MCH-013

GENERAL PHYSICAL
Indira Gandhi National CHEMISTRY
Open University
School of Sciences

Thermodynamics and
Statistical Thermodynamics
Block- 1
 MCH-013
GENERAL PHYSICAL
Indira Gandhi National CHEMISTRY
Open University
School of Sciences

VOL

1
THERMODYNAMICS, SOLID STATE AND
REACTION KINETICS-I

BLOCK 1
THERMODYNAMICS AND STATISTICAL
THERMODYNAMICS 5

BLOCK 2

SOLID STATE AND REACTION RATE

THEORIES 128
GENERAL PHYSICAL CHEMISTRY: INTRODUCTION
Dear learners, welcome to the first Core Course in Physical Chemistry titled, “General
Physical Chemistry”. This course is worth four credits and requires an average of about 120
hours of study time at your end. As is evident from the name of the course, it deals with
general aspects of Physical Chemistry. The contents of the course assume a basic
knowledge of physical chemistry at undergraduate level. It builds on that and focuses on the
concepts and provides opportunity to master them and suitably apply them. Wherever
required the concepts learnt earlier are suitably recalled. The course content has been
divided into a total of sixteen units that are organised into four blocks. The first two blocks
contain eight units and are organised into Volume-1. The remaining eight units are
distributed amongst two blocks which are organised into Volume-2.
The first block titled, Thermodynamics and Statistical Thermodynamics contains a total of
four units. The first two units deal with important areas in thermodynamics viz., Gibb’s energy
and Helmholtz energy and systems of variable composition. The next two units deal with
introductory aspects of statistical thermodynamics.
The second block titled, Solid State and Reaction Rate Theories again has four units, the
first two units are on solid state while the remaining two units deal with reaction rate theories.
The first unit concerns the fundamental aspects of solid state whereas the second unit deals
with different aspects of crystal symmetry. The third units deals with collision and transition
state theories of reaction rate in details whereas the fourth unit is dedicated to the theories of
unimolecular reactions.
Objectives
After studying the contents of this volume, you should be able to:
• derive expressions for the criterion for spontaneity for processes occurring under
constant temperature and constant volume / constant pressure conditions and predict the
spontaneity of reaction using them;
• define partial molar quantities, outline their significance, and describe the method for the
determination of partial molar volume;
• explain the term ‘chemical potential’, give its significance, and discuss the effect of
temperature and pressure on it;
• define permutation, configuration, microstate, weight of a configuration and dominant
configuration and calculate them for a system;
• derive an expression for the Boltzmann distribution function and discuss its constraints;
• explain the principle and describe the experimental set up of flow methods and relaxation
methods used for the determination of the rate of fast reactions;
• discuss the role of solvent and describe the process of reaction in reactions; derive the
expression for the rate of a reaction in terms of the encounter pair description;
• define and differentiate between the primary and secondary salt effects; and discuss the
effect of ionic strength on the rate of reaction;
• give Michaelis–Menten mechanism of enzyme catalysed reactions; derive an expression
for the rate of enzymatic reaction based on it and explain the terms involved; and
• define enzyme inhibition, explain its different mechanisms, and derive rate equations for
them.
2
 MCH-013
GENERAL PHYSICAL
Indira Gandhi National CHEMISTRY
Open University
School of Sciences

Block

1
THERMODYNAMICS AND STATISTICAL
THERMODYNAMICS
UNIT 1
Gibbs and Helmholtz’s Functions 7
UNIT 2
Systems of Variable Composition 43
UNIT 3
Introduction to Statistical Thermodynamics-I 73
UNIT 4
Introduction to Statistical Thermodynamics-II 94

3
Programme Design Committee
Prof. H. Surya Prakash Rao (Retd.) Prof. Tabrez Alam
Dept. of Chemistry Department of Chemistry
Pondicherry University Jamia Millia Islamia
Pondicherry – 605014 New Delhi –110025

Prof. Ashok Kumar Mishra Dr. Pritam Mukhopadhyay

Professor and Dean Academic Research Associate Professor,
Department of Chemistry School of Physical Sciences,
Indian Institute of Technology Madras Jawaharlal Nehru University.
Chennai – 600036, New Mehrauli Road, New Delhi- 110067
Prof. Ghanshayam Bez
Chemistry Department School of Sciences,
NEHU, IGNOU
Shillong-2722624 Prof. Sujatha Varma (Director)
Prof. Sunita Malhotra
Prof. Sunil Sharma Prof Bharat Inder Fozdar
Department of Chemistry Prof. Javed A. Farooqi
University of Delhi Prof. Sanjiv Kumar
Delhi-110007 Prof. Lalita S. Kumar
Prof. Kamalika Banerjee

Block Preparation Team

Prof. Sanjiv Kumar Prof. Ashok Kumar Mishra (Editor)
School of Sciences, Professor and Dean Academic Research
IGNOU , New Delhi-110068 Department of Chemistry
Indian Institute of Technology Madras
Dr. Neeti Misra Chennai – 600036
Associate Professor,
Acharya Narayan Dev College
University of Delhi
Kalka ji, New Delhi-110019
Course Coordinator: Prof. Sanjiv Kumar
Production
Mr.
Assistant Registrar (Pub.)

Acknowledgements: Sh. Deepak Kumar for word processing and Sh. Sarabjeet Singh for CRC
preparation.
April 2024
ÓIndira Gandhi National Open University, 2024
ISBN:
Disclaimer: Any material adapted from web-based resources in this volume are being used only for educational
purposes and not for commercial purposes.
All rights reserved. No part of this work may be reproduced in any form, by mimeograph or any other means, without
permission in writing from Indira Gandhi National Open University.
Further information on Indira Gandhi National Open University courses may be obtained from the University’s
office at Maidan Garhi, New Delhi-110 068 or IGNOU website [Link].
Printed and published on behalf of Indira Gandhi National Open University, New Delhi by Prof. Poornima Mital,
Director, School of Sciences.
Printed at:

4
 Thermodynamics and Statistical Thermodynamics
The first block of volume-1 is titled “Thermodynamics and Statistical Thermodynamics” and
contains four units. The first two units deal with two important areas of thermodynamics viz.,
Gibbs energy and Helmholtz energy and systems of variable composition. The remaining two
units are dedicated to the introductory aspects of statistical thermodynamics.

The first unit titled, ‘Gibbs energy and Helmholtz energy” deals with two important concepts
in thermodynamics viz., Gibb’s energy and Helmholtz energy. It strives to demonstrate the
inadequacy of entropy as a criterion of spontaneity and argues, introduces, and establish
Gibb’s energy and Helmholtz energy as the requisite criteria under the conditions of constant
pressure and constant volume respectively.

The second unit concerns ‘systems of variable composition’ and covers partial molar
properties, their determination and significance. Herein it takes up the issue of the effect of
change in composition of the system on its thermodynamic properties. Besides partial molar
properties, the unit also deals with the concept and significance of chemical potential.

The third unit titled, “Introduction to statistical thermodynamics-I” serves to delineate the
relationship between classical and statistical thermodynamics followed by introductory
aspects of probability theory. It covers the concepts of probability, permutations,
configurations, and probability distribution functions.

The fourth unit titled “Introduction to statistical thermodynamics-II” focuses on Boltzmann

distribution function -the cornerstone of statistical thermodynamics, partition function and
canonical ensemble.

Objectives
After studying this block, you should be able to:

• derive an expression for the criterion for spontaneity for processes occurring under
constant temperature and constant volume conditions and predict the spontaneity of
reaction using them;

• derive an expression for the criterion for spontaneity for processes occurring under
constant temperature and constant pressure conditions and predict the spontaneity of
reaction using them;

• state the significance of Gibbs energy and Helmholtz energy;

• define partial molar quantities, outline their significance, and describe the method for the
determination of partial molar volume;

• explain the term ‘chemical potential’, give its significance, and discuss the effect of
temperature and pressure on it;

• discuss the relationship between classical and statistical thermodynamics;

• differentiate between permutation and configuration and calculate them for systems with
small number of objects;

• state Stirling’s approximation, give its significance and establish its validity for systems
with large number of particles; 5
• define microstate, configuration, weight of a configuration and dominant configuration and
calculate them for a system;

• derive an expression for the Boltzmann distribution function give its significance and
outline the constraints operative in it;

• briefly describe the Lagrange’s method of undetermined multipliers; and

• derive expressions for translational, rotational, vibrational, and electronic partition

6
Unit 1 Gibbs and Helmholtz Functions

UNIT 1

GIBBS AND HELMHOLTZ

FUNCTIONS

Structure
1.1 Introduction 1.4 Properties of Helmholtz energy
Expected Learning Outcomes and Gibbs energy
1.2 Helmholtz function Temperature and Volume
Dependence of Helmholtz Energy
Helmholtz Energy Change: Criteria
for Spontaneity and Equilibrium Temperature and Pressure
Dependence of Gibbs Energy
Physical Significance of Helmholtz
Energy Maxwell’s Relations
1.3 Gibbs Function 1.5 Gibbs Energy Change for a
Reaction
Gibbs Energy Change: Criteria for
Spontaneity and Equilibrium 1.6 Summary
Physical Significance of Gibbs 1.7 Terminal Questions
Energy
1.8 Answers
Relation between Helmholtz
Energy and Gibbs Energy

1.1 INTRODUCTION
In your earlier classes you would have learnt about energetics and must be
familiar with its meaning, importance and terminology. We are sure that you
know about spontaneous reactions and their importance in chemistry
laboratory and industry. In this unit we will recall spontaneity and take up two
important thermodynamic properties that serve as a criterion for the
spontaneity and equilibrium of thermodynamic processes. You may recall that
the two state functions viz., internal energy, and enthalpy that arise in the
context of First law of thermodynamics do not serve as a criterion for
spontaneity. On the other hand, the state function, entropy that arises while
dealing with spontaneous processes does provide a criterion for spontaneity
but is inconvenient to use.

We will begin the unit by discussing the difficulty in using entropy change as a
criterion for spontaneity of a thermodynamic process. Herein, we will establish
the need of new state functions for predicting the spontaneity based on the 7
Block1 Thermodynamics and Statistical Thermodynamics

properties of the system only. Thereafter, we will take up the processes

occurring under constant temperature and constant volume conditions and
work out the criterion for spontaneity and equilibrium in terms of a new state
function viz., Helmholtz energy. Then, we will work out the criterion for
spontaneity for the processes occurring under constant temperature and
constant pressure conditions. In this context, we will define a yet another state
function viz., Gibbs energy. Having introduced two important thermodynamic
functions, we will discuss their physical significance and the interrelationship
between the two.

This will be followed by a discussion on the properties of Helmholtz energy

and Gibbs energy. Herein, we will take up the temperature and volume
dependence of Helmholtz energy and temperature and pressure dependence
of Gibbs energy. We will also take up Maxwell’s relations and discuss their
significance. Towards the end of the unit, we will be defining and evaluating
the Gibbs energy change for a chemical reaction.

In the next unit we will take up systems of variable compositions.

Expected Learning Outcomes

After studying this unit, you should be able to:

❖ explain why entropy change is not a good criterion for spontaneity of a

thermodynamic process?

❖ derive Clausius inequality and outline its significance;

❖ derive Helmholtz function as a criterion for spontaneity for processes

occurring under constant temperature and constant volume conditions;

❖ give the criteria for spontaneity and equilibrium, for a process occurring
under constant temperature and constant volume conditions, in terms of
the sign and magnitude of ∆U and ∆S;

❖ predict the spontaneity of processes occurring under constant

temperature and constant volume conditions;

❖ derive Gibbs function as a criterion for spontaneity for processes

occurring under constant temperature and constant pressure conditions;

❖ give the criteria for spontaneity and equilibrium, for a process occurring
under constant temperature and constant pressure conditions, in terms
of the sign and magnitude of ∆H and ∆S;

❖ predict the spontaneity of a process occurring under constant

temperature and constant pressure conditions;

❖ check under what conditions a given thermodynamic system will be able

to attain equilibrium;

❖ derive the relation between Gibbs energy and Helmholtz energy;

❖ explain the change in Helmholtz function with the change in temperature

and volume;
8
Unit 1 Gibbs and Helmholtz Functions

❖ explain the change in Gibbs function with the change in temperature and
pressure;

❖ give the expression for Gibbs-Helmholtz equation and state its

significance;

❖ derive Maxwell’s relations and outline their significance; and

❖ define and calculate the Gibbs energy change for a reaction.

1.2 HELMHOLTZ FUNCTION

In your earlier classes you have learnt about various aspects of energetics.
You would recall that the First Law of thermodynamics concerns the
conservation of energy. In the course of its study, we come across two
important state functions viz., internal energy and enthalpy. You have also
learnt about spontaneous processes i.e., the processes that occurs on their
own under a given set of conditions without the aid of any external agency. As
a chemist it is of interest to us to be able to predict the feasibility of a chemical
reaction. However, the internal energy and enthalpy cannot predict the
spontaneity of a process. This led to the introduction of a new state function
viz., entropy that could serve as a criterion for spontaneity but could not be
used effectively. Let us see why? However, before that let us derive two
important relations that would help us in defining new state functions that may
serve as criteria for spontaneity.

Clausius inequality

Let us consider two processes taking place between same initial and final
states i.e., we go from state 1 to state 2 by two different routes. We assume
that one of these processes takes place under reversible condition and the
other one takes place under irreversible condition. Since internal energy is a
state function, the change in internal energy for these processes will be equal
i.e.,

dUrev = dUirr …(1.1)

You would recall that according to the first law of thermodynamics

dU = q + w …(1.2)

In terms of the first law, we can write Eq. (1.1) as

qrev + wrev = qirr + wirr …(1.3)

Rearranging the expression, we get

qrev − qirr = wirr − wrev …(1.4)

Here, w represents the work done on the system and as you know that the
work done on the system is positive and its magnitude during reversible
process is greater than that during irreversible process, we can write,

dw rev dw irr …(1.5) 9

Block1 Thermodynamics and Statistical Thermodynamics

Rearranging, we get

dw irr − dw rev 0 …(1.6)

Substituting in Eq. (1.4), we get

dqrev−dqirr> 0 …(1.7)

Dividing the Eq. (1.7) throughout by T, we get

dqrev dqirr
− 0 …(1.8)
T T

You would recall that for a reversible process, the change in entropy
qrev
dS = …(1.9)
T

So, we can write Eq. (1.8) as

dqirr dqirr
dS − 0 or dS …(1.10)
T T
The sum of entropy
changes for the Combining the expression for the entropy change during reversible, Eq. (1.9)
system and the and irreversible, Eq. (1.10) processes, we get
entropy change of the
surroundings dq
dS …(1.11)
increases in case of a T
spontaneous process.
The Eq. (1.11) is known as Clausius inequality. This expression will be used
later to arrive at the expressions for the spontaneity of thermodynamic
processes under different conditions.

Combined First and Second Law: The Fundamental Thermodynamic

Equation

Once again, we take the expression for the First law of thermodynamics
dU = q + w …(1.2)

The fundamental and consider a case of a closed system containing gas having a constant
thermodynamic composition which undergoes a reversible change in such a way that only
equation, Eq. (2.14) work of expansion is permitted. Since we are considering a reversible
applies to any change expansion, the opposing pressure ( pext ) would be equal to the pressure of the
(reversible or
irreversible) of a gas. Therefore, we can write,
closed system that dwrev = −pdV …(1.12)
does not involve any
non-expansion work. From Eq. (1.9) for a reversible process we can write
qrev = TdS …(1.13)

Substituting from Eq. (1.12) and Eq. (1.13) in Eq. (1.2), we get
dU = TdS − pdV …(1.14)

This combined expression of the First and Second Laws is called the
fundamental thermodynamic equation. It applies to any change of a closed
10 system that does not involve any non-expansion work. Since internal energy is
Unit 1 Gibbs and Helmholtz Functions

a state function, dU is an exact differential. That is, its value is independent of

the path. In other words, the value of dU is same whether the change is
brought about irreversibly or reversibly.

Limited Utility of Entropy Change as a Criterion for Spontaneity.

Now, let us take up the question of limited utility of entropy change as a

criterion for spontaneity. You would have studied about entropy and
spontaneity in your earlier classes. You would recall that the direction of an
arbitrary spontaneous (irreversible) process is predicted by the following
expression.
Ssystem + Ssurr = Suniverse 0 …(1.15)

Whereas, for reversible processes, the following is true.

Ssystem + Ssurr = Suniverse = 0 …(1.16)

In other words, we can say that for an isolated system, the total entropy of the For an isolated
universe increases in an irreversible process and remains constant for a system the total
reversible processes; it never decreases. Thus, we see that entropy change entropy of the
universe increases in
does provide a criterion for establishing the spontaneity of a process.
an irreversible
However, there is a limitation in applying this criterion to different systems as it
process and remains
requires calculating the entropy changes for the system as well as for the constant for
surroundings. Calculation of the entropy change for the system is relatively reversible processes.
straightforward, but it is quite tedious (sometimes not practical) to calculate the However, it never
entropy change for the surroundings. Therefore, we need a criterion for decreases.
spontaneity that depends only on the properties of the ‘system’. We would
introduce such criterion in the next subsection.

1.2.1 Helmholtz Energy Change: Criterion for

Spontaneity and Equilibrium
You know that in chemistry the processes (chemical reactions) are carried out
under constant temperature and constant pressure (or constant volume)
conditions. Accordingly, we have two functions to predict the spontaneity of a
process depending on the reaction conditions. Let us begin with the processes
carried out under constant temperature and constant volume conditions and
deduce a function that can act as the criterion of spontaneity in such cases.

You know that according to the first law of thermodynamic, a process involving
only mechanical work, (i.e., p-V work) the change in internal energy is given by
the following expression
dU = q − pdV …(1.17)

Let us consider the process wherein only heat flow is taking place between the
system and the surroundings, and no mechanical work is involved. It implies
that in this case, the volume is kept constant, that is, dV = 0. Therefore, we Herman von Helmholtz
can write (courtesy Wikipedia)
dU = q …(1.18)

According to Clausius inequality for a spontaneous flow of energy between the

q
system and the surroundings, dS . i.e.,
T 11
Block1 Thermodynamics and Statistical Thermodynamics

Td S q …(1.19)

Comparing Eqs.(1.18 and 1.19) we can write

dU TdS or dU − TdS 0 …(1.20)

Since, U and S are state functions, a combination of two state functions will
also be a state function. We can introduce this new state function as,

A = U − TS …(1.21)

A is known as Helmholtz function in honour of German physicist Hermann

This Helmholtz Ludwig Ferdinand von Helmholtz who had developed this concept. Since the
energy was earlier
internal energy and the term TS has the dimensions of energy (K JK −1 = J) ,
known as Helmholtz
free energy. Helmholtz function A will have the dimensions of energy, therefore it is
referred to as Helmholtz energy and its SI unit is Joule. When the process is
carried out under constant temperature condition, the change in Helmholtz
energy is given by the expression

dA = dU − TdS …(1.22)

(dA)T ,V 0 serves as Comparing Eqs. (1.20) and (1.22) we get

the criteria for (dA)T ,V 0 …(1.23)
spontaneous change
and equilibrium for a We had initially assumed a spontaneous heat flow at constant volume and
process taking place
have applied the constant temperature condition on Eq. (1.21). Thus, the
at constant
temperature and Eq. (1.23) provides a criterion for the spontaneity and equilibrium for a process
volume. taking place at constant volume and constant temperature. Thus, under the
conditions of constant volume and constant temperature, a process will be
spontaneous if it is accompanied by a decrease in Helmholtz energy (∆A<0)
and would continue to be so till there is a decrease in Helmholtz energy. When
it reaches a state of minimum Helmholtz energy such that there is no further
The spontaneous
change i.e., A = 0 , the system will be at equilibrium.
processes are always
irreversible in nature For a finite change, the Eq. (1.22) can be written as
and have A 0 .
Whereas for a A = U −T S …(1.24)
reversible process,
the system is always For a finite change taking place at constant temperature and volume, the
at equilibrium, and criterion of spontaneity and equilibrium is given as
A=0 .
( A)T ,V 0 …(1.25)

Conditions for Spontaneity and Equilibrium

We have shown that for a process to be spontaneous under constant

temperature and constant volume conditions the Helmholtz energy should
decrease(∆A<0). This implies that for a non-spontaneous process the
Helmholtz energy will increase. This means that the non-spontaneous process
will be spontaneous in the reverse direction. The question then arises that
what happens when there is no change in Helmholtz energy. In such cases,
the process will neither go in the forward direction nor in the backward
direction. That is, the system will be at equilibrium when there is no change in
the Helmholtz energy i.e., ( A)T ,V = 0
12
Unit 1 Gibbs and Helmholtz Functions

Combining Eq. (1.24) and the criteria of spontaneity (∆A<0) we find that the
spontaneity of the process at constant temperature and constant volume
depends on the combination of change in internal energy and the change in
entropy of the system. That is, for a spontaneous process
U −T S 0 …(1.26)

There are several different combinations of internal energy change and

entropy change possible for a given process. Let us now see whether and
under what conditions the process will be spontaneous for each such
combination of U and S .
Case 1: U 0 and S 0
In this case, the value of U − T S will be negative for all temperatures; hence,
the process will be spontaneous at all the temperatures.
Case 2: U 0 and S 0
In this case the value of U − T S will always be positive; hence, the process
will never be spontaneous.
Case 3: U 0 and S 0
In this case, U will be negative and T S will be positive. At a certain
temperature the magnitude of these will be equal i.e., U = T S and A = 0 .
At this temperature the system will be at equilibrium. This temperature may be
called as equilibrium temperature and can be calculated as
U
T = …(1.27)
S

Such a process can be spontaneous only when the magnitude of U is more

than T S , i.e., U T S . This condition will be met at temperatures lower
than the equilibrium temperature and the process will be spontaneous. At
temperatures higher than the equilibrium temperature, T S will be greater
than U , and the process will be non-spontaneous.

Case 4: U 0 and S 0
In this case, both U and T S terms will be positive. Here again, at a certain
temperature the magnitude of these terms will be equal i.e., U = T S and
A = 0 . At this temperature the system will be at equilibrium and this
temperature can be calculated using Eq. (1.27)
Such a process can be spontaneous only when the magnitude of U is less
than T S . That is
U T S …(1.28)

This condition will be met at temperatures higher than the equilibrium

temperature and the process will be spontaneous. At temperature lower than
the equilibrium temperature, U will be greater than T S , and the process
will be non-spontaneous.
Case 5: S = 0
In such cases, the process will be spontaneous only for exothermic processes,
that is, U 0 .Such a process will be at equilibrium when U = 0 . 13
Block1 Thermodynamics and Statistical Thermodynamics

Case 6: U = 0

In such cases, the process will be spontaneous only when S 0 .Such a

process will be at equilibrium when S = 0 .

The conditions for spontaneity and equilibrium for different combinations of

U and S are compiled in Table 1.1

Table 1.1: The conditions for spontaneity and equilibrium for different
combinations of ∆U and ∆S for a process.

S. Combination of Condition for Condition for Remark

No. U and S . spontaneity equilibrium

1 U 0 and U −T S 0 A=0 Spontaneous at all

S 0 temperatures; equilibrium
not observed

2 U 0 and U −T S 0 A=0 Nonspontaneous at all

S 0 temperatures; equilibrium
not observed

3 U 0 and U T S U =T S Equilibrium at
S 0
U
T =
S

Spontaneous at
temperatures lower than
the equilibrium temperature

4 U 0 and U T S U =T S Equilibrium at
S 0
U
T =
S

Spontaneous at
temperatures higher than
the equilibrium temperature

5 S =0 U 0 U =0 Spontaneous only for

exothermic processes

6 U =0 S 0 S =0 The entropy must increase

for the process to be
spontaneous

Let us take an example to see how we can predict the spontaneity of a

process by computing its Helmholtz energy change by using the temperature
and the data on change in internal energy and change in entropy.

Example1: During an isochoric, isothermal process taking place at 25°C, the

internal energy of the system decreased by 20 kJ mol-1 and the
entropy increased by 50 J K−1 mol−1 . Calculate the change in the
Helmholtz energy. Will the process be spontaneous?
14
Unit 1 Gibbs and Helmholtz Functions

Solution: The process is taking place at constant volume and constant

temperature.

We are given: U = − 20kJ mol−1 = − 20 103 J mol−1

S = 50 J K −1 mol−1

T =25oC=298K

We know that the change in Helmholtz energy = A = U − T S

Substituting the values of T , U and S in the above expression, we get

A = − 20000 J mol−1 = − (298K 50 JK −1 mol−1 )

= − 34900 J mol−1 = −34.9 kJK −1 mol−1 )

Since the Helmholtz energy change has a negative sign, that is Helmholtz
energy decreases in the process, therefore, the process will be spontaneous.

Having learnt about the criterion of spontaneity for the processes occurring under
constant temperature and constant volume conditions in terms of Helmholtz
energy change. Let us now see the physical significance of Helmholtz energy.

1.2.2 Physical Significance of Helmholtz Energy

As per Eq. (1.22) the change in Helmholtz energy is given as

dA = dU − TdS …(1.22)
The symbol A for
We can recall the expressions for the first law of thermodynamics as under Helmholtz energy is
derived from German
dU = q + w …(1.2)
word, ‘Arbeit’
meaning work.
As per Eq. (1.13) for a system undergoing a reversible process according to
the second law of thermodynamics, we can write
qrev = TdS …(1.13)

Combining the expressions for the first and second law we can write

dU = TdS + wrev …(1.29)

Substituting the expression for dU in Eq. (1.22) we get,

dA = TdS + wrev − TdS …(1.30)

dA = wrev …(1.31)

According to Eq. (1.31), the change in Helmholtz energy is equal to the work
done during isothermal reversible process. When the Helmholtz energy
decreases, the value of w rev will be negative indicating that the work is done by
the system. Therefore, the decrease in Helmholtz energy is equal to the work
done by the system during isothermal reversible process. This work includes
all the types of work that the system can do. Thus, Helmholtz energy change
is related to the maximum amount of work that can be obtained from a system,
it is also known as work function. 15
Block1 Thermodynamics and Statistical Thermodynamics

In the previous section we have learnt that the criterion for spontaneity and
equilibrium for a process under constant temperature and constant volume
conditions is given in terms of change in Helmholtz energy. However, we
generally study the process under constant temperature and constant
pressure conditions as these are easier to maintain in the laboratory. Let us
take up the case of spontaneity of a process under constant temperature and
constant pressure conditions and determine a suitable criterion for the same.

1.3 GIBBS FUNCTION

Let us consider spontaneous heat flow taking place between the system and
the surrounding at a constant pressure. You know that such a heat flow is
equal to the change in enthalpy of the system, i.e.,

dH = qp …(1.32)

Using Clausius inequality, Eq. (1.11) we can write,

Td S q …(1.33)

Combining Eq.(1.32) and Eq. (1.33) we can write

TdS dH …(1.34)

It can be rearranged as

dH − TdS 0 …(1.35)

The Eq. (1.35) holds for a process taking place under constant temperature
and constant pressure conditions. Since the left-hand side of the equation is a
combination of two state functions, H and S, it will also be a state function.
This combination is defined as a new state function called Gibbs function or
Gibbs energy in honour of American chemist J. W. Gibbs who first came up
with a relation to predict spontaneity of a process. It is represented as G and
mathematically defined as:

G = H − TS …(1.36)

Since enthalpy ,H and TS term (K JK−1=J) , both have units of energy, G also
will also have the units of energy, hence the name Gibbs energy. The SI units
of Gibbs energy is Joule. When the temperature is kept constant, the
infinitesimal change in Gibbs energy can be written as

dG = dH − TdS …(1.37)

Eq. (1.35) is valid for the processes taking place under constant temperature
and constant pressure conditions. Comparing Eq. (1.35) and Eq.(1.37) we can
write that for a process taking place at constant temperature and pressure, the
Gibbs energy change of the system is less than or equal to zero. That is,

16 (dG)T ,p 0 …(1.38)
Unit 1 Gibbs and Helmholtz Functions

1.3.1 Gibbs Energy: Criteria for Spontaneity and Equilibrium

While deriving the criterion for spontaneity of a process occurring under
constant temperature and constant pressure conditions in terms of Gibbs
energy we used Clausius inequality. It implies that we had taken into account
both reversible and irreversible processes. Let us now take the reversible and
irreversible processes separately. For reversible process, we can write from
Eq. (1.13)

TdS = qrev …(1.13)

And as per Eq. (1.32) we have,

dH = qp …(1.32)

Comparing Eq. (1.13) and Eq. (1.32) we can write,

dH = TdS. or dH − TdS = 0. …(1.39)

This implies that, for a reversible process.

dG = 0 …(1.40)

Now, for an irreversible process

TdS qirr . …(1.41)

Substituting in Eq. (1.37) we get

dG 0 . …(1.42)

Since all spontaneous reactions are irreversible, Eq. (1.42) serves as a criteria
for spontaneity for a process occurring under constant temperature and
constant pressure conditions.

Relation between Spontaneity, Gibbs Energy and Entropy of Universe

You would recall that according to the second law of thermodynamics, the
entropy change for an isolated system is equal to or greater than zero for a
spontaneous process. That is,

dS 0 …(1.43)

In an isolated system, dS = 0 for a reversible process, and dS 0 for an

irreversible process. We know that generally a system interacts primarily with
its immediate surroundings. Thus, the system and its immediate surroundings
can be assumed to be isolated from the rest of the universe. Such a system is
said to be a composite isolated system and the total entropy change for
such a system is given by the expression.

dStotal = dSsystem + dSsurroundings …(1.44)

Since the surroundings are much larger than the system, the gain or loss of
energy as heat by the surroundings is taken as reversible process irrespective
of whether the actual process is reversible or not. Therefore, we can write, 17
Block1 Thermodynamics and Statistical Thermodynamics

qsurroundings qsystem
dSsurroundings = =− …(1.45)
T T

We know that at constant pressure, qsystem = dHsystem

Therefore, we can write Eq. (1.44) as

dHsystem
dStotal = dSsystem − …(1.46)
T

Simplifying,

TdSsystem − dHsystem
dStotal = …(1.47)
T

dGsystem
dStotal = − …(1.48)
T

Since, dStotal increases for a spontaneous irreversible process, therefore, for a

spontaneous process at constant temperature and pressure, we can say that

dGsystem
− 0 or dGsystem 0 …(1.49)
T

It is important to note This is the desired criteria for the spontaneity of a process under constant
that ∆G depends only temperature and constant pressure conditions. Therefore, for a spontaneous
on the initial and final process, Gibbs energy decreases and Gibbs energy change for the system is
states of the system.
It does not provide
negative. Thus, we can predict the spontaneity of a process based on the
any information about thermodynamic properties of the system. For a finite change, the criterion for
the rate or the spontaneity of a process at constant temperature and pressure is given as
kinetics of the
process. ( G)T ,p 0. …(1.50)

If the Gibbs energy decreases for a spontaneous process, it implies that for a
non-spontaneous process the Gibbs energy will increase. If the change in
Gibb energy (dG) is negative for the process in the forward direction, then dG
will be positive for the process in the backward direction. At equilibrium,

( G)T ,p = 0. …(1.51)

Using Eq. (1.35) we can say that both dH and TdS contribute to the
spontaneity of a process. For a finite change at constant temperature and
pressure, the change in Gibbs energy is written as

( G)T ,p = H − T S. …(1.52)

There are different combinations of enthalpy change and entropy change

possible for different processes. As discussed in case of Helmholtz energy, we
can consider different cases and see whether and under what conditions the
process will be spontaneous for each such combination. The conditions for
spontaneity and equilibrium for different combinations of ∆H and ∆S are
18 compiled in Table 1.2
Unit 1 Gibbs and Helmholtz Functions

Table 1.2: The conditions for spontaneity and equilibrium for different
combinations of ∆H and ∆S.

S. Combination of Condition for Condition for

Remark
No. ∆Hand S spontaneity equilibrium
1 H 0 and H −T S 0 G=0 Spontaneous at all temperatures;
S 0 equilibrium not observed
2 H 0 and H −T S 0 G=0 Nonspontaneous at all
S 0 temperatures; equilibrium not
observed
3 H 0 and H T S T S = H Equilibrium at
S 0 H
T =
S

Spontaneous at temperatures lower

than the equilibrium temperature
4 H 0 and H T S T S = H Equilibrium at
S 0 H
T =
S

Spontaneous at temperatures
higher than the equilibrium
temperature
5 S =0 H 0 H =0 Spontaneous only for exothermic
processes

6 H =0 S 0 S =0 The entropy must increase for the

process to be spontaneous

Let us take an example to see how we can predict the spontaneity of a

process by computing its Gibbs energy change.

Example 1.2: For a system undergoing isothermal and isobaric process, the
changes in enthalpy and entropy were found to be 58 kJ mol-1
and 185 J K-1mol-1, respectively.

a) Will the process be spontaneous at 330K?

b) Calculate the temperature at which the system will attain equilibrium

assuming that the enthalpy and entropy changes are not affected by the
change in temperature.

Solution: a) For the process to be spontaneous ∆G must be negative. We

know that ( G)T ,p = H − T S

−1 −1
Given: H = 58 kJ mol = 58000 J mol

S = 185 JK −1 mol−1 and T = 330 K

Substituting the values, we get

−1 −1 −1
58000 J mol − (330K ×185 J K mol )
=58000 − 61050 J mol−1 = −3050 Jmol−1

Since the computed value of the Gibbs energy change is negative, the
process will be spontaneous. 19
Block1 Thermodynamics and Statistical Thermodynamics

b) The system will attain equilibrium when | H | = | T S |

Since the temperature is constant, we can write

| H|
T =
| S|

Substituting the values

| H| 58000 Jmol-1
T = == = 313.51K
| S| 185 JK −1 mol-1

Thus, the system will attain equilibrium at 313.51 K.

Having learnt about the criterion for spontaneity of the processes occurring under
constant temperature and constant pressure conditions in terms of Gibbs energy
change, let us now see the physical significance of Gibbs energy.

1.3.2 Physical Significance of Gibbs Energy

Let us start with the expression for Gibbs energy, G = H − TS and substituting
As the change in
Gibbs energy is equal
for enthalpy, H = U + pV we can write.
to the energy that is G = U + pV − TS …(1.53)
available with the
system to do useful The change in Gibbs energy can then be written as
work, Gibbs energy
dG = dU + pdV + Vdp − TdS − SdT …(1.54)
was earlier known as
free energy or Gibbs
You would recall from above that by combining the expressions for the first
free energy.
and second law we can write

The decrease in dU = TdS + wrev …(1.29)

Gibbs energy is equal
to the maximum Substituting the expression for dU in Eq. (1.54) and simplifying we get,
amount of useful work
(i.e., non-expansion dG = +TdS + w rev + pdV + Vdp − TdS − SdT
work) that can be …(1.55)
obtained from a dG = w rev + pdV + Vdp − SdT
system undergoing a
reversible process
At constant temperature and constant pressure, i.e., dT = 0 and dp = 0, the
under constant
temperature and
Eq. (1.55) becomes
pressure conditions.
dG = wrev + pdV …(1.56)
The increase in free
energy of the system wrev is the total work done on the system under reversible conditions at
is equal to the work constant temperature and pressure. This can include all types of work like
done on the system electrical work, chemical work, or mechanical work. You know that the
over and above the mechanical work is the work of expansion or compression and is given by the
mechanical work. following expression,

w mechnical = − pdV …(1.57)

Substituting in Eq. (1.56) we get

20 dG = wrev − wmechanical …(1.58)

Unit 1 Gibbs and Helmholtz Functions

Here, w rev − w mechnical is the non-expansion work. We can call it the net work
done under reversible condition and write.

dG = w rev − w mechanical = w net , rev …(1.59)

As you know that the work done by the system is (− w), therefore,
−dG = ( − w net ,rev ) …(1.60)

Recall that the system does maximum work under reversible conditions, i.e.,
| wnet,rev| is the maximum amount of work that a system can do. Therefore, we
can say that the decrease in Gibbs energy is equal to the maximum amount of
useful work (i.e., non-expansion work) that can be obtained from a system
undergoing a reversible process under constant temperature and pressure
conditions.

1.4.1 Relation between Gibbs Energy and Helmholtz

Energy
Starting with Eq. (1.36) and the relation between enthalpy and internal energy
H = U + pV, we can write
G = H − TS = U + pV − TS …(1.53)

We know from Eq. (1.21), that A = U −TS. Substituting it in Eq. (1.53), we get
G = A + pV …(1.61)

Let the initial state of the system be G1, A1, p1, V1and T1 and the final state of
the system be G2, A2, p2, V2and T2. We can then write
G1 = A1 + p1V1 and G2 = A2 + p2V2 …(1.62)

The change in Gibbs energy will be given by the following expression

G2 − G1 = ( A2 − A1 ) + ( p2V2 − p1V1 )

G = A + ( p2V2 − p1V1 ) …(1.63)

You know that for an ideal gas

p2V2 = nRT2 and p1V1 = nRT1 …(1.64)

Therefore, by substituting the above expressions, the Eq. (1.63) can be

modified for an ideal gas as
G = A + nR (T2 − T1 ) …(1.65)

If the temperature is kept constant during the change i.e.,T2 = T1. We get
For an isothermal
G= A …(1.66)
process involving an
Thus, for an isothermal pressure-volume change process involving an ideal ideal gas, the change
gas, the change in Gibbs energy and change in Helmholtz energy are equal. in Gibbs energy and
change in Helmholtz
Having learnt about the Helmholtz and Gibbs energy, their physical energy are equal.
significance and interrelationship we will take up the properties of Helmholtz
energy and Gibbs energy. However, before that try to solve following simple
questions to assess your understanding. 21
Block1 Thermodynamics and Statistical Thermodynamics

SAQ 1
Predict whether the process will be spontaneous or non-spontaneous for the
following cases at 300 K:
(i) U = 35kJ mol−1 and S = 180J K−1 mol−1

(ii) U = 15.6kJ mol−1 and S = − 210J K−1 mol−1

(iii) U = 78.5kJ mol−1 and S = 80.5J K −1 mol−1

SAQ 2
The change in internal energy for a system undergoing isothermal process at
constant volume was found to be 25kJ mol−1 and the entropy change was
found to be 90J K−1 mol−1 .Calculate the temperature at which the system will
attain equilibrium assuming that the internal energy change and entropy
change do not vary with temperature.

SAQ 3
Based on the given values of ∆H and ∆S predict which one of the following
processes will be spontaneous at 300 K?
(i) H = −2.5kJ mol−1 and S = 150J K −1 mol−1

(ii) H = 18 kJ mol−1 and S = − 25J K−1 mol−1

(iii) H = 18 kJ mol−1 and S = 0J K−1 mol−1

(iv) H = 0 kJ mol−1 and S = 25J K−1 mol−1

1.4 PROPERTIES OF GIBBS ENERGY AND

HELMHOLTZ ENERGY
In previous section you have studied the importance of Gibbs energy change
and Helmholtz energy change in predicting spontaneity and equilibrium when
certain parameters were kept constant. In this section we will further elaborate
their properties which can be used for calculating various thermodynamic
properties of the system. We begin with the temperature and pressure
dependence of Gibbs energy.

1.4.2 Temperature and Pressure Dependence of

Gibbs Energy
In the process of deriving the expressions for Gibbs energy we have
considered the temperature and pressure to be constant. This implies that
while the process is taking place, we do not change the temperature and
pressure of the system. However, the process can be carried out at different
sets of temperatures and pressures. The question is what will be the effect of
performing the process at a different temperature (s) or pressure (s). Further,
22 you may know that the standard tables of thermodynamic data on Gibbs
Unit 1 Gibbs and Helmholtz Functions

energy, enthalpy or entropy are compiled for the standard state. If we have a
system at a temperature different from standard state then we need to
consider the temperature dependence of G to get the required data.

In other words, how does the Gibbs energy of the system change on changing
the temperature or pressure. Here we will consider two possibilities viz.,

i. Variation of Gibbs energy with temperature at constant pressure, and

ii. Variation of Gibbs energy with pressure at constant temperature,

Let us consider a system undergoing reversible p-V change. Using Eq. (1.53)
we write,
G = U + pV − TS …(1.53)

The change in Gibbs energy is given by the expression

dG = dU + pdV + VdP − TdS − SdT …(1.54)

For a reversible p-V change, we can write from Eq. (1.14)

dU = TdS − pdV …(1.14)

Substituting in Eq. (1.54) and simplifying the expression, we get

dG = TdS − pdV + pdV + Vdp − TdS − SdT

dG = −SdT + Vdp …(1.67)

Thus, we can say that for a system undergoing reversible p-V change the
Gibbs energy change depends upon temperature and pressure. This can be
mathematically expressed as
G = f (T , p ) …(1.68)

The change in Gibbs energy in terms of change in temperature and pressure

can be written as

G G
dG = dT + dp …(1.69)
T p p T

The derivative in first term on the right of Eq. (1.69) gives the rate of change of
Gibbs energy with temperature at constant pressure and dT is the total change
in the temperature. Similarly, the derivative in second term on right gives the
rate of change of Gibbs energy with pressure at constant temperature and dp
is the total change in the pressure. Comparing the coefficients of dT and dp in
Eqs. (1.67) and (1.69) we get

G G
= −S and =V …(1.70)
T p p T

This implies that the rate of change of Gibbs energy with temperature at
constant pressure equals the decrease in the entropy. Also, the rate of change
of Gibbs energy with pressure at constant temperature equals increase in the
volume.
23
Block1 Thermodynamics and Statistical Thermodynamics

Let us now take up the variation of Gibbs energy with temperature at constant
pressure.

i. Variation of Gibbs energy with temperature at constant pressure:

Gibbs-Helmholtz equation

The Eq.(1.69) gives the total change in the Gibbs energy when the
temperature and pressure change by dT and dp respectively. However, it is
always easier to study the variation of thermodynamic property with change in
any one parameter, keeping the rest of the parameters constant. Therefore,
while studying the changes in Gibbs energy due to change in temperature and
pressure, we keep one variable constant and change the second variable.
According to Eq. (1.67), at constant pressure, we can write

(dG)p = −SdT …(1.71)

Now, to find the total change in Gibbs energy with the change in temperature
we have to integrate the above expression. That is,
G2 T2

d G = − SdT …(1.72)
G1 T1

As the entropy is temperature dependent, it is taken inside the integral.

Integrating the above expression, we get
T2

G2 − G1 = − SdT …(1.73)
T1

Since the entropy increases with the increase in temperature, the right-hand
side of the above expression will be less than zero. Therefore, we can say that
Gibbs energy decreases with the increase in temperature.

Using Eq. (1.36) and Eq. (1.70) we can write

G
G = H +T …(1.74)
T p

Rearranging, we get

G
H = G −T …(1.75)
T p

Dividing Eq. (1.75) throughout by T2 we get

G
G −T
H T p
= …(1.76)
T2 T 2

To simplify the above expression, we take a function G/T and differentiate it

with respect to T at constant p. This gives

G
T −G
(G / T ) T p
=
24 T p T2 …(1.77)
Unit 1 Gibbs and Helmholtz Functions

Comparing Eqs. (1.76) and (1.77) we get

(G / T ) H
=−
T p T2 …(1.78)

The Eq. (1.78) is known as Gibbs-Helmholtz equation. According to Gibbs-

Helmholtz equation, if the enthalpy of the system is known, we can calculate
the variation of G/T with the change in temperature. This relation is very useful
while studying chemical equilibrium.

ii. Variation of Gibbs energy with pressure at constant temperature

According to Eq. (1.70), the change in Gibbs energy with pressure at constant
temperature is given by the following expression

G
=V
p T …(1.70)

Therefore, we can write

(dG )T = Vdp …(1.79)

To find the total change in Gibbs energy with the change in pressure we need
to integrate the above expression. That is,
G2 p2

dG = Vdp …(1.80)
G1 p1

Since the volume varies with the change in pressure, it is taken inside the
integral. However, for solids and liquids, we can assume that the volume is
constant for small changes in pressure. Therefore, on integration we get,
G2 − G1 = V ( p2 − p1 ) …(1.81)

From this expression we can conclude that for solids and liquids there is an
increase in Gibbs energy with the increase in pressure. On the other hand, in
case of gases we cannot assume that the volume is constant. So, we need to
use a relation between volume and pressure of the gas. Assuming that the gas
behaves ideally, we can write
nRT
V= …(1.82)
p

Substituting in Eq. (1.80) and integrating within limits, we get

p2
nRT p
G2 − G1 = dp = nRT ln 2 …(1.83)
p1
p p1

As the pressure is increased, (p2 p1 ) the right-hand side of the above

equation will be greater than zero. Hence, for ideal gas, the Gibbs energy
increases with the increase in pressure. We know that for an ideal gas under
isothermal condition, p2V2 = p1V1.

Substituting in Eq. (1.83), we get

V1
G = G2 − G1 = nRT ln …(1.84)
V2
25
Block1 Thermodynamics and Statistical Thermodynamics

Eq. (1.84) can be used to calculate the Gibbs energy change during
isothermal reversible expansion or compression of an ideal gas. Let us take an
example to understand it.

Example 1.3: One mole of an ideal gas undergoes isothermal expansion from
10 L to 100 L at 300 K. Calculate the change in Gibbs energy
for the process.

Solution: As per Eq. (1.84)

V1 V
G = nRT ln = 2.303 nRT log 1
V2 V2

Given: n =1 mol, R = 8.314JK−1 mol−1,T =300K,V1 = 10 L,V2 = 100L

Substituting the values,

10
G = 2.303 1mol 8.314 JK −1 mol−1 300K log
100

G = −5744.14 J

Thus, the change in Gibbs energy for the process will be

G = −5744.14 J or − 5.744 kJ

Having learnt about the temperature and pressure dependence of Gibbs

energy let us now take up the temperature and volume dependence of
Helmholtz energy. However, before that answer the following simple questions
to assess your understanding.

SAQ 4
Two moles of an ideal gas are compressed isothermally at 298 K from a
pressure of 2.5 bar to 1 bar. Calculate the change in Gibbs energy and
change in molar Gibbs energy.

SAQ 5
The Gibbs energy of one mole of an ideal gas at 300 K and 1 bar is
−3.6 kJ mol−1. Calculate the Gibbs energy of this gas at 300 K and 2 bar.

1.4.3 Temperature and Volume Dependence of

Helmholtz Energy
In the process of deriving the expressions for Helmholtz energy we have
considered the temperature and volume to be constant. This implies that while
the process is taking place, we do not change the temperature and volume of
the system. However, the process can be carried out at different sets of
temperatures and volumes. The question is what will be the effect of
performing the process at a different temperature (s) or volume (s). In other
words, how does the Helmholtz energy of the system change on changing the
temperature and volume of the system. Here we will consider two possibilities
26 viz.,
Unit 1 Gibbs and Helmholtz Functions

i. Variation of Helmholtz energy with temperature at constant volume, and

ii. Variation of Helmholtz energy with volume at constant temperature,
Let us start with Eq. (1.21)

A = U − TS …(1.21)

The change in Helmholtz energy can be expressed as

dA = dU − TdS − SdT …(1.85)

Since for a reversible p-V change, we can write from Eq. (1.14)

dU = TdS − pdV …(1.14)

Substituting in Eq. (1.85) and simplifying, we get

dA = TdS − pdV-TdS − SdT

dA = −SdT − pdV …(1.86)

Thus, we can say that for a system undergoing reversible p-V change the
Helmholtz energy depends upon temperature and volume. This can be
mathematically expressed as
A = f (T ,V ) …(1.87)

The change in Helmholtz energy with the change in temperature and volume
is given as
A A
dA = dT + dV …(1.88)
T V V T

In this expression, Eq. (1.88), the derivative in first term on the right gives the
rate of change of Helmholtz energy with temperature at constant volume and
dT is the total change in the temperature. Similarly, the derivative in second
term gives the rate of change of Helmholtz energy with volume at constant
temperature and dV is the total change in the volume. Comparing the
coefficients of dT and dV in Eqs. (1.86) and (1.88) we get

A A
= − S and = −p …(1.89)
T V V T

This implies that the rate of change of Helmholtz energy with temperature at
constant volume equals the decrease in the entropy and the rate of change of
Helmholtz energy with volume at constant temperature equals the decrease in
the pressure.

Let us now take up the variation of Helmholtz energy with temperature at

constant volume.

i. Variation of Helmholtz energy with temperature at constant volume:

Gibbs-Helmholtz equation

The Eq.(1.88) gives the total change in the Helmholtz energy when the
temperature and volume change by dT and dV respectively. However, it is
always easier to study the variation of thermodynamic property with change in 27
Block1 Thermodynamics and Statistical Thermodynamics

any one parameter, keeping all other parameters constant. Therefore, while
studying the changes in Helmholtz energy due to change in temperature and
volume, we keep one variable constant and change the second variable.
According to Eq. (1.86), at constant volume, we can write
(dA)v = −SdT …(1.86)

To find the total change in Helmholtz energy with the change in temperature
we need to integrate the above expression. That is,
A2 T2

dA = − SdT …(1.90)
A1 T1

The entropy is taken inside the integral as it is temperature dependent.

Integrating the left side of the equation and applying limits, we get
T2

A = A2 − A1 = − SdT …(1.91)
T1

Since absolute entropy is always greater than zero, this implies that Helmholtz
energy decreases with the increase in temperature at constant volume.
The variation of Helmholtz energy with temperature can be used to derive
another form of Gibbs-Helmholtz equation. Starting from Eq. (1.21) and using
Eq. (1.89), we can write
A
A = U +T …(1.92)
T V

Rearranging the equation, we get

A
U = A −T …(1.93)
T V

Dividing Eq. (1.93) throughout by T2 we get

A
A −T
U T
= V
…(1.94)
T2 T2
To simplify the above expression, we take a function A/T and differentiate it
with respect to T at constant V. This gives
A
( T)
A
=
T
T V
−A
…(1.95)
T T2
V

Comparing the above expression with Eq. (1.94), we get

( AT ) =−
U
…(1.96)
T T2
V

Eq. (1.96) is the Gibbs-Helmholtz equation in terms of Helmholtz energy.

According to this form of Gibbs-Helmholtz equation, if the internal energy of
the system is known, we can calculate the variation of A with the change in
T
28 the temperature.
Unit 1 Gibbs and Helmholtz Functions

ii. Variation of Helmholtz energy with volume at constant temperature

To understand the variation of Helmholtz energy with volume at constant

temperature we start with the following expression obtained from Eq. (1.86),
under constant temperature conditions,
(dA)T = − pdV …(1.97)

To find the total change in Helmholtz energy with the change in volume at
constant temperature we need to integrate the above expression. That is
A2 V2

dA = − pdV …(1.98)
A1 V1

nRT
We know that for an ideal gas p = . Substituting in Eq. (1.98) and
V
integrating we get
V2
1 V
A2 − A1 = −nRT dV = −nRT ln 2
V1
V V1
…(1.99)
V
or A = nRT ln 1
V2

In terms of pressure, Eq. (1.99) can be written as

p2
A2 − A1 = nRT ln …(1.100)
p1

According to Eq. (1.100), during the expansion of an ideal gas under

isothermal conditions, the Helmholtz energy of the system decreases. It is
interesting to note that the expressions for the change in Gibbs energy for an
isothermal expansion of an ideal gas (Eq. (1.67)) and the one for the change
in Helmholtz energy for isothermal expansion (Eq. (1.99)) are identical. This
can be rationalised in terms of Eq. (1.66) where we showed that the change in
Gibbs energy and Helmholtz energy for an isothermal process are equal. Let
us take the data of Example 3 discussed above and calculate the change in
Helmholtz energy for the process.

Example 1.4: One mole of an ideal gas undergoes isothermal expansion from
10 L to 100 L at 300 K. Calculate the change in Helmholtz
energy for the process.

Solution: As per Eq. (1.99)

V1
A = 2.303 nRT log
V2

Given: n =1 mol, R = 8.314JK−1 mol−1,T =300K,V1 = 10 L,V2 = 100L

Substituting the values in the above equation we get
10
A = 2.303 1mol 8.314 JK −1 mol−1 300K log
100
A = −5744.14 J

Thus, the change in Helmholtz energy for the process will be

29
Block1 Thermodynamics and Statistical Thermodynamics

G = −5744.14J or − 5.744kJ mol-1

1.4.4 Maxwell’s Relations

We have so far defined four different state functions viz., U, H, A, and G. Each
one of these has dimensions of energy. A question comes up is that in what
way are these different? You would recall that internal energy (U) represents
the sum of the kinetic and potential energies of all the components of the
system. On the other hand, the enthalpy, H equals the heat flows into or out of
the system for a thermodynamic process carried out under constant pressure
conditions. Further, as discussed above, the Helmholtz energy A represents
the maximum total work that can be produced in a process whereas, the Gibbs
energy G gives the maximum non-expansion work that the process can
produce. In addition, the change in Helmholtz energy and change in Gibbs
energy provide criteria for the spontaneity of a process under constant T & V
and constant T & p conditions respectively.

We know that the macroscopic variables i.e., T (temperature), p (pressure), V

(volume) and m (mass) of the system can be measured directly. Further, some
properties like, density and specific volume can be determined easily by using
these. On the other hand, the properties like internal energy, enthalpy, and
entropy cannot be measured directly or determined by using the above
properties. It is, therefore, desirable that we develop some fundamental
relations between the thermodynamic properties that cannot be measured
directly in terms of easily measurable properties.

Therefore, the next issue we take up is to develop the relations for the
dependence of state functions on the macroscopic variables i.e., T
(temperature), S (entropy), p (pressure), and V (volume) of the system. For a
simple compressible system, the state functions are related to the derivatives
of properties p, V, T and S with respect to each other. These are very
important relations and are known as Maxwell’s relations or Maxwell’s
equations. These can be derived from the expressions corresponding to the
exact differentials of the state functions. Let us first work out these
expressions.

We begin by recalling Eq. (1.14)

dU = T dS − pdV …(1.14)

As mentioned above, this equation is referred to as the fundamental

equation of thermodynamics and represents the exact differential of the
state function U. We can use this expression to workout similar expressions
for the exact or total differentials of the other state functions. We begin with
enthalpy and recall its definition, i.e.,
H = U + pV …(1.101)

Differentiating it, we get

dH = dU + pdV + VdP …(1.102)

Substituting Eq. (1.14) into Eq. (1.102), we get

dH = TdS − pdV + pdV + Vdp …(1.103)
30
Unit 1 Gibbs and Helmholtz Functions

Simplifying, we get the desired expression as

dH = T dS + V dp. …(1.104)

Again, for Helmholtz energy we begin with the definition of Helmholtz energy,
i.e.,

A = U − TS …(1.21)

Differentiating it, we get

dA = dU − TdS − SdT …(1.105)

Substituting Eq. (1.14) into Eq. (1.105), gives

dA = TdS − pdV − TdS − SdT …(1.106)

Simplifying, we get the desired expression as

dA = − pdV − SdT …(1.107)

Similarly, for Gibbs energy the desired expression can be determined to be

dG = VdP − SdT …(1.108)

The expressions [(1.14), (1.104)(1.107)and (1.108)] derived above represent

the exact differentials of the state functions. Sometimes these are called the
Gibbs equations. These show how the dependent variable on the left side
varies as a function of changes in two independent variables on the right side.
For example, U is dependent on S and V. These independent variables are
the natural variables of the corresponding dependent variable. Different state
functions and the corresponding exact differentials are compiled in Table 1.1,
along with the thermodynamic functions indicating the independent variables
of these state functions.

Table 1.1: The state functions, expression (fundamental equations) for

their exact differentials and the corresponding thermodynamic
functions

The expression or fundamental Thermodynamic

State Function
equations for exact differentials function

U dU = TdS − pdV U = f (S,V )

H dH = TdS + Vdp H = f (S, p )

A dA = − pdV − SdT A = f (V ,T )

G dG = Vdp − SdT G = f ( p,T )

There are two important aspects of these fundamental equation:

• Each of these fundamental equations is equally basic and is always valid.

In a way, these give four equivalent ways of conveying the same
information. The choice to use them depends on the set of independent
variables (S,V ) , (S, p ) , (T ,V ) , or (T , p ) appropriate for a given system.
31
Block1 Thermodynamics and Statistical Thermodynamics

• The fundamental equations are derived assuming a reversible process.

However, since these represent total differentials, these apply equally well
to the irreversible processes also.

We are now equipped to derive the Maxwell relations. As stated above, these
give the variation of one thermodynamic property with respect to the other
thermodynamic property under certain conditions. Starting with the combined
form of first and second laws of thermodynamics for a reversible p-V change,
we can write

dU = TdS − pdV ...(1.14)

Differentiating Eq. (1.14) with respect to S at constant V we get

U
=T …(1.109)
S V

Differentiating Eq. (1.109) with respect to V at constant S we get

U T
= …(1.110)
V S V S
V S

Similarly, differentiating Eq. (1.14) with respect to V at constant S we get

U
= −p …(1.111)
V S

Differentiating with respect to S at constant V we get

U p
=− …(1.112)
V S S V
S V

Since U is a state function, Euler’s reciprocity relation is valid. Therefore,

T p
=− ...(1.113)
V S S V

Eq. (1.13) is the first Maxwell relation. For deriving the second Maxwell
relation, we start with the Gibbs equation related to enthalpy, i.e.,

dH = TdS + Vdp ...(1.104)

Differentiating with respect to p at constant S we get

H
=V …(1.114)
p S

Differentiating the above equation with respect to S at constant p we get

H V
= …(1.115)
S p S p
S p

32 Similarly, differentiating Eq. (1.104) with respect to S at constant p gives

Unit 1 Gibbs and Helmholtz Functions

H
=T …(1.116)
S p

Differentiating the above equation with respect to p at constant S we get

H T
= …(1.117)
p S p S
p S

Since H is a state function, using Euler’s reciprocity relation we can write,

V T
=− ...(1.118)
S p p S

Eq. (1.118) is the second Maxwell relation. Similarly, we can derive Maxwell
relations by starting with the Gibbs equations related to Gibbs energy, and
Helmholtz energy respectively i.e.,

dA = −PdV − SdT …(1.107)

dG = −SdT + Vdp …(1.108)

The Maxwell relations so obtained are,

S p
= …(1.119)
V T T V

V S
=− …(1.120)
T p p T

Significance of Maxwell Relations

As stated above, Maxwell equations relate the state functions to the Maxwell relations
derivatives of properties p, V, T and S with respect to each other. Therefore, provide relationships
these can be used to calculate the variations in those properties which cannot between seemingly
unconnected
be measured directly through experiments. For example, the variation of
thermodynamic
entropy with volume cannot be determined experimentally. However, variation functions.
of pressure with temperature at constant volume can be experimentally For example, these
studied. Therefore, (∂S/∂V)T can be calculated with the help of Eq. (1.119). can be used to
Similarly, the variation of entropy with pressure can be determined with the determine the change
help of Eq. (1.120) by studying the variation of volume with temperature at a in entropy by simply
measuring the
constant pressure.
changes in easily
Having learnt about Maxwell relations, let us discuss about the Gibbs energy measurable
properties like p,V
change for a reaction. However, before that answer the following simple
and T.
question to assess your understanding.

SAQ 6
Derive the following expression for Maxwell relation by using Gibbs equation
corresponding to Helmholtz function.

S p
=
V T T V

33
Block1 Thermodynamics and Statistical Thermodynamics

1.5 GIBBS ENERGY CHANGE FOR A REACTION

To calculate the Gibbs energy change for a reaction we assume that at the
start of the reaction only the reactants are present in their respective standard
states and when the reaction is complete, only products are present in their
respective standard states. It is also assumed that the temperature and the
pressure remain constant during the reaction. Under these conditions the
standard Gibbs energy change for the reaction can be calculated in terms of
the Gibbs energy of formation of all the reactants and the products. Standard
Gibbs energy of formation of a compound ( Gfo ) is defined as the Gibbs
energy change taking place during the formation of a compound in its standard
state from its constituent elements in their respective standard states at
constant temperature. For each element in its standard state, the Gibbs
energy is assigned a zero value. The standard Gibbs energy change for a
reaction is represented as, = G o . Mathematically, we can write the following
expression for the standard Gibbs energy change for a reaction,

Go = Gfo ( products ) − Gfo (reac tan ts ) …(1.121)

Alternatively, the standard Gibbs energy change of a reaction can also be

calculated from the change in standard enthalpy of formation and standard
entropy change values of a reaction at a given temperature.

Go = H o − T S o …(1.122)

Where, the change in standard enthalpy of formation and standard entropy

change values of a reaction are as given below

Ho = Hfo ( products ) − Hfo (reactants ) …(1.123)

So = S o ( products ) − S o (reactants ) …(1.124)

Let us take some examples to learn the calculation of standard Gibbs energy
change for a reaction.

Example 1.5: Calculate the standard Gibbs energy change for the following
reaction at 300 K using the data given below.

C6H12O6(s) + 6O2 (g) → 6CO2(g) + 6H2O(l)

Substance C6H12O6(s) CO2(g) H2O(l) O2(g)

Gfo (kJmol−1 ) −913 −395 −237 0

Solution: We know that according to Eq. (1.123) the standard Gibbs energy
change for a reaction is given as

Go = Gfo (products) − Gfo (reactants)

The standard Gibbs energy change for the formation of reactants can be
34 calculated as
Unit 1 Gibbs and Helmholtz Functions

= Gfo (reactants) = ( Gfo (C6H12O6 )) + (6 Gfo (O2 ))

= ( −913 + (6 0) )kJ = −913kJmol−1

Similarly, for products the standard Gibbs energy change can be calculated as

Gfo ( products ) = (6 Gfo (H2O)) + (6 Gfo (CO2 ))

= (( −237 6) + ( −395 6))kJ = −3792kJmol−1

Substituting in the Eq. (1.101) we get,

Go = −3792kJ − (−913kJ) = −2879kJmol−1

Thus, the Gibbs energy change for the given reaction will be, = −2879kJ mol-1

Example 1.5: Calculate the standard Gibbs energy change for the following
reaction at 300 K from the data given in the table below.
N2 (g) + 3H2 (g) → 2NH3 (g)

Substance N2 H2 NH3

Hfo (kJmol−1 ) 0 0 -46

So (JK−1mol−1 ) 192 131 193

Also, predict whether the reaction will be spontaneous under the given
conditions.

Solution: We know that according to Eq. (1.122) the standard Gibbs energy
change for a reaction is given as

Go = H o − T S o

The change in standard enthalpy of formation is given as

Ho = Hfo ( products ) − Hfo (reactants )

Substituting the values for the standard enthalpy changes for the products and
the reactants, we get

H o = (2 Hfo (NH3 )) − { Hfo (N2 ) + 3 Hfo (H2 )}

= {(−46 2) − (0 + (0 3))}kJmol−1 = −92kJmol−1

The standard entropy change for the reaction is given as

So = S o ( products ) − S o (reactants )

Substituting the values for the standard entropies for the products and the
reactants, we get

So = 2 So (NH3 ) − {So (N2 ) + 3 S o (H2 )}

= {(2 193) − (192 + (3 131))} JK −1mol−1

= (386 − 585)JK−1mol−1 = −199JK −1mol−1

35
Block1 Thermodynamics and Statistical Thermodynamics

Substituting the values of the change in standard enthalpy of formation and

standard entropy change of the reaction in the expression for standard Gibbs
energy change, we get

Go = {( −92.0kJmol−1 ) − (300K −199 JK −1mol−1 )}

= −32300 Jmol−1 = −32.3kJmol−1

Since, the standard Gibbs energy change for the reaction is negative, the
reaction will be spontaneous.

Let us now sum up what have we learnt in this unit. However, before that
answer the following simple question to assess your understanding.

SAQ 7
Calculate the standard Gibbs energy change for the following reaction at
300Kfrom the data given in the table below.

2N2O5 (g ) → 4NO2 (g ) + O2 (g )

Substance N2 O 5 NO2 O2
o −1
H (kJmol )
f
11.3 33.2 0

So (JK−1mol−1 ) 355.7 240 205.1

Also, predict whether the reaction will be spontaneous under the given
conditions.

1.6 SUMMARY
In this unit we have discussed about the need, significance and properties of
two important thermodynamic functions viz., Gibbs energy and Helmholtz
energy. These provide criterion for the spontaneity of thermodynamic
processes in terms of the properties of the system. We began the unit by
discussing the difficulty in using entropy change as a criterion for spontaneity
and outlined the need of new functions for predicting the spontaneity based on
the properties of the system only.

In this context initially we considered the processes occurring under constant

temperature and constant volume conditions and worked out the criterion for
their spontaneity in terms of a new function called Helmholtz energy. It was
demonstrated that the Helmholtz energy for a process must decrease for such
a process to be spontaneous. Also, for equilibrium the value of ∆A would be
zero. Similarly, the criterion for spontaneity in terms of Gibbs energy was
worked out for the processes occurring under constant temperature and
constant pressure conditions. Herein, it was shown that the Gibbs energy for a
process must decrease for such a process to be spontaneous and remain
unaltered for the process to be at equilibrium.

Thereafter, we discussed about the properties of Helmholtz and Gibbs

functions. It was shown that both the functions, A and G, decrease with the
increase in temperature. Under isothermal conditions, it was shown that A
36
Unit 1 Gibbs and Helmholtz Functions

increases with the increase in volume and G increases with the increase in
pressure. Further, we derived two important versions of Gibbs-Helmholtz
relation which could be used to understand the variation of G/T and A/T
respectively with temperature

This was followed by defining the Standard Gibbs energy change for a
reaction in terms of the standard Gibbs energies for the reactants and
products. It was also shown that the standard Gibbs energy change for a
reaction could be expressed in terms of the standard enthalpy change and
standard entropy changes. In the end we worked out different Maxwell’s
relations and discussed their significance.

1.6 TERMINAL QUESTIONS

1. Define spontaneity. Under what conditions of

i) ΔU and ΔS

ii) ΔH and ΔS

a reaction is always spontaneous.

2. When ΔH > 0 and ΔS < 0 a reaction is never spontaneous. Give reason.

3. What is work function? State the conditions when change in Gibbs energy
is equal to work function.

4. Calculate the change in

i) Helmholtz energy and

ii) Gibbs energy

when two moles of an ideal gas are allowed to expand isothermally at a

temperature of 300 K from 5 bar to 1 bar.

5. Derive the following relations using appropriate expressions.

A
a. = −S
T V

A
b. = −p
V T

G
c. =V
p T

G
d. = −S
T p

6. Define standard Gibbs energy change for a reaction.

7. Calculate the standard Gibbs energy change for the following reaction at
300 K using the data provided.
N 2 O(g) + 3H2 (g) → N2H4 (g)+H2O(g) 37
Block1 Thermodynamics and Statistical Thermodynamics

Substance N2O H2 N2H4 H2O

Hof (kJ mol−1 ) 82 0 50 -242

So (J K −1) 220 131 120 189

1.7 ANSWERS

Self-Assessment Questions
1. A process is spontaneous if ∆U−T∆S < 0.

(i) U = 35kJ and S = 180JK-1

U − T S = 35000J − 300K×180JK-1 =-19000J.

Since the value is negative, the reaction will be spontaneous.

(ii) U = 15.6kJ and S = −210JK-1

U − T S = −15600J − 300 K −210JK-1 =47400J.

Since the value is positive, the reaction will be non-spontaneous.

(iii) U = 78.5 kJ and S = −80.5JK-1

U − T S = −78500J − (300 K −80.5JK-1 )=54350J.

Since the value is positive, the reaction will be non-spontaneous.

2. Given: U = 25 kJ, S = 90JK-1

At equilibrium, U = T S

U 25000 J
T = = = 277.78K
S 90 JK −1

Thus, the system will be at equilibrium at 277.8 K

3. The process is spontaneous when ∆G < 0 i.e, ∆H−T∆S <0.

(i) H = −2.5 kJ, S = 150JK-1

Since H 0, S 0 G 0. The process is spontaneous

(ii) H = 18 kJ and S = −25JK-1

Since H 0, S 0, G 0. The process is non-spontaneous.

(iii) H = 18 kJ and S = 0J K-1

Since H 0, S = 0, G 0. The process is non-spontaneous.

(iv) H = 0 kJ and S = 25J K-1

Since H 0, S 0, G 0. The process is spontaneous.

4. Given: n = 2mol,T = 298K, p1 = 2.5bar, p2 = 1bar, R = 8.314JK −1mol−1

38
Unit 1 Gibbs and Helmholtz Functions

p2
G = 2.303 nRT log
p1
1bar
= 2.303 2mol 8.314 JK −1mol−1 298K log
2.5bar

G = −1971.85 J

Molar Gibbs energy change = ∆G/n = −1971.85/2 = −985.9 J

5. Given: n = 1 mol,T = 300K , p1 = 1 bar, p2 = 2 bar,G1 = 3.6 kJ

p2
G = G2 − G1 = 2.303 nRT log
p1
p2
G2 = G1 + 2.303 nRT log
p1
2bar
= −3600 J + 2.303 1mol 8.314 JK −1mol−1 300K log
1bar
G2 = −3600 J + 1728.99 J = −1871.01J

6. We know that the Gibbs expression corresponding to Helmholtz energy is.

dA = −SdT − pdV

Differentiating with respect to T at constant V we get

A
= −S
T V

Differentiating the above equation with respect to V at constant T we get

A S
=−
V T V T
V T

Differentiating the Gibbs equation with respect to V at constant T we get

A
= −p
V T

Differentiating the above equation with respect to T at constant V we get

A p
=−
T V T V
T V

Now, since A is a state function, Euler’s reciprocity relation is valid.

Therefore, we can write.

S p
=
V T T V

This is the desired Maxwell relation.

7. We know that the standard Gibbs energy change for a reaction is given as

Go = H o − T S o

The change in standard enthalpy of formation is given as

39
Block1 Thermodynamics and Statistical Thermodynamics

Ho = Hfo ( products ) − Hfo (reactants )

Substituting the values for the standard enthalpy changes for the products
and the reactants, we get

H o = (2 Hfo (NH3 )) − { Hfo (N2 ) + 3 Hfo (H2 )}

= {(−46 2) − (0 + (0 3))}kJmol−1 = −92kJmol−1

The standard entropy change for the reaction is given as

So = S o ( products ) − S o (reactants )

Substituting the values for the standard entropies for the products and the
reactants, we get,

So = 2 So (NH3 ) − {So (N2 ) + 3 S o (H2 )}

= {(2 193) − (192 + (3 131))} JK −1mol−1

= (386 − 585)JK−1mol−1 = −199JK −1mol−1

Substituting the values of the change in standard enthalpy of formation

and standard entropy change of the reaction in the expression for
standard Gibbs energy change, we get

Go = {( −92.0kJmol−1 ) − (300K −199 JK −1mol−1 )}

= −32300 Jmol−1 = −32.3kJmol−1

Since, the standard Gibbs energy change for the reaction is negative, the
reaction will be spontaneous.

Terminal Questions
1. Spontaneity refers to the occurrence of a thermodynamic process on its
own under a given set of conditions without the aid of any external
agency.

I. The reaction is always spontaneous if ∆U−T∆S< 0. Therefore, the

process will be spontaneous when ∆U< 0 and ∆S >0.

II. The reaction is always spontaneous if ∆H−T∆S< 0. This will happen

when ∆H< 0 and ∆S >0.

2. When ∆H> 0 and ∆S <0; ∆G = ∆H −T∆S will always be positive. Since the
Gibbs energy change under the given set of conditions is always positive,
the reaction will always be non-spontaneous.

3. The decrease in Helmholtz energy in an isothermal reversible process is

called work function as it is equal to all the types of work that the system
can do.

Gibbs energy is equal to work function, that is, ∆G = ∆A for an ideal gas at
constant temperature.

40 4. We know that:
Unit 1 Gibbs and Helmholtz Functions

I. Helmholtz energy change for an isothermal reversible process is

given as
p2
A = 2.303 nRT log
p1

Given: n = 2mol,T = 300K, p1 = 5bar, p2 = 1bar, R = 8.314J K −1mol−1

Substituting the values, we get.

5bar
A = 2.303 2mol 8.314 JK −1mol−1 300K log = 8030 J = 8.03kJ
1bar

Since, for an ideal gas at constant temperature, ∆G = ∆A, the change

in Gibbs energy will be =8030 J =8.03 kJ

5. a. For a reversible process

dA = −SdT − pdV

i.e., A = f (T ,V )

A A
dA = dT + dV
T V V T

A
Comparing the coefficients of dT we get = −S
T V

A
b. Comparing the coefficients of dp we get = −p
V T

c. For a reversible process

dG = −SdT + Vdp

G = f (T , p )

G G
dG = dT + dp
T p p T

G
Comparing the coefficients of dp we get =V
p T

G
d. Comparing the coefficients of dT we get = −S
T p

6. Standard Gibbs energy of a reaction is the difference between the total

standard Gibbs energy of formation of the products and the total standard
Gibbs energy of formation of the reactants. Mathematically,

Go = Gfo (products) − Gfo (reactants)

7. N2O(g)+ 3H2 (g) → N2H2 (g)+H2O(g)

o
= H o − T So
Given: T = 300 K 41
Block1 Thermodynamics and Statistical Thermodynamics

Ho Hfo ( products ) − Hfo (reac tan ts )

H o = Hfo (N2H4 ) + Hfo (H2O) − { Hfo (N2O) + 3 Hfo (H2 )}

= {(50 + −242) − (82 + 0 3)} kJ = −274kJ

So = S o (products) − S o (reactants)

So = So (N2H4 ) + So (H2O) − {So (N2O) + 3 So (H2 )}

= {(120 + 189) − (220 + 3 131)} JK−1

= (309 − 613)JK−1 = −304JK −1

Go = −274000J − (300K −304JK−1 ) = −182800J = −182.8kJ

42
Unit 2 Systems of Variable Compositions

UNIT 2

SYSTEMS OF VARIABLE
COMPOSITION

Structure
2.1 Introduction 2.4 Chemical Potential
Expected Learning Outcomes
Temperature and pressure
2.2 Partial Molar Quantities and Dependence of Chemical Potential
Their Significance
Significance of Chemical Potential
Relationship Between
Thermodynamic Property and 2.5 Mixture of Gases
Partial Molar Properties
Fugacity and Its Significance
2.3 Partial Molar Volume
2.6 Gibbs–Duhem Equation
Volume of Mixing
2.7 Summary
Experimental Determination of
2.8 Terminal Questions
Partial Molar Volume
2.9 Answers

2.1 INTRODUCTION
You have learnt earlier that the state of a thermodynamic system can be
described in terms of four thermodynamic parameters. These are temperature,
pressure, volume, and composition. You have also studied that various
thermodynamic properties of the system change when the temperature,
pressure and volume of the system are changed. Further, you also know that
the extensive properties of the system depend on the amount (moles) of the
substance present, for example, when the amount of the substance is
doubled, the value of the extensive properties of the system like internal
energy, enthalpy, entropy, etc. gets doubled. Have you wondered how the
thermodynamic properties will be affected if the composition of the system is
changed? In this unit, we will take up the issue of the effect of change in
composition of the system on its thermodynamic properties. This aspect is
particularly important in the application of thermodynamics to solutions and
gaseous mixtures which in turn have a significant role to play in chemical
process industries. 43
Block1 Thermodynamics and Statistical Thermodynamics

We will begin the unit by initiating a discussion on partial molar properties and
understanding their significance. This will be followed by a discussion on the
experimental determination of partial molar volume–an important partial molar
property of the system. Thereafter, we will delve on an important quantity viz.,
chemical potential and outline its significance. We will also discuss the effect
of pressure and temperature on this fundamental property. We will then take
up systems consisting of mixtures of gases. In this context we will bring in the
concept of fugacity and explain its significance. Towards the end of the unit,
we would derive Gibbs–Duhem equation and discuss its importance.

In the next unit we will take up fundamental aspects of statistical

thermodynamics.

Expected Learning Outcomes

After studying this unit, you should be able to:

❖ define partial molar quantities and outline their significance;

❖ differentiate between molar and partial molar properties;

❖ define and explain the term, partial molar volume;

❖ describe the method of experimental determination of partial molar

volume of a system;

❖ calculate partial molar volume of a component in the mixture by using

the mean molar volume versus mole fraction data;

❖ explain the term ‘chemical potential’ and state its significance;

❖ discuss the effect of temperature and pressure on chemical potential;

❖ explain the term ‘fugacity’ and state its significance; and

❖ derive Gibbs–Duhem equation and explain its importance.

2.2 PARTIAL MOLAR QUANTITIES AND THEIR

SIGNIFICANCE
As stated above, we wish to understand the effect of change in the
composition of the system on the thermodynamic properties of the system. As
you are aware the composition of a system comes into picture only when the
system consists of more than one substance. Such a system having two or
more substances is called a mixture. When the substances do not react with
each other, we call them components of the system. In this unit we will discuss
the systems having two components. You would learn that the familiar
thermodynamic properties of the system will need to be modified to include the
effect of change in the composition of the system.

We are sure that you are familiar with the terms like solute, solvent, and
solution. Here, we need to define a few more terms which will be useful in
44 understanding the concepts being taken up later. A solution having one
Unit 2 Systems of Variable Compositions
solvent and one solute has two components and is known as a binary
solution. So, when we dissolve sodium chloride in water, we get a binary
solution in which sodium chloride is the solute and water is the solvent. A
solution having one solvent and two solutes that do not react has three
components, therefore, it is known as a ternary solution. Similarly, a solute
and two immiscible solvents also make a ternary solution.

The solute can be in the solid, liquid, or gaseous state. A solution formed by
dissolving two solutes, say sodium chloride and potassium nitrate in water is
an example of a ternary solution. In case both the components are liquid, the
mixture is known as a solution only if both the liquids are completely miscible
e.g., a mixture of alcohol and water. When the system consists of two or more
gases, it is known as a mixture of gases. A mixture of two gases is known as a
binary mixture of gases.

To determine various properties of the mixture, we need to know how many

components are present in the mixture and what is the amount of each one of
them. You know that the amount of each component in the mixture is
expressed as the concentration of that component. You would have learnt
various ways to express the concentration of the components in a solution. In
this unit, we will mainly use molarity, molality, and mole fractions. Let us
quickly recall them.

Molarity, c, is defined as the moles of the solute per dm3 (litre) of the solution.
The concentration of component i is given as
ni 1000
ci = …(2.1)
V

Here, ni is the amount of component i in moles and V is the volume of the

solution in cm3. The unit of molarity is mol dm–3. Since, volume is temperature
dependent, molarity changes with the change in temperature.

Molality, m is defined at the amount of solute per kilogram of the solvent. The
molarity of component i is given as
ni 1000
mi = …(2.2)
w

Here, w is the mass of the solvent in gram. The unit of molality is mol kg–1. It is
important to note that contrary to molarity, the molality is not affected by the
change in temperature.

Mole fraction, xi of a component i is given as

ni
i = … (2.3)
ntotal

Here, ntotal is the total amount of all the components in the mixture in moles.
Mole fraction is unitless and it does not depend on the temperature of the
system. For a two–component system having nA moles of component A and nB
moles of component B, we can write,
nA nB
A = and B = …(2.4)
nA + nB nA + nB 45
Block1 Thermodynamics and Statistical Thermodynamics

Also,

A + B =1 …(2.5)

Solve the following simple questions to assess your understanding of

concentration terms defined above.

SAQ 1
Calculate the following:
a) the molarity of a solution whose 100 cm3 contain 0.630 g of oxalic acid
b) the molality of a solution prepared by dissolving 0.630 g of pure oxalic acid
in100 g of water.
c) You may use the molar mass of oxalic acid as 126 g mol–1.

SAQ 2
Calculate the mole fraction of ethanol in the following solutions.
a) a solution prepared by mixing 23 g of ethanol and 180 g of water, and
b) a solution prepared by mixing 18 g of water in 100 g of ethanol.
c) You can use the molar mass of ethanol and water as 46 g mol–1 and 18.0 g
mol–1, respectively.

2.2.1 Partial Molar Properties

You know that the extensive properties of a system depend on the amount (in
moles) of the substance. If we divide such a property by the number of moles,
we will get the corresponding molar properties i.e., property per mole of the
substance. For example, if we represent an extensive thermodynamic property
(V, U, H, A or G) by X, then the molar value of the property X will be given as

X Valueof thermodynamic property of the system

Xm = = …(2.6)
n amount of the system (inmoles)

You may note that the molar properties are represented by suffixing m to the
symbol of the property. For example, the volume of one mole of a substance,
is known as the molar volume and is represented as:

V
Vm = …(2.7)
n
Here, V is the volume of the system having n moles of the substance. Now,
you may ask that molar properties are fine but what is a partial molar property
and why do we need it? Also, when do we use molar properties and when do
we use partial molar properties? Let us answer the second question first. We
use molar properties when the system consists of a pure substance and partial
molar properties when the system has two or more components. This will get
clearer as we move ahead. Let us now try to understand what are partial molar
properties and why do we need them in the systems having two or more
46 components?
Unit 2 Systems of Variable Compositions
Let us write the mathematical expression for the change in thermodynamic
property, X, with the change in the composition of the system. First, we will do
so for the change in the thermodynamic property for a pure substance or a
one component system. You know that for a one component system, the
extensive properties like internal energy, enthalpy, entropy, and Gibbs energy
depend upon the temperature, pressure, and the amount of the substance.
This fact is represented mathematically as

X = f (T , p, n ) …(2.8)

That is, the property is a function of T, p, and n. Here, X is said to be a

dependent variable which depends on the values of T, p and n, which can be
varied independent of each other and are called independent variables. We
can take a case where, we keep two variables (say p and n) constant and vary
one of the variables (say T). This will bring in certain change in the property.
Similarly, we can change p or n while keeping (T & n)and (T & p) respectively
as constant. The total change in the thermodynamic property can then be
expressed as The terms

X X X
dX = dT + dp + dn …. (2.9) X
,
X
T p,n p T ,n
n T ,p T p,n p T ,n

X
Where, the differential in the first term on the right, represents the and
T p ,n

change in the property X when the temperature of the system is changed, X

while keeping the pressure p and amount n constant. In other words, this term n T ,p
represents the rate of change of the property, X as a function of temperature
when p and n are kept constant. Whereas dT represents the total change in are known as the
the temperature. The differential in the second term on right in Eq. (2.9) viz., partial derivatives and
X indicate how the
represents the change in the property X when the pressure is property X changes
p T ,n
when one of the
changed, while the temperature T and amount n are kept constant and dp is variables is changed
X while keeping the
the total change in pressure. Similarly, the third term, represents the other two variables
n T ,p
constant.
change in property X when the amount, n is changed while the temperature
and pressure are kept constant and dn is the total change in the amount.

Let us now extend the argument to a binary mixture which has n A moles of
component A and nB moles of component B and work out corresponding
expression for the change in the property X. The thermodynamic property in
this case will depend upon temperature, pressure, n A and nB . Therefore, the
expression for the dependent variable, X becomes.

X = f (T , p, nA , nB ) …. (2.10)

Where, T, p, n A and nB are the independent variables. Using similar method

as above, we can write the expression for the total change in X as follows

X X X X
dX = dT + dp + dnA + dnB …(2.11)
T p,nA ,nB p p,nA ,nB
nA T , p,nB
nB T , p,nA
47
Block1 Thermodynamics and Statistical Thermodynamics

You can note that the thermodynamic property X changes with the changes in
T, p, n A and nB . Now, if we wish to know the change in the property on
changing n A and nB while keeping the temperature and pressure as constant.
In such a case since dT = 0 and dp = 0 ; the Eq. (2.11) would get modified as

X X
dX = dn A + dnB …(2.12)
nA T , p,nB
nB T ,p,nA

X X
The terms and are the partial derivatives of the property
nA T , p,nB
nB T , p,nA

with respect to the number of moles of the components i.e., n A and nB ,

respectively. These terms are known as partial molar properties. Here,
X
represents the partial molar property for the component A while
nA T ,p,n
B

X
represents the partial molar property for the component B.
nB T , p,nA

To understand the physical significance of the term partial molar property, let
us consider a situation wherein we take a large volume of the binary mixture
and add one mole of component A to it i.e., (dnA = 1) .Since we have taken a
large volume, the mole fraction of component B practically does not change
i.e., (dnB = 0) . Substituting in Eq. (2.12) we get,

X
dX = …(2.13)
nA T , p,nB

Thus, the partial molar property of component A in a mixture can be defined as

the change in the property X of the mixture when one mole of component A is
added to a large volume of mixture at constant temperature and pressure
while keeping the amounts (number of moles) of remaining components
constant. The partial molar properties are represented either as X i ,m or X i
where the subscript i represents the component whose amount is being
varied and the subscript m (as stated before) or a bar on top of the symbol of
the property indicates the property to be molar. Using such representation for
The partial molar
partial molar properties, Eq.(2.12) can be written as
quantities are
intensive properties dX = X A,m dnA + X B,m dnB …(2.14)
and depend upon the
temperature,
Here, dX is the total change in property X due to change in composition at
pressure, and the
composition.
constant temperature and pressure. Similarly, other extensive thermodynamic
properties can also be written in terms of respective partial molar properties for
systems having more than one component. Various partial molar properties of
a binary mixture are summarised in Table 2.1.

It is important to note here that the partial molar quantities are intensive
properties. These depend upon the temperature, pressure, and composition.
In other words, the partial molar quantity will not change if the amount of a
component is changed in such a way that the concentration / composition of
48 the mixture does not change.
Unit 2 Systems of Variable Compositions
Table2.1:Expressions and symbols for different Partial Molar Properties for a
binary mixture

Thermodynamic Partial Molar Property

Property
Name Symbol Expression

Volume Partial molar volume V

V = f (T , p, n A , nB ) Vi ,m or Vi
ni T , p ,n j

Internal energy Ui ,m or Ui U
U = f (T , p, nA , nB ) Partial molar internal energy
ni T , p ,n j

Enthalpy H
H = f (T , p, nA , nB ) Partial molar enthalpy H i ,m or H i
ni T , p ,n j

Entropy S
S = f (T , p, nA , nB ) Partial molar entropy Si ,m or Si
ni T , p ,n j

Gibbs energy G
Partial molar Gibbs
G = f (T , p, nA , nB ) Gi ,m or Gi
energy(Chemical potential) ni T , p ,n j

For example, in a binary solution, if the number of moles of A as well as that of

B is doubled then the composition of the solution will not change. Therefore,
the partial molar properties will also not change. However, if the moles of A
are doubled and that of B are not changed, then the composition of the
solution changes. In this case the partial molar properties of the solution will
also change.

If we extend the argument to a multicomponent mixture, the thermodynamic

property will depend upon temperature, pressure, n A , nB and the amounts of
other components. That is,
X = f (T , p, nA , nB ,......) …. (2.15)

For an infinitesimal change in the temperature, pressure, and the amounts of

all the components, the change in the property X will be given as
X X X
dX = dT + dp + dn A
T p,nA ,nB ,n... p p,nA ,nB ,n...
nA T , p,nB ,n...
…. (2.16)
X X
+ dnB + dni + .....
nB T , p,nA ,n ...
ni T , p,nA ,nB ,n j ......

Under conditions of constant temperature and constant pressure, the

expression becomes
X X X
dX = dn A + dnB + dni + .... …. (2.17)
nA T , p,nB ,n ...
nB T , p,n A ,n...
ni T , p,n A ,nB ,n j ......

Using the representation of partial molar quantities, the expression becomes

dX = X A,m dnA + X B,mdnB + X i ,mdni + ..... …(2.18) 49
Block1 Thermodynamics and Statistical Thermodynamics

Having learnt about the partial molar quantities and their representation let us
learn about the use of partial molar properties. Partial molar properties are
quite important in thermodynamic studies of the systems having variable
compositions. The values of partial molar properties help us in deducing the
type of interactions between various components in a mixture. If the
composition of the mixture and the partial molar quantities of all the
components of the mixture at that composition are known, the value of any
thermodynamic property can be calculated at a given temperature, pressure,
and composition of the mixture. Let us derive a general mathematical
expression for calculating the thermodynamic property from the partial molar
properties in a binary mixture.

2.2.2 Relationship between Thermodynamic Property

and Partial Molar Properties
As you already know, the mole fraction for component A in a mixture
containing n A moles of A and nB moles of B can be written as

nA n
A = = A nA = A .n …(2.19)
nA + nB n

Here, n is equal to the sum of the number of moles of A and B. Similarly, for
component B, we can write.
nB = B .n …(2.20)

To keep the composition of the mixture constant, n A and nB should be

changed in such a way that A and B do not change. Let us increase the
total amount of A and B (i.e., n A + nB ) from n moles to kn moles. If we
represent the change in moles of A as dnA ; the change in moles of B as dnB
and the change in total moles as dn , then we can write
dn A = A dn …(2.21)

dnB = B dn …(2.22)

dn = kn − n = n(k − 1) …(2.23)

Here A and B are not changing as we change the amount of A and B.

Substituting the values of d n A and dnB in Eq. (2.14) we get,

dX = X A,m A dn + XB,m B dn …(2.24)

Now, to find the total value of X, Eq. (2.24) needs to be integrated from the
initial amount (n) of the mixture to the final amount (kn). Since, X is an
extensive property, if n is changed by k times, X will also change by k times.
Therefore, the integration of Eq. (2.24) will have the following expression
kX kn
dX = ( X A,m A + X B,m B )dn …(2.25)
X n

You would recall that we have changed the composition in such a way that the
mole fractions of the two components do not change. Since, the partial molar
quantities are also constant, the term in the bracket on the right side of the
50 equation will be constant.
Unit 2 Systems of Variable Compositions
So, we can write
kn
(kX − X ) = ( X A,m A + X B,m B ) dn
…(2.26)
n

Simplifying, we get

X (k − 1) = ( X A,m A + XB,m B )(k − 1)n …(2.27)

Substituting for A & B and simplifying, we get

X = X A,mnA + XB,mnB …(2.28)

This is the desired relationship between thermodynamic property, and the

partial molar properties of the components in a binary mixture. For a mixture
having more than two components this expression can be generalised as
follows
X= X i ,m ni …(2.29)
i

To have a clearer understanding of partial molar properties we take the case

of partial molar volume. The volume is a good choice as it is an extensive
property that can be easily visualised. However, before that answer the
following simple question to assess your understanding.

SAQ 3
Partial molar temperature for a mixture is irrelevant. Explain.

2.3 PARTIAL MOLAR VOLUME

We know that if we add 100 cm3 of water to one litre (1000 cm3) of water, the
total volume of water increases and becomes 1100 cm3. However, it is
observed that if 100 cm3 of water is added to one dm3of a binary aqueous
solution (say of ethanol and water) at a constant temperature and pressure,
the total volume of the solution generally increases, but it may or may not
become 1100 cm3. That is, the increase in volume of the mixture may not be
equal to 100 [Link] molar volume of water i.e., the volume of 1 mole of pure
water at 298 K and 1bar is 18.07 cm3 mol–1. So, when one mole of water is
added to a large volume of pure substance B, we expect the volume of the
resulting solution to increase by the molar volume of water (i.e.,18.07 cm3). In
some cases, the increase in the volume of the solution is equal to the molar
volume of the added component. However, it has been observed that
generally the increase in the volume of the solution is different from the molar
volume of the added component.

There are two questions now,

I. Why the total volume of the solution is not equal to the sum of the initial
volume of the solution and the volume of the added component?
II. Can we calculate or predict the change in the total volume of a solution
when a given volume of one component is added to it at constant
temperature, pressure while keeping the amount of the rest of the
components constant? 51
Block1 Thermodynamics and Statistical Thermodynamics

Let us try to find answers to these

In a pure substance (one component system), a given molecule is surrounded

by same type of molecules, and therefore, the intermolecular forces between
them are same. These intermolecular forces do not change on further addition
of one mole of the pure substance to it as there are no new interactions. As a
consequence, the change in volume is equal to the molar volume of the added
component. On the other hand, in a mixture (say a binary mixture of A and B)
a given molecule is surrounded by different types of molecules. Consequently,
different types of intermolecular forces exist between the molecules in the
mixture. This affects the arrangement of molecules with respect to each other.

In such a case, when we add a mole of one of the components then the
increase in volume will depend on the nature of interaction between the
molecules. If the nature of intermolecular forces in the pure solvent and in the
solution are similar, then the increase in the volume of the mixture will be
similar to the molar volume of the added component. On the other hand, if
these forces are of different nature, then the increase in the volume of the
mixture will be different from molar volume of the added component in the
pure form.

The answer to the second question is ‘yes’. For this, let us derive a general
expression for the change in the volume of the mixture in terms of partial molar
volumes of its components. We will proceed in the same way as we did while
deriving the general expression for partial molar properties. The volume of a
binary solution which depends on temperature, pressure, amount of A and the
amount of B can be expressed as
V = f (T , p, n A , nB ) …(2.30)

Using Eq. (2.12) we can write the expression for change in the total volume of
the binary mixture with the change in the amounts of components A and B at
constant temperature and pressure as follows.
V V
dV = dnA + dnB …(2.31)
nA T , p,nB
nB T , p,nA

V V
Here, is the partial molar volume of A and is equal to
nA T , p,nB
nB T , p,nA

partial molar volume of B.

Now, suppose we add 1 mole of the component A to the mixture at constant
temperature, pressure, and the number of moles of B. In such a case, dnA = 1
and dnB = 0 . Substituting in Eq. (2.31), we get.

V
dV = …(2.32)
nA T , p,nB

This expression provides us the definition of partial molar volume. Thus, the
partial molar volume of component A in a binary mixture may be defined as
the change in the volume of a mixture when one mole of component A is
added to a large volume of mixture at constant temperature, pressure while
keeping the number of moles of component B constant. The unit of partial
52 molar volume is dm3 mol–1.
Unit 2 Systems of Variable Compositions
Using the notation of partial molar properties as given above, we can write the
expression for the change in volume of the binary mixture in terms of partial
molar volumes of the components A and B as follows:

dV = VA,mdnA + VB,mdnB … (2.33)

The total volume of a binary mixture when n A moles of component A are

mixed with nB moles of component B at constant temperature and pressure is
given by the following expression.

V = Vm,AnA + Vm,B nB …(2.34)

Partial molar volume: a physical insight

Let us try to have an insight into the concept of partial molar volume by taking
the example of binary solution containing water (substance A) and methanol
(substance B). The partial molar volume of the methanol here will be the rate
at which the volume of the binary solution changes with the amount of
methanol added to a large volume of the mixture at constant temperature and
pressure i.e.,

V
Vm,B = …(2.35)
nB T , p,nA

−1
At 25°C and 1 bar, the molar volume of pure water is Vm,A = 18.07cm mol
* 3

−1
and that of pure methanol is Vm,B = 40.75cm mol . When we mix 100.0 cm3
* 3

of water at 25°C with 100.0 cm3 of methanol at same temperature, it is found

that the volume of the resulting mixture is 193.1 cm3and not equal to the sum
of the volumes added i.e., 200.0 cm3. Let us first calculate the composition of
this mixture:

VA 100.0 cm3
nA = = = 5.53 mol …(2.36)
Vm, A 18.07 cm3 mol −1

VB 100.0 cm3
nB = = = 2.45 mol …(2.37)
Vm, B 40.75 cm3 mol −1

nB 2.45 mol
xB = = = 0.307 …(2.38)
nA + n B 5.53 mol + 2.45 mol

Now, if we prepare a mixture by mixing 5.0 dm3of water (277.78 moles) and
5.0 dm3 of methanol (122.5 moles) i.e., 50 times the volumes used in above
case, the volume of the mixture will also be 50 times i.e., 9.655 dm3 and, it
will have the same mole fraction of methanol i.e., 0.307.

On adding 40.75 cm3 (one mole) of pure methanol to the above solution it is
observed that the volume of the solution increases by only 38.8 cm3. Similarly,
on adding one mole (40.75 cm3) of pure methanol to the solution prepared by
mixing 10.0 dm3 of water and 10.0 dm3 of methanol also the volume increases
by only 38.8 cm3. Let us calculate the mole fractions of these solutions and
compare as given in Table 2.2. 53
Block1 Thermodynamics and Statistical Thermodynamics

Table 2.2: Data pertaining to the example being discussed.

S. Volume (in cm3) Volume (in cm3) Total observed Mole Mole fraction
No. (No. of moles) of (No. of moles) volume fraction of of methanol
water (in cm3) water (XA) (XB)
of methanol

1 100.0 (5.53) 100.0 (2.45) 193.1 0.693 0.307

5000 (277.78) 5000 (122.5) 9655 0.693 0.307

2 Adding 40.75 cm3 (one mole) of pure methanol

5000 (277.78) 5040.75 (123.5) 9693.8 0.6922 0.3078

10000 (555.56) 10000 (245.0) 19310.0 0.693 0.307

3 Adding 40.75 cm3 (one mole) of pure methanol

10000 (555.56) 10040.75 (246.0) 19348.8 0.693 0.307

The amount of methanol added (40.75 cm3, 1 mole) in the examples above is
not infinitesimal, but it is quite small as compared to the amounts in the initial
mixture. Therefore, as shown in Table 2.1 there is very little change in the
composition of the mixture. We can say that in these additions the mole
fraction of water ( n A ) remains practically constant. The observed change in
the volume (38.8 cm3) of the mixture on adding 1 mole of methanol is the
partial molar volume of methanol as it corresponds to the change in volume
per mole of methanol at constant T, p and n A . In other words,

V V
Vm,B = = = 38.8cm3 mol−1 .
nB T , p,nA
nB

Thus, as defined above, the partial molar volume of component B is the

change in volume per mole of the component B added to such a large volume
of the mixture that its composition is not appreciably affected. Now if instead of
methanol, we had chosen to add a mole of water (18.07 cm3) to different
volumes of the mixture then the change in the volume of the mixture would
have been 17.74 cm3 in each case. This is the partial molar volume of water at
this composition. Thus, the partial molar volumes of methanol and water in a
solution having a mole fraction of methanol = 0.307 are 38.8 cm3 mol–1 and
17.74 cm3 mol–1respectively.

Imp: The data and the calculations in this example are for a mixture of water
and methanol having the mole fractions as 0.693 and 0.307 respectively. Had
we used any other composition, the corresponding partial molar volumes
would have been different.

Let us take some examples to learn about the applications of the expressions
derived for partial molar volumes.

Example 2.1: The partial molar volumes of water and methanol in a solution
having 5.53 mol of water and 2.45 mol of methanol at 298 K
and 1 bar pressure are 17.74 cm3 mol–1 and 38.8 cm3 mol–1
54 respectively. Calculate the total volume of the solution
Unit 2 Systems of Variable Compositions
Solution: Using Eq. (2.34) we can write, total volume,
V = Vm,W nW + Vm,M nM

Given: nW = 5.53 mol & nM = 2.45 mol

Vm,W = 17.74 cm3 mol–1& Vm,M = 38.8 cm3 mol–1

Substituting the values in the expression for total volume, we get

Vt = (17.74cm3 mol-1 5.53mol )+(38.8cm3 mol-1 2.45mol)=193.16cm3

Thus, the volume of the solution would be = 193.16 cm3

Example 2.2: Calculate the volumes of pure methanol and pure water at 25 C
that must be mixed to form 1000 cm3 solution of methanol in
water having a mole fraction of 0.3. The molar volumes of pure
methanol and water and their partial molar volumes for this
composition and temperature are given as under

Vm,M = 38.63 cm3mol−1 & Vm*,M = 40.73 cm 3mol −1

Vm,W = 17.77 cm3mol−1 & Vm*,W = 18.07cm3mol−1

Solution: According to Eq. (2.34) we can write the expression for the molar
volume of the mixture as
V = Vm,W nW + Vm,M nM

Substituting the values of mole fractions and the partial molar volumes, we get
the molar volume of the mixture as
V = (17.77 0.7) + (38.63 0.3) = 24.03 cm3mol−1

Now, since we need 1000 cm3 of the mixture, the total number of moles of the
solution required will be:

1000 cm3
n= = 41.61 mol
24.03 cm3 mol−1

We are given that the mole percentage of methanol is 30%. It implies that the
number of moles of methanol in the mixture will be
nM = (0.3)(41.61) = 12.48 mol

and that of water will be nW = (0.7)(41.61) = 29.13

The volume of each pure species required will be given as: Vi = ni Vm* ,i cm3

thus, VM = 12.48 40.73 cm3 = 508.31cm3 and

VW = 29.13 18.07 cm3 = 526.38cm3

It is interesting to note here that the total volume of the pure components
added is (508.31cm3 + 526.38cm3 = 1034.69 cm3 ) is more than the total
volume of the solution. This is due to the fact that the partial molar volumes at
this composition are smaller than the molar volumes of the pure species.
55
Block1 Thermodynamics and Statistical Thermodynamics

2.3.1 Volume of Mixing

From the discussion on partial molar volumes, you now understand that when
we mix 50 cm3 of component A and 50 cm3 of component B, the volume of the
binary solution formed may or may not be equal to 100 cm3. Let us now try to
learn how much could be the change in the volume of solution from the total
volume of combining components. This difference is called volume of mixing
(∆mixV) and is equal to the difference between the volume of the solution and
the sum of the volumes of the pure components. That is, we can write the
expression for volume of mixing as:
V = VA,m nA + VB,m nB − (VA*,mnA + VB*,mnB )
mix …(2.39)
* *
Here the sum of first two terms gives the volume of the solution. VA,m and VB,m
are the molar volumes of pure A and pure B components respectively and the
term in the brackets (VA,m nA + VB,m nB ) is the sum of the volumes of pure A and
* *

pure B components.
The volume of mixing of two components can be zero or non–zero depending
on the nature and magnitude of intermolecular interactions between similar
and dissimilar molecules. When the interactions between the molecules of
two components (A–B) are equal to the A–A and B–B interactions then the
volume of mixture is equal to the total volume of the pure components. This
means that the volume of mixing (∆mixV) is zero. Such a solution or a mixture is
known as an ideal solution or an ideal mixture. The total volume of an ideal
solution can therefore be written as
V = VA*,m nA + VB*,m nB …(2.40)

Dividing Eq. (2.40) throughout by the total amount (moles) of the solution, n
we get the expression for mean molar volume of the solution (Vm). That is,
V n n
Vm = = VA*,m A + VB*,m B … (2.41)
n n n
Or we can write,
Vm = VA*,m A + VB*,m B … (2.42)

If we plot a graph between mean molar volume of the solution versus mole
fraction of component B, ( B ), we will get a straight line with the intercept as
VA*,m at B = 0 . On the other hand, the intercept at B
*
= 1 will be equal toVB,m .
Thus, for an ideal two–component solution, the volume of the solution varies
linearly with the concentration of the solution as shown in Fig.2.1.

Fig.2.1: Variation of the volume of an ideal two–component solution with the

56 concentration of the solution.
Unit 2 Systems of Variable Compositions
When the interactions between the unlike molecules of the two components (A
and B) are not equal to the interactions between like molecules (A–A and
B–B) of the pure components, then the volume of the solution is not equal to
the sum of the volumes of the pure components. Hence, the ∆mixV is not equal
to zero. Such a solution is known as a non–ideal solution. That is,
VA,m nA + VB,m nB (VA*,m nA + VB*,m nB ) … (2.43)

In case of non–ideal solutions, the volume of the solution varies non–linearly

with concentration. The nature of variation of volume of the solution will
depend upon the nature and magnitude of A–A, B–B and A–B intermolecular
interactions between the two components. These interactions will manifest in
terms of the difference in volume of mixing. Accordingly, the nature of the
variation of volume of the non–ideal solution with concentration will be of any
one of the four possible types. These types along with their representative
plots for variation are given in Fig. (2.2)

Fig. 2.2: Different types of plots of the variation of molar volume of the solution
as a function of mole fraction. The dotted lines represent ∆mixV =0

Let us now take up the experimental determination of partial molar volumes.

However, before that answer the following simple question to assess your
understanding.

SAQ 4
2.0 moles of pure ethanol is added to 2.0 moles of pure water. If the partial
molar volume of ethanol and water for A = B =0.5 are 57.4 cm3mol–1 and
16.9 cm3mol–1respectively, calculate the total volume of the solution.

2.3.2 Experimental Determination of Partial Molar

Volume
Determination of partial molar volume helps in understanding the nature of
interactions between the components of the solution. Such determinations of
partial molar volume are also of interest because of their thermodynamic
relationship with other partial molar quantities like partial molar Gibbs energy
i.e., chemical potential. 57
Block1 Thermodynamics and Statistical Thermodynamics

As explained above, the mean molar volume for a non–ideal binary solution is
given as

Vm = VA,m A + VB,m B …(2.44)

Since A + B = 1 , we can substitute ( B) by (1 − A) to get

Vm = VA,m A + VB,m (1 − A ) …(2.45)

Rearranging, the equation, we get

Vm = (VA,m − VB,m ) A + VB,m …(2.46)

Thus, according to Eq. (2.46) a plot of mean molar volume as a function of the
mole fraction of component A will give a straight line with slope and intercept
as (VA,m − VB,m ) and (VB,m ) respectively.

In a typical experimental determination of partial molar volume, a series of

binary solutions of A and B are prepared with varying mole fractions of A.
Then the volume of each solution is measured, and the mean molar volume is
calculated for each composition. A graph of Vm is then plotted against the mole
fraction of component A ( A ). This graph may or may not be a straight line
depending upon the nature of the two components as discussed above. A
linear plot suggests that the partial molar volumes are independent of the
composition and are expected to be equal to the molar volumes of respective
pure components.

In case of non–linear variation of mean molar volume with mole fraction of

component A, the molar volumes of the components ( VA,m and VB,m ) are
dependent upon the composition of the solution. That is, at each value of A

the partial molar volumes are different. However, for a given composition, VA,m
and VB,m are constant. Therefore, the slope of the tangent drawn at any given
point on the curve along with the corresponding intercept can be used to
calculate the partial molar volumes of the two components for the composition
corresponding to the point. The slope of the tangent can be obtained by
differentiating Eq. (2.46) with respect to A

Vm
The slope of the tangent = = (VA,m − VB,m ) …(2.47)
xA

Further, as per Eq. (2.46) the intercept for the tangent at A is equal toVB,m .

Therefore, to determine the partial molar volumes at a given composition, A a

tangent to the curve is drawn at A and the slope and the intercept are
measured. These can then be used to determine the partial molar volumes.

Partial molar volume for component B for composition, A = VB,m = intercept

Partial molar volume for component A for composition, A = VA,m

= slope + intercept = (VA,m − VB,m ) + VB,m

58
Unit 2 Systems of Variable Compositions
Having learnt about partial molar volume and its determination, let us now take
up a very important partial molar property, viz., Chemical potential.

2.4 CHEMICAL POTENTIAL

In this section you are going to learn about probably the most important partial
molar property with a special name–chemical potential. You have studied in
earlier classes and in the previous unit that Gibbs energy plays an important
role in predicting the feasibility of a process. In a system having many
components, instead of Gibbs energy of pure components, partial molar Gibbs
energy values of all the components are used to predict the feasibility and the
direction of the process. The partial molar Gibbs energies are also important in
the study of chemical equilibrium and phase equilibrium. Due to the
importance of partial molar Gibbs energy in thermodynamic studies, it was Josiah Williards
given a special name chemical potential by American scientist Josiah Williards Gibbs (1839 –1903).
Gibbs (1839 –1903). The chemical potential is represented by the symbol Source: Wikipedia

and is defined as

G
i = …(2.48)
ni T , p,n j

That is, the chemical potential of the i th component ( i ) refers to the rate of
change of Gibbs energy with the change in the number of moles of i th
component when temperature, pressure, and the number of moles of all other
components are kept constant.

As discussed above, Gibbs energy, G depends upon temperature, pressure,

and composition i.e.,

G = f(T , p, nA , nB ) ….(2.49)

Using Eq. (2.11) we can write the expression for total change in Gibbs energy
as

G G G G
dG = dT + dp + dnA + dnB ….(2.50)
T p,nA nB p T ,nA nB
nA T , p,nB
nB T , p,nA

From the previous unit you will recall that

G G
= −S and =V ….(2.51)
T p,nA ,nB p T ,nA ,nB

And by using Eq. (2.48), we can write

G G
= A and. = B ….(2.52)
nA T ,nA ,nB
nB T , p, nA

Therefore, Eq. (2.50) can be written in the short form as

dG = −SdT + Vdp + A dn A + B dnB ...(2.53) 59

Block1 Thermodynamics and Statistical Thermodynamics

At constant temperature and pressure, dT = 0 and dp= 0, therefore, Eq. (2.53)

can be written as

dG = A dn A + B dnB ...(2.54)

This expression gives the change in Gibbs energy with the change in the
composition of a binary mixture at constant temperature and pressure. Using
Eq. (2.28) and replacing X by G, the total Gibbs energy of a binary mixture can
be given as follows

G= A nA + B nB …(2.55)

This gives the total Gibbs energy of a binary mixture at a fixed temperature,
pressure and composition. For a pure component, Eq. (2.55) can be written as

G= A nA …(2.56)

Rearranging, we get

G
A = . …(2.57)
nA

This expression implies that chemical potential of a pure substance is equal to

the molar Gibbs energy of that substance. Let us learn about the significance
of Chemical potential.

2.4.1 Significance of Chemical Potential

The chemical potential of a pure substance is same as the molar Gibbs energy
of the substance. The chemical potentials of the pure substances can be
compared to determine the relative stability of the substances at a given
temperature and pressure. Lower is the value of chemical potential more
stable is the substance. Chemical potential plays a significant role in
thermodynamic studies of system of variable composition also. Therefore,
chemical potential can be applied to study a wide range of systems.

The significance of chemical potential is similar to the potential in physics. The

system tends to move spontaneously from a higher chemical potential to lower
chemical potential state. When two systems are at equilibrium, this means that
their temperature, pressure, and chemical potential are equal. Therefore,
chemical potential is also used in studying chemical equilibrium and phase
equilibrium.

Chemical Potential: A Physical Insight

The chemical potential can be seen as ‘escaping tendency’ of a substance in a

mixture. The molecules of a component spontaneously move from regions of
high chemical potential to the regions of low chemical potential in a mixture.
This continues till the chemical potential of the component becomes same
throughout the system. For example, if we add a drop of ink in some water
taken in a beaker then the molecules of ink in the region of drop gradually but
60 spontaneously spread out. There is a concentration gradient and the
Unit 2 Systems of Variable Compositions
molecules at higher chemical potential move spontaneously to the regions of
low chemical potential i.e., they diffuse.

Fig. 2.3: Schematic representation of a concentration gradient.

Fig. (2.3) gives a schematic representation of a concentration gradient. The

species inside the porous membrane (at higher chemical potential) diffuse out
to the region of lower chemical potential outside the membrane. The process
continues till the concentrations become equal throughout the system.

The temperature dependence of chemical potential (discussed in next sub

section) helps us in understanding how the phase of the system will change
under different temperature and pressure conditions. For pure substances,
these relations can also predict whether a given substance can exist in solid,
liquid, or gaseous state under certain conditions of temperature and pressure.
The concept of chemical potential is used to derive the expressions for
colligative properties, equilibrium constants and other thermodynamic
properties of the system. Chemical potential expressions of half cells are used
to derive the Nernst equations of the Galvanic cells.

2.4.2 Effect of Temperature and Pressure on

Chemical Potential
As seen above, the Chemical potential is a fundamental thermodynamic
property that provides an insight into the behaviour of chemical systems. Since
chemical potential plays a crucial role in determining the conditions for
chemical equilibrium viz., the chemical potentials of reactants and products
must be equal. Studying the temperature and pressure dependence of
chemical potential helps in predicting the conditions under which chemical
reactions will reach equilibrium.
Further, an understanding of temperature and pressure dependence of
chemical potential, can help us gain a deeper understanding of the
thermodynamic stability and behaviour of substances. This aspect of chemical
potential is particularly relevant to the fields like geology where it helps in
explaining geological processes. In environmental science it helps us
understand the behaviour of fluids in subsurface reservoirs; the movement of
contaminants in the environment etc. Let us take up the effect of temperature
on chemical potential.
Effect of temperature on chemical potential
To study the effect of temperature change on chemical potential for a system
we need to differentiate the expression for chemical potential with respect to
temperature. You will recall the expression for chemical potential from above, 61
Block1 Thermodynamics and Statistical Thermodynamics

G
i = …(2.48)
ni T , p,n j

Euler’s theorem of
Differentiating the expression with respect to temperature while keeping the
exactness pressure and composition constant we get

A brief to come G
i
= …(2.58)
T p,ni ,n j T ni T , p,n j
p,ni ,n j

According to Euler’s theorem of exactness

G G
= …(2.59)
T ni p,T ,n j
ni T p,ni ,n j
p,ni ,n j T , p,n j

We know that,

G
= −S
ni p,ni ,n j

Therefore,

G G S
i
= = =− …(2.60)
T p,ni ,n j T ni T , p,n j
ni ni p,ni n j
ni T ,p,n j
p,ni n j T , p,n j

From the definition of partial molar properties, Eq. (2.13) we can write,

S
= Si ,m …(2.61)
ni T , p,n j

ui
= −Si ,m …(2.62)
T p,ni ,n j

As you know that absolute entropy is always positive, therefore, partial molar
ui
entropy also has a positive value. The Eq. (2.62) indicates that is
T p,ni ,n j
always negative. It is clear from these relations that chemical potential will
decrease with the increase in temperature. On rearranging Eq. (2.62), we get.

d i = −Si ,mdT …(2.63)

On integrating between the limits of temperature, we get

i (T2 ) T2

d i = −Si ,m dT …(2.64)
i (T1 ) T1

Simplifying,
T2

i (T2 , p ) − i (T1, p ) = −Si ,m dT …(2.65)

62 T1
Unit 2 Systems of Variable Compositions
The extent of change in chemical potential with temperature will depend upon
the partial molar entropy of the component. For a pure substance, the extent
of change in chemical potential with temperature will depend upon the molar
entropy of that substance. You know that the molar entropy of a substance in
vapour phase is maximum and in solid phase is minimum, that is,
SV ,m Sl ,m Ss,m …(2.66)

Thus, with the increase in temperature, the entropy of the substance will
decrease the most if the substance is in the gaseous state. This expression is
particularly useful in explaining the colligative properties like, depression in
freezing point and elevation in boiling point.

Effect of pressure change on chemical potential

As for temperature change, to understand the effect of pressure on chemical

potential we differentiate Eq. (2.48) with respect to pressure keeping the
temperature and composition constant, i.e.,

G
i
= …(2.67)
p T ,ni ,n j
p ni T , p,n j
T ,ni ,n j

According to Euler’s theorem of exactness

G G
= = Vi ,m …(2.68)
p ni T ,ni ,n j
ni p T ,ni ,n j
T ,ni ,n j T , p,n j

Therefore, the expression becomes,

i
= Vi ,m …(2.69)
p T ,ni ,n j

The above equation can be written as

d i = Vi ,m dp …(2.70)

It implies that the chemical potential increases with an increase in the

pressure. The extent of increase in the chemical potential for a given increase
in pressure will depend upon the value of partial molar volume.

2.5 MIXTURE OF GASES

As you have studied above, the chemical potential of a pure substance is
same as its molar Gibbs energy. Further, you would recall from previous unit
that Gibbs energy change for a pure substance is given as
dG = −SdT + Vdp …(1.67)

Using Eq. (2.63) and Eq. (2.70) we can write the Eq. (1.67) as
d i = −Si m dT + Vi m dp …(2.71)

For a pure substance at constant temperature, the above equation becomes

d = Vm dp …(2.71(a)) 63
Block1 Thermodynamics and Statistical Thermodynamics

This expression can be integrated to get the chemical potential values for
different pressures at a given temperature.
(T , p2 ) p2

d i = Vm dp …(2.72)
(T , p1 ) p1

However, since the molar volume is pressure dependent the integration of

right side is not straight forward. In the case of solids and liquids, they are not
easily compressed, so for small pressure range the molar volume of a pure
solid or a pure liquid substance can be assumed to be constant. In such
cases, the molar volume term can be taken out of the integral.

Therefore, for pure solid and pure liquid

(T , p2 ) p2

d = Vm dp …(2.73)
(T , p1 ) p1

If p1 is taken as standard pressure (1 bar), then the chemical potential at 1 bar

is the standard chemical potential of the substance, μo. Therefore, the
chemical potential at any other pressure, p2=p is written as
(T , p) = o
(T ) = Vm ( p − 1) …(2.74)

Chemical potential of an ideal gas in terms of pressure

Now let us take the case of gases. In this case the molar volume changes with
the change in pressure. So, for gases, we have to write molar volume in terms
of pressure. There are different equations of state which relate molar volume
with pressure. We will take the ideal gas equation.
RT
For an ideal gas, V m = …(2.75)
p

Therefore, Eq. (2.72) can be written as

(T , p2 ) p2
RT
d = dp …(2.76)
(T , p1 ) p1
p

Integrating between the limits, we get

p2

(T , p2 ) − (T , p1 ) = RT dln p …(2.77)
p1

For initial pressure (p1=p0) equal to 1 bar, the chemical potential is equal to the
standard chemical potential of the gas, the equation for any other pressure,
p2=p will be

(T , p) − o
(T ) = RT In ( pp )
0
…(2.78)

Since po is 1 bar, the equation is written as

(T , p) − o
(T ) = RT In p …(2.79)

As can be seen from the above expression, the chemical potential of an ideal
gas increases with the increase in pressure. If the standard chemical potential
64 of an ideal gas is known at a given temperature, then the chemical potential
Unit 2 Systems of Variable Compositions
can be calculated at any other pressure and given temperature. Remember
that standard conditions do not include the temperature. Hence, standard
chemical potential at different temperatures will be different. Wherever the
temperature is not mentioned for the standard values, it is taken as 298 K.

You know that the gases move from high pressure region to low pressure
region. Now let us state the same fact in terms of chemical potential. At high
pressure, chemical potential is more as compared to the chemical potential at
lower pressure. Hence, gases move spontaneously from higher chemical
potential state to the lower chemical potential state. Therefore, chemical
potential is also a measure of escaping tendency of the gas (recall the
physical insight on chemical potential, given above).

Chemical potential of a real gas in terms of fugacity

The Eq. (2.78) gives the chemical potential of an ideal gas. However, a real
gas is expected to have a different behaviour depending upon the size of its
molecules and intermolecular interactions present in it. You have studied
various equations of state for real gases in your earlier classes. However,
these equations show deviations under certain conditions of temperature and
pressure. Writing different equations of state under different conditions makes
the expression complicated. To simplify things, in the case of real gases,
pressure is replaced by the term fugacity (f). All the deviations arising due to
intermolecular interactions are included in it. Therefore, fugacity is known as
corrected pressure, and it has the unit of pressure.

f = p …(2.80)

Here, is known as fugacity coefficient whose value is equal to 1 for an ideal

gas. At very low pressures, fugacity coefficient can be approximately taken as

1 for real gases. That is, lim → 1. For real gases, Eq. (2.78) is modified as
p→0

(T , p) − o
( )
(T ) = RT In f0 = RT In 0
p p( )
p
…(2.81)

Since p0 = 1 bar, the equation is written as

(T , p) − o
(T ) = RT In p - RT In …(2.82)

If we compare Eq. (2.78) for ideal gases and Eq. (2.82) for the real gases we
find that there is an addition term, viz., RT In . This term accounts for all the
deviations from ideal behaviour.

Chemical potential of an ideal gas in a mixture of ideal gases

Whenever two or more gases are present in a system it is known as a mixture

of gases. During the study of thermodynamic properties of the gaseous
mixture, it is easier to assume that all the components of the mixture are ideal
gases. The thermodynamic expressions obtained for a mixture of ideal gases
can then be further modified to get the expressions for the mixture of real
gases. 65
Block1 Thermodynamics and Statistical Thermodynamics

It is assumed that when the gases are mixed then the temperature and the
total pressure are fixed. So, what happens when the gases are mixed and how
do the thermodynamic property of the mixture differ from the thermodynamic
properties of the pure components.

Recall that in a mixture of gases, the contribution of each gas to the total
pressure of the mixture (p) is known as the partial pressure of that component
(pi) and it depends upon the mole fraction of the component (xi) in the mixture
as stated by Dalton’s law of partial pressures.

pi = xi p …(2.83)

Let us find how does the chemical potential change when a pure ideal gas is
added to the mixture of ideal gases. The change in chemical potential of a
component at constant temperature is given by the expression

d i = Vi ,mdp …(2.84)

For ideal gas, since the partial molar volume is equal to its molar volume, i.e.,
Vi ,m = RT / p. , the expression for dμi is written as

RT
d i = dp = RTd ln p …(2.85)
p

To find the total change in the chemical potential on mixing, we have to

integrate the above equation. Let us take the initial state as pure gas at the
pressure of 1 bar, that is the standard pressure (po) and any temperature T.
The chemical potential under these conditions is equal to the standard
chemical potential of the pure component i . The final state is the gas in the
mixture at temperature T and partial pressure pi. Let the chemical potential of
the component in the mixture be μi. Upon integrating the expression, we get,
i (T ) pi

d = RT d ln dp …(2.86)
o
i (T ) p0

pi
i (T ) = i
o
(T ) = RT ln …(2.87)
p0

Since the partial pressure of a gas is always less than the pressure of pure
gas, the right-hand side of the above equation is negative. Thus, it can be
seen from Eq. (2.87) that the chemical potential of an ideal gas is less in a
mixture as compared to the pure gas. Since, there is a decrease in the
chemical potential of the gas on mixing, the mixing of the ideal gases is a
spontaneous process.

2.6 GIBBS–DUHEM EQUATION

As shown earlier, the change in the extensive property X at constant
temperature and pressure is given by Eq.(2.24)

dX = Xm,AdnA + Xm,B dnB …(2.24)

66
Unit 2 Systems of Variable Compositions
Whereas the total extensive property at a given temperature, pressure and
composition is given by Eq.(2.28)

X = Xm,AnA + Xm,B nB …(2.28)

The differential form of Eq.(2.28) will give the total change in X as

dX = Xm,AdnA + nAdXm,A + Xm,B dnB + nB dXm,B ...(2.88)

Since, both Eqs. (2.24) and (2.88) represent total change in the state variable
X, the equations are equal. Therefore,

Xm,AdnA + Xm,B dnB = Xm,AdnA + nAdXm,A + Xm,B dnB + nB dXm,B …(2.89).

Simplifying, we get

nAdXm,A + nB dXm,B = 0 ...(2.90)

The Eq. (2.90) is the general form of a very important relation which is known
as Gibbs–Duhem equation.

Gibbs–Duhem equation can be written in terms of volume as

nAdVm,A + nB dVm,B = 0

or nAdVm,A = −nB dVm,B ...(2.91)

This relation implies that for a binary system, the partial molar volume of a
component cannot change independent of the partial molar volume of the
other component. If partial molar volume of a given component is increased by
some method, then the partial molar volume of the second component will
decrease.

This is true for all the types of partial molar quantities. The Gibbs–Duhem
equation in terms of chemical potential is written as

nA d A + nB d B =0 ...(2.92)

If the changes in temperature and pressure are also considered, then the
Gibbs–Duhem equation in terms of chemical potential is written as

−SdT + Vdp + nA d A + nB d B =0 ...(2.93)

SAQ 5
For a binary mixture in which nA = 3nB if the chemical potential of A changes
by 2 J mol–1 what is the change in the chemical potential of B?

SAQ 6
In a binary mixture of ideal gases in which nA = 5nB , the partial molar volume
of B was decreased by 0.01 m3 mol–1. What would be the change in the partial
molar volume of A?
67
Block1 Thermodynamics and Statistical Thermodynamics

2.7 SUMMARY

In this unit, we have discussed the issue of the effect of change in composition
of the system on its thermodynamic properties. We began the unit by initiating
a discussion on partial molar properties and attempted to understand their
significance. Herein, we briefly recalled the commonly used concentration
expressions like, molarity, molality, and mole fraction. Then, we defined molar
properties as the thermodynamic properties per mole of the system. We stated
that the thermodynamic properties of a system of variable composition are a
function of the pressure, temperature, and the composition of the system. The
change in each one of these parameters causes a change in the property and
the total change in the property is equal to the sum of these changes.

The partial molar property of component A in a mixture has been defined as

the change in the property X of the mixture when one mole of component A is
added to a large volume of mixture at constant temperature and pressure
while keeping the amounts of remaining components constant. Having defined
partial molar properties, we developed the relationship between the
thermodynamic properties and the corresponding partial molar properties.

We then took up a specific partial molar property viz., partial molar volume as
it is easier to visualise the changes. We defined partial molar volume and
provided an insight into its physical meaning using an example and explained
its experimental determination. Thereafter, we took up an important quantity
viz., chemical potential i.e., Gibbs energy per mole. We discussed the effect of
pressure and temperature on this fundamental property. It has the same
significance in systems with variable composition as Gibbs energy has for
pure components. We then took up systems consisting of mixtures of gases.
In this context we brought in the concept of fugacity and explained its
significance. Towards the end of the unit, we derived the Gibbs–Duhem
equation and explained that it states that partial molar properties of various
components in a mixture cannot be changed independent of each other.

2.8 TERMINAL QUESTIONS

1. Define a partial molar property.

2. Calculate the partial molar volume of water in an equimolar mixture of

ethanol and water for which the partial molar volume of ethanol is found
to be 57.4 cm3 mol–1. Given that the density of the mixture is 0.80 g cm–3.

3. Define chemical potential. How will the chemical potential of an ideal gas
change if the pressure of the gas is increased at constant temperature.

4. Show that for an ideal gas in a mixture, the partial molar volume is equal
to RT/p.

5. Show that in a binary mixture when the chemical potential of component

A decreases then the chemical potential of component B will increase if
68 the temperature and pressure are fixed.
Unit 2 Systems of Variable Compositions

2.9 ANSWERS
Self–Assessment Questions
1. i) We are given a solution containing 6.30 g of oxalic acid per 100 cm3
of the solution.

We know that molarity (ci) of a solution is given as.

ni 1000
ci = mol dm–3
V
Now ni = 6.30 g /126 g mol–1= 0.05 mol ; V= 100 cm3

Substituting, we get
0.05 1000
ci = = 0.5
100

ii) We are given a solution containing 6.30 g of oxalic acid per 100 g of
water
We know that molality ( mi ) of a solution is given as.
n 1000
mi = i mol kg–1
w
We have calculated ni as 0.05; w= 100 g (Given)

Substituting, we get
ni 1000
mi = = 0.05 1000 /100=0.5
w

2. i) Given: mass of ethanol = 23 g. mass of water 180 g.

We know that the mole fraction is given as
ni
i =
ntotal

ne
Mole fraction of ethanol xe =
ne + nw

23 g
➔ number of moles of ethanol = ne = = 0.5mol
46 gmol−1

180 g
number of moles of water = nw = = 10mol
18 gmol−1

0.5
xe = = 0.0476
0.5 + 10
100 g
ii) ne = = 2.17mol
46 gmol−1

18 g
nw = = 1.0mol
18 gmol−1

2.17
xe = = 0.685
2.17 + 1.0 69
Block1 Thermodynamics and Statistical Thermodynamics

3. The partial molar properties are relevant only for those thermodynamic
properties that are extensive, and we know that temperature is an
intensive property. Further, the temperature of a system will not change
by changing composition of the system. Therefore, partial molar
temperature for a mixture is irrelevant.

4. According to Eq. (2.34) we can write, the expression for total volume as

V = Vm,w nw + Vm,e ne

We are given: nw = 2.0 mol & ne = 2.0 mol

2 mol
Thus, the mole fraction of ethanol = = 0.5
2 mol + 2mol

Now for the mole fraction equal to 0.5 the partial molar volumes are given
as

Partial molar volume of water, Vm,w = 16.9 cm3 mol–1

Partial molar volume of ethanol, Vm,e = 57.4 cm3 mol–1

Substituting the values in the expression for total volume, we get

V = (16.9cm3 mol-1 2.0mol )+ (57.4cm3 mol-1 2.0mol)=148.6cm3

Thus, the volume of the solution would be = 148.6cm3 .

5. According to Gibbs–Duhem equation

nA d A + nB d B =0

nA = 3nB ,d A = 2Jmol−1

nA 3nB
d B =− d A =− 2Jmol−1 = −6 Jmol−1
nB nB

6. According to Gibbs–Duhem equation nAdVA,m = −nB dVB,m

nA 5nB , VB,m = −0.01m3 mol−1

nB n
dVA,m = − dVB,m = − B ( −0.01)m3mol−1 = 0.002m3 mol−1
nA 5nB

Partial molar volume of A increases by 0.002 m3 mol–1

Terminal Questions
1. Partial molar property gives the change in thermodynamic property of the
mixture if one mole of a component is added to the mixture at constant
temperature, pressure and keeping the amount of the rest of the
components constant. For any thermodynamic property X, partial molar
X
property of component i is written as
ni T ,p,n
70 j
Unit 2 Systems of Variable Compositions
2. Let us assume the partial molar volumes of ethanol and water to be Vm ,e
and Vm,w respectively

Given: Vm,e = 57.4cm3 mol−1 and Vm,w = ?

Using Eq. (2.34) we can write, the expression for total volume as
V = Vm,e ne + Vm,w nw

We know that Me = 46.4gmol−1 and Mw = 18.0gmol−1

Given: e = 0.5 and w = 0 .5

Therefore, the molar mass of the equimolar mixture of ethanol and water
can be estimated as
M = Me e + Mw w
= [46.4gmol−1 0.5] + [18 .0gmol−1 0.5] = 32 gmol−1

Molar volume of the mixture is given by.

Mass 32 g mol−1
Vm = = = 39.99 gmol−1
Density 0.8002 gcm−3

We know that the molar volume of the mixture, V = Vm,w nw + Vm,e ne

Substituting
39.99 = (0.5 57.4) + (0.5 Vm,w )

(39.99 − 28.70)
Solving, Vm,w = = 22.58 gcm−3
0.5
Hence, the partial molar of water in the given mixture,
Vm,w = 22.58 gcm−3

3. Chemical potential is equal to partial molar Gibbs energy. For any

G
component i, the chemical potential is written as i =
ni T ,p,n
j

The variation of chemical potential with pressure for an ideal gas is given
RT
by the expression d i = dp. Therefore, as the pressure increases,
p
chemical potential of the gas will also increase.
V
4. Partial molar volume is equal to . and for an ideal gas V =nRT/p.
ni T , p,n j

n = n1+ n2 +n3 +…. n = n1 + n2 + n3 + ....

RT
V= (n1 + n2 + n3 + ...)
p

V RT
Therefore, = .
ni T , p ,n j
p

This will be same for all the ideal gases in the mixture. 71
Block1 Thermodynamics and Statistical Thermodynamics

5. In a binary mixture at constant temperature and pressure

dG = m,A dnA + m,B dnB ...(A)

We know that at constant temperature and pressure dG = n +

m,A A n
m,B B

dG = m,A dnA + m,B dnB + nAd m,A + nBd m,B ...(B)

Comparing right hand sides of Eqs (A) and (B) we get

nAd m,A + nBd m,B = 0 or nAd m,A = − nB d m,B

It can be seen from the above equation that as chemical potential of one
component increases, the chemical potential of the second component
will decrease.

72
Unit 3 Introduction to Statistical Thermodynamics-I

UNIT 3

INTRODUCTION TO
STATISTICAL
THERMODYNAMICS–I

Structure
3.1 Introduction Stirling Approximation
Expected Learning Outcomes 3.4 Probability Distribution
functions
3.2 Classical vs Statistical
Thermodynamics 3.5 Summary
3.3 Basic Probability Theory 3.6 Terminal Questions
Permutations and 3.7 Answers
Configurations

3.1 INTRODUCTION
You know that thermodynamics refers to the study of heat and its relation to
other forms of energy. The term ‘thermodynamics’ was coined by Lord Kelvin
(1824–1907) to signify the dynamic nature of heat, which was then considered
to be a kind of fluid. If we take literal translation, the term, thermodynamics
conveys the dynamics (or motion) of heat. In your earlier classes you have
learnt about thermodynamics from a phenomenological perspective wherein
the observation of macroscopic phenomenon is expressed in terms of
mathematical relationships. The generalities emanating from the extensive
observation of macroscopic world over hundreds of years are stated in terms
of the laws of thermodynamics. These laws are applicable to all macroscopic
systems, irrespective of their physical state, whether solid, liquid, gas, or any
combination of these states. These laws have withstood the test of time and
have proved quite useful in understanding / determining the thermodynamic
properties of a system.

In this and the next unit we will introduce you to an alternative approach to
understand thermodynamic properties of a system. It is based on the
application of statistics to large collections of microscopic systems like atoms
and molecules. In fact, it provides the link between the microscopic 73
Block 1 Thermodynamics and Statistical Thermodynamics

perspective of matter (the quantised energy levels at the atomic / molecular

level) and its macroscopic or bulk (observable) properties. This approach can
be used to calculate the bulk thermodynamic properties such as internal
energy, enthalpy, entropy, and Gibbs energy of the system from structural and
spectroscopic data of microscopic species constituting the thermodynamic
system. It is important to note that though quantum mechanics is at the
foundation of statistical thermodynamics, we need not have the knowledge of
the intricate details of quantum mechanics. Just the basic ideas of energy
levels, energy states, and intermolecular forces are sufficient to develop a
satisfactory theory of statistical thermodynamics.

In a way, statistical thermodynamics can be considered as a mathematical tool

that bridges the gap between the properties of individual atoms / molecules in
a system and the macroscopic thermodynamic properties of the bulk matter.
Alternatively, statistical thermodynamics can be visualised as a vehicle that
serves to provide means to extend the properties of individual atoms and
molecules to a large collection of them.

Fig. 1.1: Schematic representation of statistical thermodynamics being a bridge

between quantum mechanics and classical thermodynamics.

We will begin the unit by delineating the relationship between classical and
statistical thermodynamics. This will be followed by a discussion on basic
probability theory. Herein, we will introduce the concept of probability and
explain the meaning of permutations and configurations with the help of
suitable examples. This will be followed by introducing an important
approximation i.e., Stirling approximation. We will discuss the need for and
importance of this approximation. Then we will take up the probability
distribution functions.

In the next unit we will continue with the discussion on statistical

thermodynamics.

Expected Learning Outcomes

After studying this unit, you should be able to:

❖ discuss the relationship between classical and statistical

thermodynamics;

❖ explain the basic terms related to probability theory;

❖ calculate the probability of randomised experiments;

❖ explain the difference between permutation and configuration;

❖ calculate the permutations and configurations for systems with small

74 number of objects;
Unit 3 Introduction to Statistical Thermodynamics-I

❖ state Stirling’s approximation and outline its significance;

❖ establish the validity of Stirling’s approximation for systems with large
number of particles; and
❖ define and classify probability distribution functions.

3.2 CLASSICAL VERSUS STATISTICAL

THERMODYNAMICS
In any chemical system we deal with many species e.g., a sample of one mole
of a hydrogen gas contains as many as 6.022 1023 molecules of hydrogen.
The experimentally determined macroscopic properties of the system,
correspond to a kind of an average of the molecular properties. In your earlier
study of thermodynamics, you would have calculated the macroscopic
properties of the system without paying any attention to the properties of the
individual atoms / molecules. On the other hand, in quantum mechanics, you
had studied the properties of a single atom / molecule. This was done by
solving the Schrodinger wave equation to calculate the different energy states
available in a microscopic system. You would recall that for a molecule, the
quantised energy states include translational, electronic, rotational, and
vibrational energy states. These all contribute to the total energy of the
molecule. These discrete energy states are represented in terms of energy
expressions and are identified by the respective quantum numbers.

You may argue that if we know the energy of one molecule, then we can
calculate the total energy of one mole of these molecules simply by multiplying
the energy of the single molecule by Avogadro’s constant. This simplified view
could be applicable if all the molecules in the sample had equal energies, but it
is not so. To understand this, let us consider an ideal gas at equilibrium i.e.,
the total energy of the gas does not change with time. Now, since the
individual molecules are in a constant state of random motion and keep
colliding with each other, some molecules will gain momentum and others will
lose it during these collisions. Therefore, the kinetic energies of individual
molecules will not be same i.e., the energies are distributed over a range.
Recall, that you have already studied this while discussing the Maxwell
distribution of velocities in the kinetic theory of gases.

Besides the kinetic energy, we also know that there are various rotational and
vibrational energy states available which the molecules can occupy. Though In a macroscopic
under low temperature conditions all the molecules may have same ground system the molecules
state electronic energy, they may still have different translational, rotational, are distributed over
many available
and vibrational energies i.e., these molecules will be occupying vibrational and
energy states at the
rotational states of different energy. Thus, we can say that the total energy is microscopic level.
not equally distributed among the molecules. There are different combinations
of energy states which the molecules can occupy in such a way that the total
energy of the system remains constant. The statistical treatment of such
systems allows us to calculate the fraction of molecules that occupy a given
energy state. Once we know the fractions of the molecules occupying different
energy states, we can estimate the macroscopic properties of the system. In
75
Block 1 Thermodynamics and Statistical Thermodynamics

other words, the properties of the individual molecules are scaled to

macroscopic level using theory of probability and suitable statistical tools. In
the next unit we will study how the macroscopic properties of a system can be
calculated from the properties of a single particle using the concept of
statistics.

We can say that both the approaches to the understanding of thermodynamic

systems are based on experimental data, however the type of data involved in
the two cases is different. The classical thermodynamics uses the data based
on the bulk properties of matter i.e., density, pressure, heat capacity etc.,
whereas the statistical thermodynamics uses the molecular properties like,
bond lengths, quantised energies, vibration frequencies etc. As an example,
the experimental value of enthalpy of formation of H2O(g) is – 241.8 kJ mol–1 at
298 K and 1atm. It refers to the enthalpy change associated with the formation
of one mole of water vapours from its constituent elements in their standard
states under standard conditions. Classically speaking it has nothing to do with
the microscopic nature of the species involved and what determines the
observed value. On the other hand, the statistical thermodynamics, deals with
what in the individual atoms and molecules determines the observed value.

In conclusion we can say that the classical thermodynamics ignores molecular

complexity i.e., the quantum mechanical nature of the molecules whereas the
statistical thermodynamics deals primarily with the quantum mechanical nature
of atoms and molecules. As stated above, statistical thermodynamics
represents a probabilistic approach to study the thermodynamic systems.
Therefore, a basic understanding of probability and related concepts is a must
to learn the statistical approach to thermodynamics. Let us learn about the
same. However, before that answer the following simple question to assess
your understanding.

SAQ 1
Statistical mechanics provides a theoretical basis for the experimental science
of thermodynamics. Comment.

3.3 BASIC PROBABILITY THEORY

You might have learnt about the term, ‘probability’ in your earlier classes or
maybe you are aware of it otherwise. The probability refers to the likelihood of
a given event to occur. The outcome of an experiment is represented in terms
of a random variable (X). The same experiment on being repeated many times
can give rise to (slightly) different values which can be discrete or continuous.
For example, when we throw a dice, the random variable is the number of dots
on the top face of the dice; it has discrete values.

If an experiment can have more than one outcome, then probability refers to
the chances of getting a particular (desired) outcome. The most common
example of using probability is in tossing of a coin. A coin has two faces, head,
76 and tail. In an experiment involving, tossing of an unbiased (fair) coin, the
Unit 3 Introduction to Statistical Thermodynamics-I

outcome can be either a head or a tail. Head and tail are referred to as the
possible events or outcomes of tossing a coin. The probability of getting a
head is mathematically written as
1
p(H ) = P (H ) = PH = … (3.1)
2

Thus, there is 50% chance of getting a head and 50% chance of getting a tail
in a throw of the coin. The total probability of all the events / outcomes in an
experiment is always equal to one, e.g.,

1 1
p(H ) + p(T ) = + =1 … (3.2)
2 2

If two coins are tossed simultaneously, then the possible outcomes will be:

First coin H H T T
Second coin H T H T

The collection of all possible outcomes of an experiment is known as the

sample space. An event refers to any outcome or collection of outcomes in
the sample space and may include the null set as well as the set of all possible
outcomes. For example, the sample space for tossing two coins has four
events or outcomes, i.e., HH, TT, HT, and TH. These constitute the sample
space for the experiment.
Let us take an example of travelling from city A to city C via city B. If there are
X ways of reaching city B from city A and Y ways of going from city B to city C,
then the total number of ways one can go from city A to city C will be X.Y.
Therefore, the sample space for the event (going from city A to city C) will
consists of all the X.Y elements. You may note here that in probability an
event may consist of a series of actions, so the sample space consists of all
possible outcomes.
For N mutually
If the sample space has N mutually exclusive events and an event of interest exclusive, equally
(A) occurs I times, then the probability of that event is given as likely points in sample
space, if I of these
I
P ( A) = … (3.3) correspond to the
N random event A, then
the probability
Therefore, the probability of the event of getting one head and one tail i.e., I
P(HT) = 2/4 = ½. In this case, the event is getting one head and one tail and P ( A) =
N
includes two elements of the sample space i.e., it refers to the following subset
[HT, TH] of the sample space. This example refers to the probability
associated with a single experiment of throwing two coins.

Many a times we are interested in the probability of a particular event, that is,
the probability of a specific outcome for a series of experiments. Let us take an
example for the calculation of probability for a desired outcome in the repeated
tossing of a coin.

Example 3.1: Consider an experiment involving tossing a coin four times.

Calculate the probability of getting at least two heads in the
experiment. 77
Block 1 Thermodynamics and Statistical Thermodynamics

Solution: In this case we have a coin that is to be tossed four times. In every
toss we have two possible outcomes i.e., a head or a tail. We can
write down all the possible outcomes of the experiment as follows:

Outcome Toss 1 Toss 2 Toss 3 Toss 4 Outcome Toss 1 Toss 2 Toss 3 Toss 4
No. No.
1 H H H H 9 H T T H
2 T H H H 10 H T H T
3 H T H H 11 H H T T
4 H H T H 12 T T T H
5 H H H T 13 T T H T
6 T T H H 14 T H T T
7 T H T H 15 H T T T
8 T H H T 16 T T T T

Thus, there are a total of sixteen possible outcomes of the experiment of

tossing a coin four times i.e., the sample space has sixteen outcomes. Of the
16 possible outcomes, 11 meet our requirement of having at least two heads.
Therefore, the probability of getting at least two heads after tossing a coin four
times will be 11 i.e., the number of desired outcomes divided by the total
16
number of possible outcomes.

Similarly, when we roll an unloaded dice (the one in which the chance of any
one of the six faces to be the top face is the same as that of any other face),
the probability that we get number 6 on the top face is 1/6 as it is one of the six
possible outcomes. Similarly, the other numbers from 1 to 5 will have the
same probabilities and the sum of all the probabilities will be 1.
1 1 1 1 1 1
+ + + + + =1 … (3.4)
6 6 6 6 6 6

Now, we consider a slightly more complex case of rolling a pair of unloaded

dice. There is only one outcome in which we get the sum of the numbers on
top face to be equal to two i.e., when both dice show 1, such an event is
identified as "throws a two". Similarly, there are only two possibilities for
"throws a three" i.e., when we get the outcome as (2,1) or (1,2). All the
possibilities in the rolling of a pair of dice are as compiled below:

Outcome (as ‘throws a..’) 2 3 4 5 6 7 8 9 10 11 12 Total

Number of ways 1 2 3 4 5 6 5 4 3 2 1 36

The sample space for rolling a pair of dice has a total of 36 outcomes. You
may note that in the above example, ‘throwing a seven’ is the most probable
event i.e., the one that can be accomplished in the greatest number of ways.
Since it can occur in 6 possible ways, it has a probability of 6/36=1/6. You may
also note that the six ways of ‘throws a seven’ are [(1,6), (2,5), (3,4), (4,3),
(5,2) and (6,1)] and these have been considered as equivalent as in the
78 example of throwing two coins, HT and TH are considered equivalent.
Unit 3 Introduction to Statistical Thermodynamics-I

It is important to state that the random or randomised variables, or

quantities are those variables whose value can change throughout the course
of an experiment or series of events. In the above examples, the sides of the
coin, the number on top face of the dice are the variables that change the
outcome of the experiment. The variables can be discrete variables i.e., the
ones that can have countable number of values. For example, a coin has only
2 sides, so it has only two variables, whereas dice has six variables. In the
event of picking one student from a class of 60 students, has 60 variables. It is
important to note that the number of outcomes may be astronomical however
as long as the possible outcomes are countable it will remain discrete. In case
of discrete variables, the sample space is represented as a collection of all the
variables. For example, the sample space in the experiment involving throwing
a fair dice will be {1,2,3,4,5,6}.

On the other hand, we can have continuous variables that can have any
value in the given range of values say a real number between 0 and 1 is an
example of continuous variable. Contrasting with discrete variables, a
continuous variable has uncountable set of possible [Link] is
another example of continuous variable as the system temperature can be
anywhere from zero K onwards. A yet another example could be the distance
travelled by different people within a given time interval. In case of continuous
variables, the sample space is represented by stating set of limits in which the
variable varies i.e., the variable X may have any value in a range, a ≤ X ≤ b.
For example, if the continuous variable, X can have any value from 1 to 100 it
will be represented as 1 X 100.

It is pertinent to point out that in the chemical systems that we are interested
in, the observables are random and fluctuating in nature. For example, the
instantaneous vapour pressure of a liquid in equilibrium with its vapour
continuously fluctuates around an average value. Similarly, the energy of a
system at constant temperature or the volume of a system at constant
pressure keeps on fluctuating around an average value. However, the
outcome of the measurement may be different from the average value and
cannot be predicted. Therefore, we need some kind of probability distribution
to describe the behaviour of the random variables. We will take it up towards
the end of the unit. Having learnt about important elementary terms related to
probability, let us now take up two critical concepts viz., permutations and
configurations. However, before that answer the following simple questions to
assess your understanding.

SAQ 2
Define sample space. Write the sample space for drawing a queen of heart or
diamond from a deck of playing cards.

SAQ 3
Define the term, ‘continuous variable’. If the temperature in the capital city of
Delhi varies from 2oC to 46oC. How would you represent the sample space for
the temperature in Delhi.
79
Block 1 Thermodynamics and Statistical Thermodynamics

3.3.1 Permutations and Combinations

As shown above, to calculate the probability of an event we need to count the
total number of events. This is an easy exercise in situations with a small
number of events. However, the exercise becomes increasingly difficult when
the number of events is large, say simultaneously tossing 100 coins or rolling
20 dice. A much easier way to calculate the total number of outcomes of an
experiment is based on the fundamental counting principle. According to
this principle, if there are N1 ways of doing first thing, N 2 ways of doing the
second thing, N3 ways of doing third thing and so on, then the total number of
possible outcomes is equal to

(N1 ).(N2 ).(N3 )..... … (3.5)

Let us again calculate the total number of outcomes for rolling a pair of dice by
using the fundamental counting principle. For rolling the first dice there are six
possible outcomes, similarly for rolling the second dice again there are six
ways. So, the total number of possible outcomes is 6 6=36. Similarly, if we roll
three dice, the total number of outcomes will be 6 6 6 = 216 and so on i.e., it
N
will be 6 for N dice. It is equivalent to the number of outcomes for drawing R
objects from a total of N objects such that firstly the objects are replaced after
every pick and secondly the order in which the objects are drawn does not
matter is given by the following expression.
R
Number of permutations = N … (3.6)

In the examples of tossing two coins, we have considered certain events (e.g.,
HT and TH) as equivalent, i.e., the order of choosing the variable or object
does not affect the total outcomes.

Permutations

Let us now count the total number of outcomes when the order of choosing the
objects does affect the total number of possible outcomes with the help of
some examples.

Example 3.2: Consider a box containing four different balls labelled A, B, C

and D. Calculate the number of possible ways of randomly
selecting the four original balls out of the box without
replacement.

Solution: The first ball chosen can be any of four available, the second
pick can be any of the remaining three balls, the third ball
chosen must be one of the two remaining possibilities, and for
the fourth pick there is only one possibility. Hence, the number
of ways that the four sequential but independent choices can be
made must be 4 3 2 1 =24.

These 24 possible ways of randomly selecting the four original balls are the
number of possible arrangements or permutations of any single sequence of
the four balls, e.g., ABCD.
80
Unit 3 Introduction to Statistical Thermodynamics-I

You may note here that if the balls were not labelled, i.e., these were
indistinguishable then the 24 possible arrangements discussed above would
refer to only one combination. That is, only a single collection or combination
of the four balls would exist. Also, if we chose to disregard the ordering of
picking the balls then even if the balls are labelled, we will have only one
combination. Let us take another example.

Example 3.3: Calculate the number of possible unique nine–digit numbers

that can be formed by using the digits 1–9 just once in the
number.

Solution: While writing the desired number, there are nine choices (any
number from 1–9) available for the first digit. For the second
digit only 8 choices are available as one numeral has already
been used and cannot be used again. In the similar manner the
number of available choices decreases by one for each
subsequent digit. Thus, the total possible unique nine–digit
numbers will be 9 8 7 6 5 4 3 2 1 =362880 = 9!

Each number in the example 3.3 is made up of unique arrangement of the

numerals and is known as permutation of nine numerals. Thus, from these
The number of
examples we say that the permutation is a unique arrangement of all the
permutations of N
available objects in which the order of selection affects the arrangement. In distinct objects taken
such type of permutation, all the available objects are used in the R at a time, is given
arrangement. The number of permutations for drawing N objects from a set of by the following
N objects without replacement is equal to N ! ( N factorial). The number of expression:
permutations is written as NP or P (N ) . N
PR = N(N − 1) (N − 2) (N − 3)...(N − R + 1)
N!
=
(N − R )!
Let us take yet another example,

Example 3.4: How many five-letter codes in English alphabet can be written in
which no letter is repeated.
Solution: There are 26 letters in English alphabet. The first place of the
required code can be occupied by any letter from A to Z i.e.,
there are 26 options. Since the letters are not to be repeated,
the second place can be occupied in 25 different ways, third
place in 24 ways, fourth place in 23 ways and fifth place in 22
ways. Therefore, the total number of possible arrangements will
be P = 26 25 24 23 22=7893600
Thus, there are 7893600 unique five-letter codes that can be
created by using all 26 letters of English alphabet

The example 3.4 represents a type of permutation of objects in which out of all
available objects ( N , =26) only a subset ( R , =5) is used in the arrangement.
The expression obtained in the example 3.3 can be written in terms of N and
R as
P = N(N − 1)(N − 2)(N − 3)...(N − R + 1) … (3.6)
81
Block 1 Thermodynamics and Statistical Thermodynamics

This is the general expression for the possible number of permutations of R

N
objects out of a set of N objects. It is represented as PR or P (N, R ) .
Multiplying and dividing the Eq. (3.6) by (N − R )! we get

(N − R )(N − R − 1).....(3)(2)(1)
N
PR = N(N − 1)(N − 2)...(N − R + 1) … (3.7)
(N − R )!

The numerator in Eq. (3.7) becomes N ! .Thus, the total number of

A probability permutations for choosing R objects from N objects without repetition is
distribution function is given by Eq. (3.8)
a mathematical
N!
function that
N
PR = N (N − 1)(N − 2)(N − 3)...(N − R + 1) = for 0<R≤N. … (3.8)
expresses the (N − R )!
distribution (variation)
of the probability of This refers to the number of permutations of N objects taken R at a time.
the outcomes of an You may note here that when R = N , the denominator becomes 0! Which is
experiment. equal to 1. Therefore, the number of permutations P(N, R ) = N !

However, if the repetition of the objects is allowed then for every object to be
chosen all the N objects will be available. Thus, the number of permutations
R
for taking R objects from N objects with repetition, will be N .Similarly, the
number of permutations of N objects, where R1 objects are of one kind, R2 are
of second kind, ..., RK are of kth kind is given by the expression.
N!
… (3.9)
R1 ! R2 ! R3 !...Rk !

The proof for this expression can be had from any basic book on statistics.

Configuration

There are certain conditions in which the order of choosing the objects is not
important. For example, if you want to select two books out of three available
books A, B and C, then you can select the books in the following ways:

AB, AC or BC.

In this case whether you choose A first and then B or choose B first and then
A, you have the same set of books. Therefore, the order of choosing the object
does not affect the outcome. You have three different combinations of two
books out of three books. When the order of arranging the objects is not
important, then the collection of R objects out of N objects is known as
combination or configuration. Therefore, configuration is an unordered
arrangement of R objects out of N objects. The key difference between a
permutation and a combination is that the order of the elements in a
combination does not matter, whereas it does so in case of a permutation. In
simple words, there is no first place, second place, and so on, in case of
combination. The combinations pqrs and prsq are identical and are not
counted twice whereas these represent two of the permutations of alphabets
p, q, r, and s.

Thus, the number of combinations represents all the different subsets

containing R items that can be sampled from N distinct objects. Here, the
specific arrangement of R objects within a subset is irrelevant. The number of
82
Unit 3 Introduction to Statistical Thermodynamics-I

combinations can then be obtained by dividing the number of permutations

P (N, R ) by the number of permutations of R distinct objects. That is, the
number of combinations equals the number of permutations of R objects
selected out of a total of N objects divided by the number of permutations of
the selected R objects. Mathematically,
N
Number of combinations= PR / R ! … (3.10)

Substituting for N PR we get,

N
PR N!
C(N, R ) = NCR = = 0 R N … (3.11)
R! R !(N − R )!

This can be seen as the number of combinations of N objects taken R at a

time. You may note that C (N, R ) can also be visualised as the number of
different arrangements of N objects when R of these are of one distinct type
and (N − R ) are of a second type. Let us take up some examples to learn about
the calculation of permutations and configurations for a given system.

Example 3.5: Calculate the number of permutations and configurations

possible while selecting three balls out of five different balls
taken in a box.
Solution: In this case, N = 5 and R = 3.
Total number of permutations
5! 5! 5 4 3 2!
= 5P3 = = = = 5 4 3 = 60
(5 − 3)! 2! 2!

Total number of configurations

5! 5! 5 4 3!
= 5C3 = = = = 10
3!(5 − 3)! 3! 2! 3! 2 1

The number of permutations is 60 whereas the configurations are just 10. This
implies that for each configuration, there can be more than one permutation
possible.

The expressions for number of possible outcomes for picking R objects out of
a total of N objects under different conditions as discussed in the examples
above are summarised in Table 3.1.
Table 3.1: The expressions for number of possible outcomes for picking
R objects out of a total of N objects under different
conditions.
Replacement? Ordered? Expression
N!
No Yes
N
PR =
(N − R )!
N!
No No
N
CR =
R !(N − R )!
Yes Yes NR
(N − R − 1)!
Yes No
R !(N − 1)!
We have so far considered situations for picking up a certain number of
objects out of a given total under different conditions. Let us now take up 83
Block 1 Thermodynamics and Statistical Thermodynamics

examples of distributing objects with or without restrictions. Let us take three

identical boxes and consider the distribution of a red–coloured ball amongst
them. We know that it can be accomplished in three different ways, Fig. 3.1(a)
i.e., the ball is in either first, second, or the third box. Now if we take two
identical red balls (i.e., the balls are indistinguishable) and try to distribute
them into the three identical boxes with the condition that each box can have
only one ball. We find that again there are three possibilities as shown in
Fig.3.1(b).

Fig. 3.1: Possible ways of distributing (a) one and (b) two identical balls
amongst three identical boxes.

However, if we take a red and a blue ball (i.e., the balls are distinguishable
now) and distribute amongst three boxes again with the condition that each
box can have only one ball. We find that now there are six possible ways of
doing so as shown in Fig. (3.3).

Fig. 3.2: Possible ways of distributing two distinguishable balls amongst three
identical boxes.

We can say that when distinguishable objects are distributed into separate
subsystems there are more possible ways of doing so as compared to the
indistinguishable objects. Let us take yet another example.

Example 3.6: a) Calculate the number of ways in which three balls of different
colours can be distributed amongst four identical boxes
without any restriction.
b) Calculate the number of ways if we put a condition that box 1
can have only one ball and box 2 can have two balls
whereas the third and the fourth boxes have no ball at all.
Solution: a) Let us label the balls as R (red), B (blue) and G (green). The ball
R can be placed in any of four boxes and for each of these, the
ball B can also be placed in any of the four boxes. Thus, there
are 4 X 4 =16 ways of placing the R and B balls. Again, for each
of these there are 4 ways of placing the ball G, so the total
84 number of ways is 4 4 4 = 64.
Unit 3 Introduction to Statistical Thermodynamics-I

b) Under the conditions of given restrictions, the number of ways reduces to

just three. These are as shown below

In general, when a total of N distinguishable objects are distributed in k

boxes such that they can have N1 , N2 , N3 , Nk objects respectively, then
according to combinatorial mathematics the number of ways W is given
by the following expression.
N!
W=
N1 ! N2 ! N3 ! Nk !

In the present case, N =3, N1 =1, N 2 =2, N3 =0 and N 4 =0

3!
W= = 3 (Note 0!=1)
1!2!0!0!

Why don’t you answer the following simple questions to assess your
understanding?

SAQ 4
How many 3–card arrangements can be drawn from a deck of 52 playing
cards?

SAQ 5
For an international cricket series, a group of 16 players are picked up. On the
day of the match a team of 11 players is to be selected. If the captain of the
team is fixed, how many 11 membered teams could be selected?

3.3.2 Stirling Approximation

You have learnt above that the calculation of permutation and combination (or
configurations) involves the calculation of factorials. It is easier to calculate
factorials for small numbers as used in the examples above. However, in
chemistry we deal with very large numbers, say of the order of 1022, whose
factorials are beyond the range of most computing devices. You may have
noticed that common handy scientific calculator indicates ‘out of range’ even
for 70! as the result is greater than 10100, the limit of the calculator. In such
cases, we can use an expression called Stirling’s approximation to calculate
the factorials. Let us derive the expression for the same.
The factorial of number N is defined as
N ! = N(N − 1)(N − 2)(N − 3)...(3)(2)(1) … (3.12)
85
Block 1 Thermodynamics and Statistical Thermodynamics

Taking natural logarithm of the above expression we get,

ln N ! = ln N(N − 1)(N − 2)(N − 3)...(3)(2)(1) … (3.13)

= ln N + ln(N − 1) + ln(N − 2) + ln(N − 3)... + ln(3) + ln(2) + ln(1) … (3.14)

We can write the expression as a summation of the terms on right as

N =N
ln(N !) = ln(N ) … (3.15)
N =1

Since, N is very large, summation can be replaced by integration within the

limits. Therefore, we can write.
N
ln(N !) = ln N dN = [N ln N − N ]1N
1 … (3.16)
= ( N ln N − N ) − (1ln1 − 1) N ln N − N

Thus, according to Stirling’s approximation

ln N ! ~ N ln N − N … (3.17)
You may note that Stirling’s approximation in fact approximates ln N ! and not
N ! . The factorial is a product of numbers whereas ln N ! is a sum which is
easier to work with. The actual and approximated values of ln N ! for different
values of N are compiled below.

N ln N ! N ln N − N Percent error

5 4.79 3.47 27.56

10 15.10 13.03 13.71
100 363.74 360.52 0.89
1000 5912.13 5907.70 0.07
100,000 1051299.00 1051292.55 0.00061

You can note that the Stirling Approximation gives good results for very large
values of N ,however, for smaller values of N the results don’t match the
actual value. Let us learn about the probability distribution functions. However,
before that answer the following simple question to assess your understanding
of Stirling approximation.

SAQ 6
Calculate the percent error introduced in computing the value of 50! using
Stirling approximation. (You may use a scientific calculator for the purpose).

3.4 PROBABILITY DISTRIBUTION FUNCTIONS

As shown above, when an experiment is carried out, there are different
possible outcomes, and each outcome has some probability of occurrence. To
characterise a random variable, we first need the information about all the
86 possible outcomes of the elementary events of the random variable and then
Unit 3 Introduction to Statistical Thermodynamics-I

get the probability of a given value as a possible outcome. Recall that in the
case of tossing of a coin twice, there were four possible outcomes, and we
could calculate the probability of getting a particular outcome. The probability
distribution refers to the variation of probability of occurrence of all possible
outcomes of an experiment. Each outcome has a its own probability of
occurrence, however, the total probability of occurrence of all the events is
always one.

Let us recall the experiment given above in which two dice are rolled
simultaneously. The minimum value of the sum of the numbers on the top
faces of the two dice is 1+1=2 and the maximum value of the sum is 6+6=12;
the other possible sums being 3,4, 5, 6, 7, 8, 9, 10, and 11. These sums can
be obtained with different permutations as shown in Table 3.2.

Table 3.2: Number of permutations and the probabilities of different outcomes of

rolling two dice.

Sum Permutations Number of Probability

permutations
2 1,1 1 1/36=0.028
3 2,1; 1,2 2 2/36=0.056
4 2,2; 1,3; 3,1 3 3/36=0.083
5 2,3; 3,2; 1,4; 4,1 4 4/36=0.111
6 1,5; 2,4; 3,3; 4,2; 5,1 5 5/36=0.139
7 1,6; 2,5; 3,4; 4,3; 5,2; 6,1 6 6/36=0.167
8 2,6; 3,5; 4,4; 5,3; 6,2 5 5/36=0.139
9 3,6; 4,5; 5,4; 6,3 4 4/36=0.111
10 4,6; 5,5; 6,4 3 3/36=0.083
11 5,6; 6,5 2 2/36=0.056
12 6,6 1 1/36=0.028

In this case the possible outcomes for the two dice are 6 each and are
independent of each other. Then, according to the fundamental counting
principle, given above, the total number of possible outcomes, i.e., the number
of permutations is 62 =36. The last column of the table gives the probability of
each possible set of permutation for the sum of the numbers on the top faces
of the dice. You can clearly see that the probability depends upon the
outcome. A plot (variation) of probability against the outcome is referred to as
probability distribution.
The probability distribution is generally expressed in terms of a mathematical A probability
function of outcome i.e., the variable of interest. If X is the variable of interest, distribution function is
then the probability distribution function is written as P(X). For a specific a mathematical
outcome, the function is written as P(Xi). In the above example, the variable of function that
interest is the sum of the digits on the top face of the two dice with discrete expresses the
values from 2 to 12. Probability of different outcomes lies between zero and distribution (variation)
of the probability of
one and the sum of probabilities of all the outcomes is always one.
the outcomes of an
You must have realised by now that if the number of possible arrangements is experiment.
small, we can easily determine them by counting. For example, in the above
case the possible arrangements (permutations) or probabilities for different 87
Block 1 Thermodynamics and Statistical Thermodynamics

outcomes can be easily determined and represented by a histogram as shown

in Fig. (3.3(a)).

Fig. 3.3: (a) For small number of possible outcomes the data can be represented
in terms of a histogram and the total probability is the sum of the
individual probabilities (b) A smooth function representing the
distribution of probabilities of different outcomes; the total probability
is given by the area under the curve. The shaded portion represents
the possible arrangements in the selected range of outcomes.

Since the outcomes in this experiment have discrete values the probability
distribution function is called discrete probability distribution function. In
such a case, the total number of possible arrangements can be obtained by
adding the values represented by the bars in the histogram. Mathematically,
N
P ( X1 ) + P ( X 2 ) + P ( X 3 ) + ... = P( X i ) … (3.18)
i =1

The probability of a variable having the given value fi in the sample space is
defined as
fi
P( X i ) = N … (3.19)
f
i =1 i

The average value of the function is equal to the sum of the value of the
function for all values of X i multiplied by the probability of the variable with the
corresponding value i.e.,
N
f(X) = f ( X i )P ( X i ) … (3.20)
i =1

However, when the number of possibilities increases, it becomes increasingly

difficult and time–consuming process to count the possible number of
permutations. After a certain stage it becomes impossible to determine the
total number of possibilities by adding the possible permutations for each
outcome. In such cases where the possible outcomes do not form a discrete
set but a continuum in R ,the distribution of probabilities can be expressed by
a smooth function (Fig. (3.3(b)) and is called continuous probability
distribution function. The total number of possible arrangements in such
cases equals the area under the curve. This can be determined by suitably
integrating the probability function. In probability density function, the
probability of an outcome in a small range of outcomes dX is given by the
expression p(X)dX. Here, p(X) is called probability density and represents
the probability of an outcome to have a value between x and x + dx. The total
88 probability in the range is given by the following integral.
Unit 3 Introduction to Statistical Thermodynamics-I
X2

p( X )dX … (3.21)
X1

The probability distributions are usually normalised. It implies that these satisfy
the following criteria
p( X )dX = 1 …(3.22)

The probability of a variable having the given value f ( X ) in the ranged X is

defined as
f ( X )dX
P ( X )dX = X2 … (3.23)
f ( X )dX
X1

The probabilities can be used to determine the average values of some

variable. Let us consider variable X that can have certain possible individual
values Xi. If we represent the probability that a given value Xi exists as Pi .
Then the average value of the variable X is given as
N
X i Pi
X = i =1 … (3.24)
Pi
i

Let us sum up what have we learnt in this unit. However, before that answer
the following simple questions to assess your understanding.

SAQ 7
a) Calculate the number of permutations and configurations possible while
selecting three toys out of four different toys labelled A, B, C, and D taken
in a box.
b) Identify explicitly each combination and permuation for the three chosen
toys.

3.5 SUMMARY
In this unit we have introduced you to an alternative approach to understand
thermodynamic properties of a system by applying statistics to large
collections of microscopic systems like atoms and molecules. We began the
unit by discussing the relationship between classical and statistical
thermodynamics. Herein we stated that whilst the classical thermodynamics
ignores molecular complexity i.e., the quantum mechanical nature of the
molecules, the statistical thermodynamics is based primarily on the quantum
mechanical description of atoms and molecules.
Since statistical thermodynamics provides a probabilistic approach to
understand the thermodynamic properties, we took up basic aspects of
probability and related concepts. Herein, we introduced the concept of
probability and explained the meaning of randomised variables, their types,
sample space, permutations, and configurations with the help of suitable
examples. Thereafter, we took up an important approximation i.e., Stirling
approximation and discussed its need and importance. Then we discussed
about probability distribution functions. 89
Block 1 Thermodynamics and Statistical Thermodynamics

3.6 TERMINAL QUESTIONS

1. What is the probability of selecting a card of spade from a standard deck
of 52 playing cards?

2. What are randomised variables? Name their types.

3. Consider that an unbiased coin is tossed thrice. List all possible outcomes
and calculate the probabilities of i) getting heads exactly twice and ii)
getting tails at least once.

4. Consider that a fair dice is rolled twice and the numbers on the top face
are designated as A and B. Assuming that all outcomes are equally likely.

i) Write down the sample space.

ii) Calculate the probability of the event defined as A+B=8

5. Write the expression for permutation for choosing when R objects out of
N objects when(i) repetition is not allowed (ii) repetition is allowed (iii)
where R1 objects are of one kind, R2 are of second kind.

6. There are a set of 20 natural amino acids. Calculate the number of

tripeptides that can be created by using them under the following
conditions:

(i) The amino acids are replaced.

(ii) The amino acids are not replaced.

3.7 ANSWERS
Self–Assessment Questions
1. Thermodynamics is a phenomenological science that is based on the
observations of the phenomenon in the macroscopic world. The science of
thermodynamics was developed before an understanding of the nature of
atoms and molecules was developed. The thermodynamics properties of a
system can either be directly measured experimentally or are determined
by the experimental data on other thermodynamic properties. The laws are
universal.

Statistical mechanics applies statistical tools to the quantum mechanical

description of matter at microscopic i.e., atomic, and molecular level and
provides a framework which predicts the macroscopic properties of the
thermodynamic system.

2. The collection of all possible outcomes of an experiment is known as the

sample space. In a standard deck of playing cards, there are three types of
photo cards viz., Jack, Queen, and King. Each one of these types of photo
cards can have four variants viz., hearts, spades, clubs, and diamonds.
The desired outcome is the photo cards of heart of diamond. Therefore,
the sample space would be:

[Jack of heart, Queen of heart, King of heart, Jack of diamond, Queen of

90 diamond, King of diamond.]
Unit 3 Introduction to Statistical Thermodynamics-I

3. The continuous variables are the ones that can have any value in the
given range of values. The sample space for the temperature of Delhi as
given will be represented as

[ 2oC T 46oC]

4. Each 3-card hand is a configuration obtained from the deck of 52-cards.

Thus, the values of N and R are as follows:
N =52 and R =03
Using Eq. (3.11) we can calculate the number of possible permutations as
N!
N
CR =
R !(N − R )!

52! 52 51 50 49!
N
CR = =
3! 49! 3 2 1 49!

Solving, we get the number of permutations as 795, 600.

5. We have a group of 16 players of which we need to pick up 11 members

for the playing team. Since the choice for the captain of the team is fixed,
we are left with fifteen players from which we need to pick up 10 players.
15! 15 14 13 12 11 10!
The number of possible teams = =
10!5! 10! 5 4 3 2 1

15 14 13 12 11
The number of possible teams = = = 3003
5 4 3 2 1

6. The calculated value of 50! is = 3.014 1064

The corresponding value of In 50! = 148.478.

As per Stirling approximation In 50! = (50 ln50) − 50 = 145.601= 145.601

Difference in the calculated and estimated values of ln 50!

= 148.478 -145.601= 2.877
Percentage error = (2.877 x 100)/148.478=1.94%
7. a) In this case, N= 4 and R = 3.
4! 4!
Total number of permutations = 4P3 = = = 4 3 2 = 24
(4 − 3)! 1!

4! 4! 4 3!
Total number of combinations = 4C3 = = = =4
3!(4 − 3)! 3! 1! 3! 1

b) The four combinations are ABC, ABD, ACD, and BCD. Each of these
four combinations can be permuted in 3! = 6 ways. For these
combinations the permutations are as under

ABC: ABC, ACB, BAC, BCA, CAB, and CBA

ABD: ABD, ADB, BAD, BDA, DAB, DBA
ACD: ACD, ADC, CAD, CDA, DAC, DCA
BCD: BCD, BDC, CDB, CBD, DBC, DCB 91
Block 1 Thermodynamics and Statistical Thermodynamics

Terminal Questions

1. In a standard deck of cards, there are 4 suits (hearts, spades, clubs, and
diamonds) of13 cards each. Thus, the sample space consists of the 52
cards. Of these, 13 correspond to the event of interest (selecting a
13
spade). Therefore, the required probability will be, = 0.25
52

2. The random or randomised variables, or quantities are those variables

whose value can change throughout the course of an experiment or series
of events. There are two types of random variables viz., discrete variables
and continuous variables.
3. In this case we have a coin that is to be tossed three times. In every toss
we have two possible outcomes i.e., a head or a tail. We can write down
all the possible outcomes of the experiment as follows:

Outcome Toss 1 Toss 2 Toss 3 Outcome Toss 1 Toss 2 Toss 3

No. No.

1 H H H 5 H T T

2 T H H 6 T T H

3 H T H 7 T H T

4 H H T 8 T T T

Thus, there are a total of eight possible outcomes. The desired outcomes
and the corresponding probabilities are as under

i) Getting heads exactly twice: You may note that the outcome number
2,3 and 4 i.e., three outcomes out of a total of eight meet the
requirement. So, the probability will be 3/8=0.375

ii) Getting tails at least once: You can note that all the listed outcomes
except at number 1 meet the requirement. So, the probability will be
7/8=0.875

4. The Since the dice is fair, the possible outcomes on each rolling of the
dice are 1,2,3,4,5, and 6. Therefore the sample space will be.

[(1,1), (1,2), (1,3), (1,4), (1,5), (1,6), (2,1), (2,2), (2,3), (2,4), (2,5), (2,6),
(3,1), (3,2), (3,3), (3,4), (3,5), (3,6), (4,1), (4,2), (4,3), (4,4), (4,5), (4,6),
(5,1), (5,2), (5,3), (5,4), (5,5), (5,6), (6,1), (6,2), (6,3), (6,4), (6,5), (6,6)]

There are five outcomes corresponding to the desired event: (2,6), (3,5),
(4,4), (5,3), (6,2). Therefore, the probability of the event = 5/36 = 0.139

(It is important to note that the outcomes (2,6), & (6,2) are different. The
outcome, (2, 6) implies that the first roll of dice gives a 2 whereas the
second roll gives a 6. On the other hand (6, 2) implies that the first roll of
dice gives a 6, whereas the second roll gives a 2. Similarly, (3,5), & (5,3)
are also different.
92
Unit 3 Introduction to Statistical Thermodynamics-I

5. The required expressions are as under

n!
(i) repetition is not allowed n Pr =
(n − r )!

(ii) repetition is allowed and r1 objects are of one kind, r2 are of second
n!
kind P =
r1! r2!

6. We are given a set of 20 amino acids, and we know that a tripeptide

contains three amino acids. Now

i) In the case when the amino acids are replaced, 20 amino acids are
available each time to be picked up for creating the tripeptide. That is
the possible number of tripeptides that can 02 be formed will be.

20 x 20 x 20 = 8000

ii) In this case since the amino acids are not replaced, we will have a
choice of 20 amino acids at the first instance but subsequently we will
have respectively 19 and 18 amino acids available. That is, the
possible number of tripeptides that can be formed will be.

20 19 18 = 6840

93
Block 1 Thermodynamics and Statistical Thermodynamic

Unit 4

INTRODUCTION
TO STATISTICAL
THERMODYNAMICS-II

Structure
4.1 Introduction 4.4 Canonical Ensemble

Expected Learning Outcomes 4.5 Thermodynamic Properties

from Partition Function
4.2 Boltzmann’s Distribution
Average energy of a molecule
Microstates and Configurations
4.6 Summary
Derivation of Boltzmann’s
Distribution 4.7 Terminal Questions
4.3 Molecular Partition Functions 4.8 Answers

4.1 INTRODUCTION
In the previous unit you were introduced to the basic aspects of the statistical
approach to the understanding of thermodynamic properties of a system. You
have learnt that the statistical thermodynamics provides the link between
microscopic perspective of matter (the quantised energy levels at the atomic /
molecular level) and its macroscopic or bulk (observable) properties. You have
also learnt about the basic ideas of probability theory wherein you have learnt
the concepts of probability and the meaning and significance of permutations
and configurations, and probability distribution functions.

In this unit we will continue our discussion on statistical thermodynamics and

begin the unit by taking up Boltzmann distribution function. It is the
cornerstone of statistical thermodynamics and tells us about the frequency of
occurrence of the individual microscopic states of a system at a given
temperature. In the process of deriving expression for the Boltzmann’s
distribution function we will introduce the crucial concept of partition function.
94 This will be followed by formulating expressions for molecular partition
Unit 4 Introduction to Statistical Thermodynamics-II

functions corresponding to different types of quantised energies. We will then

take up the idea of canonical ensemble and outline its significance. Towards
the end of the unit, we will demonstrate how statistical mechanics helps us
determine the thermodynamic properties of the bulk matter by taking the
example of average energy of a molecule.
In the next unit we will take up the fundamentals of solid state.

Expected Learning Outcomes

After studying this unit, you should be able to:
❖ define and differentiate between microstates and configurations;
❖ define and calculate the weight of a configuration;
❖ demonstrate that for extremely large number of particles, the system
predominantly exists in the dominant configuration;
❖ derive an expression for the Boltzmann distribution function;
❖ outline the constraints operative while deriving Boltzmann distribution
function;
❖ briefly describe the Lagrange’s method of undetermined multipliers;
❖ outline the significance of Boltzmann’s distribution function;
❖ define and explain molecular and canonical partition functions;
❖ derive expressions for translational, rotational, vibrational, and electronic
partition functions;
❖ calculate electronic partition functions for a given system;
❖ explain the concept of canonical ensemble and outline its significance;
and
❖ derive the expression for average energy of a molecule using partition
functions.

4.2 BOLTZMANN’S DISTRIBUTION

In a system containing many molecules, the available quantised energy states
are occupied by different fractions of molecules. That is, the entire population
of molecules is distributed among different energy states. Alternatively, we can
say that the total energy of the system is distributed amongst the constituent
molecules. Ludwig Boltzmann, an Austrian mathematician and physicists was
the first person to propose an expression to calculate the population of
different energy states. The expression is known as Boltzmann distribution
function and is essentially a probability distribution function that allows to
calculate the probability of finding the system in a certain energy state. It is Ludwig Eduard
central to the statistical approach to the [Link] Boltzmann Boltzmann
distribution function is a function of the energy of the state and the (1844-1906),
Austrian physicist
temperature of the system. The study was later extended by William Gibbs.
Before going into the details of Boltzmann distribution function and its
significance, it is important to understand a few related terms. In the section on
basic probability theory, in the previous unit, you have learnt about 95
Block 1 Thermodynamics and Statistical Thermodynamic

permutation and configuration. You are aware that each configuration is

associated with one or more permutations. The probability of a configuration
depends upon the number of permutations associated with it.

4.2.1 Microstates and Configurations

Let us now apply the concept of permutation and configuration to the chemical
systems. Here, instead of placing balls in boxes we would consider assigning
particles (molecules or atoms) to different energy states. Let us take a simple
“molecular” system consisting of three quantum particles that collectively have
an energy equal to say three quanta. The energy states available for the
particles have energies of 0, 0 ,2 0 ,3 0 … and so on; where 0 represents a
quantum of energy. Now, we ask a question, “What is the most probable
distribution of energy?” i.e., what is the most likely way in which the total
energy (3 quanta) is distributed amongst the particles of the system?

To begin with, we assume that the particles are indistinguishable and try to
work out possible distributions. There are three possibilities as shown below in
Fig. 4.1. in the first case all the molecules are in 0 state i.e., having energy
equal to 1 quantum each. In second case, the levels corresponding to
energies of 0, 0 and 2 0 have a molecule each whereas in the third case all
the energy is with one molecule in the state of energy 3 0 . The other two
molecules have zero energy i.e., they are in energy level of energy=0.

Fig. 4.1: Three possible configurations for the distribution of three quanta of
energy amongst three indistinguishable particles.

Thus, we see in the above example, there are three different ways of
arranging three particles (molecules) in various energy states such that the
total energy of the system is 3 0 . Such arrangements of molecules in different
energy states are called configurations. You may note that here we have
assumed that the molecules are indistinguishable as indicated by their colour.
Let us take one more example.

Example 4.1: Consider a system consisting of five indistinguishable particles

(N = 5) which can occupy any of the available energy levels
having energies equal to 0 , 2 0 , 3 0 , 4 0 and 5 0 . If the total
energy of the system is 10 0 , identify different possible ways in
which the molecules can occupy various energy states.
Solution: As discussed above, in this case also there are several constraints
such as the total energy of the system is fixed, the total number of molecules
cannot change, and the available energy levels are also fixed. Based on these
restraints, there are six different ways of arranging five particles in various
96 energy states such that the total energy of the system is 10 0 i.e., there are six
Unit 4 Introduction to Statistical Thermodynamics-II

configurations. Different possible ways of distributing five indistinguishable

molecules in the available energy states are shown below:

Let us see what happens if the molecules are distinguishable. You would The distinguishable
recall from previous unit, that when distinguishable objects are distributed into here implies, that
separate subsystems there are more possible ways of doing so as compared each particle in the
to that for the indistinguishable objects. In such cases each configuration can system can be readily
have one or more than one permutation possible. identified.

Let us reconsider the example of distribution of three quanta of energy

amongst three quantum particles discussed above but now we take these
particles to be distinguishable. We find that the three configurations identified a microstate is
for the case with indistinguishable particles can have one or more equivalent of
permutations as shown in Fig. 4.2. The distinguishability of the particles is permutation in
indicated in terms of the colour of the spheres representing particles. The probability theory.
configurations 1, 2 and 3 have 1,6 and 3 permutations respectively as shown.
The specific arrangement (permutations) of particles in a configuration is
known as a microstate. In other words, we can say that a microstate is
equivalent of permutation in probability theory.

Fig. 4.2: The configurations and the corresponding microstates for the
distribution of three quanta of energy amongst three distinguishable
particles.

A configuration in this case is represented as { N0 , N1, N2 , N3 } where

N0 , N1, N2 , N3 refer to the number of particles in the energy states 0, 0 , 2 0 , 3 0 97
Block 1 Thermodynamics and Statistical Thermodynamic

respectively. That is, the three configurations given above will be represented
as {0,3,0,0}; {1,1,1,0}; and {2,0,0,1} respectively. It is important to note that the
total energy of all the arrangements is the same and that the only difference is
the distribution of the energy over the particles.

The total number of microstates associated with a given configuration is

known as the weight of the configuration and is represented by W. Now, a
question comes up is that ‘which of the possible configurations is likely to be
observed for the system at equilibrium? Since all the microstates have an
equal a priori probability of being observed, the probability of observing the
{1,1,1,0} configuration with maximum number of microstates (W=6), will be
maximum in this case. The probability of observing this configuration is simply
the number of microstates associated with this configuration divided by the
total number of microstates possible, for all the configurations put together i.e.,
6
= 0.6 .
10
The three particles case discussed above provides a general approach to
determine the most probable configuration of energy for a chemical system.
That is, we work out all the possible configurations of energy and determine
the corresponding microstates. The configuration with the greatest number of
microstates is the likely configuration. Such an approach is fine for systems
with small number of particles. However, it is a difficult task to do similar
exercise for any practical chemical system having astronomical number of
particles. For such cases, we need to find a way to obtain the number of all the
microstates associated with a configuration without any counting. Fortunately,
it is possible to work out an expression for the same. However, before that
answer the following simple question to assess your understanding.

SAQ 1
What is the relationship between microstates and configuration?

The problem in hand is to work out the number of possible ways of distributing
N distinguishable particles amongst different energy states i.e., 0 , 1 , 2 …such
that there are N0 , N1, N2, N3 ... particles in them respectively. To begin with, we
assume that a given energy state can be had only by a single particle. Now,
since there is a total of N particles, we can assign an energy state to the first
particle in N different ways. Once an energy state is assigned to a particle,
the next energy state can be assigned to any one of the remaining particles;
and there are (N − 1) different ways to do so, and so on. Thus, there are
N(N − 1)(N − 2)...3 2 1 i.e., N ! ways of assigning the energy states to N
distinguishable particles.
You would recall from your earlier classes that in a chemical system all the
particles (atoms and molecules) do not have same energy and a given energy
(or velocity) can be possessed by many particles. You may like to revisit
Maxwell’s distribution of molecular velocities (energies) of a gaseous sample.
It implies that more than one molecule can have the same energy i.e., N
distinguishable particles are so assigned to different energy states, that
98 N0 , N1, N2, N3 ... particles respectively have energies of 0 , 1 , 2 …Thus, by
Unit 4 Introduction to Statistical Thermodynamics-II

considering that a given energy state can be had only by a single particle we
have overestimated the possible microstates and we need to make necessary
changes by correcting for the overcounting of identical permutations.
Now since each of N0 particles can have energy equal to 0 , there are N0 !
different ways in which the distinguishable particles could have been selected
to have energy state, 0 . Similarly, N1 particles could be assigned 1 energy state
in N1 ! ways and so on. Therefore, the total number of distinguishable ways of
distributing N particles to different energy states as required, regardless of the
order in which the particles were chosen i.e., the weight of a configuration is
given by the following expression.

N!
W = … (4.1)
N0 ! N1 ! N2 ! N3 !...

In the three-particle example, N = 3 and we can calculate the number of

microstates for different configurations as given in Table 4.1.
Table 4.1: Number of permutations (microstates) for each configuration for a
system containing 3particles with total energy equal to 3 0 .

Configuration Number of particles in No. of Probability

different energy levels Permutations
(weight)

I N0 = 0, N1 = 3, N 2 = 0 3! 1/10=0.1
=1
3!
{0,3,0}

II N0 = 1, N1 = 1, N 2 = 1 3! 6/10=0.6
=6
1!1!1!
{1,1,1}

III N0 = 2, N1 = 0, N 2 = 1 3! 3/10=0.3
=3
2!1!
{2,0,1}

Total 10 1.0

th th
The probability of observing i configuration is given by the weight of i
configuration divided by the total weight of all the configurations put together,
i.e.,

Wi W
Pi = = N i … (4.2)
W1 + W2 + W3 + ...WN
Wi
1

Similarly, we can calculate the number of microstates and their probabilities for
each of the configurations in Example 4.1. The results are summarised in
Table 4.2. The configuration I has only one microstate, so it has the least
probability (0.008) while configurations III and IV with thirty microstates each,
have a probability of 0.248 each.
99
Block 1 Thermodynamics and Statistical Thermodynamic

Table 4.2: Number of permutations (microstates) for each configuration for a

system containing 5 particles with total energy equal to 10 0 .

The possible
Configuration Number of particles in No. of Probability
distributions are
different energy states Permutations
referred to as
(weight)
macrostates or
configurations 5!
I N2 = 5 =1 1/121=0.008
whereas, different 5!
{0,5,0,0,0}
ways of getting the
same macrostates II N1 = 1, N 2 = 3, N3 = 1 5! 20/121=0.165
= 20
are called 1!3!1!
{1,3,1,0,0}
microstates or
complexions. III N1 = 2, N 2 = 1, N3 = 2 5! 30/121=0.248
= 30
2!2!
{2,1,2,0,0}

IV N1 = 2, N 2 = 2, N 4 = 1 5!
= 30 30/121=0.248
The configuration 2!2!
{2,2,0,1,0}
having the maximum
weight is known as V N1 = 3, N 2 = 1, N5 = 1 5! 20/121=0.165
= 20
dominant 1!3!1!
{3,1,0,0,1}
configuration.
VI N1 = 3, N3 = 1, N 4 = 1 5! 20/121=0.165
= 20
1!3!1!
{3,1,0,1,0}

Total 1 1.0

You would recall from the previous unit that such a distribution is dictated by
probability and the most likely distribution will correspond to a configuration
with the highest number of associated permutations. Therefore, in the current
example, the configurations III and IV are most likely i.e., these are dominant
configurations. In other words, the configuration having the maximum weight is
known as dominant configuration.

Let us take an example, to learn about the idea of dominant configuration.

Example 4.2: a) Consider a system consisting of five molecules which can

occupy any of the available energy states having energies
equal to 0, , 2 , 3 , 4 and 5 . If the total energy of the
system is only 5 , calculate different possible ways in which
the molecules can occupy various energy levels.

b) Repeat the exercise with

i) 50 ii) 500 and iii) 1000

molecules to be distributed amongst the same energy levels with the total
energy still being 5

c) Comment on the results

Solution: a) The possible distributions and the corresponding weights and

100 probabilities for five molecules under given set of conditions are
Unit 4 Introduction to Statistical Thermodynamics-II

Distribution / N0 N1 N2 N3 N4 N5 W Probability
configuration =W
Wtot

1. 4 0 0 0 0 1 5 0.04
2. 3 1 0 0 1 0 20 0.16
3. 3 0 1 1 0 0 20 0.16
4. 2 2 0 1 0 0 30 0.24
5. 2 1 2 0 0 0 30 0.24
6. 1 3 1 0 0 0 20 0.16
Total 125 1.00

Here, the configuration number 4 and 5 are the dominant configurations

with a probability of 0.24 each.

b) i) The possible distributions and the corresponding weights and probabilities

for fifty molecules under same set of conditions (total energy = 5 ) are

Distribution N0 N1 N2 N3 N4 N5 W Probability =
W
Wtot

1. 49 0 0 0 0 1 50 0.00
2. 48 1 0 0 1 0 2450 0.00
3. 48 0 1 1 0 0 2450 0.00
4. 47 2 0 1 0 0 58800 0.02
5. 47 1 2 0 0 0 58800 0.02
6. 46 3 1 0 0 0 921200 0.29
7. 45 5 0 0 0 0 2118760 0.67

Total 3162510 1.00

Here, the configuration number 7 is the dominant configuration, with the

probability of 0.67.

ii) The possible distributions and the corresponding weights and probabilities
for five hundred molecules under conditions of total energy = 5 are

Distribution N0 N1 N2 N3 N4 N5 W Probability
=W
Wtot
1. 499 0 0 0 0 1 500 1.182x10-9
2. 498 1 0 0 1 0 249500 9.392x10-7
3. 498 0 1 1 0 0 249500 9.392x10-7
4. 497 2 0 1 0 0 62125500 0.00023
5. 497 1 2 0 0 0 62125500 0.00023
6. 496 3 1 0 0 0 10292124500 0.0387
7. 495 5 0 0 0 0 255244687600 0.9608
Total 265661562600 1.00 101
Block 1 Thermodynamics and Statistical Thermodynamic

Here again, the configuration number 7 is the dominant configuration, but

As the number of
molecules increases, with a probability of 0.96; it has increased substantially.
the weight of the most
ii) The possible distributions and the corresponding weights and probabilities
probable configuration/
for one thousand molecules under conditions of total energy = 5 are
probability of the
dominant configuration
also increases. Distribution N0 N1 N2 N3 N4 N5 W W
Wtot

1. 999 0 0 0 0 1 1000 1.188x10-10

2. 998 1 0 0 1 0 999000 1.868x10-7
3. 998 0 1 1 0 0 999000 1.868x10-7
4. 997 2 0 1 0 0 498001500 5.916x10-5
5. 997 1 2 0 0 0 498001500 5.916x10-5
6. 996 3 1 0 0 0 165668499000 0.01968
7. 995 5 0 0 0 0 8250291250200 0.9802
Total 8416957751200 1.00
Here also the configuration number 7 is the dominant configuration. Now,
it has a calculated probability of 0.98.

c) On the basis of the calculations given above, we note that as the number
of molecules increases, the weight of the most probable configuration or
the probability of the dominant configuration also increases.

From the above example, we conclude that as the size of the system
increases, the weight / probability of the dominant configuration goes on
increasing and at a certain stage the dominant configuration outweighs the
rest of the configurations. That is, the probability of dominant configuration is
so high that the system predominantly exists in the dominant configuration. All
microstates have equally a priori probabilities i.e., these are equally probable.
However, large number of them being associated with the most probable
macrostates makes them much more probable than other possible
microstates. In such cases, W becomes the number of microstates for a given
macrostates or configuration. The energy distribution in the dominant
configuration is referred to as the most probable energy distribution.

In real systems we deal with astronomically high number ( N ~1022-23) of

particles. The available numbers of energy levels are also very large. From the
above examples we can say that in real systems, W for the most probable
distribution is so close to Wtot that its probability is ~1. Thus, for real systems
we need to find Wtot which is almost same as W for the most probable
distribution. In other words, we need to find the distribution of energy amongst
the particles in the most probable state.

You may recall that we raised a question above, that ‘what is the most
probable distribution of the particles in a system at thermal equilibrium?’ We
are now equipped to take up this question for a chemical system. We can
visualise such a system at equilibrium in terms of many particles that are so
102 arranged that they can exchange energy with one another. The particles in a
Unit 4 Introduction to Statistical Thermodynamics-II

group have a certain average energy, but at any instant, the energy content of
each particle fluctuates around the average value due to the energy exchange
with other particles. Also, we need to note that the particles are independent of
each other, implying that at a given instant the energy of a given particle is
independent of the energy of all the other particles. Thus, the energies of the
particles are distributed according to a specific probability distribution function
that depends on the equilibrium temperature. Boltzmann’s distribution function
is such a distribution function.

So, the issue boils down to determining the most probable distribution without
having to find any of the less probable distributions. As stated above, the
dominant configuration has the largest weight. If the system changes to any
other configuration, it’s weight will decrease. Thus, we can say that the
dominant configuration corresponds to the peak or maxima of the plot of the
weight as a function of some identifier of the configurations. Let us derive an
expression for Boltzmann’s distribution.

However, before that answer the following simple questions to assess your
understanding.

SAQ 2
Determine the possible ways of distributing 7 distinguishable particles among
4 energy levels with energies of 0, 1,2, and 3 quanta such that the total energy
is equal to 3 quanta.

SAQ 3
Repeat the calculation in Example 4.2 for 100 molecules under same
conditions and calculate the W / WTot ratio for the dominant configuration.

4.2.2 Derivation of Boltzmann Distribution

We have shown above (Eq. 4.1) that the weight of a configuration depends on
the values of all the Ni ' s in it. So, to determine the dominating configuration
we need to find the value of N i that maximises the function W . This can be
achieved by varying the values of N i and looking for the values that
correspond to dW = 0 .You would recall from your earlier classes that to
determine the maxima for a function we have to equate its first derivative to
zero. However, in case of systems with many particles it is mathematically
convenient to maximise the function lnW instead of W . It is so because firstly
for large value of N , the value of W will be very large and secondly, for lnW
we can employ Stirling’s approximation discussed above. Let us take natural
logarithm of Eq. (4.1), and simplify, we get.

ln N !
lnW = … (4.3)
ln(N0 ! N1 ! N2 ! N3 !...)

lnW = ln N !− ln(N0 ! N1 ! N2 ! N3 !...) … (4.4)

lnW = ln N !− ln N0 !− ln N1 ! − ln N2 !... … (4.5) 103

Block 1 Thermodynamics and Statistical Thermodynamic

Putting the ln N i together, we can write.

lnW = ln N !− ln Ni ! … (4.6)
i

Using Stirling’s approximation, the above equation is written as

lnW = (N ln N − N ) − {Ni ln Ni − Ni } … (4.7)
i

Since, Ni = N, we can write,

i

lnW = N ln N − N + N − Ni ln Ni = N ln N − Ni ln Ni … (4.8)
i i

Let us note that a change in the configuration of a system is accompanied by

the change in the population of each state i from N i to Ni + dNi . The
corresponding change in lnW can be given as

lnW
d lnW = dNi = 0 … (4.9)
i
Ni

The term on right hand side gives the summation of the rate of change of ln W
with N i multiplied by the total change in Ni (dNi ). For ln W to be maximum, this
sum must be equal to zero.

So far, we have considered all the states to have same energy, implying that
all the configurations have the same total energy. Let us extend the argument
now to different states having different energies i.e., 0 , 1 , 2 ... Since an
isolated system has a constant total energy so we can retain only those
configurations that have this total energy i.e., the configurations must satisfy
the following condition.

Ni i =E … (4.10)
i

In other words, one particle in the collection is free to gain or lose energy, but
a corresponding amount of energy must decrease elsewhere in the system.
Thus, the following condition should hold.

i dNi = 0 … (4.11)
i

The Eq. (4.10) and Eq. (4.11) represent the energy constraint on the system.
Now, since the total number ( N ) of particles in the system are fixed, we
cannot vary the populations of all the energy states independently. A change
in the population of any state must be accompanied by the corresponding
change (s) in the population of other state (s), i.e., the decrease in the
population of a state must be balanced by an increase in the population of
another state. Mathematically

Ni = N … (4.12)
i

It demands that the sum of all the changes in dNi must be zero. That is.

dNi = 0 … (4.13)
104 i
Unit 4 Introduction to Statistical Thermodynamics-II

The Eq. (4.12) and Eq. (4.13) represent the number constraint of the system.
Thus, the problem reduces to solving the Eq. (4.9) under the constraints as
given in Eqs. (4.11) and (4.13). For this we will use Lagrange’s method of
undetermined multipliers discussed later. This method allows to determine the
maximum value of a function that depends on many variables. In this method
each constraint is multiplied by a coefficient and the resultant is added to the
main equation. The variables are assumed to vary independent of each other
and the values of coefficients are then determined. However, before using this
method we simplify the expression to be maximised.
For this, we rewrite Eq. (4.8) by labelling the states by j in place of i as

lnW = N ln N − N j ln N j … (4.14)
j

Differentiating the equation w. r. t., N i , we get

d ln W dN ln N N j ln N j
= − … (4.15)
dNi dNi j dNi

Solving the terms of right-hand side using differentiation by parts, we get

d lnW dN d ln N dN j d ln N j
= ln N + N − ( )ln N j + N j ( ) … (4.16)
dNi dNi dNi j dNi dNi

We know that.
N = N0 + N1 + N2 + ... + Ni + ... … (4.17)

Since N j is independent of N i , when we differentiate Eq. (4.17) w. r. t., N i ,

the derivative for the i th term on the right will be unity whereas for all the other
dN j dN j
terms ( j terms) it will vanish. That is = 1 when i = j . and , = 0 when
dN i dNi
i j . This fact can be represented in terms of a delta function as

dN j
= ij … (4.18)
dNi

where ij = 1 when i = j and ij = 0 when i j . Thus, we can write.

dN
=1 … (4.19)
dNi

Also, we can write.

d ln N 1
= … (4.20)
dNi N

d ln N j 1 dN j
and = … (4.21)
dNi N j dNi

Using Eq. (4.18) to Eq. (4.21) we can write Eq. (4.16) as

d ln W 1 1 dN j
= [(1 ln N ) + (N )] − (0 ln N j ) + N j … (4.22)
dNi N j N j dNi
105
Block 1 Thermodynamics and Statistical Thermodynamic

d lnW dN j dN j
= (ln N + 1) − ln N j + … (4.23)
dNi j dNi dNi

d lnW
= (ln N + 1) − ij (ln N j + 1) … (4.24)
dNi j

Simplifying we get
d lnW
= (ln N + 1) − (ln Ni + 1) … (4.25)
dNi

d lnW N
= ln N − ln Ni = − ln i … (4.26)
dNi N

d lnW
Since for the dominant configuration, = 0 , for all values of Ni we write.
dNi
Ni
d lnW = − ln dNi = 0 … (4.27)
i N

This is the simplified version of Eq. (4.9). Let us come back to Lagrange’s
method of undetermined multipliers and apply it to the simplified expression,
Eq. (4.27). For this we multiply Eq. (4.11) by an unknown coefficient α and the
Eq. (4.13) by an unknown coefficient, −β and add these terms to Eq. (4.27),
we get.
Ni
d lnW = − ln dNi + dNi − i dNi = 0 … (4.28)
i N i i

Using Eq. (4.26) we can write the right-hand side of the Eq. (4.28) as
d lnW
dNi + dNi − i dNi = 0 … (4.29)
i d Ni i i

Rearranging the terms, we get

Ni
− ln + − i dNi = 0 … (4.30)
i N

Since we have started the derivation by making a small change in Ni so dNi is

not zero. Therefore, Eq. (4.30) will be valid when the term in the brackets is
equal to zero, i.e.,
Ni
− ln + − i =0 … (4.31)
N
d lnW
or + − i =0 … (4.32)
dNi

Rearranging Eq. (4.31) we get.

Ni
ln = −( − + i )= − i
… (4.33)
N

Taking the antilogarithm, we get

Ni
= e .e − i … (4.34)
N
Ni = Ne .e− i
… (4.35)
106
Unit 4 Introduction to Statistical Thermodynamics-II

This equation is the mathematical form of Boltzmann distribution. It gives the

number of molecules in the ith energy state of the dominant configuration.

Ni
is also equal to the probability of occupation of the given energy level. To
N
appreciate the significance of Eq. (4.35) we need to determine the unknown
multipliers i.e., the coefficient α and

Determination of α

We know that the sum of the populations N i of the states is equal to the total
number of molecules, N i.e.,

Ni = N … (4.12)
i

Using Eq. (4.35) and substituting for N i , we can write,

Ne .e − i
=N … (4.36)
i

Simplifying and cancelling N on both sides we get

Ne e− e i
=N … (4.37)
i

1
e e− i
=1 e = … (4.38)
i e− i

The term in the denominator is called partition function and is represented as

q i.e.,

Partition function = q = e− i
… (4.39)
i

Pending the determination of coefficient , we can visualise the partition

function to describe the distribution of energy in a system. For example, using
the partition function with Eq. (4.35), we find that the probability Pi of
occupying a given energy level i becomes

Ni e− i N e− i

pi = = e .e− i
= or i = … (4.40)
N q N q

Eq.(4.40) is the an important result. It quantifies the probability of occupying a

given energy for the dominant configuration of energy. This well-known and
important result is referred to as the Boltzmann distribution. Comparing
Eq.(4.40) to the expression for probability involving discrete variables, viz.,

fi
P( X i ) = … (4.41)
M
fj
j =1

we find that the Boltzmann distribution is nothing more than a statement of

probability, with the partition function serving to normalize the probability
distribution. 107
Block 1 Thermodynamics and Statistical Thermodynamic

Determination of β

In Eq. (4.40), since the exponent (βεi) should be unitless, the coefficient β
must have unit of inverse of energy. This implies that β must be related to
energy. We can derive β by studying the variation of weight of the dominant
configuration with energy. Let us recall Eq. (4.14),

ln W = N ln N − N j ln N j … (4.14)
j

The differential form of the equation can be written as

d ln W = d (N ln N ) − d (N j ln N j ) … (4.42)
j

Since N is the total population of the system and does not change, the first
term of Eq. (4.42) will be zero (i.e., d (N ln N) = 0). Solving the second term on
the right side of the equation we can write,

d ln W = − ln N j dN j + N j d ln N j … (4.43)
j

Simplifying using Eq. (4.21), we get

d ln W = − ln N j dN j + dN j = − ln N j dN j − dN j … (4.44)
j j j

Since, dN j = 0 , we get d lnW = − ln N j dN j … (4.45)

j j

From Eq. (4.35) we have

−
N j = Ne .e j
… (4.35)

For the lowest energy state with energy ε0, the population is given by the
expression.

N0 = Ne .e − 0
… (4.46)

Therefore, the population of any state with respect to the ground state is.
−
Nj e j
− ( j 0)
= =e … (4.47)
N0 e− 0

By convention, the ground state energy is taken as zero and rest of the energy
states are scaled from it. Therefore,
−
N j = N0 e j
… (4.48)

Taking logs,
ln N j = ln N0 − j … (4.49)

Substituting the above expression in Eq. (4.45) we get

d ln W = (ln N0 − j )dN j … (4.50)

j

d ln W = − ln N0 dN j + j dN j … (4.51)
108 j j
Unit 4 Introduction to Statistical Thermodynamics-II

Since dN j = 0 and j dN j = dE … (4.52)

j j

We write, d lnW = dE … (4.53)

You can note that β has the unit of inverse of energy. Let us now consider two
systems, 1 and 2 having distinguishable particles with the weight of dominant
configurations as W1 and W2 respectively. Let us assume that these are
brought together to form an isolated composite system. At the time of mixing,
the weight of the composite configuration will be W1.W2. The system will then
attain equilibrium with some change in the configuration. Let the change in
configuration = d (W1.W2).

Since equilibrium results in a configuration with greater weight,

d (W1.W2) ≥ 0 or W2dW1 +W1dW2 ≥ 0 … (4.54)

Dividing the Eq. (4.54) by W1.W2 we get

dW1 dW2
+ 0 … (4.55)
W1 W2

Writing in the derivative form

d ln W1 + d ln W2 0 … (4.56)

Substituting from Eq. (4.53)

1 dE1 + 2 dE2 0 … (4.57)

Here, β1 and β2 are values of β for systems before mixing. dE1 and dE2 are the
changes in the energies of the two system on mixing. Since the composite
system is isolated

dE1 + dE2= 0 … (4.58)

Let dE1> 0, then dE2< 0. This implies that the initially E2 was greater than E1.
Therefore, since temperature is a measure of energy, T2>T1. In this case, for
Eq. (4.57) to be valid, β1≥ β2. Since β is inversely related to energy, it must
also be inversely related to temperature.
1
… (4.59)
T

Removing the proportionality sign, we get

1
= … (4.60)
kT

Here, k is known as Boltzmann constant and has a value of 1.38 10−23 JK−1

The Boltzmann distribution, Eq. (4.40) can be written as

i
−
Ni e kT N − i kT
= or Ni = e … (4.61)
N q q

The denominator of Eq. (4.61) is termed particle partition function or

molecular partition function and is represented as q . It is one of the most
important quantities in statistical thermodynamics and will be taken up in 109
Block 1 Thermodynamics and Statistical Thermodynamic

details in the next section. We can write the Eq. (4.61) by using the expression
for partition function from Eq. (4.39) as
− i

According to the e kT
Ni = N … (4.62)
Boltzmann distribution − i

e kT

law if the energy i

associated with some
energy state of a This is the mathematical expression for the Boltzmann distribution law and
system is ε then the signifies that if the energy associated with some state or condition of a system
probability of its is ε then the frequency with which that state or condition occurs, or the
occurrence, is probability of its occurrence, is proportional to
−
proportional to e kT
−
e kT
… (4.63)

where T is the system’s absolute temperature and k is the Boltzmann

constant. You may note here that at a given temperature, since the energies of
all the levels are positive, k is also a positive constant and T is always positive,
the population of the levels decreases exponentially with an increase in the
energy of the level. In other words, the ground state has the largest
population, and the population of excited states goes on decreasing; higher
the energy, lower the population. Let us take an example to understand its
application.

Example 4.3: A molecule can be in any of the two states. It’s excited state is
separated from its ground state by 8.0 10−21J . If the ground state
has energy = 0, calculate the fraction of molecules in the
excited state at 300 K.

Solution: we know that according to Boltzmann’s law

− i

Ni e kT
=
N − i

e kT

Now let us first calculate the denominator by substituting the exponents.

8.0 10−21 J
−
− i − 1
(1.381 10−23 JK −1 )(300 K )
e kT
=e +e 0 kT
= 1+ e
i

Solving, we get
8.0 10−21 J
−
− i
(1.381 10−23 JK −1 )(300 K )
e kT
= 1+ e = 1 + e −1.931 = 1.145
i

The the fraction of molecules in the excited state can then be calculated as
8.0 10−21 J
−
(1.381 10−23 JK −1 )(300 K )
Ni e 0.145
= = = 0.127
N 1.145 1.145

Thus, fraction of molecules in the excited state at 300 K would be = 0.127 i.e.,
about 12.7% of the molecules will be in excited state.
110
Unit 4 Introduction to Statistical Thermodynamics-II

The Eq. (4.62) gives the population of i th level w.r.t., the ground state. The
ratio of population of two different energy levels ( i th and jth) can be determined
by eliminating the partition function between them as follows.
i
−
e kT
− i
( i− )
Ni q e kT −
j

= j
= =e kT … (4.64)
Nj − −
j

e kT
e kT
q

Degeneracy and Boltzmann Distribution

So far, we have considered a case where each energy level is associated with
one energy state. You would have learnt in earlier classes that the quantised
energy levels are composed of a finite number of allowed energy states.

(a) (b)
Fig. 4.1: Schematic representation of energy levels and their degeneracies.
(a) both the levels are singly degenerate (b) The 0 level is singly
degenerate whereas level 1 has a degeneracy of 4.

The energy states having same energy are said to be degenerate. The
number of such energy states for i th energy level is termed as its degeneracy
and is denoted as gi. We can visualise such an energy level, as a shelf with a
known number of boxes/ trays equal to the value of the degeneracy, gi as
shown in Fig. 4.1(b). The height of each shelf represents the energy of the
level, and the trays indicate the possible locations for the molecules in the
energy level. Now the question is, how does degeneracy influence probability?
Let us consider two systems the first with single states at energy 0 and i as
shown in Fig. 4.1(a) and the second system in which four states are present at
energy level, i , Fig. 4.1(b). To simplify the calculations, we write the
expression for the partition function as

qsystem1 = g i e −Ei where Ei = i … (4.65)

i
kT

The partition function for the first system will be

qsystem1 = g i e −Ei = 1e0 + 1e −1 = 1 + 0.37 = 1.37
i

Whereas for the second system

qsystem1 = g i e −Ei = 1e0 + 4e −1 = 1 + 1.48 = 2.48
i

It is important to note that all the degenerate energy states will have equal
probability of occupation. That is, the total probability of occupation of a given
energy level with degeneracy g i is equal to
− i
ge kT
pi = i … (4.66)
q 111
Block 1 Thermodynamics and Statistical Thermodynamic

The probabilities of energy level i being occupied in the two cases would be.

0.37
psystem1 = = 0.27
1.37

1.48
psystem 2 = = 0.60
2.48
It is important to note that depending on their nature, some microscopic
particles can be occupied in a way that there is only a single particle per
energy state whereas some other particles, do not have such a limit on their
occupancy. Thus, the particles could either be distinguishable or
indistinguishable and may singly occupy an energy state or any number of
these can be present in an energy state. Let us now take up partition function
in somewhat details. However, before that answer the following simple
questions to assess your understanding.

SAQ 4
A molecule can be in any of the two states. Its excited state is separated from
its ground state by 1.60 x 10−22 J. If the ground state has energy = 0, calculate
the fraction of molecules in the excited state at
i) 300 K ii) 1000 K
Compare the results with the one obtained in Example 4.3.

4.3 MOLECULAR PARTITION FUNCTIONS

In the beginning of the unit, we stated that statistical thermodynamics acts as
a bridge between molecular (microscopic) properties and the bulk properties of
the macroscopic system. The partition function is the tool that relates the
properties of molecule to the macroscopic properties of the system. As
discussed above, the partition function is mathematically defined as
− i
q= e kT … (4.67)
i

This gives the sum over all the energy states in Boltzmann distribution. Recall
that Boltzmann distribution gives the distribution of molecules over different
energy levels in the dominant configuration. Therefore, the partition function is
also known as molecular partition function.

While considering Boltzmann distribution, we had assumed that the

degeneracy of all the energy levels is one. However, since molecules have
many degenerate energy levels, it is important to include degeneracy in the
expression for partition function. Therefore, the molecular partition function
can also be written as
− i
q= gi e kT … (4.68)
i

Here, gi is the degeneracy of the ith energy level. Until now we have discussed
the total energy of a molecule for a given energy state. You know that the total
energy of the molecule has contribution from translational energy, rotational
112 energy, vibrational energy, and electronic energy of the molecule i.e.,
Unit 4 Introduction to Statistical Thermodynamics-II

i ,total = i ,trans + i ,rot + i ,vib + i ,el … (4.69)

Further, you would recall that according to Born-Oppenheimer approximation

these energies can be taken as independent of each other. Thus, the
molecular partition function without considering degeneracy can be written as
( i ,trans + i , rot + i ,vib + i ,el )
−
q= e kT
… (4.70)
i
( i ,trans ) ( i , rot ) ( i ,vib ) ( i , el )
− − − −
q= e kT
e kT
e kT
e kT
… (4.71)
i

Separating each term and introducing degeneracy for each type of energy
level we get
( i ,trans ) ( i , rot ) ( i ,vib ) ( i ,el )
− − − −
q= g i ,trans e kT
g i ,rot e kT
g i ,vib e kT
g i ,el e kT
… (4.72)
i i i i

q = qtrans qrot qvib qel … (4.73)

Thus, the total partition function is the product of partition functions of each
type of the energies of the molecule. This is known as factorization of
partition function. We can calculate the contribution of each type of partition
function to the total partition function separately and put them together to get
the total partition function. Let us learn how do we calculate the partition
function of each type of energy. For this, we need to know the respective
expressions for quantised energy. Here, we will be using the expression
without deriving them. You must have learnt about their derivation during your
earlier studies on quantum mechanics. We will take up the following four types
of partition functions.

I. Translational partition functions

II. Rotational partition functions
III. Vibrational partition function
IV. Electronic partition function
Let us derive expressions for these partition functions, one by one.
I. Translational partition functions
You know that translational energy has three components, one each in x, y,
and z directions. Each of these contribute to the translational partition function.
To work out the expression, we will consider the motion of particle in one-
dimension and use the energy expression from particle in a one-dimensional
box system which you would have studied in your earlier classes. You would
recall that for a particle of mass m confined into a one-dimensional box of
length l , the energy of the particle is given as

n2 h2
n ,trans = n = 1, 2,3... … (4.74)
8 ml 2

Here, n is the quantum number. The energy of the lowest energy state n =1, is
equal to
h2
n,trans = … (4.75)
8ml 2 113
Block 1 Thermodynamics and Statistical Thermodynamic

In case of Boltzmann distribution, the energy of the lowest level is taken as

zero. Therefore, the energy of the nth energy level must be scaled relative to
the lowest level. This implies that the energy of nth energy level relative to the
lowest energy state is equal to

n ,trans = ( n 2 − 1) 1,trans
… (4.76)

The translational partition function in x-direction can thus be written as

( n2 −1) 1,trans
−
qx ,trans = e kT … (4.77)
n

The summation is over all the energy levels. Since the translational energy
levels are very close to each other, they can be taken to vary continuously.
Hence, we can replace summation by integration. Therefore,
( n2 −1) 1,trans
−
qx ,trans = e kT
dn … (4.78)
1

To solve the above integral, we use the approximations that n2 −1≈n2 and the
lower limit of integral is changed to zero. Therefore,
n2 1,trans
−
qx ,trans = e kT
dn … (4.79)
0

n2
Let us substitute, 1,trans
kT
= x2 … (4.80)

Differentiating and simplifying, we get

1/2
kT
dn = dx … (4.81)
x ,trans

Substituting in Eq. (4.79) we get

1/2
kT 2
qx ,trans = e− x dx … (4.82)
x ,trans 0

2
The integral here is a standard form of the integral, viz., e − x dx which is
0
1/ 2
equal to . Substituting in Eq. (4.82) we get.
2
1/2
1/2
kT
qx ,trans = … (4.83)
x ,trans 2

Substituting the expression for εx,trans from Eq. (4.74) in Eq. (4.83) we get
1/2
kT 8ml 2 1/2
qx ,trans = … (4.84)
h2 2

On simplification, we get.
1/2
2 kTm
qx ,trans = l … (4.85)
h2
114
Unit 4 Introduction to Statistical Thermodynamics-II

This is the expression for the translational partition function for x- direction. We
will have similar expressions for the y- and z- direction. The total translational
partition function is equal to the product of the partition functions in three
directions, viz.,
qtrans = qx,[Link],[Link],trans … (4.86)

Let us assume that the translational length in the x, y and z directions be

l , b, and h respectively. Substituting the expression for the translational
partition function for different directions, we get
1/2 1/2 1/2
2 kTm 2 kTm 2 kTm
qtrans = l b h … (4.87)
h2 h2 h2

Simplifying, we get
3/2
2 kTm
qtrans = lbh … (4.88)
h2

Since, lbh is equal to the volume, V, we can write

3/2
2 kTm
qtrans = V (4.89)
h2

Thus, the translational partition function depends on the mass of the particle,
temperature, and the volume of the system. Let us take an example

Example 4.4: 1 mole of Neon gas (molar mass 20.1797 g mol-1) occupies a
volume of 0.0245 m3 at 298 K. Calculate the translational
partition function for the system.

Solution: The translational partition function is given by the following

expression.
3/2
2 kTm
qtrans = V
h2

Most of the parameters are given or known, however we need to calculate the
value of mass of the neon atom to use the expression.

Molar mass 0.0201797kg mol −1

Mass = = = 3.351 10−26kg
Avagadro' s number 23 −1
6.022 10 mol

Substituting the values in the expression, we get

3/2
2(3.1419) (1.381 10−23 JK −1) (298 K ) (3.351 10−26 )
qtrans = 0.0245 m3
-34 2
(6.626 10 J.s )

Solving the expression we get

qtrans = 2.148 1024

115
Block 1 Thermodynamics and Statistical Thermodynamic

II. Rotational partition functions

i ,rot
−
qrot = gi ,rot e kT … (4.90)
i

Recall that the rotational energy for linear molecule is given as

J (J + 1) 2
Erot = … (4.91)
2I

Here J = 0, 1, 2, …are the rotational quantum numbers, cross is action

constant ( h / 2 ) and I is the moment of inertia. Since the lowest rotational
energy is zero, there is no need to scale i , rot with respect to lowest energy
level. Further, you know that the degeneracy of rotational level J is given as

gJ = 2 J + 1 … (4.92)

Substituting the values of J , and g J ,the expression for qrot becomes

J ( J + 1) 2
/ 2I
qrot = (2 J + 1) exp − … (4.93)
J =0 kT

Simplifying, we get.

J ( J + 1) 2
qrot = (2 J + 1) exp − … (4.94)
J =0 2 IkT
You may note that all the constants except T and J in the exponent can be
combined to give a constant having units of temperature. This constant is
called rotational temperature r and is mathematically defined as
2
r … (4.95)
2Ik
Substituting the new constant, the equation becomes

qrot = (2 J + 1) e − J ( J +1) r /T
… (4.96)
J =0

At high temperatures, the fraction r/T is small. Also, the difference between
the consecutive energy levels is very small at room temperature. For most of
the cases, kT is much larger than the energy difference between the rotational
levels, thereby many rotational levels are occupied. Therefore, we can replace
the summation in Eq. (4.96) by integration i.e.,

qrot = (2J + 1) e −J (J +1) r /T

dJ … (4.97)
J =0

Solving this integral is not straight forward as the quantum number J appears
in body as well as the exponent part in the integral term. To solve it we make a
small substitution as
x = J2 + J … (4.98)

Taking derivative, we get

dx = (2J + 1) dJ … (4.99)

116 Rearranging Eq. (4.97) we get

Unit 4 Introduction to Statistical Thermodynamics-II

qrot = e− J ( J +1) r /T
[(2 J + 1) ] dJ … (4.100)
J =0

Making the substitution we get

qrot = e − x r / T dx … (4.101)
J =0

This is a standard integral of the type e −ax dx having a known solution

0
(1a ). Using this solution, we get
T
qrot = … (4.102)
r

Substituting the value of r we get

2
8 IkT
qrot = … (4.103)
h2

III. Vibrational partition function

We can write the general expression for vibrational partition function as

i ,vib
−
qvib = gi ,vib e kT … (4.104)
i

Let us take only the case of harmonic vibrations. The energy of a given
vibrational level is.
1
v = v+ h … (4.105)
2

Here, v = 0, 1, 2, 3…. and the energy levels are singly degenerate. The energy
1
of the lowest level is, 0 = h . and the difference between two consecutive
2
energy levels is h .
1 1 1
0+ h 1+ h 2+ h
2 2 2
− − −
qvib = e kT
+e kT
+e kT
+ ... … (4.106)

Simplifying, we get
h 3h 5h
− − −
qvib = e 2 kT
+e 2 kT
+e 2 kT
+ ... … (4.107)

This can be written as

h h h h
− − −2 −3
qvib = e 2 kT
1+ e kT
+e kT
+e kT
+ ... … (4.108)

h
Let = x , we can write.
kT 117
Block 1 Thermodynamics and Statistical Thermodynamic
h

(1 + e + e− 2 x + e−3 x + ...)
−
−x
qvib = e 2 kT
… (4.109)

1
The term in the bracket is a series expansion for
1 − e− x
Substituting back and replacing x we get
h
−
1 e 2 kT
qvib = h
= h … (4.110)
− −
1− e kT
1− e kT

In this expression we have not scaled the energy of the upper level to ε0. Thus,
Eq. (4.94) includes zero-point energy. When the zero-point energy is scaled to
zero as we had done for translational energy, vibrational partition function
becomes
1
qvib = … (4.111)
−(h / kT )
1− e

IV. Electronic partition function

The electronic partition function is given by the following expression.

ei ,el
−
qel = g i ,el e kT … (4.112)
i

Let us learn about electronic partition function with the help of an example.

Example 4.5: The energies and degeneracies of first three electronic states of
an atom from the first period are given below:

State Energy (J) Degeneracy

1 0 1
2 3.26 10 −22 3
3 8.64 10 −22 5

Calculate the electronic partition function at 300 K.

Solution: As per Eq. (4.112) the electronic partition function is given as

ei ,el
−
qel = g i ,el e kT

Expanding the expression for the first three energy levels we can write
1,el 2,el 3,el
− − −
qel = g1,el e kT
+ g 2,el e kT
+ g3,el e kT

Let us first calculate the value of kT that appears in all the terms

kT = 1.381 10−23 J K −1 300K = 4.143 10−21 J

Substituting the values of degeneracies, energies and the kT term, we get

0 3.26 10−22 J 8.64 10−22 J
− − −
qel = 1e 4.143 10−21 J
+ 3e 4.143 10−21 J
+ 5e 4.143 10−21 J

118
Unit 4 Introduction to Statistical Thermodynamics-II

Simplifying, we get

qel = 1e0 + 3e −0.079 + 5e −0.209

Solving we get,

qel = 1 + (3 0.924) + (5 0.811) = 1 + 2.772 + 4.055 = 7.827

Thus, the electronic partition function would be =7.827.

Generally, the gap between the electronic ground state and first excited state
is large and most of the molecules exist in ground electronic energy state.
That is, in the dominant configuration, only one electronic state is present.
Therefore, we can write the partition function as

qel = gel … (4.113)

Here gel is the degeneracy of the ground electronic energy state. If the
degeneracy of the ground state happens to be one, the electronic partition
function simply is:

qel = 1 … (4.114)

Having learnt about partition function, its significance and different

components let us take up the important concept of canonical ensemble.
However, before that answer the following simple questions.

SAQ 5
Calculate the rotational partition function at 500 K for a diatomic molecule
having moment of inertia of 2.69×10⁻⁴³ kg m².

SAQ 6
The volume of 1 mole of a gas (molar mass 39.947 g mol-1) at 105 Pa and
400K is 0.033258m3. Calculate the translational partition function for the
system.

4.4 CANONICAL ENSEMBLE

In the previous sections we have discussed statistical treatment of a single
system which is a collection of many identical particles. A system is also
known as assembly. A group or collection of assemblies or systems is known
as an ensemble. We had considered the total energy of the system to be
fixed. However, in case of ensemble, the energy of the systems in contact with
the surroundings will fluctuate. Since different parts of the ensemble can also
exchange energy with each other, the energy of the ensemble will fluctuate.
So, in such cases we cannot assume that the energy is fixed. However, we
can fix the temperature of the ensemble by keeping it in contact with the large
heat reservoir. The word canonical means according to the rule. The
ensemble in which by rule number of molecules (N), volume (V) and the
temperature (T) are kept constant is known as canonical ensemble. In 119
Block 1 Thermodynamics and Statistical Thermodynamic

canonical ensemble, all the systems are at the same temperature, pressure,
and volume. The total energy of the canonical ensemble is constant. However,
the energies of the individual systems can be different. The energy of
individual systems will be close to the average energy of the ensemble. In this
section we will extend the statistical treatment of a system to that of an
ensemble.

The total number of particles in an ensemble = N

The number of particles in an ensemble with energy Ei = Ni

The weight of a configuration of ensemble with population N 0 , N1 , N 2 , N 3 ,... is

given as
~
~ N!
W = ~ ~ ~ ~ … (4.115)
N0 ! N1! N2! N3 !...

Like Boltzmann distribution, canonical distribution for the dominant

configuration of the ensemble with the constraints that the total energy of the
~
ensemble, E is constant and N is also fixed can be written as
~
Ni − Ei e − Ei
~ = e .e = … (4.116)
N Q

Q= e − Ei … (4.117)
i

Here, Q is the canonical partition function,

Relation between molecular partition function and canonical partition

function
Let us consider a collection of N molecules that are not interacting with each
other, that is, they are independent of each other. The total energy of the
collection of molecules in a state is the sum of the energies of individual
molecules in that state. The total energy of the collection in ith state is written
as
E i = i (1) + i (2) + i (3) + ... + i (N ) … (4.118)

The partition function of the collection is written as

Q= e − { i (1)+ i (2)+... + i (N )} … (4.119)

i
This expression can be expanded for each energy state as

Q = e − { 1(1)+ 1( 2)+...+ 1(N )} + e − { 2 (1)+ 2 ( 2)+...+ 2 (N )} + ... … (4.120)

Expanding the above equation further and taking all the terms of one molecule
together we get

Q = {e − { 1(1) + e − { 2 (1) + ...} . {e − { 1( 2) + e − { 2 ( 2) + ...}...{e − 1( N ) + e− 2 (N ) + ...

Q= e − { 1(1) . e − { i ( 2) . e − { i (3) ... e − { i (N ) … (4.121)

120 i i i i
Unit 4 Introduction to Statistical Thermodynamics-II

The terms inside the curly brackets are the partition functions of the individual
molecules. Therefore,
Q = q(1). q(2)...q(N ) … (4.122)

For distinguishable molecules of the same type, all the molecular partition
functions will be same. Therefore,
Q = q N for distinguishable particles … (4.123)

When the particles are indistinguishable, the above expression is modified as

qN
Q= for indistinguishable particles … (4.124)
N!

4.5 THERMODYNAMIC PROPERTIES AND

PARTITION FUNCTION
You may recall that we stated that statistical thermodynamics acts as a bridge
between the microscopic properties of atoms and molecules and the
thermodynamic properties of bulk matter. Also, the major aim of statistical
thermodynamics is to be able to calculate the thermodynamic properties of
systems using the mathematics of statistics. In the previous sections, you
have studied about the molecular partition function, and canonical partition
function. We are now equipped to take up the calculation of thermodynamic
properties of a system from partition functions. However, we will not be taking
these in detail here. We will just demonstrate the approach by taking the case
of calculation of average energy of a molecule in terms of molecular partition
function.
4.5.1 Average Energy of the Molecule
In this section, we will write the energy of the molecule in terms of partition
function. For a collection of N identical molecules, if the energy of the system
is E, then the average energy of one molecule is obtained by dividing the
energy of the system by the total number of molecules in the system i.e.,
E
= … (4.125)
N
You have learnt above that at equilibrium the energy distribution is given by
Boltzmann distribution. The total energy is written as,
Ni i = E … (4.126)
i

− i
Ne kT
And Ni = … (4.127)
q
Therefore,
Ni 1 − i
= i i
= i e kT
… (4.128)
N q i

Using some mathematical rearrangement, we can show that

−
de i
−
= ie i
… (4.129)
d 121
Block 1 Thermodynamics and Statistical Thermodynamic

Therefore,
−
1 de i 1 d −
=− =− e i … (4.130)
q i
d qd i

1 dq d ln q
=− = … (4.131)
qd d

Recall that β = 1/kT. So, the differential in Eq. (4.131) is showing the variation
of partition function with temperature. During the differentiation, other
parameters which may affect the partition function, for example, volume are
kept constant. Also, in partition function expression we take the lowest energy,
ε0 (which is equal to the zero-point energy) as zero. Therefore, the zero-point
energy has to be included in the calculation of average energy of the
molecule. After applying these modifications, Eq. (4.131) is written as

d ln q
= 0− … (4.132)
d V

Let us sum up what all we have discussed in this unit.

4.6 SUMMARY
In this unit we continued our discussion on statistical thermodynamics and
began the unit by taking up Boltzmann distribution function. The discussion
was initiated by extending the concept of permutation and configuration to the
chemical systems. In this process we introduced and explained the concepts
of microstates and configurations. The total number of microstates associated
with a given configuration is known as the weight of the configuration and is a
measure of the probability of the configuration to be observed. The
configuration with maximum weight is known as dominant configuration. With
an increase in the size of the system the dominant configuration becomes the
equilibrium configuration.

In the process of deriving expression for the Boltzmann’s distribution function

we introduced the crucial concept of partition function and outlined its
significance. Thereafter we took up the formulation of partition functions for
different types of energies of the system. It was followed by the introduction of
the idea of canonical ensemble. Having learnt Boltzmann’s distribution and
partition functions in details we took up the calculation of average energy as
an example for getting the thermodynamic properties of a system from
partition functions.

4.7 TERMINAL QUESTIONS

1. Define the following:

a) Microstates

b) Weight of a configuration

122 c) Dominant configuration

Unit 4 Introduction to Statistical Thermodynamics-II

2. Consider the distribution of five quanta (E0) of energy amongst three

distinguishable quantum particles having four available energy states of
energy equal to 0, E0, 2E0 and 3E0. Identify different configurations and
calculate their weights.

3. Give the constraints applied for Boltzmann distribution.

4. Which among N2 and H2 will have higher translational partition function?

Give reason.

5. Calculate the electronic partition function at 300 K for a system using its
first three doubly degenerate energy levels with energies as 0, 1x10-21 J
and 2.5x10-21 J respectively.

6. The rotational temperature for HI molecule is 9.43 K. Calculate the

moment of inertia for HI.

4.8 ANSWERS
Self-Assessment Questions
1. The arrangements of particles in different energy states are called
configurations whereas the specific arrangements (permutations) of
particles in each configuration are known as microstates.

2. The possible distributions for 7 distinguishable particles among 4 energy

levels under the constraint of total energy being 3 quanta along with their
weights and ratios W are as under.
Wtot

Dist. N0 N1 N2 N3 W W
Wtot
1. 6 0 0 1 7 0.08
2. 5 1 1 0 42 0.50
3. 4 3 0 0 35 0.42
Total 84 1.00

3. The possible distributions and the corresponding weights and probabilities

for one hundred molecules under given set of conditions are.

Dist. N0 N1 N2 N3 N4 N5 W W
Wtot

1. 99 0 0 0 0 1 100 1.088x10-8
2. 98 1 0 0 1 0 9900 1.077x10-4
3. 98 0 1 1 0 0 9900 1.077x10-4
4. 97 2 0 1 0 0 485100 5.27x10-3
5. 97 1 2 0 0 0 485100 5.27x10-3
6. 96 3 1 0 0 0 15654900 0.1703
7. 95 5 0 0 0 0 75287520 0.8189
Total 91932520 1.00
123
Block 1 Thermodynamics and Statistical Thermodynamic

4. i) We know that according to Boltzmann’s law

− i
Ni e kT
=
N − i
e kT
i

Substituting the values in the expression for denominator gives

1.6 10−20 J
−
− i − 1
(1.381 10−23 JK −1 )(300 K )
e kT
=e +e 0 kT
= 1+ e
i

Solving, we get
1.6 10−20 J
−
− i
(1.381 10−23 JK −1 )(300 K )
e kT
= 1+ e = 1 + e −3.862 = 1.021
i

The fraction of molecules in the excited state can then be calculated as

1.6 10−20 J
−
(1.381 10 −23 JK −1 )(300 K )
Ni e 0.021
= = = 0.0206
N 1.021 1.021

Thus, fraction of molecules in the excited state at 300 K = 0.0206. This

means about 2.06% of the molecules will be in excited state.

ii) For 1000 K

Again, substituting the values in the expression for denominator gives

1.6 10−20 J
−
− i − 1
(1.381 10−23 JK −1 )(1000K )
e kT
= e0 + e kT
= 1+ e
i

Solving, we get
1.6 10−20 J
−
− i
(1.381 10−23 JK −1 )(1000 K )
e kT
= 1+ e = 1 + e −1.1586 = 1.3139
i

The fraction of molecules in the excited state can then be calculated as

8.0 10−21 J
−
(1.381 10−23 JK −1 )(300 K )
Ni e 0.3139
= = = 0.2389
N 1.3139 1.3139

That is, the fraction of molecules in the excited state at 1000 K =

0.2389 implying that 23.89 % of the molecules will be in excited state.

iii) You may note that the given system is like the one in the example but
with the energy of the excited state is double of that in the example.
We find that there is a significant decrease (from 12.7% to 2.06 %) in
the population of the excited state. It is in accordance with the
Boltzmann’s law as the energy of excited state increases its population
decreases.

In the second case the temperature was increased, and this led to an
increase in the population (from 2.06 % at 300K to 23.89% at 1000K)
124 of the excited state.
Unit 4 Introduction to Statistical Thermodynamics-II

5. As per Eq. (4.103) the rotational partition function is given as

2
8 IkT
qrot =
h2

Given: I = 2.69 10−43 kg m2 and T = 500K

Substituting the values, we get

8(3.1419)2 (2.69 10−43 kg m2 )(1.381 10−23 J K −1 )(500K)

qrot =
(6.626 10−34 J.s)2

Solving we get

qrot = 3.34 105

6. The translational partition function is given by the following expression.

3/2
2 kTm
qtrans = V
h2

Most of the parameters are given or known, however we need to calculate

the value of mass of the gas molecule to use the expression.

Molar mass 0.039947 kg mol −1

mass = = = 6.634 10−26 kg
Avagadro's number 6.022 1023 mol −1

Substituting the values in the expression, we get

3/2
2(3.1419)(1.381 10−23 J K −1 )(400K)(6.634 10−26 )
qtrans = 0.03326 m3
(6.626 10−34 J.s)2

Solving the expression we get

qtrans = 2.154 1026

Terminal Questions

1. a. Microstates are the permutations possible for a given configuration

b. Weight of a configuration is the number of microstates possible in each

configuration

N!
W=
N 0 ! N1 ! N 2 ! N 3 !...

c. Dominant configuration is the configuration of a system having maximum

weight.

2. Under given conditions, three different configurations are possible for the
given case. These are {1,0,1,1}, {0,1,2,0}, and {0,2,0,1} with respective
number of microstates as 6,3 and 3. The three configurations and
corresponding microstates are given diagrammatically as 125
Block 1 Thermodynamics and Statistical Thermodynamic

3. Constraints applied for Boltzmann distribution are as under

(i) the total population remains constant, if the population of one state
increases, then the population of another state will decrease.

(ii) if the population of a state changes, the total energy is not affected.

4. Translational partition function is given by the expression.

3/2
2 m
qtrans = V
h

For equal volumes of different gases, qtrans m3 / 2

Since, m(H2) <m(N2), qtrans(H2) <qtrans(N2)

5. As per Eq. (4.112) the electronic partition function is given as

ei ,el
−
qel = g i ,el e kT

Expanding the expression for the first three energy levels we can write
1,el 2,el 3,el
− − −
qel = g1,el e kT
+ g 2,el e kT
+ g3,el e kT

Let us first calculate the value of kT that appears in all the terms

kT = 1.381 10−23 J K −1 300K = 4.143 10−21 J

Substituting the values of degeneracies, energies and the kT term, we get

0 1 10−21 J 2.5 10−21 J
− − −
qel = 2e 4.143 10−21 J
+ 2e 4.143 10−21 J
+ 2e 4.143 10−21 J

Simplifying, we get

126 qel = 2e0 + 2e−0.2414 + 2e−6.034

Unit 4 Introduction to Statistical Thermodynamics-II

Solving we get,

qel = 1 + (2 0.7855) + (2 0.0019) = 2 + 1.571 + 0.0038 = 3.5748

Thus, the electronic partition function would be=3.5748.

2
6. As per Eq. (4.95), the rotational temperature is given as, r
2Ik
2
Rearranging we get, I
2 rk

Substituting the values and solving gives

(6.626 10−34 J.s)2

I 2 −23 −1
= 4.33 10−47 kg m2
(2 3.1419) 2(9.3K )(1.381 10 J K )

127