Statistical Physics

We consider a physical system composed of N identical particles confined to a space of volume V. In

a typical case, N would be an extremely large number. When N→∞ and V→∞ system is said to be in
thermodynamic limit.
Microstate and Macrostate
Microstate is a specific microscopic configuration of a thermodynamic system that the system may
occupy with a certain probability. Macrostate of a system refers to its macroscopic properties, such as
its temperature, pressure, volume and density. Microstates and macrostates of flipping a coin twice.
All microstates are equally probable, but the macrostate (H, T) is twice as probable as the macrostates
(H, H) and (T, T).

Each microstate is defined by its values of position (x) and momentum (p). This space is known as
phase space. A system of N particles in d dimensions has a 2dN dimensional phase space.

A given macrostate of the system, corresponds a large number of microstates and at any time the
system is equally likely to be in any one of these microstates. This is generally referred to as the
postulate of “equal a priori probabilities”.

The number of microstates corresponding to any given macrostate is called its thermodynamic
Probability. Consider four particles a,b,c and d to be arranged in two boxes. The possible arrangements
are

Macrostate Compartment 1 Compartment 2 Number of microstate or

thermodynamic Probability
(0,4) abcd 1
a bcd
b acd
(1,3) c abd 4
d abc
ab cd
ac bd
(2,2) ad bc 6
bc ad
 bd ac
cd ab
bcd a
acd b
(31,) abd c 4
abc d
(4,0) abcd 1

Entropy and thermodynamic probability

The thermodynamic probability W is related to the entropy (S) by the relation S=kBln(W), the universal
constant kB is referred to as the Boltzmann constant. This related microscopic quantity W to
macroscopic quantity entropy.

Statistical ensembles
“Ensembles” can be 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 Volume, Energy, Pressure,
Temperature and Total number of particles etc. 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 quantum states. There are three types of ensembles: (1) Micro-canonical Ensemble, (2)
Canonical Ensemble (3) Grand Canonical Ensemble.

1. Micro-canonical Ensemble: It is the collection of a large number of essentially independent

systems having the same total number of particles N, volume V and energy E. The systems of
a micro-canonical ensemble are separated by rigid impermeable and insulated walls, such that
the values of E, V and N are not affected by the mutual pressure of other systems. This
ensemble is as shown in the figure below:

2. Canonical Ensemble: It’s the collection of a large number of essentially independent systems
having the same number of particles N, volume V and the temperature T. The equality of
temperature of all the systems can be achieved by bringing all the systems in thermal contact.
Hence, in this ensemble the systems are separated by rigid impermeable but conducting walls,
so that there is an exchange energy but not particles. This ensemble is as shown in the figure
below:
 3. Grand Canonical Ensemble: It is the collection of a large number of essentially independent
systems having the same chemical potential μ, volume V and temperature T. The systems of
a grand canonical ensemble are separated by rigid permeable and conducting walls. There is
exchange of particles as well as energy. This ensemble is as shown in the figure:

Maxwell Boltzmann distribution

The Maxwell-Boltzmann equation forms the basis of the kinetic theory of gases and defines the
distribution of speeds for a gas at a certain temperature. From this distribution function, the most
probable speed, the average speed, and the root-mean-square speed can be derived.

Consider a system that consists of identical but distinguishable particles. Let the total number of
particles in the system be N. Now we distribute the N particles over the m cells, such that nj particles
are allotted to cell number j. In a closed system with total energy E must fulfil the condition
∑ 𝑛𝑖 𝐸𝑖 = 𝐸 ---- (1). The other condition is the conservation of the number of particles,
∑ 𝑛𝑖 = 𝑁 --- (2)
We want to know the number of particles at a given energy level. The energy levels available within
the system are ε0, ε1, ε2………., εr. The number of particles sitting at each energy level is variable and
𝑁!
the number of particles is given by n1, n2, n3……..,nr. Thermodynamic probability, = 𝑛 . Since N
1 𝑛2 …𝑛𝑚
is a very large number, it is better to take the logarithms,
𝑁!
∴ ln 𝑊 = ln (𝑛 ) = ln 𝑁! − ln(𝑛1 𝑛2 … 𝑛𝑚 ) = ln 𝑁! − ∑𝑖 ln(𝑛𝑖 )
1 𝑛2 …𝑛𝑚
According to Sterling’s approximation ln 𝑛! = 𝑛 ln 𝑛 − 𝑛
Therefore we can rewrite above eqn. as
ln 𝑊 = 𝑁ln 𝑁 − 𝑁 − (∑𝑖 𝑛𝑖 ln 𝑛𝑖 − 𝑛𝑖 ) = 𝑁ln 𝑁 − 𝑁 − ∑𝑖 𝑛𝑖 ln 𝑛𝑖 + 𝑛𝑖
We need to know the combination of n1, n2, n3……..,nr that maximizes the value of W along with ∑ 𝑛𝑖 =
𝑁 and ∑ 𝑛𝑖 𝐸𝑖 = 𝐸. This combination results in the most probable microstate.

By taking the derivative of the above equation, we get δln 𝑊 = −δ ∑𝑖(−𝑛𝑖 ln 𝑛𝑖 + 𝑛𝑖 ) = 0

1
∑𝑖 (−δ𝑛𝑖 ln 𝑛𝑖 − 𝑛𝑖 δ𝑛𝑖 + δ𝑛𝑖 ) = 0
𝑛𝑖
∑𝑖(ln 𝑛𝑖 )δ𝑛𝑖 = 0 --- (3)
The proven method for maximizing a function of many variables, allowing for additional constraints,
is the variational method with Lagrange multipliers. Taking the derivative of eqn. (2) and (3) result in
∑ 𝐸𝑖 δ𝑛𝑖 = 0 --- (4)
∑ δ𝑛𝑖 = 0 -- (5)
Combing eqn. (3), (4) and (5)
∑(ln 𝑛𝑖 )δ𝑛𝑖 + 𝛼 ∑ δ𝑛𝑖 + 𝛽 ∑ 𝐸𝑖 δ𝑛𝑖 = 0
𝑖

∑(ln 𝑛𝑖 + 𝛼 + 𝛽𝐸𝑖 )δ𝑛𝑖 = 0

𝑖
This is true for any combination of 𝑛𝑖 , hence each term in the summation should be separately zero.
∴ (ln 𝑛𝑖 + 𝛼 + 𝛽𝐸𝑖 ) = 0 or ln 𝑛𝑖 = −𝛼 − 𝛽𝐸𝑖
𝑛𝑖 = 𝑒𝑥𝑝−𝛼−𝛽𝐸𝑖 ---- (6)
1
For an ideal gas, 𝐸𝑖 = 2 𝑚𝑣𝑖 2 where vi is the velocity of the particle in cell i.
the mean number of molecules with positions between r and r+dr, and velocities in the range v and
v+dv is f(r,v)d3rd3v.
1 2
𝑓(𝑟, 𝑣)𝑑3 𝑟𝑑3 𝑣 = 𝐶𝑒𝑥𝑝−𝛽𝐸𝑖 𝑑3 𝑟𝑑3 𝑣 = 𝐶𝑒𝑥𝑝−𝛽2𝑚𝑣 𝑑3 𝑟𝑑3 𝑣
Where C is a constant of proportionality which includes, 𝑒𝑥𝑝−𝛼 . The sum over molecules with all
possible positions and velocities gives the total number of molecules N. Hence constant C can be
determined by ∫𝑟 ∫𝑣 𝑓(𝑟, 𝑣)𝑑3 𝑟𝑑3 𝑣 = 𝑁
The integral over the molecular position coordinates just gives the volume, V, of the gas, because the
Boltzmann factor is independent of position. The integration over the velocity coordinates can be
reduced to the product of three identical integrals (one for vx, one for vy, and one for vz), so we have
1 2 3
∞
𝐶𝑉 (∫−∞ 𝑒 −𝛽2𝑚𝑣𝑧 𝑑𝑣𝑧 ) = 𝑁 --- (7)
𝛽𝑚 𝛽𝑚 2
Substituting = √ 𝑣
2 𝑧
, hence 𝑑𝑦 = √ 2
𝑑𝑣𝑧 or 𝑑𝑣𝑧 = √𝛽𝑚 𝑑𝑦
1 2
∞ ∞ 2 2 2 ∞ 2
∴ ∫−∞ 𝑒 −𝛽2𝑚𝑣𝑧 𝑑𝑣𝑧 = ∫−∞ 𝑒 −𝑦 𝑑𝑦 √𝛽𝑚 = √𝛽𝑚 √𝜋 where we used the relation, ∫−∞ 𝑒 −𝑦 𝑑𝑦 = √𝜋
Combining the above results, eqn. (7) becomes,
3 3 3⁄
2𝜋 𝑁 𝛽𝑚 ⁄2 𝛽𝑚 2
𝐶𝑉√𝛽𝑚 = 𝑁 or 𝐶 = (
𝑉 2𝜋
) = 𝑛 ( 2𝜋 )
3
𝛽𝑚 ⁄2
∴ 𝑓(𝑟, 𝑣)𝑑 𝑟𝑑 𝑣 = 𝑛 ( ) 𝑒𝑥𝑝−𝛽𝐸𝑖 𝑑3 𝑟𝑑3 𝑣
3 3
2𝜋
1
Using zeroth law of thermodynamics, we can calculate 𝛽 = 𝑘 𝑇
𝐵
⁄2 𝑚𝑣 3 2
𝑚 −
∴ 𝑓(𝑣)𝑑3 𝑟𝑑3 𝑣 = 𝑛 ( ) 𝑒𝑥𝑝 2𝑘𝐵𝑇 𝑑3 𝑟𝑑3 𝑣
2𝜋𝑘𝐵 𝑇
The average number of molecules per unit volume with a speed 𝑣 =|v| in the range of 𝑣 and 𝑣 + 𝑑𝑣
is obtained by summing over all molecules with speeds in this range, irrespective of the direction of
their velocities and integral over spherical shell of radius 𝑣 and thickness 𝑑𝑣 in velocity space. This is
𝑓(𝑣) multiplied by the volume of the spherical shell in velocity space. So 𝐹(𝑣)𝑑𝑣 = 𝑓(𝑣)4𝜋𝑣 2 𝑑𝑣
3⁄ 𝑚𝑣2
𝑚 2 −
Hence, 𝐹(𝑣)𝑑𝑣 = 𝑛 (2𝜋𝑘 𝑇) 𝑒𝑥𝑝 2𝑘𝐵𝑇
4𝜋𝑣 2 𝑑𝑣
𝐵
This is the famous Maxwell distribution of molecular speeds.
The speed at the top of the curve (vp) is called the most probable speed because the largest number
of molecules have that speed.
Below shows the Maxwell-Boltzmann distribution with temperature. At lower temperatures, the
molecules have less energy. Therefore, the speeds of the molecules are lower and the distribution has
a smaller range. As the temperature of the molecules increases, the distribution widens. Because the
molecules have greater energy at higher temperature, the molecules are moving faster.

3⁄ 𝑚𝑣2
1 ∞ 1 ∞ 𝑚 2 −
The mean molecular speed is given by 𝑣𝑎𝑣𝑔 = ∫0 𝐹(𝑣)𝑣𝑑𝑣 = ∫0 𝑛 ( ) 𝑒𝑥𝑝 2𝑘𝐵 𝑇 4𝜋𝑣 3 𝑑𝑣
𝑛 𝑛 2𝜋𝑘 𝑇 𝐵
3⁄ ∞ 𝑚𝑣 2
𝑚 2 −
2𝑘𝐵 𝑇 𝑣 3 𝑑𝑣
𝑣𝑎𝑣𝑔 = 4𝜋 ( ) ∫ 𝑒𝑥𝑝
2𝜋𝑘𝐵 𝑇 0
𝑚𝑣 2 2𝑘 𝑇 𝑚𝑣 2 2𝑘 𝑇
Substituting 𝑦 = √2𝑘 𝑇 𝑣 and hence 𝑣 = √ 𝐵2 𝑦 𝑑𝑦 = √ 𝑑𝑣 or 𝑑𝑣 = √ 𝐵2 𝑑𝑦
𝐵 𝑚𝑣 2𝑘𝐵 𝑇 𝑚𝑣
3⁄ 3 ⁄2 1
𝑚 2 ∞ 2 2𝑘𝐵 𝑇 2𝑘𝐵 𝑇 ⁄2
𝑣𝑎𝑣𝑔 = 4𝜋 ( ) ∫ 𝑒𝑥𝑝−𝑦 ( ) 𝑦 3
( ) 𝑑𝑦
2𝜋𝑘𝐵 𝑇 0 𝑚𝑣 2 𝑚𝑣 2
3⁄ 2 ∞
𝑚 2 2𝑘𝐵 𝑇 2
𝑣𝑎𝑣𝑔 = 4𝜋 ( ) ( ) ∫ 𝑒𝑥𝑝−𝑦 𝑦 3 𝑑𝑦
2𝜋𝑘𝐵 𝑇 𝑚 0
3⁄
𝑚 2 2𝑘𝐵 𝑇 2 ∞ 2
𝑣𝑎𝑣𝑔 = 4𝜋 (2𝜋𝑘 𝑇) ( 𝑚 ) × 1⁄2 Because ∫0 𝑒𝑥𝑝−𝑦 𝑦 3 𝑑𝑦 = 1⁄2
𝐵

8𝑘𝐵 𝑇
𝑣𝑎𝑣𝑔 = √
𝜋𝑚

3𝑘𝐵 𝑇
A similar calculation gives 𝑣𝑟𝑚𝑠 = √̅̅̅
𝑣2 = √
𝑚
The most probable speed is the maximum value on the distribution plot. This is established by finding
𝑑𝐹(𝑣)
the velocity when the following derivative is zero, 𝑑𝑣
= 0. This will give

2𝑘𝐵 𝑇
𝑣𝑝 = √
𝑚
 Comparison between Maxwell-Boltzmann, Bose-Einstein and Fermi-Dirac statistics
These three statistics concern how particles occupy a system which consists of several energy levels.
A particle in this system can be in one of those energy levels depending on the energy particle has.

1. Particles which obey Maxwell-Boltzmann Statistics have to be distinguishable and one energy
state can be occupied by one or more particles. Distinguishable means that if we have 2
particles, let say A and B, also two states, 1 and 2. We put A to state 1 and B to state 2, it will
be different with the distribution A to state 2 and B to state 1. It means that A and B are
distinct. So let’s enumerate the states of the system
Single particle states 1 2
AB
A B
B B
AB
We get a total of 4 states of the system as a whole.

Particles which obey Bose-Einstein Statistics have to be indistinguishable and one energy state
can be occupied by one or more particles. So instead of saying it as particle A or B, we call it
as just “particle” since they are the same thing. So let’s enumerate the states of the system:
Single particle states 1 2
AA
A A
AA
We get a total of 3 states of the system as a whole

Particles which obey Fermi-Dirac Statistics have to be indistinguishable each other and one
energy state can be occupied by only one particle. So we have to fill it to another state when
a state has just been occupied by another particle. So let’s again enumerate the states of the
system:
Single particle states 1 2
A A
We get a total of 1 state of the system as a whole.

2. Maxwell Boltzmann statistics is applicable to identical, distinguishable particle of any type

spin. The molecules of gas are of this type

Bose Einstein statistics is applicable to the identical, indistinguishable particles of zero or

integral spin. These particles are called Bosons. Example photons, Helium atom.

Fermi Dirac statistics is applicable to the identical, indistinguishable particles of half integral
spin. These particles obey Pauli Exclusion Principle. E.g. Electron, proton etc.

3. Maxwell–Boltzmann statistics apply to “classical” particles with non-quantized energy levels.

There aren’t any truly classical particles, but M-B statistics is still a good approximation in
many cases.
 Bose-Einstein statistics apply to quantum particles with the property that any number of
particles can occupy any level. It turns out these are the particles with integral “spin” such as
photons and Helium-4 (but not Helium-3) atoms.

Fermi–Dirac statistics apply to quantum particles with the property that only one particle can
occupy any particular level. It turns out these are the particles with half-integral “spin” such
as electrons, neutrons and Helium-3 atoms.

4. The mean number of particles in each statistics is:

1
Maxwell–Boltzmann: 𝑛𝑖 =
𝑒 𝛽(𝜖𝑖 −𝜇)
1
Bose-Einstein: 𝑛𝑖 =
𝑒 𝛽(𝜖𝑖 −𝜇)−1
1
Fermi–Dirac: 𝑛𝑖 = where 𝛽 = 1⁄𝑘𝑇 and 𝜖𝑖 is the energy of each levels, μ
𝑒 𝛽(𝜖𝑖 −𝜇)+1
is the chemical potential.

Bose–Einstein condensate
Bose–Einstein condensate (BEC) is a state of matter which is typically formed when a gas of bosons at
low densities is cooled to temperatures very close to absolute zero (-273.15 °C). Under such conditions
a significant fraction of the particles to occupy a single quantum state. No analogous phenomenon
occurs for two or more fermions, which are prohibited by the Pauli Exclusion Principle from occupying
the same quantum state. First experimentally observed in 1995 by cooling about 2000 atoms of
rubidium gas to 170 nanokelvin.