STATISTICAL THERMODYNAMICS

(Module - 2: Maxwell – Boltzmann Statistics)

(M.Sc. Sem – II/ Chemistry, Paper III, Unit – II)

v v v

v v v
Statistical Thermodynamics
 Contents:
• Recapitulation of last lecture
– Statistical Thermodynamics
– System, Assembly & Ensembles
– Essential Condition for Distribution
– Macrostate and Microstate
– Principle of equal a priori probabilities
– Thermodynamic Probability
– Most Probable Distribution
– Stirling's Approximation

• Maxwell – Boltzmann Statistics S Mathew

 v

v
STATISTICAL THERMODYNAMICS

v
Classical Thermodynamics deals with the macroscopic matter
and describes the behaviour of large no. of molecules in terms

v
of properties like P, V, T.

Quantum Mechanics deals with matter at microscopic level and

v
the state is described by a wave function.

v
Statistical Mechanics provides the link between the microscopic
properties of matter and its bulk/ macroscopic properties.

Statistical Thermodynamics thus acts as a bridge between

Microscopic world & Macroscopic world
Quantum Mechanics and Classical Thermodynamics

S Mathew
 SYSTEM, ASSEMBLY & ENSEMBLES

• System: If there is a collection of particles, each single particle is referred to as

System

• Assembly: Collection of systems (particles) as a whole forms Assembly

• Ensemble: Collection of very large number of assemblies which are independent of
each other but which have been made macroscopically as identical as possible is
known as Ensemble introduced by W. Gibbs
Ensembles

Micro-Canonical Canonical Grand Canonical

Ensemble Ensemble Ensemble

✓ Same E , V, N ✓ Same T , V, N ✓ Same T , V, 

✓ Rigid, Impermeable, ✓ Rigid, Impermeable, ✓ Rigid, Permeable,
Insulated Walls Conducting Walls Conducting Walls
S Mathew
MACROSTATE , MICROSTATE , THERMODYNAMIC PROBABILITY & DEGENERACY
MACROSTATE
The number of ways of obtaining a distribution by just specifying the
number of molecules (or phase points) in each energy state (or cell of
phase space)
MICROSTATE
The number of possible ways a particular distribution can be achieved
if molecules (or phase points) are distinguishable i.e. by specifying to
which energy state (or cell of phase space) molecules (or phase points)
temporarily belongs
THERMODYNAMIC PROBABILITY
The number of microstates corresponding to a particular macrostate
and is represented by W
DEGREE OF DEGENERACY
The number of energy states of an energy level having the same energy,
represented by Statistical Weight Factor , gi
S Mathew
 IMPORTANT CONCEPTS & FORMULAS

ESSENTIAL CONDITIONS FOR DISTRIBUTION OF MOLECULAR STATES

Ni = N and  Nii = E (N & E must remain constant)

PRINCIPLE OF EQUAL A PRIORI PROBABILITIES

All possibilities for the distribution of energy are equally probable

THERMODYNAMIC PROBABILITY
W = N! / N1! N2!.... = N! /  Ni!

STERLING'S APPROXIMATION
ln N!  N ln N – N

MOST PROBABLE DISTRIBUTION

W (N) /  Ni = 0 or  ln W (N) /  Ni = 0
S Mathew
 Results from
Quantum Mechanics
(Property of individual Microscopic matter)

Average property using Probability theory Different

& Statistical methods from

STATISTICS
Statistical Mechanics (Bulk matter)

Applied to
Classical Thermodynamics
Nature of microscopic
(Bulk/ Macroscopic System) Treatment constituents
& Occupancy
Gives Property of STATISTICAL
Bulk/Macroscopic System THERMODYNAMICS
S Mathew
 PARTICLE STATISTICS
CLASSICAL QUANTUM
STATISTICS STATISTICS
Internal structure ignored Internal structure Particles
Particles distinguishable considered indistinguishable

MAXWELL – BOSE – FERMI –DIRAC

BOLTZMANN EINSTEIN
STATISTICS STATISTICS STATISTICS

Restriction
No restriction
on
on occupancy
occupancy
Integral Spin Half Integral Spin
S Mathew
• No. of ways of arranging N distinguishable particles W = N! PROBABILITY
DISTRIBUTIONS
• No. of ways in which n distinguishable particles can be selected from N
distinguishable particles is W = N!/(N-n)! n!

• No. of ways of arranging N distinguishable particles into groups without restriction

– Boltzmannons eg: Gas molecules
W = N!/N1! N2!.... (Boltzmann distribution)

• No. of ways in which N indistinguishable particles can be arranged in g states

(g>>N) with not more than one particle in each state – Fermions (those with half
integral spin) eg: Electron
W = g!/(g-N)! N! (Fermi –Dirac distribution)

• No. of ways in which N indistinguishable particles can be arranged in g states

(g>>N) with no restriction – Bosons (those with integral spin) eg: Photons
W = (g + N -1)!/(g-1)! N! (Bose-Einstein distribution) S Mathew
 MAXWELL – BOLTZMANN DISTRIBUTION

• Suppose we have a bulk/ macroscopic system with the following conditions:

– Consist of N distinguishable particles with total energy E at temperature T
– Isolated system with no interaction between particles
– No restriction on Occupancy of energy levels
– Total number of particles and energy must remain conserved
Ni = N and  Nii = E (N & E must remain constant)
 Ni = 0 and  i Ni = 0
----------- 3 ------------ 3 ----------- 3
----------- 2 ------------ 2 ------------ 2
----------- 1 ----------- 1 ------------ 1
----------- 0 ----------- 0 ------------ 0
• Such particles are called Boltzmannons e.g. System composed of gas
S Mathew
• Most Probable Macrostate ???
– The one with maximum no. of microstates - gives Maximum Thermodynamic
Probability (Wmax)
– W (N) /  Ni = 0 or  ln W (N) /  Ni = 0

• Consider the distribution of Total energy E among various energy levels -

0, 1, 2 ……………… of N distinguishable particles at temperature T

• Total number of particles N and Total energy E remains constant

Ni = N and  Nii = E …….. (1)

• N0 particles are present in level with 0 energy level, N1 in level with 1 energy, N2 in
level with 2 energy, …..

• No. of ways of achieving : N!

W = ………. (2)
N 0! N 1! N 2!.....
S Mathew
• Taking logarithm on both sides of eqn. 2
W = N! / N0! N1! N2!....
ln W = ln N! - (ln N0! + ln N1! + ln N2! + ……)
= ln N! - ln Ni! .......................... (3)
• According to Stirling’s Approximation
ln N ! = N ln N – N ........................ (4)
ln Ni! = Ni ln Ni – Ni .
 ln Ni! =  Ni ln Ni – Ni (But Ni = N)
=  Ni ln Ni – N ..........................(5)
• Substituting eqn. 4 & eqn. 5 in eqn. 3, we get
ln W = N ln N – N –  Ni ln Ni + N
ln W = N ln N – N –  Ni ln Ni + N
ln W = N ln N –  Ni ln Ni .......................... (6) S Mathew
• Most Probable Distribution of particles
– The one for which W is maximum (Wmax)
– Condition for maxima: W and  ln W will have to be zero
W =  ln W = 0 ................... (7)
• On differentiating eqn. 6, ln W = N ln N –  Ni ln Ni results in
 ln W = N ln N –  Ni ln Ni
 ln W = –  Ni ln Ni ................. (8) [N ln N = 0 , (constant) = 0]
• Putting the condition of eqn. 7
 ln W = –  Ni ln Ni = 0 ................... (9)
= – [ Ni ln Ni +  ln Ni  Ni ] = 0
1
= – [ Ni 1 Ni +  ln Ni  Ni ] = 0 where  ln Ni = Ni
Ni Ni
= – [  Ni +  ln Ni  Ni ] = 0 [  Ni = 0]
= – [  ln Ni  Ni ] = 0 ...................... (10)
S Mathew
• Distribution must satisfy the condition: N & E must remain constant,
– N & E must be equal to zero
N =  Ni = 0 ........................ (11)
E =  i Ni = 0 .......................... (12)

• Using Lagrange's Method of Undetermined Multipliers

– multiplying eqn. 11 by  and eqn. 12 by 
 N =  Ni = 0 ......................... (13)
 E =   i Ni = 0 ......................... (14)
– and adding to eqn. 10, – [  ln Ni  Ni ] = 0, we get

( +  i - ln Ni ) Ni = 0 .......................... (15)

• N0, N1, N2, N3,…….Ni are independent of each other, so each term in
summation must be zero Ni  0
 +  i - ln Ni = 0 .......................... (16)
ln Ni = ( +  i ) ........................... (17)

S Mathew
• Removing logarithm from eqn. 17
ln Ni = ( +  i ) ............................ (17)
N = e ( + i )
i where  = - 1/kT [k = Boltzmann constant]

Ni = e e-i/kT ............................. (18)

• For getting General Distribution Law, Degeneracy of energy states has to be incorporated.
For this a Statistical Weight factor gi is introduced for each energy level i

............................ (19)
Ni = gi e e-i/kT

This equation is known as Maxwell – Boltzmann Distribution

S Mathew
• Maxwell – Boltzmann Distribution Law is given by
Ni = gi e e -i/kT .......................... (19)

• Taking summation over all energy levels from zeroth level (For all the particles present
in the system)
Ni = gi e e -i/kT [Ni = N ]
N = gi e e -i/kT .......................... (20)

• Dividing eqn. 19 by eqn. 20, we get

Ni gi e e -i/kT
............................ (21)
N  g e e -i/kT
i

S Mathew
 This gives the general form of Maxwell – Boltzmann Distribution Law

Ni gi e -i/kT
.......................... (22)
N  g e -i/kT
i

• Fraction of total molecules in the ith energy level

• The quantity in the denominator is represented by Q and is called Partition

Function

Q = gi e i/kT
- .......................... (23)

• It is a great analytical tool in Statistical Thermodynamics

S Mathew