ(V, T, µ) = e−[EN,j (N,V )−µN ]/kT .
N,j

For systems of non-interacting identical particles, the microstates can be expressed

as
{n1 , n2 , . . . , nk , . . . },

3
 Homework 1 - Statistical Mechanics

where ni is the occupied number of ith single-particle state. In this way, we can
rewrite the grand partition function as

e−[ k nk εk −µN ]/kT

X P
Ξ(V, T, µ) =
microstates of sys.
X Y
= e−[nk εk −µnk ]/kT
microstates of sys. k
YX
= e−[nk εk −µnk ]/kT
k nk
Y
= Ξk (V, T, µ).
k

The first two equal signs are obvious, the third one is easy to understand as
the ”microstates of sys.” just means {n1 , n2 , . . . , nk , . . . }, which provides every
single-particle state with energy εk a number nk that varies form 0 to N , and thus
X Y
Ξ(V, T, µ) = e−[nk εk −µnk ]/kT
{εk ; nk } k
YX
= e−[nk εk −µnk ]/kT .
k nk
Q P
The k can be understood as ”for each εk ” now, the nk means in the subsystems
only have single-particle state εk we sum over all allowed occupied number.
As for Ξk (V, T, µ), it’s the grand partition function of the subsystem which share
the volume and temperature with the total system. It only has single-particle states
with energy εk , but the particles number nk can change.
From the view of subsystems, we can give another simple explaination. According
to the Gibbs-Duhem relation (in the Appendix of the book), the grand potential is
−pV = E − T S − µN = −kT ln Ξ,
P
no doubt it’s an extensive quantity, thus pV = k (pV )k , where (pV )k is the grand
potential for the subsystems, and therfore
X
pV = kT ln Ξk
k
Y
= kT ln Ξk
k
= kT ln Ξ.

4. Problem 1.14
For a single-component system with a variable number of particles, the Gibbs
free energy is a function of temperature, pressure and number of particles: G =

4
 Homework 1 - Statistical Mechanics

G(T, p, N ). Since N is the only extensive variable upon which G depends, show
that the chemical potential for this system is equal to the Gibbs free energy per
particle: G = N µ

Solution: For systems with fixed particle numbers, their Helmholtz free energy and
Gibbs free energy are defined as

F = E − TS
G = F + pV.

The chemical potential µ is defined such that the Helmholtz free energy F changes
by µN .
dF = −pdV − SdT + µdN,
and thus
dG = V dp − SdT + µdN,
it should be noted that dN here is actually different from differential in the mathe-
matical sense. In fact, N is not a continuous variable. Generally speaking, we will
define the chemical potential as
   
∂F ∂G
µ= = .
∂N V,T ∂N p,T

From a traditional functional perspective, G(p, T, N ) will increase the corresponding

share of µ as N grows. As N is a discrete variable, this means:
 
∂G
G(p, T, N + 1) = G(T, p, N ) + · 1.
∂N p,T

From the above equation, it’s clear that both F and G are extensive quantities, and
we can conclude that  
∂G
= G(T, p, 1).
∂N p,T
Therefore,
G(p, T, N ) = N · G(T, p, 1) = N µ.

5. Problem 1.16
The energy of a harmonic oscillator may be written as mω 2 x2 /2 + p2 /2m so
it is quadratic in both position and momentum - thus, classically, equipartition
should apply.

The energy levels of the quantum harmonic oscillator are given by
εn = 12 + n h̄ω . Show that the partition function of this system is given by

1 h̄ω
Z = cosech
2 2kT

5
 Homework 1 - Statistical Mechanics

and that the internal energy is given by

1 h̄ω h̄ω h̄ω

E = h̄ω coth = h̄ω/kT + .
2 2kT e −1 2
Show that at high temperatures E may be expanded as

h̄2 ω 2
E = kT + + ···
12kT
Identify the terms in this expansion.

Solution: For single harmonic oscillator system, the partition function is

∞
X
Z= e−εn /kT
n=0
∞
h̄ω
e(− 2 −nh̄ω)/kT
X
=
n=0
∞  n
h̄ω X h̄ω
− 2kT
=e e− kT .
n=0

In mathematics, a typical geometric sequence is

∞
X
1 + x + x2 + x3 + · · · = xn ,
n=0

its radius of convergence is |x| < 1, at this time, this infinite series has a compact
expression
∞
X 1
xn = , (|x| < 1).
1−x
n=0
h̄ω h̄ω
− kT
Now, we identical e h̄ω
= x, as − kT < 0, 0 < e− kT < 1, within the convergence
radius of the series. Therefore, the partition function can be simplified as
h̄ω 1
Z = e− 2kT h̄ω
1 − e− kT
1
= h̄ω h̄ω
e 2kT − e− 2kT
1 h̄ω
= csch .
2 2kT
1
Where ”csch” is the hyperbolic cosecant function defined as cschx = sinh x =
2
ex −e−x
.

6
 Homework 1 - Statistical Mechanics

The intenel energy of the sysytem can be derived from F = E − T S = −kT ln Z.

∂ ln Z
E = F + T S = kT 2
∂T V,N
h̄ω h̄ω
= coth
2  2kT 
h̄ω 2
= h̄ω +1
2 e− kT −1
h̄ω h̄ω
= + h̄ω .
2 e− kT − 1
To get the expansion of E with the high temperature limit, we discuss the expansion
h̄ω h̄ω
of coth 2kT . Taking y = 2kT from now on. When we take the high temperature limit,
saies T → ∞, the variable y → 0. The Taylor series expansion of coth(y) at the
neighbor of y = 0 is

y y3 2x5
coth(y) = y −1 + − + + ··· .
3 45 945
Thus, the intenel energy is
(  3 )
h̄ω h̄ω 1 h̄ω
E = kT + − + ··· .
2 6kT 45 2kT