Statistical Mechanics and Thermodynamics
Statistical Mechanics and Thermodynamics
Subject Physics
TABLE OF CONTENTS
1. Learning Outcomes
2. Introduction
3. Thermodynamic Limit
9. Summary
Appendix
A1 Particle in a one dimensional box
A2 Particle in a cubical box
A3 Quantum degeneracy of a level
1. Learning Outcomes
After studying this module, you shall be able to
Introduction
2. Introduction
In this module from the inherent statistical nature of macroscopic systems, we underline
the statistical formulation of thermodynamics. We will see how this approach provided
microscopic basis to thermodynamics by successful application of this approach linking
entropy with the probability of microstates overcoming the lack of information about the
system made up of a large number of particles of the order of avogadro’s number. We
shall see that the statistical-mechanic meaning of temperature, pressure and chemical
potential is the same as thermodynamic meaning of temperature, pressure and chemical
potential. Las of all we shall later try to get a prescription to derive thermodynamic
functions related to the system under study. System of study shall always be treated as a
system in equilibrium.
3. Thermodynamic Limit
Realizing the fact that macroscopic systems are made up of typically 𝟔. 𝟎𝟐𝟑 × 𝟏𝟎𝟐𝟑
points to the enormity of the systems involved. Thermodynamic limit is a way to express
this enormity and a natural way to express this by carrying out the limit number of
particles 𝑵 → ∞ and volume 𝑽 → ∞, such that the particle density 𝒏 remains fixed.
Taking this limit amounts to saying that extensive properties of the system varies
proportional to the size of the system i.e. 𝑵 or 𝑽 where as intensive properties remain
independent of the size of the system. The process can be start with a finite system with a
fixed density and keep on adding more and more such identical finite systems to it such
that number density remains the same causing increase only in extensive parameters
without change in the intensive parameters. Not going to thermodynamic limit in the
mathematical sense can lead to serious problems with the strict compliance of the
extensive nature of extensive parameters such as entropy.
A macroscopic state also called a macrostate is described in terms of macroscopic variables such
as total energy 𝑬, volume 𝑽 and number of particles 𝑵. In describing a macrostate we are not at
all bothered by the description of the properties of individual particles constituting the system
such as their momentum, spin etc.
A microscopic state also called microstate is specified by describing the state of each particle in a
given macrostate. Counting the number of microstates in a given macrostate is the most basic task
to indulge in statistical mechanics
To further appreciate the difference, let us take a system m ade up of N number of non-interacting
particles with total energy 𝑬. This energy is the sum of the energies 𝝐𝒊 of individual particles,
such that
𝑬 = ∑ 𝒏𝒊 𝝐𝒊 (1)
𝒊
𝑵 = ∑ 𝒏𝒊 (2)
𝒊
It may be recalled (see appendices A1 and A2) that 𝝐𝒊 are the single particle energies which
have two distinct properties:
This macrostate can be realized at the microscopic level in many different possible ways
known as microstates. In simple words, it implies that total energy 𝑬 can be distributed
among 𝑵 particles constituting the system in large numer of different ways.
One of the aims of statistical mechanical formulation is count the number of all the possible
microstates. In this course this number shall be denoted by 𝛀, which is a function of
macroscopic variables of the system, which in the present case are 𝑵, 𝑽 and 𝑬. Therefore,
𝟏
this count is a number denoted by the function 𝛀(𝑵, 𝑽, 𝑬). The value 𝛀, gives the pro bability
of ocuurance of any one of the states, which is as per the principle of equal apriori probability
𝟏
the same for each microstate. This 𝛀, is also known as thermodynamic probability. 𝑷𝒓 .
The equilibrium state of the system then is the state for which function 𝛀(𝑵, 𝑽, 𝑬) maximizes
for each of its arguments. It is interesting to note that though a system corresponding to a
given macrostate remains in a state of equilibrium, its microscopic states may keep on
changing continuously.
imposed by equations (1) and (2) or any other constaint on macroscopic variables involved) are
equally likely.
Allowing possibility of all microstates of a given macrostate has resulted in derivation of ideal
gas laws and derivation of Maxwell Boltzmann velocity distribution law which were later
confirmed experimentally. Justification of this principle, therefore, lies in the results derived from
it and observed experimentally. This principle is treated as one of the postulates of statistical
mechanics.
For this we consider two physical systems: system 1 and system 2 which are separately in
equilibrium, described by the macroscopic variables 𝑵𝟏 , 𝑽𝟏 , 𝑬𝟏 and 𝑵𝟐 . 𝑽𝟐 , 𝑬𝟐 . System 1 has
𝛀𝟏 (𝑵𝟏 , 𝑽𝟏 , 𝑬𝟏 ) number of microstates and system 2 has 𝛀𝟐 (𝑵𝟐 , 𝑽𝟐 , 𝟐) as shown in figure
System 1 System 2
𝑁1 , 𝑉1 , 𝐸1 𝑁2 , 𝑉2 , 𝐸2
Figure 1 Two systems ready for making contact in three ways (a) when the partition is
conductiong (b) when the partition is made conducting as well as movable and (c) when
the partition also allows for exchange of particles.
𝑬 = 𝑬𝟏 + 𝑬𝟐 = 𝑪𝒐𝒏𝒔𝒕𝒂𝒏𝒕 (3)
𝛀 number of microstates of the combined system here vary with 𝑬𝟏 till equilibrium is
established, and this happens until 𝛀 maximizes. The combined system then attains a
macrostate which has the most number of microstates. The important point to note is that
this macrostate of the system is that state in which the system spends the largest portion
of its time and presents itself as an equilibrium state.
𝝏𝛀(𝑬,𝑬𝟏 ) (6)
= 𝟎, when 𝑬𝟏 = ̅𝑬̅̅𝟏̅, the equilibrium value of 𝑬𝟏 and 𝑬𝟐 = ̅𝑬̅̅̅𝟐
𝝏𝑬𝟏
Note 𝑬𝟐 = 𝑬 − 𝑬𝟏
𝝏𝛀𝟏 (𝑬𝟏 ) 𝝏𝛀 (𝑬 )
( ) ̅̅̅𝟐̅) − ( 𝟐 𝟐 )
𝛀𝟐 (𝑬 ̅̅̅𝟏̅) = 𝟎
𝛀𝟏 (𝑬
𝝏𝑬𝟏 ̅̅̅̅
𝑬=𝑬
𝝏(𝑬𝟐 ) 𝑬 ̅̅̅̅
𝟏 𝟐 =𝑬 𝟐
This (7) is the condition for equilibrium of the system 1 and system 2. Where
𝝏 𝐥𝐧 𝛀(𝑵,𝑬,𝑽) (8)
We define 𝜷 = ( 𝝏𝑬
)
̅ ,𝑽,𝑵
𝑬=𝑬
𝜷𝟏 = 𝜷𝟐 (9)
When two systems are brought in contact and allowed to exchange energy
through conducting partition, the exchange of energy continues till equilibrium
values 𝑬𝟏 = ̅𝑬̅̅𝟏̅ and 𝑬𝟐 = ̅𝑬̅̅𝟐̅ of the two systems is not attained. Once this happens
systems come to thermal equilibrium
This stage of thermal equilibrium is indicated when the respective values of the
parameter 𝜷, 𝜷𝟏 and 𝜷𝟐 become equal, pointing towards zeroth law of thermodynamics
which defines temperature as that parameter. It is, therefore, natural to look for this
connection between 𝜷 and temperature 𝑻.
If we recall that the temperature is defined by the derivative of the entropy with respect to
energy 𝑬:
𝝏𝑺 𝟏 (10)
( ) =
𝝏𝑬 𝑵,𝑽 𝑻
Equation (11) was first established by Ludwig Boltzmann. He also underlined the
connection between the two approaches i.e. thermodynamic approach, equation
(10) and statistical approach equation (8) and further concluded that the constant
in equation (11) should be a universal constant, which was later named as
Boltzmann constant by Planck as
𝑺 = 𝒌𝑩 𝐥𝐧 𝛀 (12)
Also,
𝟏 𝟏 (13)
= 𝒌𝑩 𝒐𝒓 𝜷 =
𝜷𝑻 𝒌𝑩𝑻
According to formula (12), one can get ent ropy of a system by knowing the total number
of microstates of a macrostate for a given system.
If 𝛀 = 𝟏, this corresponds to a unique completely ordered state, with entropy 𝑺 = 𝟎. As
𝛀 increases, the number of microstates increase and more is the number of microstates
more is the entropy. Therefore, in a given system disorder is manifested through the
count of the microstates of a given macrostate of the system, larger this number larger is
the disorder and less the degree of predictability.
Furthermore, maximization of the number of possible microstates or probability amounts
to maximizing the entropy for equilibrium.
If we extend the considerations of the case I and further relax the constraint on the
partition separating system 1 from system 2 by making it movable as well as conducting
figure 1(b), then the volumes of the two systems also become variables, though the total
volume𝑽 = 𝑽𝟏 + 𝑽𝟐 , remains constant and besides equation (7) corresponding to thermal
equilibrium
𝝏 𝐥𝐧 𝛀𝟐 (𝑬𝟏 ) 𝝏 𝐥𝐧 𝛀𝟐 (𝑬𝟐 )
( ) =( )
𝝏𝑬𝟏 𝑵 ̅̅̅̅ 𝝏𝑬𝟐 ̅̅̅̅
𝟏 ,𝑽𝟏 ,𝑬𝟏 =𝑬𝟏 𝑵 𝟐 ,𝑽𝟐 ,𝑬𝟐 =𝑬𝟐
We have another equation which can be obtained in an analogous way as in case I giving
another condition for the system to be in mechanical equilibrium as well
𝝏 𝐥𝐧 𝛀(𝑵, 𝑬, 𝑽) (15)
𝜼=( )
𝝏𝑽 ̅ ,𝑬,𝑵
𝑽=𝑽
𝝏𝑺 𝑷 (16)
( ) =
𝝏𝑽 𝑵,𝑬 𝑻
𝚫𝑺 𝑷 (17)
= = 𝐂𝐨𝐧𝐬𝐭𝐚𝐧𝐭
𝚫(𝐥𝐧 𝛀) 𝜼𝑻
The equilibrium condition during thermal and mechanical contacts, such that
𝑷 𝑷
= 𝒌𝑩 or 𝜼 = 𝑲 𝑻becomes
𝜼𝑻 𝑩
𝑷𝟏 𝑷𝟐 (18)
=
𝑻𝟏 𝑻𝟐
i.e. 𝑻𝟏 = 𝑻𝟐 and 𝑷𝟏 = 𝑷𝟐 .
Case III: Chermical Contact Between Two Systems: Notion of Chemical Potential in
Statistical Mechanics :
Next if we consider besides thermal contact, partition such that it not only allows thermal
contact but also allows exchange of particles between system 1 and system 2, but keep
the volume fixed. Then only total Number of particles 𝑵 = 𝑵𝟏 + 𝑵𝟐 and total Energy
𝑬 = 𝑬𝟏 + 𝑬𝟐 remain fixed but 𝑽𝟏 , 𝑽𝟐 , 𝑬𝟏 and 𝑬𝟐 of the two systems can change. So
besides equation (7)
𝝏 𝐥𝐧 𝛀𝟏 (𝑬𝟏 ) 𝝏 𝐥𝐧 𝛀𝟐 (𝑬𝟐 )
( ) =( )
𝝏𝑬𝟏 𝑵 ̅̅̅̅ 𝝏𝑬𝟐 ̅̅̅̅
𝟏 ,𝑽𝟏 ,𝑬𝟏 =𝑬 𝟏 𝑵 𝟐 ,𝑽𝟐 ,𝑬𝟐 =𝑬 𝟐
We have another equation which can be obtained in an analogous way as in case I giving
another condition for the system to be in chemical equilibrium as well
𝝏 𝐥𝐧 𝛀(𝑵, 𝑬, 𝑽) (20)
𝝁=( )
𝝏𝑵 ̅ ,𝑬,𝑽
𝑵=𝑵
𝝏𝑺 𝝁 (21)
(𝝏𝑵) = −𝑻
𝑵,𝑬,𝑽
We get
𝚫𝑺 𝝁
= − = 𝐂𝐨𝐧𝐬𝐭𝐚𝐧𝐭
𝚫(𝐥𝐧 𝛀) 𝑻
Therefore,
𝝁
The equilibrium condition during Chemical contact, such that − 𝑻 = 𝒌𝑩 or 𝝁 =
−𝒌𝑩 𝑻 becomes
𝝁𝟏 𝝁𝟐 (22)
=
𝑻𝟏 𝑻𝟐
i.e. 𝑻𝟏 = 𝑻𝟐 and 𝝁𝟏 = 𝝁𝟐 .
If we combine the three case discussed above such that if we allow exchange of energy, change of
volume and exchange of particles simultaneously, the conditions of equilibrium are
𝑻𝟏 = 𝑻𝟐 , 𝑷𝟏 = 𝑷𝟐 and 𝝁𝟏 = 𝝁𝟐 (23)
For the macrostate (𝑵, 𝑽, 𝑬) of a given system count the number of all possible
microstates accessible to the macrostate denoted by 𝛀(𝑵, 𝑽, 𝑬)
The entropy of the system can be calculated by equation the equation (12)
𝑺 = 𝒌𝑩 𝐥𝐧 𝛀
The intensive parameters, 𝑻, 𝑷 and 𝝁 can then be calculated by the following
thermodynamic relations (See Module 6)
𝟏 𝝏𝑺 (24)
=( )
𝑻 𝝏𝑬 𝑵,𝑽
𝑷 𝝏𝑺 (25)
=( )
𝑻 𝝏𝑽 𝑵,𝑬
𝝁 𝝏𝑺 (26)
= −( )
𝑻 𝝏𝑵 𝑽,𝑬
Or alternatively
𝝏𝑺 (27)
(𝝏𝑽) 𝝏𝑬
𝑵,𝑬
𝑷= 𝝏𝑺
= −( )
( ) 𝝏𝑽 𝑵,𝑺
𝝏𝑬 𝑵,𝑽
And
𝝏𝑺 (28)
(𝝏𝑵) 𝝏𝑬
𝑽,𝑬
𝝁=− 𝝏𝑺
=( )
(𝝏𝑬) 𝝏𝑵 𝑽,𝑺
𝑵,𝑽
Enthalpy 𝑯 = 𝑬 + 𝑷𝑽 = 𝑮 + 𝑻𝑺 (31)
Furthermore , specific heat at constant volume, 𝑪𝑽 and Specific heat at constant pressure, 𝑪𝑷 are
given by
𝝏𝑬 (32)
𝑪𝑽 = ( )
𝝏𝑻 𝑵,𝑽
𝝏𝑯 (33)
𝑪𝑷 = ( )
𝝏𝑻 𝑵,𝑷
In nut shell,once one knows the entropy 𝑺(𝑵, 𝑬, 𝑽) of the system, equilibrium behavior of the
system can be known by using thermodynamics.
9. Summary
In this module we have learnt
𝑺 = 𝒌𝑩 𝐥𝐧 𝛀
That equilibrium state of the system is that state for which 𝛀(𝑵, 𝑬, 𝑽) maximizes, which
further means maximization of entropy 𝑺.
That the equilibrium conditions for two systems under different types of contacts are
𝑷𝟏 𝑷 (35)
𝑻𝟏
= 𝑻𝟐 under mechanical and thermal contact
𝟐
If there is thermal equilibrium this amounts to
𝑷𝟏 = 𝑷𝟐
𝝁𝟏 𝝁
𝑻𝟏
= 𝑻𝟐 under chemical contact (36)
𝟐
If there is thermal equilibrium this amounts to
𝝁𝟏 = 𝝁𝟐
Bibliography
1. Pathria R.K. and Beale P. D., Statistical Mechanics, 3rd ed. (Elsevier, 2011).
2. Landau L.D., Lifshitz E.M., “ Statistical Physics Part 1,” 3rd Edition, Oxford:
Pergamon Press., 1982
3. Pal P.B., “An Introductory Course of Statistical Mechanics”, New Delhi: Narosa
Publishing House Pvt. Ltd., 2008.
4. Panat P.V., “Thermodynamics and Statistical Mechanics,” New Delhi: Narosa
Publishing House Pvt. Ltd., 2008
5. Ma S.K., “ Statistical Mechanics,” Singapore: World Scientific Publishing Co. Pte.
Ltd., 1985.
6. Greiner W., Neise L., Stocker H., “ Thermodynamics and Statistical Mechanics,”
New York, Springer Verlag, 1995.
Appendices
Following appendices provide a recapitulation of how quantization of energy levels arises
and how the value of single particle energy in a confined system is volume dependent.
We shall try to elaborate on the concept of degeneracy which is auseful idea in counting
the microstates of a physical system
A1 Particle in a one dimensional box:
0 L
Figure 2 Particle moving on a line along x-axis
The motion of this particle is described by one dimensional schrodinger equation, whose
solution 𝝍, assumes the form of a stationary wave, with the boundary condition that
𝟎 𝒂𝒕 𝒙 = 𝟎 (37)
𝝍={
𝟎 𝒂𝒕 𝒙 = 𝑳
Or
𝝏𝟐 𝝍
= −𝒌𝟐 𝝍
𝝏𝒙𝟐
Where 𝒉 is planks constant, m and 𝝐 respectively are the mass and energy of the particle.
The solution of this wave equation then is
𝝍 = 𝑨 𝐬𝐢𝐧 𝒌𝒙 + 𝑩 𝐜𝐨𝐬 𝒙 (39)
𝒏𝝅
Putting in boundary conditions we obtain 𝑩 = 𝟎 and 𝒌 = , where 𝒏 = 𝟏, 𝟐, 𝟑 … … …
𝑳
Knowing 𝒌 yields momentum from 𝒑 = ℏ𝒌 i.e.
𝒏𝒉 (40)
𝒑 = 𝟐𝑳
𝟐
Which gives 𝑨 = √𝑳 .
L
x
Figure 3 Particle in a cubical box
𝝏𝟐 𝝍 𝝏𝟐 𝝍 𝝏𝟐 𝝍 (42)
+ + = −𝒌𝟐 𝝍
𝝏𝒙𝟐 𝝏𝒙𝟐 𝝏𝒙𝟐
⃗ ; 𝒏𝒙 , 𝒏𝒚 and
Where 𝒌𝒙 , 𝒌𝒚 and 𝒌𝒛 can be taken as the components of the wave vector 𝒌
𝟑
𝟐 𝟐
𝒏𝒛 are three quantum numbers taking integer values 𝟏, 𝟐, 𝟑 … … … … and 𝑨 = (𝑳) .
𝒏 𝒉
𝒙 𝒏𝒚 𝒉 𝒏𝒛 𝒉
Therefore, 𝒑𝒙 = 𝟐𝑳 , 𝒑𝒚 = 𝟐𝑳 𝒑𝒛 = and hence momentum p is quantised and the
𝟐𝑳
energy of the energy levels is
𝒑𝟐 𝒉𝟐 𝒉𝟐 (44)
𝝐= = (𝒏𝟐𝒙 + 𝒏𝟐𝒚 + 𝒏𝟐𝒛 ) = 𝒏 𝟐
𝟐𝒎 𝟖𝒎𝑳𝟐 𝟐
𝟖𝒎𝑽𝟑
Here note that 𝑽 is the volume of the cube. 𝝐 explicitily depends on 𝒏 and 𝑽.
A careful look at equation (44) shows that, a quanum state of the particle in a cube is
designated by a set of three quantum numbers and 𝒏𝒙 , 𝒏𝒚 and 𝒏𝒛 . For the lowest level this
set is (𝒏𝒙 = 𝟏, 𝒏𝒚 = 𝟏, 𝒏𝒛 = 𝟏) with 𝝐𝟏 = 𝝐𝟏,𝟏,𝟏 = 𝟑𝝐.This level is said to have
degeneracy 𝒈𝟏 = 𝟏.
However, for the next level any one of the quantum numbers can be 2 leading to three
possibilities (2,1,1), (1,2,1) and (1,1,2) having the same energy 𝝐𝟐 = 𝝐𝟐,𝟏,𝟏 = 𝝐𝟏,𝟐,𝟏 =
𝝐𝟏,𝟏,𝟐 = 𝟔𝝐. So this level is said to have degeneracy 𝒈𝟐 = 𝟑 and so on.
There are many other examples where gegeneracy manifests itself in physical systems
𝟏
quantum mechanically. If a particle has spin 𝟐, it introduces a double degeneracy because
of two orientations of the spin.