56 Introduction to Statistical Mechanics

Chapter 5

Introduction to Statistical Mechanics

Thermodynamics limit :
If N   V
N
Particle density n  is fixed at a preassigned value.
V
Extensive property : In the thermodynamic limit the properties that become directly proportional to the size of
the system ( or proportional to the extensive parameters A, N, V, S, E) i.e N  N , V  V S  S then
such properties are called extensive properties. They have an additive nature.
Intensive property: In the thermodynamic limit the properties that become independent of the size of the
system.
Example: T, P,  (chemical potential)
Accessible microstate : The microstate which are permitted under the constraints imposed upon the system
are called accessible microstate.
Thermodynamic Probability : For a given macrostate the total number of microstate is called thermodynamic
Probability of that microstate.
S  kB log  or S  kB log 
where or   Total no. of microstate for a given macrostate
(thermodynamic Probability)
Fundamental Postulate of statistical mechanics :
(equal a priori a probability) :
For a system in equilibrium all accessible microstate corresponding to a given macrostate are equally probable
i.e. probability of finding the system in all microstate for a given macrostate are equal
S  k B log   Boltzman entropy probability relation
This eqn. shows the relation between macroscopic & microscopic properties or relation between thermody-
namics and statistical mechanics
Random Walk: Let n1 denote the number of steps to the right and n2 the corresponding number of steps to the
left.

Total number of steps N = n1 + n2

Net displacement m = n1 – n2 = n1 – (N – n1) = 2n1 – N (towards right)
If P = Probability that the step is to the right, q = 1 – p = Probability that step is to the left.
Probability WN(n1) of taking (in a total of N steps) n1 steps to the right and n2 steps to the left is given
N ! n1 n2
by WN (n1 )  p q .
n1 ! n2 !
Introduction to Statistical Mechanics 57
Mean Free Path: Mean free path  is the average distance traversed by molecule between collisions.
Mean free path of a molecule is related to its size, larger its size, shorter its mean free path.
1

2 Nd 2
where d  diameter of the gas molecule, N  Number of molecules per unit volume.
Q. 54. A system of N particles has only two allowed state A and B. The probability for A is P and for B is
1 – P? What is the probability for the system to be in macrostate defined by the distribution of
(r, N – r)?
Ans. The probability of finding r particle in state A = Pr
The probability of finding N – r particle in state B = (1–P)N–r
The Total no. of ways in which r particle can be choosen from N – particle is N Cr  N!r!(N– r)!

The probability in which r particle are in state A and N – r particle in state B is =

Q. 55. A one dimensional random walker takes step to left or right with equal probability. The probability
that the random walker starting from orign is back to orign after N even number of step, is
N
N! 1 N! 2N N
(i)   (ii) 1 1
N N
    2 N N (iii) 2N!   (iv) N!  
!
   !   !  ! 2 2
22 22
N! N–r
Ans. Probability P  P r 1 – P 
r! N – r  !
N N
N– N
N! 12  1 2 N! 1
   1 –    
N N
    2  2 N N
    2
  !  !   ! !
22 22

(i) is correct
Q. 56. Calculate the no. of microstates for a configuration of a system of N distinguishable particles in
which there are n1 particles in a particle state 1 & n2 particle in state 2, n3 particle........ ni particle in
the ith state.
Ans. Total No. of particle = N
No. of microstate for state –1 = N C
n 1
No. of microstate for state –2 = N  n1
C n2

No. of microstate for state ith = N  n1  n 2 .....n i  1 C n

i

So total number of microstate is

N Cn  N – n1 C n 2  N – n1 – n 2 Cn 3  ......... N – n1 – n 2 – n 3 – ni  1 Cn
1 i

N!  N – n !  N – n1 – n 2 ...ni !
   .......
n1 ! N – n1 ! n 2 ! N – n1 – n 2 ! ni !  N – n1 – n 2 ...n i !

N!
 – For distinguishable particle
ni ! N – n1 – n 2 .....n i !

1
  For indistinguishable particle
ni ! N – n1 – n 2 .....n i ! !
 58 Introduction to Statistical Mechanics

Q. 57. Four distinguishable coins are tossed a large no. of time write down the different microstate which
may be observed & the macrostate into which they would fall. Give the probability of the most
probable macrostate.
Microstate coins Microstate coins
Macrostate No. of microstate Pr obability
having head up having tail up
1
n1  4, n 2  0 abcd  1
16
abc d
bcd a 4
n1  3, n 2  1 4
cda b 16
dab c
ab cd
ac bd
ad bc 6
n1  2, n 2  3 6
bc ad 16
bd ac
cd ab
a bcd
b acd 4
n1  1, n 2  3 4
c abd 16
d abc
1
n1  0, n 2  1 – abcd 1
16
Total no. of microstate = 16
The probability of most probable state = 6/16
Q. 58. An isolated system consist of two non-interacting Spin – 1 2 particles a & b fixed in space & kept in
magnetic field B. Find out the total no. of microstates allowed to the system.
System state or Particle state or Magnetic moment energy no. of microstate
Ans. macrostate Microstate
1 U U 2 0 –2 0B 1
2 U D 0 0 }2
D U 0 0
3 D D – 2o + – 2oB 1
1
Total no. of microstate = 4 for Spin  two particle system no. of accessible microstate corresponding to
2
energy
E  0 is 2 .
Introduction to Statistical Mechanics 59
The no. of microstate for N no. of Spin S particle
N
 2 S  1
The probability of getting a macrostate in which there are r particle out of N Spin 1/2 particle in spin
up state

CONTACT BETWEEN STATISTICS & THERMODYNAMICS

Now we consider two physical systems A1 and A2 which are A1 A2

separately in equilibrium. Let the macrostate of E1 be N1 , V1 , E1 N 2 , V2 , E 2
represented by parameter N1 , V1, E1 so that it has possible microstate 1  N1 , V1, E1 
Similarly system A2 by '' '' '' '' N2 ,V2 ,E2 '' '' '' ''  2  N 2 , V2 , E 2 
Now bring the two system into contact with each other through a wall (conducting, rigid & impenetrable) such
that N1 ,N2 and V1 ,V2 remain fixed but energies E1 and E2 become variable such that
E1+E2 = E(0) = constant.
The no. of microstate at any time t of the composite system is represented by

  
  E1, E2   1  E1  2  E2   1  E1  2 E o – E1   E o , E1 
The microstate of composite system is also represented by
o  E1, E 2   1  E1  2  E 2 

 1  E  E2   2  E2 

d  2 E 2 
  2 1  1
d dE 1 E 1 E 2 E1  E2  E 0  E1
For maximum Number of microstate 0  dE
dE1 d1 d 2 2
0  2  1   1
dE1 dE 2  dE1

1  1  1  2 
     ( at equilibrium E1 and E 2  E)
1  E1  N ,V ,E  E  2  E 2  N ,V , E  E
1 1 1 1 2 2 2 2

  n  
Let   
 E  N1,V1 ,E1  E 1

Under equilibrium 1  2
1  S  1 1 1 ... (1)
   so  
k B  E  N,V k B T k BT1 k B T2

This is the condition for thermal equilibrium

If the wall is movable and conducting
1  S  1 P
     
k B  V  k B  T 
60 Introduction to Statistical Mechanics

1 2 P P
under equilibrium 1  2  k T  k T
B 1 B 2

for conducting medium T1  T2  ... (2)

Eqn (2) is the condition for mechanical equilibrium and thermal equilibrium.
If the wall is conducting. non-rigid, movable and penetrable
V1  V2  V o E1  E 2  Eo , N1  N 2  N o (fixed)
1  2  T1  T2
1  2  P1  P2

  
Now, let    n 
 N E,V

1  S  
  –
k B  N E1 kBT

... (3)

1  2  system is in chemical equilibrium

If T1 = T2, P1 = P2 and 1  2 then system is thermodynamic equilibrium.

Classical ideal gas : Now we discuss the thermodynamical properties of a classical ideal gas composed of N
monoatomic (non-interacting) molecule, by statistical method the total no. of ways in which the N- particle can
be distributed in the system will be simply equal to the product of the no. of ways in which the individual
particles can be distributed in the same space independently of one another volume of phase space
4
d   d 3xd 3P   dxdydz  d3P  V  P 3 (6D phase space  3D position and 3D momentum)
3
3
Volume of each phase cell = h ( particle have 3 degree of freedom)
Volume of phase space V 4
  No of microstate   3 P 3
Volume of phase cell h 3

5 3 3 1
 !  
3 3 2 2 2 2
V P V  P
 
3
So h 3 h3 3
!
4 2

3
V  2ME  2

3 3
h
 2 !
 
No. of microstate composed of N identical particle

 
N 3N
 V    2ME  2 
 E       for sharp energy
 h 3    3N  ! 
  2  
 

If the energy lies between E and E + dE number of microstate

Introduction to Statistical Mechanics 61


  E  E     E   E higher terms neglected
E
N 3N 3N
V  2ME  2 3N –1
2 dE
  E
 h3   3N  2
 2 !
 

 
3N
  V N  2ME 2 
Now entropy S  kB log   kB n   
  h3   3N  
  ! 
  2  

  V N 3N  3N  
S  k B n    2ME  2 – n  !
  h3   2  

  V N 3N  3N 3N 3N 
 k B n    2ME  2 –  n – 
  h3   2 2 2 

(log n! = n log n – n ) = Stirliing relation

 N 3N 
  V N 3N 3N 3N 3N  V 3N  3N  2 3N 
S  k B n    2ME  2 – n    k B n    2ME  2 – n   
  h3  2 2 2    h3   2  2 
  

  3N
2
 
  N
   
V   2ME     3N 
 k B n    
 3  2 
  h   3N   
   2  


3
 V    4ME   2
3
S  N k B n  3     Nk B ... (1)
 h   3N  2

Equation (1) can also be written as

3
V  4ME  2 3
S  Nk B n 3  Nk B n    Nk B
h  3N  2

V 3  3 
Nk B n  Nk B 1  n  2Mk BT    E  Nk BT 
h3 2  2 
3 3
Nk B n V – NkB n  h 2  2
 NkB 1  n  2Mk BT  
2
3 3
Nk B n V – NkB n h 2  Nk B 1  n  2Mk BT  
2 2

3  2Mk BT 
S  Nk BnV  Nk B 1  n ... (2)
2  h 2 
62 Introduction to Statistical Mechanics
From equation (1)
3h 2 N  2S 
E  N V S  exp  –1
2
4mV 3  3Nk B 

dQ  TdS  dU  PdV – dN (U  E  internal energy  P.E  0

 E  3h 2 N  2S  2
T   exp  –1
 S  NV 2  3Nk B  3Nk B
4MV 3

2
TE 
3Nk B

3 3
E NkBT  RT
2 2

Internal energy of an classical ideal gas U  3 kBT specific heat

2

 Q  U  3
Specification CV   T     Nk B
V  T  V 2

Pressure : TdS = dE + PdV – dN

 E  2 E
P–   So
 V  N, S 3 V

2 3
PV   Nk BT  Nk BT PV  Nk BT  RT  (Ideal gas equation)
3 2

 E  PdV   E  PdV  VdP 

CP       P  constant dP  0 so we can add 
 T P  T P
 dP  0

    H 
CP    E  PV     
 T  P  T  P

 3
 E  2 NkBT
  3   5 5
CP    NkBT  NkBT   NkBT  NkB
 T  2  P T 2 2 PV  NkBT from

5
CP  NkB
2
5
Nk
CP 2 B 5 5
Now,      
CV 3 Nk 3 3
B
2
For adiabatic process PV   constant

PV5/3 = const
Introduction to Statistical Mechanics 63

2 E E
P  energydensity dQ = dE + PdV
3 V V
For adiabatic proces dQ = 0
3  3 3
0  dE  PdV  d  PV   PdV , 0 PdV  VdP  PdV
2  2 2

3 5 dP 5 dV
– VdP  PdV , – .
2 2 P 3 V
Integrating both side
5 5
n P  – nV  c   nV 3  c
3
n P  n V 5/3  c , n PV 5/3  c
5
3
PV  constant
 dE  adiabatic  – PdV ( dQ  0

2 E  V  4ME  3 2  3
– dV S  Nk B n  3     Nk B
3 V  h  3N   2
Replace N   N, V  V and E  E and if S  S
Then S is an extensive property
  V  4 M  E  3 2  3
  N  k B n  3     N  kB   S
 h  3Nk B   2
 
Since entropy doesn’t satisfy the extensive property
(additive). So we have to modify
Entropy of mixing : The entropy of mixing is defined as
N
S  ST –  Si , ST = Entropy after mixing, Si = Entropy before mixing of individual system
i 1

The entropy of an ideal gas given by

3   2mk BT  
S  NkB n V  Nk B 1  n  2 
2   h 
Entropy is not an extensive quantity of the system i.e if we increase the extensive parameter by a factor  ,
keeping the intensive variable unchanged then the entropy of the system which should also increase by same
factor  doesn’t increase. The entropy of an ideal gas is not extensive due to the term nV. This means that
entropy of the system is different from the sum of the entropies of its parts. This is called Gibbs paradox.
Gibbs visualised the mixing of two ideal gas (1) and (2) both being initially at the same temperature T. The
temperature remain same after mixing and before mixing
2
3   2mi k BT 
N i k BnVi  Ni k B 1  n  2 
i 1 2   h 

3   2m1k BT  3   2m2 k BT 
 N1k B nV1  N1k B 1  n  h 2   N2 k B n V2  N2 k B 1  n  
2    2   h2 
64 Introduction to Statistical Mechanics
Entropy after mixing  N = N1 + N2 V = V1 + V2

3   2mk BT 
ST   N1  N2  k B n  V1  V2    N1  N2  k B 1  n  h 2 
2   
Here m1 = m2 = m
Entropy of mixing
2
3   2mk BT  
S  ST –  Si   N1  N 2  k Bn  V1  V2    N1  N2  k B 1  n  h 2  
i 1 2   

3   2mk BT  3   2mk BT  
– N1k B n V1 – N1k B 1  n  2  – N2 k B n V2 – N2k B 1  n  h 2  
2   h  2   

  N1  N 2  k B n  V1  V2  – N1 k B n V1 – N 2 k B n V2

  N1 k B n  V1  V2  – N1 k B n V1    N2 k B n  V1  V2  – N2 k B n V2 

V1  V2 V  V2
S  N1k B n  N2 k B n 1
V1 V2

V1  V2 V  V2
Since  1 or 1 1
V1 V2

So S  0
N1 N 2 N  N2
If, initial density = density of mixture i.e.   1
V1 V2 V1  V2
 N  N2 N  N2 
S  k B  N1 n 1  N 2 n 1 
 N1 N2 
S   0
If we consider the mixing of two same gas with a common initial temperature T and a common initial particle
density n (reversible process) we get
S   0
But for reversible process S   0
Thus once again we are led to believe that there is something basically wrong with that expression. To avoid the
paradoxial sit uation Gibbs diminished ent ropy(S) by an another term
k B n N!, S1 by k B n N1 ! and S2 by k B n N2 !
3   2mk BT  
S  Nk B n V  Nk B 1  n  2   – k B n N!
2   h 
3   2mk BT 
 Nk B nV  Nk B 1  n  2  – NkB n N  Nk B
2   h 

V 3 5  2mk BT  
S  Nk B n  Nk B   n  2    Sackur Tetrode eqn.
N 2 2  h 
By replacing E by E, V by V and N by N we get S instead of S. So, S is an extensity property
Introduction to Statistical Mechanics 65
Chemical Potential :
 S  N  h2 32 
 E  2S 
  E – 2   k BT n    
 N V, S  3N 3N k B   V  2mk BT  
 
Gibb’s free energy :
3
 N  h2  2
G  N  Nk BT n     G = E + PV – TS, A = E – TS = G – PV
 V  2 mk BT  
 
Helmholtz free energy :
3
  2 2 
  N  h   
So A  Nk BT n     –1
  V  2mk BT   
   
A is an extensive property while  is an intensive property of the system.
Q. 59. We have equal amount of two identical ideal gases at the same temperature T but at different
pressure P1 and P2 in two containers of volume V1 and V2 respectively. The containers are con-
nected. Find the change in entropy ( The temperature after mixing also remains same)

V 3 5  2mk BT  
Ans. S  Nk B n    Nk B   n  2 
 N 2 3  h 

2
Vi 3 5  2mk BT  
Si  N k1 B n  N1k B  3  n  h 2  
i 1 N 2   

V1 3 5  2mk BT 
 Nk B n  Nk B   n  2  
N 2 3  h 
V2 3 5  2mk BT 
 Nk B n  Nk B   n  2 
N 2 3  h 
 V1 k BT 
P1V1  Nk BT  N  P 
1
 V2 k BT 
P2 V2  Nk BT   
 N P2 
2
k BT 3 5  2mkBT 
Si  Nk B n  Nk B   n  2  
i 1 P1 2 3  h 

kBT 3 5  2mkBT 
NkB n  NkB  3  n  h2 
P2 2   
Total entropy :
V1  V2 3 5  2mk BT  
ST  2Nk B n   2Nk B   n  2 
2N 2 3  h 
Change in entropy
2 V1  V2 k T k T
S  ST –  Si  2Nk B n – Nk B n B – NK n B
i 1 2N P1 P2
Let after mixing pressure = P and after mixing total No of particle = N + N = 2N
P  V1  V2   2Nk B T P  V1  V2   2Nk B T
 66 Introduction to Statistical Mechanics

 V1  V2   kBT 1 V1  V2
 
V1

V2
2N P P 2Nk B T 2Nk B T 2Nk B T
kBT kT kT 1 11 1
S  2NkB n – NkB n B – NkB n B    
P P1 P2 P 2  P1 P2 
2
 k T kT kT 2P1P2
 NkB n  B  – Nk B n B – NkB n B P
 P  P1 P2 P1  P2
2
 kBT 
 
 NkB n  P 
 kBT  kBT 
  
 P1  P2 
  kBT2  P1  P2 2 P1P2 
 NkB n  2
 2
  2 P1P2   kBT 
2

S  Nk B
P  P 
n 1 2
4 P1P2
2

S  Nk B
P  P 
n 1 2 Here we get S  0
4 P1P2
( We are considering identical particle. So it is a reversible case only when the particle density is same otherwise
for identical gas also we will get  S  0 ).

PHASE SPACE

Phase Space: d  dx dydz dPx dPy dPz

 dxdPx dydPy dzdPz  h.h.h  h 3

Volume of phase cell  h3N  h f

Dimension of phase space = 2f (f degree of freedom). If f is the degree of freedom of a single particle and N is
the total no. of particle, then dimension of phase space is = 2Nf
Total Volume of Phase Space allowed to the system
No. of microstate in phase space =
Volume of phase Cell (h 3 )

Q. 60. Calculate the number of microstate for a free particle in 3-dimension that have momentum p.
Ans. d  dx dy dz dp x dp y dpz

Allowed Phase Space Volume =  d   dxdydz  dp x dp y dp z

 2 4
 V  p 2dp  sin dq  d  V p3
0 0 3
V 4 3
No. of microstate   p
h3 3
Note : for 2. dimension.
Introduction to Statistical Mechanics 67

 dx   dx  dy  dp x  dp x
  d 2q . d 2 p  A.p 2 {whereA  area}
Since wave polarize in 2 dimension
2
 dr  2Ap
If momentum lies between p and p + dp
2A 4A
 No. of microstate = 2 2 pdp  pdp
h h2
V
If momentum lies between p and p + dp then no. of microstate = 4 p2dp
h3

no. of state no. of state

Density of state = Volume or energy int erval

PHASE – SPACE DIAGRAM

Problems that appeared in GATE from this section :

G-1. The phase diagram of a free particle of mass M and K.E, E moving in one-dimensional box with
perfectly elastic walls at x = 0 and x = L is given by [GATE-2006]

Px
2mE Px

L
(a) (b) O L

 2mE

Px Px
2mE 2mE

O
(c) O L (d) – L L

Soln. For free particle momentum Px is doubly degenerate. For each energy E
P2
 E  P   2mE
2M
Which is independent of x but x can only be +ve between 0 to L.
Hence correct option is (a).
 68 Introduction to Statistical Mechanics
• If there would be some range for E from E to E + dE instead of sharp values.

P P2
E  P2  2mE  P  2mE
+dP 2m
1
dP  2m dE
x 2 2mE
mdE
–dP dP 
2mE

• If the origin of phase space is the origin of box

Px

L
2

For the above problem we now calculate the no. of phase cell or accessible microstate. If E lies between E
and E + dE
As it is 1 dimension case the area of phasecell in phase space is
x Px  h
2L dPx  this is area of region accessible to free space

1 1
2L d  
2mE  2L 
2mE
2m dE  2m L E 2 dE

Density of state g(E)= number of states per unit energy range

1 1
L  2M  2 E 2

h
G-2. The dimension of phase space of ten rigid diatomic molecules is:
(a) 5 (b) 10 (c) 50 (d) 100
[GATE-2004]
Soln. A rigid diatomic molecule  degree of function = 5 so, each molecule will require total 5 coordinate variables
and 5 momentum variables for their complete description (3-translational, 2-rotational). Hence, each molecule
requires 10 phase space dimensions.
 Total dimension of 10 such molecules in the phase space = 10 × 10 = 100.
Hence correct option is (d).
(5 degree of freedom for a rigid diatomic molecule can also be understood by the fact that there are 2 particles
each with 3 degree of function but there is one constraint that being rigid, the distance between them is fixed,
therefore, total degree of function = 3 + 3 – 1 = 5).
Introduction to Statistical Mechanics 69

PHASE SPACE DIAGRAM OF AN OSCILLATOR

Energy of a one dimensional harmonic oscillator whose position co-ordinate is x and momentum co-ordinate is
Px is given by
Px 2 1
E  Kx 2 K  m2  = angular frequency
2m 2
Where the 1st term is the K.E and the 2nd term is P.E. What is the trajectory of oscillator in phase space?

Q. 61. Find the accessible region & the no. of accessible microstate of oscillator?
Px
P2 1 Px 2 Kx 2
Ans. E  x  Kx 2 or  1
2m 2 2mE 2E

Px 2 x2 x
 1
2 2
 2mE   2E 
  ... (1)
 K 

For constant energy E eqn.(1) describes an ellipse in phase space i.e. x – px plane
2E
The phase points are those lying on the elliptical path having a semimajor axis 2mE and semiminor axis
K
At any particular instant the phase space of the oscillator is represented by some point on the ellipse.
The area of the ellipse in phase space is ab where a is the semimajor axis and b is the semiminor axis.
2E m 2 E
A  ab   2mE  2 E 
K K 
It represents the phase space available to the oscillator having energy between o & E or this is the accessible
region.
Area of accessible region
Now the accessible no. of microstates 
Area of a po int in phase space (i.e.cell)
2E E
 
 h 
The phase space available to the oscillator having energy between E and E + dE is

m 2
dA  2 dE  dE
K 
Q. 62. Calculate the number of microstates accessible to the photon having frequency between  and  + d
confined to a 3 dimentional cavity of volume V.
Ans. No. of microstate in frequency range 0 to 
V 4 3 P2
  3
  2mE  2 E  P  2mE
h 3 2m
3
V 4 E h

h3 3
 P2 2  P 
C C
70 Introduction to Statistical Mechanics

3
V 4 3 V 4  h  4 V3
  3  P  3   
h 3 h 3  c  3 c3
No. of microstates in the frequency range  and  + d  is
4 V 2 4V
d  3 d  3 2 d
3 c3 c
Density of state
g(E) = no. of states per unit energy range
V 4 3
we have   3 3
  2mE  2
h

V 4 3 1
d  .   2mE  2 2m dE
3 3 2
h

V 3 1
 3
.2  2m  2 E 2 dE
h

 P2 P2 E2 E
E   2m    
 2m E c E c 2 
2

3
V E 2 1
 2   E 2 dE
3
h  c2 

V 2
 3 3
E2d E
h c

d 2VE 2
g(E)   .
dE h 3c 3