Statistical Mechanics

“Statistical Mechanics is branch of science which

establishes the relation between the macroscopic
behavior (bulk properties )of the system in terms
of its microscopic behavior (individual properties).
So , Statistical Mechanics, as its the name
implies ,is not concerned with the actual motion
of particles of the system but investigates the
most probable behavior or Statistical properties
of the assembly of particles”.
The Subject which deals with the relationship
between the overall behavior of the system
and the properties of the particles is called
Statistical Mechanics .
 Thestudy of statistical mechanics can be
classified mainly in the following in the two
categories.

statistical
mechanics

Quantum
Classical
Maxwell-Boltzmann Fermi-Dirac,Bose-Einstein
 Statistical Mechanics

Boltzmann hypothesis: the entropy of a system is related to

the probability of its state; the basis of entropy is statistical.

Statistics + Mechanics

Statistical Mechanics

Thermal Properties
 Statistical Mechanics – ideas
A quantum state, or microstate
• A unique configuration.
• To know that it is unique, we must specify it as
completely as possible...
e.g. Determine: Position
Momentum
Energy
Spin

............
of every particle, all at once!!!!!
THIS IS ACTUALLY IMPOSSIBLE FOR ANY REAL SYSTEM
 Macrostates and Microstates of systems
• Any state of a system as described by actual or
hypothetical observations of its Macroscopic statistical
properties is known as ‘ Macro state ’ and it is specified by ‘ (
N, V and E ) ’.

• The state of a system as specified by the actual properties

of each individual, elemental components and it is
permitted by the uncertainty principle is known as ‘
Microstate ’.
 quantum states (microstates)
Volume V 10 particles, 36 cells

10
 1 
pi =  
 36 
≈ 3 × 10−16

Cell volume, ∆ V
quantum states (microstates)
quantum states (microstates)
quantum states (microstates)
quantum states (microstates)
quantum states (microstates)
 quantum states (microstates)
Volume V 10 particles, 36 cells

10
 1 
pi =  
 36 
−16
≈ 3 × 10

Cell volume, ∆ V
 quantum states (microstates)
Many more states look like this, but no more probable than the last one

Volume V
10
 1 
pi =  
 36 
−16
≈ 3 × 10

Cell volume, ∆ V
Phase space

1. The three dimensional space in which the location of a

particle is completely specified by the three position
coordinates, is known as ‘Position space’.

small volume in a Position space dV = dxdydz

2. The three dimensional space in which the momentum of a
particle is completely specified by the three momentum
coordinates Px, Py and Pz is known as ‘Momentum space’.
small volume in a momentum space dΓ = dp x dp y dp z

 The combination of the position space and momentum space

is known as ‘Phase space’.
small volume in a phase space dτ = dVd Γ
 Phase space volume:

 Consider
Let ‘ pm’ be the maximum value of the
momentum of the particles in the system.

 letpx ,py, pz represents the three mutually

perpendicular axes in the momentum space as
shown in figure.

 Draw a sphere with an origin ‘ O ’ as centre and

the maximum momentum ‘ pm’ as radius.

 All
the points within this sphere will have their
momenta lying between ‘ 0 ’ and ‘ pm’.
 pz
The momentum space volume =
volume of the sphere of radius p m .
4 3
momentum volume Γ = πpm pm
3 0
py
phase space volume τ = VΓ
4 3
τ = πpmV .....
3 px
( where V is a position space volume )
Cell in Phase Space

1.The small volume dτ in phase space is called a cell .

2.The total number of cells in phase space =

Total available volume in phase space
volume of one cell

VΓ
3.Total number of cells n =
dτ
Ensemble:
If we have a collection of particle, then
each particle referred as system.Further,the
collection of particle as a whole is called
assemly.Moreover,the collection of a large
number of assemblies is known as Ensemble.

All the members of an ensemble ,which are

identical in feature like volume,energy,total
number of particles are referred to as element.
A large number of such similar,non-
interacting,independent systems when
imagined to be grouped together is called an
ensemble.

 So all members of an ensemble are identical in

features like N,V,E etc.. And are called
elements.

 According ( N,V,E etc..) these parameters

Ensemble further classified three types.
 Ensemble

Micro Canonical Canonical Grand Canonical

 Micro Canonical Ensemble:
The Micro Canonical
ensemble is a collection of essentially
independent assemblies having the same
energy E, Volume V,and the no of systems N
,all the systems are same type.

The individual assemblies are separated by

rigid ,impermeable and well insulated walls
such that the values are E,V,N are not effected
by the presence of other systems
E,V,N E,V,N E,V,N

E,V,N E,V,N E,V,N

E,V,N E,V,N E,V,N

 Canonical ensemble:
The Canonical ensemble is collection
of essentially independent assemblies having
the same temperature T,volume V ,and number
of identical particles N.

In this ensemble all the assemblies have the

same temperature ,we could simply bring all
of the assemblies in thermal contact with each
other.
T,V,N T,V,N T,V,N

T,V,N T,V,N T,V,N

T,V,N T,V,N T,V,N

 Grand Canonical Ensemble:
The Grand Canonical Ensemble is a
collection of essentially independent assemblies
having a same temperature T,volume V,and the aµ
chemical potential .

In Grand Canonical Ensemble we than effectively

have a collection of assemblies. each occupying a
separate volume V,but which can exchange energy
and molecule as well with each other.

The individuals assembly systems are separated by

rigid,permeable,diathermic walls
T ,V , µ T ,V , µ T ,V , µ

T ,V , µ T ,V , µ T ,V , µ

T ,V , µ T ,V , µ T ,V , µ
 Chemical Potential:
 Concept of Probability:

 The Probability of an event may be defined as

the ratio of the number of cases in which the
event occurs to the total number of cases
,provided the total number of cases is very
large
 Probability of Distribution:

 Let us consider the case of a large box which is

divided in k cells of different areas a1,a 2,a 3………ak.
Now suppose that we throw N identical
distinguishable balls in the box in a completely
random manner .

 The experiment is repeated a large no of times it

is observed that particular distribution of balls
among the cells occurs often than any one .
 This particular distribution of balls called as most
probable distribution.
 Probabilities are of two types:
 a priori probability distribution
 thermodynamic probability distribution.

**a priori probability distribution:

A priori probability gi that a ball falls into the ith cell is the
ratio of the total area ai of the cell to the total area A of
entire box i.e
ai
gi =
A
 The Probability of two balls fall simultaneously
In the cell is 2
gi × gi = gi
*** Now a priori probability G of any particular distribution
of N balls among the k cells such that n1 balls fall in
first cell ,n2 balls in second cell,..etc is given by

G = ( g ) .( g ) ,....., ( g i )
n1 n2 ni

condition ∑ ni = n1 + n2 + ......ni = N
WhentheCel lsAreIdent ical
G=g ∑ ( n1 + n2 +.......)

G = gN
 Thermodynamic Probability:
 The Thermodynamic Probability of distribution is the
number of different sequences in which the balls may
be distributed among the cells with out changing the
number in each cell .
 consider the case of n1 balls out of total N balls that
fall in first cell. this can be achieved in the following
way

N!
N Cn1 =
( N − n1 )!n!!
 N!
w=
n1!n2 !......nk
Thetotal Pr obabilityP = G ×W
N!
× ( g1 ) ( g 2 ) .....( g k )
n1 n2 nk

n1!n2 !......nk

P = N !Π
( gi )
k ni

i =1 ni !
 Statistical distribution:

 Statistical mechanics determines the most probable way

of distribution of total energy ‘ E ’ among the ‘ N ’
particles of a system in thermal equilibrium at absolute
temperature ‘ T ’.

 In statistical mechanics one finds the number of ways ‘ W

’ in which the ‘ N ’ number of particles of energy ‘ E ‘ can
be arranged among the available states is given by.

N(E) = g(E) f(E)

 Where ‘ g(E) ’ is the number of states of energy ‘ E ’ and

‘ f(E) ’ is the probability of occupancy of each state of
energy ‘ E ’.
 Distribution
Laws
in

statistical
mechanics

Quantum
Classical
Maxwell-Boltzmann Fermi-Dirac,Bose-Einstei
 MAXWELL-BOLTZMANN STATISTICS:

** Identical but distinguishable

particles:

* Molecules of a gas are the particles of

this kind The particles obey MAXWELL-
BOLTZMANN Distribution Law
** The particles are distinguishable .
Thus each or both the particles can occupy
any one the two cells.
 Consider a system of N similar but
distinguishable molecules are freely within
vessel in all possible directions. at certain
temperature ,the entire gas has a certain
average energy but different gas molecules
have different energies.

 In1854 Maxwell considered that how a fixed

total amount of energy is distributed among the
various member of an assembly. The
expression concerning this idea is known as
M-B.
 Consider a system having n distinguishable particles .

 Let these particles divided into quantum groups such

that n1,………….ni particles lie in groups having
energies
ε1 , ε 2 ..........ε i respectively.

 Let gi be the statistical weight of ith quantum group.

 Condition in Maxwell-Boltzmann:

** Particles are distinguishable.i.e there are no symmetry

restrictions.

** The total number of particles in the entire system is

always constant i.e.,
n = n1 + n2 + n3 .......ni = ∑ ni

** The sum of energies of all the particles in different

quantum groups taken together constitutes the total
energy of the system
**
 In M-B the particles are distinguishable and
distinguishable element are divided into groups such
that n1+n2+n3……..+ni=n.
 suppose first of all we choose n1 particle which are to
be first state this can be done by in ncn1 ways

n!
n1!( n − n1 )!

the remaining particle are (n-n1) and we want to n2

particle in second state which can be done by in
n − n1C ways
n2
 Therefore the number of ways in which particles in
different states may be exchanged ,leaving the total
number in each state fixed ,is

n! (n − n1 )! n!
× × .....
n1!(n − n1 )! n2!(n − n1 − n2 ) n1!n2!n3!...ni !
 consider the ith group , the particles are
accommodated in gi different groups ,the first
particle of ni can be accommodated in any of
the gi groups which can be done by in gi
ways .the second particle can also be
accommodated in gi ways .in this manner all the
ni particles can be occupied ,gi groups or
giniways.

 total no of eigen state for the whole system is

given by
 n1 n2 ni
n! g1 g 2 g i
G=
n1!n2 !.......ni !
n1 n2 ni
g1 g 2 gi
G = n! . ....
n1! n2 ! ni !
ni
gi
G = n!Π
i n!
i

the Pr obablityof thegivenst ateis

n
gi i
ω = n!Π × const
i n!
i

ApplyingLo g
log ω = n log n + ∑{ni log g i − ni log ni !} + const
i

UseStirlin gsApproxim ationLogn != n log n − n

log ω = n log n + ∑{ni log g i − ni log ni } + const
i

log ω = const − ∑{ni log ni − ni log g i }

i

Diff .W .r.t
ni wehave
δ log ω = −∑δ {ni log ni − ni log g i }
i

ni
δ log ω = −∑{δ ni log ni + δni − δni log g i }
i ni
ni
δ log ω = −∑[log + 1]δni
i gi
theConditi onofMaximu mprobabili tyGives
ni
∑i [log g + 1]δni = 0
i

Introducin gTheAuxili aryConditi ons δn = ∑δni = 0 : δE = ∑ ε iδni = 0 :

 Applying the lagrange method of undetermined
multipliers.i.e Multiplying auxiliary conditions one with
α and second with β and adding the we get
ni
∑i {log g + α + β εi }δni = 0
i

TheConditi onOfMaximu m Pr obabilityG ives

ni
{log + α + β εi }δni = 0
gi
δni canBetreat edasArbitr ary
n
log i + α + β εi = 0
gi
n
log i = −(α + β εi )
gi
gi
ni = α + β εi
e
Thisisthem ost Pr obableDist ributionAc cordingtoM − B
 BOSE-EINSTEIN STATISTICS:

 Identical but indistinguishable particles of zero

0r integral spin

 Photon are the particles of this kind. The

particles Bose –Einstein statistics and hence
know as Bose particles (Bosons)
In Bose-Einstein Statistics, the Particles are
indistinguishable and hence lose their
individuality. In this case ,each cell may have
any number of particles
 Consider the case of a box which is divided into
gi sections (cells) .Let ni indistinguishable
particles are to be distributed among these
section. The choice that which of the section
will head sequence can be made in gi ways
.After doing this ,the total number or
permutation of ni particles and remaining (gi-1)
section will be (ni+gi-1).Therefore the total
number of possible ways in which ni particles
can be distributed in gi sections is given by1
gi[(ni+gi-1)]
 The particles are indistinguishable and
therefore rearrangement of particles will not
give rise to any distinguishable arrangement.
There are ni! Permutations which corresponds
to the same configuration .hence term indicated
by (1) should be divided by ni!
 Secondly the distributions which can be derived
from one another by mere permutation of the
cells among themselves, do not produce
different states the eq(1)should also be divided
by gi!
 gi(ni+gi-1)! OR (ni+gi-1)!
gi ! n i ! ni!(gi-1)!
 SimilarExpression will be for the various other
quantum states. Therefore the total number of
ways in which n1 particles can be assigned to
the level with energy ε1, n2 to ε2 ………..and so
on is given by the product of such expression as
given below:

G=(n1+g1-1)! . (n2+g2-1)! … (ni+gi-1)!

n1!(g1-1)! n2(g2-1)! ni(gi-1)!

=π (ni+gi-1)!
ni!(gi-1)!
 According to the postulate of a priori probability .

ω= π(ni+gi-1)! * constant
ni!(gi-1)!
So to obtain the condition of maximum probability

log ω = ∑{log( ni + g i − 1)!− log ni !− log( g i − 1)!} + const

u sin gTheStirli ngsApproxi mation
log ω = ∑{( ni + g i ) log( ni + g i ) − ni log ni − g i log g i ) + const
i

Here we have neglected 1 in comparison to ni and gi as they are very large number
δ log ω = ∑ δ [(ni + g i ) log(ni + g i ) − ni log ni − g i log g i ]
i

(ni + g i ) ni
∑{δni log(ni + gi ) + (n + g ) δni − δni log ni − n δni }
i i i

∑ [δn log(n + g ) − δn log n ]

i
i i i i i

ni
− ∑ [log ]δni
(ni + g i )
theconditi onofMaximum Pr obabilityGives
ni
∑ [log (n + g ) ]δni = 0
i i

TheAuxilla ryConditio nδn = ∑ δni = 0

δE = ∑ ε iδni = 0
 Applying the ni
Lagrange method ∑i {log (n + g ) + α + β iε}δni = 0
i i
of undetermined
multipliers.i.e δni canbeTreatedasArbitrary
Multiplying auxiliary ni
conditions one with log + α + β iε = o
ni
∑ [log (n + g ) + α + β ε]δn =0

(ni + g i )
i i
i i

α and second with

n
∑ [log (n +i g ) + α + β εi ] = 0
i i

gi
1+ = eα + β εi

β and adding the

ni

gi
gi
ni =
eα + β εi − 1
α + β iε
we get 1+ = e
This Re presentsmost Pr obableDistributionOfTheElementsAmongTheVariousEnergyLevelsFoeSystemObeyingTheB − E

ni
gi α + β iε
=e −1
This represents the most probable
ni
Distribution of the element among the
gi
Various energy levels for system ni = α + β iε
Obeying B-E.
e −1
STATISTICS
1 ,− 1
2 2
 Identical but indistinguishable of particles of
spin1/2:
 Electrons,Protons,neutrons are the particle of
this kind. they obey the Pauli’s exclusion
principle and Fermi-Dirac Statistics. The
particles are know as Fermions.
 The particles are indistinguishable, So we have
concentrate on the cells rather than on
individual particles. The particles obey pauli’s
exclusion principle,i.e it is not possible to have
more than one representive point in any one
cell, so each cell will contain only one particle.
 Letus consider the having the n identical but
indistinguishable particles having half integral
spin.

 Though the particles are indistinguishable, the

restriction imposed that only one particle will be
occupied by single cell. The situation of
distribution is as follows
 Energy Level ε1, ε2, ε3,………εi
 Statistical weight g1,g2,g3,……gi
 Occupation no n1,n2,n3,……..ni
 Fermi –Dirac Statistics conditions;
** The particles are indistinguishable from each
other

** Each sublevel or cell may contain 0 or one

particle .

** The sum of energies of all the particles in the

different quantum groups taken together
constitutes the total energy of the system .
 The distribution of ni particles among the gi states.
 Wecan easily find the first particle can be put in any
one of the ith level in gi ways.

g i C n ways
i

gi!
ni !( g i − ni )!
 The total no of eigen gi
states for whole G=Π
i n !( g − n )!
i i i
system is given by
 ToObtain the
condition of the
maximum Probability
log ω = ∑ {log g i !− log ni !− log( g i − ni )!} + CONST
i

useStirling ' s
log ω = ∑ {( ni − g i ) log( g i − ni ) + g i log g i − ni log ni } + const
i

Diff .W .r.t
ni

ni − g i n
δ log ω = ∑ {δni log( g i − n) + (−δni ) − δni log ni − i δni }
i ( g i − ni ) ni
δ log ω = ∑ − {log(ni ) − log( g i − ni )}δni
i

(ni )
δ log ω = −∑ {log }δni
i ( g i − ni )
ConditionOfMaximum Pr obabilityGives
( ni )
∑{log ( g
i − n )
}δni = 0
i i
 Introducing the auxiliary condition
δn=Σδni=0
δE=Σεiδni=0

 Applying the Lagrange method of undetermined

multipliers.i.e Multiplying auxiliary conditions one
with α and second with β and adding the we get
 (ni )
∑{log
i ( g i − ni )
+ α + β εi }δni = 0

δni canBeTreat edAsArbitr ary

(ni )
log = −(α + β εi )
( g i − ni )
g − ni
log( i ) = (α + β εi )
ni
g
log( i −1) = (α + β εi )
ni
gi
− 1 = e (α + β εi )
ni
gi
= 1 + e (α + β εi )
ni
g
ni = (α + β εii )
e +1
ThisisMost Pr obableDist riutionAcc ordingToF − D
What is my position

What is my energy

Statistical Mechanics

Please read

What is my momentum