MME215 Thermodynamics of Materials

Lecture 08
A. K. M. Bazlur Rashid
Professor, Dept. of Materials and Metallurgical Eng.
Bangladesh Univ. of Eng. and Tech., Dhaka-1000
The Laws of Thermodynamics
5 – Statistical Interpretation of Entropy

© Copyright Protected

Lecture Outcome (LO)

At the end of this lecture, students should be able to

• differentiate statistical thermodynamics with classical thermodynamics,
• explain the concept of microstate and microstate, and
• explain the concept of probability and entropy in statistical thermodynamics

2/21
 1. Introduction

❑ Classical thermodynamics is phenomenological ; ❑ In this lecture, we will use the understanding

that is, it deals with matter as it appears to our senses. of matter consisting of atoms and molecules
and introduce the use of statistics to the
❑ In this regard, we use our senses to describe the approach of the topic of thermodynamics.
properties of matter such as pressure, volume, and
temperature. ❑ This starting place moves us from the realm
of classical thermodynamics to statistical
❑ Classical thermodynamics does not delve into the thermodynamics.
more fundamental aspect, of what does matter
consist? ❑ Our excursion will be brief, but it is hoped that
it will give us a better physical understanding
of entropy.

3/21

❑ The introduction of entropy as a ❑ In this respect, entropy differs from internal

thermodynamic state function was facilitated energy in spite of the fact that, within the scope
(1) by the realization that there exist possible of classical thermodynamics, both properties are
and impossible spontaneous processes, and extensive thermodynamic functions of the state
(2) by the examination of the thermal energy of a system.
and work effects occurring during these
processes. ❑ The ready acceptance of the First Law of
Thermodynamics was due to the easily
understood physical significance of internal
❑ From the formal statements of the Second energy, whereas the lack of a corresponding
Law of Thermodynamics, as developed from understanding of entropy caused the
classical thermodynamics arguments, it is acceptance of the Second Law to be slower.
difficult to assign a physical significance or a
physical quality to entropy.
❑ A more physical interpretation of entropy
had to await the development of statistical
thermodynamics and the subsequent
development of quantum mechanics.
4/21
 2. The Concept of Microstate

❑ In classical thermodynamics, the state of a single-component isolated system,

which in this chapter is denoted as its macrostate , is fixed with the knowledge
of two of its intensive thermodynamic variables of pressure and temperature.

❑ In this macroscopic description of the behaviour of matter, we

• consider a thermodynamic system as continuous.
• ignore the fact that the substance is actually composed of atoms or molecules
• ignore the fact that the behaviour of the system is somehow related to the properties
of the particles that compose it.

5/21

❑ The connection between the thermodynamics behaviour of macroscopic system

and that of atoms on their sub-microscopic scale is useful for a variety of reasons :
• a new level of understanding of how matter behaves emerges from this connection.

• in the macroscopic level of description, for example, it is sufficient only to know that
the heat capacity of A is different from that of B.

• but at the atomistic level of description, it is possible not only to predict this difference
in heat capacities of A and B, but also the expected magnitude of those differences.

❑ Atomistic models provide a level of explanation for the behaviour of a substance;

phenomenological information merely provides a consistent description of how it behaves.

6/21
❑ When we consider the atomic and molecular ❑ Statistical thermodynamics considers the various
makeup of the system, there are many more ways that the total number of constituents, N, can
possible configurations of the constituents which be configured in the systems, namely :
give rise to the same macrostate of the system.
1. Isolated systems, where S′ = S′ (U′, V′, N )
or U′ = U′ (S′, V′, N)
(the micro-canonical formalism)
❑ Each constituent has three coordinates for its
position and three coordinates for its momentum. 2. Closed systems in equilibrium with a heat bath,
where S′ = S′ (T, V′, N)
❑ In addition, there is an enormous number of (the canonical formalisms)
constituents in real systems, of the order of 1024
for a mole of particles. 3. Open systems, where S′ = S′ (T, V′, μ)
(the grand-canonical formalism)

❑ In this brief introduction to statistical thermodynamics,

we will consider the microcanonical (isolated) cases.

7/21

❑ An atomistic description of the behaviour of matter begins with the idea that
each atom in the system can be assigned values of properties that describe its condition.
• if it were possible to know the masses, velocities, positions, and all modes of motion of all the
constituent particles of a system, then this mass of knowledge would serve to describe the
microscopic state of the system, which, in turn, would determine all of the properties of the system.

❑ However, such a description of the state of system by specifying the property

(i.e, the energy level) of all the atoms is impossible !!

8/21
❑ To analyse very large collection of numbers, ❑ Description of the behaviour of matter in terms of
the concept of the distribution function of the distribution of particles over their allowable
statistics can be used. states is called statistical thermodynamic.
• atoms that have similar values of properties • specification of the thermodynamic state of a system
are lumped together into classes in terms of distribution function is called the
(in this case, energy levels) macrostate for the system.

• the distribution function simply reports • so, statistical thermodynamics is not concerned with
the number of particles in each class. the behaviour of individual particles of a system –
only the statistically average behaviour of the system.
• in this way the quantity of information required to
specify the condition of the system is greatly reduced.

❑ Statistical thermodynamics asks the following questions:

• In how many ways can the N particles be distributed over the available energy levels
such that the total energy of the crystal (i.e., Uʹ ) remains the same?
• Of the possible distributions, which is the most probable?
9/21

❑ The characteristics of the particles and the crystal structure determine the quantization
of the allowed energy levels, in which the lowest energy level, or the ground state, is
designated ε0 , and the succeeding levels of increasing energy are designated ε1 , ε2 ,
ε3 , and so on.

❑ The effect of the quantization of energy and its resulting distribution can be illustrated
by considering a hypothetical system comprising a perfect crystal in which all of the
distinguishable sites are occupied by identical particles.

❑ The crystal contains N particles and has the fixed energy Uʹ and fixed volume Vʹ . The
system is therefore considered to be an isolated one: neither energy nor particles can
enter or leave the system.

10/21
❑ Consider that a crystal contains four List of Microstates :
identical particles which are located
on two distinguishable energy levels,
e1 and e2. Microstate e1 e2 Microstate e1 e2

A abcd -- I bc ad
4 Particles: a, b, c, d B abc d J bd ac
2 Energy States: e1, e2 C bcd a K cd ab
D cda b L a bcd
•All particles are DIFFERENT. E dab c M b cda
•If any entry in this list is altered, that is, F ab cd N c dab
if any particle changes its condition, the
G ac bd O d abc
system is considered to be in a different
microstate. H ad bc P -- abcd

Total number of microstates = 16

11/21

Particles: a, b, c, d • All particles are physically IDENTICAL.

States: e1, e2 • Macroscopically observable behaviour of the system is
not dependent on which particles exists in a given state,
but merely on how many particles are in that state.
List of Macrostates:
No. of particles Corresponding
Macrostate Number Probability
e1 e2 microstates
I 4 0 A 1 1/16
II 3 1 B, C, D, E 4 4/16
III 2 2 F, G, H, I, J, K 6 6/16
IV 1 3 L, M, N, O 4 4/16
V 0 4 P 1 1/16

Total number of macrostates = 5

• Specification of a macrostate requires two numbers:

the number of particles in state e1 and the number of particles in state e2.
• For example, the specification of the macrostate labelled II is (3, 1), which means that
3 particles in state e1 and 1 in state e2.
12/21
Thus, in general,

❑ If a system composed of N number of particles distributed over r number of available

atomic states, a macrostate is specified by assigning a number of particles n1, n2, ..., nr
to each of the r states (defined by their energy levels) e1, e2, ...., er respectively.

❑ Then set of numbers (n1, n2, ...., nr) is a distribution function

specifying how the atoms are distributed over the energy levels.

❑ This distribution describes the macrostate for the system.

• for example, for a twenty-particle system with seven energy states,
the list of microstates for a particular macrostate might read as (1, 3, 4, 6, 2, 3, 1).

13/21

Number of Microstates in a Macrostate

“How many microstates correspond to the macrostate (n1, n2, .... , nr) ?”
This question is mathematically equivalent to
“How many different ways can N balls be arranged in r boxes so that there are
n1 balls in the first box, n2 in the second box and so on to nr balls in the rth box?”
𝑁!  = number of microstates
Ω= in a given macrostate
𝑛1 ! 𝑛2 ! … 𝑛𝑟 !

In the previous example,

Macrostate I: I = 4! / (4! . 0!) = 1
N = 4, r = 2
Macrostate II: II = 4! / (3! . 1!) = 4
Macrostate III: III = 4! / (2! . 2!) = 6
Macrostate IV: IV = 4! / (1! . 3!) = 4
Macrostate V: V = 4! / (0! . 4!) = 1
14/21
 3. The Concept of Probability

❑ The concept of macrostate lies within the domain of classical thermodynamics.

❑ The macrostate is fixed when the values of the independent variables (T , P ) are fixed.

❑ In the preceding example, the system was considered to be isolated; that is, the values
of U , V , and N are fixed, and hence, the macrostate of the system is fixed.

15/21

❑ During a process, the atoms redistribute themselves over their allowable microstates
and experience a change in their condition.
• this results an overall change in the macrostate of the system.

❑ Now, since the particles are identical, the average time a particle spends
in a given energy level or a microstate is expected to be the same for all particles.
• thus, each of the microstates is equally probable

• then, the time spends by a system in a macrostate is the sum of the time spends
in all the microstates correspond to the given macrostate.

16/21
 ❑ The fraction of time the system spends in any given macrostate
is thus the ratio of the number of microstates that corresponding to that macrostate
to the total number of microstates that the system is capable of exhibiting.
• this fraction of time is considered to be the probability of the system
of existing in the given macrostate at any given period of time.

❑ Thus, the probability that the system exists in the Jth macrostate, will be the ratio
of the number of microstates J corresponds to the Jth macrostate to the total
number of microstates the system may exhibit.

Ω𝐽 PJ = probability of Jth macrostate

𝑁! 1
𝑃𝐽 = = J = number of microstate in Jth macrostate
𝑟𝑁 𝑛1 ! 𝑛2 ! … 𝑛𝑟 ! 𝑟𝑁 rN = total number of microstate

17/21

N0 = 100; r = 2
Largest macrostate = 1.27x1030 microstates


(log scale)
III The probability distribution for macrostates
Macrostate I: I = 1 Most probable
macrostate has an extremely sharp peak for systems
Macrostate II: II = 4
max II IV with large numbers of particles and
Macrostate III: III = 6
macrostates. Here the most probable
Macrostate IV: IV = 4
macrostate contains the most number of
Macrostate V: V = 1 I V
microstates, max.
Macrostate

❑ For systems with large number of particles, this function has an extremely sharp peak
at the macrostate containing the maximum number of microstates, Fig.10.1.
• macrostates that are only slightly different from the maximum probability state
have a much smaller probability.

❑ The maximum probability state will be observed almost all of the time, and can be
interpreted as the macrostate that corresponds to the equilibrium state for the system.
18/21
❑ In phenomenological thermodynamics,
the equilibrium state is characterised by an extremum:
“entropy of an isolated system is the maximum
when the system is at equilibrium”

❑ So it appears that there might be a connection between the entropy and ,

the number of microstates corresponding to a given macrostate.

❑ Thus, the most probable microstate is considered to be the equilibrium

state in the macrosystem, and the total number of possible arrangements,  ,
is set equal to the number of microstates with maximum probability, max .

19/21

❑ Appropriately so, Boltzmann suggests in his hypothesis that:

S = entropy of the system
𝑆 = 𝑘 ln Ω k = Boltzmann’s constant = R/N0
N0 = Avogrado’s number.

• entropy can be regarded as the probability of attaining the state of the system

❑ A system with the highest degree of order

is an improbable state with a low value of  and, hence, of low entropy.

❑ Conversely, a state of high entropy is highly probable and will appear to be

disorganised and random.

❑ Hence, an increase in entropy attending a spontaneous process

will cause more disorder in the system.

20/21
Next Class

Lecture 09
The Laws of Thermodynamics
6 – The third law of thermodynamics; Problem solving