BY: GIDESA ABDENA

ID NO:PGR/28055/15

OUTLINES
1. ENSEMBLES
2.STATISTICAL MATRIX

1
 ENSEMBLE

 Ensemble is defined as a collection of a number

macroscopically identical but essentially
independent systems.
 Here the term macroscopically identical means, as,
each of the systems constituting an ensemble
satisfies the same macroscopic conditions, like V, E,
P, T, N and etc. 
 Here again, the term essentially independent means
the system (in the ensemble) being mutually non-
interacting to others, i.e., the systems differ in
microscopic conditions like parity, symmetry,
quantum states etc. 2
 The possible microstates a system could be in are known
as the ensemble of states for a system.
 There are three types of ensembles.
These are : 1.Micro canonical ensembles
2.Canonical ensembles and
3. Grand canonical ensembles
1. Micro canonical ensemble
 The micro canonical ensemble is a statistical ensemble in
which a system is specified by the number of particle  N,
volume of system V, and energy of system  E, and an
arbitrary microscopic state appears with the same
probability. 
3
 In this case the system is not interacting with any other
system and the energy of the system is a constant.
 The systems of a micro-canonical ensemble are separated
by rigid, impermeable and insulated walls such that the
values of E, V & N are not affected by the mutual
pressure of other systems.

4
This statistical ensemble is highly appropriate for dealing
with a physical system which is completely isolated from
the outer system; in such an isolated system, there is no
exchange of particles and energies.
 The probability density is called the micro canonical
distribution for this statistical ensemble and expressed
as : ρ(r)=C
in which r means a certain microscopic state, and C is a
constant. The above Equation means that all microscopic
states arise with the same probability.
 If the total number of microscopic states is denoted
by W, the constant C is written as: C=1/W
5
Thus ρ(r) is expressed as:
ρ(r) =1/W
Probability of finding the energy state (P) is proportional to
finding the particular energy level (E). Therefore the
probability P is given by:

Since the total probability of energy state is one, that means:

Ʃ P=1 , then 1
From the above equation A is given by:
A=1/ Ʃ
In which Ʃ is known as partition function f.
6
 In the above discussion, microscopic states have been
considered to change discretely, so that an arbitrary
microscopic state may be regarded as a quantum state.
If we assume that microscopic states change continuously
that is if a system is classically described and prescribed by
the particle positions r (which means r1, r2, …, rN) and the
particle momenta p (which means p1, p2, …, pN), then W
and ρ (r, p) are written as

7
 The partition function of the micro canonical ensemble
is a function of the extensive variables E, V,
and N given by:
Ω= Ω(E, V, N)
 The entropy of micro canonical ensemble is:
S=kB logΩ
2. Canonical Ensemble
 It is a statistical ensemble which is specified by the volume of
system V, number of particles N, and temperature T.
 We will develop the method of canonical ensembles by
considering a system placed in a heat bath at temperature T.
 The canonical ensemble is the assembly of systems with fixed N
and V. 8
 In other words we will consider an assembly of systems closed
to others by rigid, diathermal, impermeable walls.
 The energy of the microstates can fluctuate, the system is kept
in equilibrium by being in contact with the heat bath at
temperature T.
 Schematically, we can view this ensemble as
State 1.....E1,V,N…..Bath T
State 2…..E2,V,N…..Bath T
State 3…..E3,V,N…..Bath T
State v….En, V, N….Bath T
The system for which the canonical ensemble is appropriate can be
thought of as a sub-system of the system for which the
microcanonical ensemble is appropriate.
9
Isolated system with E,V,N fixed. Basically what we do is
to examine one state and consider the rest to be in the heat
bath. Thus the macroscopic system is specified by T,V, and
N as illustrated.

10
 Let the combined energy of the system and the heat bath
= E0
E1 ≤ E2 ≤ E3 ≤ ... ≤ Er ≤ ...
We’ll select δE so that we select one energy level but
several microstates.
 Let the system be in a state with energy Er, the energy
of the reservoir (heat bath) is then E0 −Er.
 What’s the probability that the system will be in a
microstate with energy Er? When we considered an
isolated system, we found the probability of it being in a
macrostate specified by (E,V,N,α) was proportional to
the multiplicity Ω(E,V,N,α). 11
 The multiplicity of the heat bath is:
Ω(E0 −Er)
We have an isolated system with two sub-systems, labeling
the heat bath as system 2, we have:
Pr = Const Ω2(E0 −Er)
The ratio of probabilities for the states Ei, and Ej is

 We could write similar expressions for all pairs of levels,

if we consider three:

12
 ,
we can add them to give

We can generalize this result to yield:

Where the sum is taken over all levels.

 Next, we rewrite the expression for pr in terms of the
reservoir entropy.

13
S = k ln Ω, so S/k = ln Ω, and Ω = eS/K. Using this we write
pr = (constant) × exp
With a large reservoir, we can assume that E0 Er. If the heat
bath is large this inequality holds for all states with a
reasonable chance of occurring. Now we’ll expand in a
Taylor series about S2(E0). Recall

so,

Now, we know that :

14
where T is the temperature of the heat bath. Thus, we have

All higher order partial derivatives are zero by assumption

of a large heat bath

Now using
Substituting and

15
We get:

The above equation refers that the probability of the

system to be in r microstate.

16
3. Grand canonical ensemble
 In microcanonical ensemble, each system
contains same energy as well as same number of
particles. Hence, the system dealt within this
ensemble is a closed isolated system.
 With microcanonical ensemble, we can not deal
with the systems that are kept in contact with a
heat reservoir at a given temperature

17
 In canonical ensemble, the condition of constant energy
is relaxed and the system is allowed to exchange energy
but not the particles with the system, i.e. those systems
which are not isolated but are in contact with a heat
reservoir.
 This model could not be applied to those processes in
which number of particle varies, i.e. chemical process,
nuclear reactions (where particles are created and
destroyed) and quantum process.
 So, for the method of ensemble to be applicable to such
processes where number of particles as well as energy of
the system changes, it is necessary to relax the condition
of fixed number of particles. 18
 Such an ensemble where both the energy (E) as well as
number of particles(N) can be exchanged with the heat
reservoir is called Grand Canonical Ensemble.
 In canonical ensemble T, V and N are independent
variables. Whereas, in grand canonical ensemble, the
system is described by its temperature (T),volume (V)
and chemical potential (μ).
 Since, the system is not isolated, its microstates are not
equally probable. So, for this we have to first construct
an isolated system where we can apply the assumption of
equi-probable microstates and then to calculate the
probability of a microstate of an open system.
19
 Let us consider a subsystem ‘s’ in contact with a heat
reservoir ‘r’ and the subsystem can exchange energy as
well as particles with the reservoir. Total energy of the
system is represented by microcanonical ensemble with
constant energy E and number of particles N.
If Es refer to the energy of subsystem, Ns refers to the
number of particles of subsystem; Er refers to the energy of
reservoir, Nr refer to the number of particles of reservoir
and Et is energy of total system and Nt is number of
particles of the total system then,
Et= Es + Er and Nt= Ns + Nr

20
 To get the entropy of grand canonical ensemble, by
using the first law of thermodynamics. we have for a
reversible process
dE= TdS - PdV+μ dN ………………………..(1)
………………………....(2)

Now,
S=S(E,V,N)

…………….(3)

21
From (2) and (3), we get

…………… ..(4)

……………….(5)

22
Now, probability of a microstate is given by:
Ps(Ns)…………………..(6)

……………………..(7)

Where and

23
where constant ‘α’can be determined by the normalization condition

and the sum is over all the microstates of the system.

………………(8)

where
.........................(9)

An ensemble characterized by the distribution function given by

equation (9) is called the grand partition function.
24
 It is the sum of canonical partition function Z for
ensembles with different N’s, with weighting factor
exp (βNμ).
…………………(10)

where λ=exp(βμ) is fugacity.

 Fugacity: is a particles tendency to escape from one
environmental compartment to another (in this case from
subsystem to reservoir or from reservoir to subsystem).
Therefore, we can write the grand canonical partition
function as: …(11)
25
 STATISTICAL MATRIX

 The Density Matrix of Ensembles

 The density matrix is of special importance in the formulation of
quantum statistics; it can also be denoted by the terms ‘statistical
operator’ and ‘density operator’.
Let a system be in the state . The observable A in this state has the
mean value or expectation value of:

The structure of the mean value makes it convenient to define the

density matrix by

then we have:
< A˃= Tr(ρA) 26
Here, the definition of the trace (Tr) is:

where {|n˃} is an arbitrary complete orthonormal basis

system. Owing to

the trace is independent of the basis used.

27
 If the systems or objects under investigation are all in
one and the same state |ψ˃, we speak of a pure
ensemble, or else we say that the systems are in a pure
state.
 Along with the statistical character which is inherent to
quantum-mechanical systems, in addition a statistical
distribution of states can be present in an ensemble. If
an ensemble contains different states, we call it a mixed
ensemble, a mixture, or a mixed state. We assume that
the state |ψ1˃ occurs with the probability p1, the state |
ψi˃ with the probability pi, and so on. with:

28
The mean value or expectation value of A is then:

This mean value can also be represented in terms of the

density matrix defined by:

29
THANK
YOU!
30