Rutgers University - Physics Graduate Qualifying Exam Thermal Physics - August 29, 2008

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Rutgers University Physics Graduate Qualifying Exam

Thermal Physics August 29, 2008

Work problems A and B and (C1 or C2) and (D1 or D2).


Work each problem in a separate blue book, labeled with your code number and the
problem number.
Each problem is worth a total of 10 points.

TP - A
Let H be the enthalpy and let it be dependent on pressure P and temperature T: H = H(P,T).

a) [4 points] How is H connected to:


i) the internal energy U and the pressure P and volume V.
ii) the Gibbs free energy G, the entropy S, and temperature T.
iii) the S, T, the chemical potential , and number of particles N

H
to the heat capacity at constant pressure P.
b) [3 points] What is the relation
T P
H
What is the connection between and V, T and the volume expansion coefficient
P T
1 V
. (The number of particles N is also held constant.)
V T P

c) [3 points] Consider a modified ideal gas equation of state that reads: P(V - bN) = NkT

where b is a constant and k is Boltzmanns constant. What is the change in H in an isothermal


expansion (at temperature T1 ) from volume V1 to volume V2 .

TP - B
Consider a system which has energy levels n , with n = 0,1,2,3..... . Six identical spin 1 particles
with mass m are placed in these levels so that the total energy is 6 .

a) [2 points] List all macrostates or configurations of this system.

b) [4 points] For each configuration give the counting factor or statistical weight associated with
it.

c) [4 points] Find the mean number of particles in each energy level assuming the system is in
equilibrium.

Page 1 of 3
TP - C1
(a) [5 points] Show by using the density of states that for a photon gas, the pressure is:
P = U/3V.

(b) [5 points] Using thermodynamic arguments (First and Second Laws), and the above
relationship between pressure and energy density, obtain the dependence of the
energy density on the temperature in a photon gas

TP - C2
We want to understand the behavior of non-ideal gases using the canonical ensemble. Let us
assume that molecules in the gas interact via inter-molecular interactions V(ri - rj), where ri
denotes a 3D radius-vector. There are no intra-molecular interactions. Each molecule has mass m
and occupies volume vc which is inaccessible to all other particles. Approximate the interactions
between the molecules by assuming that each particle feels only the average gas density :

1
U pot = V (ri r j ) Nu pot (v)
2 i j

where v = V/N = 1/ (V is the gas volume, and N is the total number of particles). This is known
as a molecular field approximation.

Calculate the Helmholtz free energy per particle f(T, v) using a canonical ensemble. Assume that
vc << V and N >> 1. Calculate the internal energy and find the equation of state as first
derivatives of f(T, v).

Hint: use Stirling's approximation: log(N!) _~ Nlog(N) N + log(2N) .

Page 2 of 3
TP - D1
A cubic box (with infinitely hard walls) of volume, V = L3, contains an ideal gas of N identical
atoms, each of which has spin S and magnetic moment . A magnetic field, B, is applied to the
system.

(a) [3 points] Compute the partition function for this system.

(b) [3 points] Compute the internal energy and heat capacity.

(c) [4 points] What is the magnetization?

TP - D2
A vessel of volume V is filled initially with N non-interacting molecules of type X. U is the total
energy of the molecules. The molecules of type X can transform spontaneously into molecules of
type Y: X Y. This transformation does not require any energy.

a) [4 points] The total number of microscopic states of this gas, (V, N, U), is given by:

(V , N , U ) Ns tot
exp M
o (V , N , U ) kB

Find the exact value of s mtot by summing over all states with different numbers of particles of
types X and Y. Show that in the limit of large N

Ns tot 1 Ns ( x )
exp M N dx exp M
kB
0
kB

where x = NX/N is the fraction of molecules of type X, is the entropy of gas when all
molecules are in state X, and sM(x) is the mixing entropy per particle:

sM(x) = -kB[xlogx + (1-x)log(1-x)]

b) [3 points] Find the value of x for which sM(x) is at its maximum. Use it to find a zero-order
approximation to s mtot . Compare with the exact result from part (a) and explain the
differences (if any).

c) [3 points] Approximate the above integral for (V, N, U) by the saddle-point method (i.e.,
compute the Gaussian integral contribution from the second derivative of sM(x) evaluated at
the maximum). Calculate the total mixing entropy s mtot of the gas under the saddle-point
approximation. What is the order of the contributions neglected in part (b)?


e x dx = 1
2
1
Useful formula:

Page 3 of 3

You might also like