INDEX

S. No Topic Page No.

Week 1
1 Introduction 1
2 Introduction (Contd.) 17
3 First Law of Thermodynamics 36
Second Law of Thermodynamics
4 51
5 Second Law of Thermodynamics (Contd.) 68
Week 2
6 Entropy Change during Spontaneous Processes 82
7 Criteria of Spontaneity 97
8 Criteria of Spontaneity (Contd.) 113
9 Thermodynamic Network 130
10 Thermodynamic Network (Contd.) 143
Week 3
11 Tutorial 1 162
12 Gibbs free energy as a function of temperature and pressure 192
13 P-v-T behaviour of gases 221
14 P-v-T behaviour (contd.) 246
15 P-v-T behaviour (contd.) 270
16 P-v-T behaviour (contd.) 293
17 Tutorial 2 309
18 Property estimation from P-v-T behaviour 323
Week 4
19 Property estimation (contd.) 344
20 Concept of chemical potential 367
21 Chemical potential (contd.) 385
22 Homogeneous open systems 403
23 Homogeneous open systems (contd.) 424
24 Heterogeneous Closed Systems 435
Week 5
25 Tutorial 3 454
26 Concept of fugacity 486
27 Fugacity (contd.) 501
28 Estimation of fugacity coefficients 517
29 Fugacity of condensed phase 533
30 Mixtures 548
31 Mixtures (contd.) 566
32 Tutorial 4 581
Week 6
33 Partial molar properties 610
34 Partial molar properties (contd.) 627
35 Partial molar fugacity 642
36 Ideal solutions 657
37 Ideal solutions (contd.) 676
38 Ideal solutions (contd.) 691
39 Ideal solutions (contd.) 708
Week 7
40 Non-ideal solutions 726
41 Non-ideal solutions (contd.) 741
42 Non-ideal solutions (contd.) 761
43 Non-ideal solutions (contd.) 784
44 Non-ideal solutions (contd.) 799
45 Deviations from ideal dilute solutions 818
46 Tutorial 824
47 Tutorial 848
Week 8
48 Lecture 48 : 878
49 Lecture 49 : 899
50 Lecture 50 : 912
51 Lecture 51 : 924
52 Lecture 52 : 937
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 01
Introduction

Hello everybody welcome to the first-class on phase equilibrium thermodynamics, now you must
have all gone through the introductory video and you more or less know what you should be
expecting from this class. Now before I go into the details concerning phase equilibrium
thermodynamics it is very important to understand exactly the individual words comprising the
title of this particular subject. What is phase? What is equilibrium? And exactly what does
thermodynamics in general deal with? And how do we use the laws of thermodynamics in order
to predict phase equilibrium or rather in order to predict the conditions of phase equilibrium and
to predict the properties of different systems when they are undergoing some change from one
equilibrium state to another.

(Refer Slide Time: 1:24)

Now let us come to the first thing regarding on the most generalized the topic what is
thermodynamics? Now if you look at the word thermodynamics it basically means the patterns of
change of the different energies, so therefore it primarily deals with energy and its
transformation. It is about understanding the patterns of energy change how such changes are

1
related to the states of matter. Now historically this particular subject developed from man’s
endeavor to convert heat into work.

Because man found that there is a lot of thermal energy all around, so he was trying to extract
useful work from the enormous amount of thermal energy around and therefore in this particular
endeavor he came across some very unique properties of heat energy and also he tried to, also he
found out how much amount of heat can be converted into useful work and the interrelations
between heat and work etc.

And since it was primarily rather it primary developed due to man's endeavor to convert heat into
work, so therefore we find that most of the developments at least for the first and second law
they are associated with the development of heat engines and heat pumps but let me tell you that
these particular whatever relations etc has been developed they are applicable for any particular
kind of work it will and for any particular kind of system which may not be only applicable for
only heat engines and heat pumps.

(Refer Slide Time: 3:01)

Now as far as the engineering aspects are concerned the primary objectives of thermodynamics
can broadly be generalized into 2 categories the first is to describe properties of matter when it
exists in an equilibrium state and to describe processes in which the properties of matter undergo
changes and to relate these changes to the energy transfer which have brought about this
particular changes. Now therefore these are the 2 forms.

2
First we have to understand rather we have to in order to find out the properties we have to first
identify one particular content of matter where they are going to concentrate and whose
properties we are going to find out and then from that particular portion of mass on which we are
concentrating when that undergoes some particular changes why will it change? Only when it
interacts with the surroundings due to some amount of energy interaction between that particular
portion of matter and the remaining portion.

(Refer Slide Time: 4:53)

Well, that portion of matter undergoes some change in its state which will be manifested by a
change of some of its characteristics which we are going to estimate experimentally or
theoretically and relate these changes of the properties or characteristics of that particular content
of matter with the energy interaction it had with its environment or with its surroundings. So
therefore the first thing in thermodynamics is we need to identify some particular quantity of
matter which is enclosed by a well-defined boundary on which we are required to focus our
attention for thermodynamics analysis.

3
(Refer Slide Time: 5:10)

Just like in physics if you want to perform any particular analysis say for example you we can
find out that maybe there is an inclined plane and the body is trying to fall across the inclined
plane. We would like to find out the acceleration of the body and the displacement of the body
may be fent it starts from the rest or fent is given some particular amount of force in the
beginning. So in order to perform any sort of analysis on this particular or in the order to perform
the physical analysis what we first need to do is we need to perform or rather we need to make a
free body diagram of the body and or in other words we need to separate the body from its
surroundings make the free body diagram and then start the physical analysis.

In the same way when we are going to define our, or rather when you are going to perform any
thermodynamic analysis we have to first identify that portion of matter which is affected by due
to some particular changes and then after we identify that particular portion of matter we perform
the or rather we try to find out how it is interacting with its environment? Anything outside what
we have considered as the system is its surroundings and the first thing which you should be
remembering is that defining a system can be completely arbitrary but its choice as we will be
seeing shortly its choice is largely dictated by convenience there is no particular principle.

4
(Refer Slide Time: 6:36)

The only requirements of selecting a system is firstly it has to be macro size it cannot be
something of a molecular level because we are dealing with classical thermodynamics and it has
to be enclosed by a well-defined boundary, the boundary can be real, the boundary can be
imaginary, the boundary can coincide with the outside boundary of the system that we have
considered, the boundary can be adiabatic, the boundary can be diathermal, it can allow the flow
of mass intake, it might not allow the flow of mass into it.

It can have all, it can be flexible, it can be rigid, it can assume a wide variety of sizes, shapes,
configurations etc but it has to be a well-defined boundary which separates the system from the
surroundings and depending upon the nature of the boundary as you all know because it assume
that you have a preliminary idea of thermodynamics, so therefore depending upon the nature of
the boundary you know a system can be isolated it can be closed it can be open.

In isolated system to be very honest is actually it is ns idealization or it is an approximation in

Cruces we hardly get isolated systems in our real life or in practical applications a Thermos flux
can be considered a close approximation of an isolated system. Mostly we deal with close
systems which interfere, the system interacts with the surroundings by flow of energy into and
out of it or open systems which fed the boundary allows flow of both mass as well as energy
across its boundary.

5
(Refer Slide Time: 8:33)

Now once you have defined system you know that you are supposed to perform your analysis on
this particular system, now how to identify this system? We have to identify it by means of a
number of properties just like if you want to describe your friend to someone you will be
describing your friend by means of some of his or her characteristics how tall, whether that
person wears a specks or not, whether he is short or tall or whether he is fair, whether he has got
any identification marks etc. In the same way when we define a particular thermodynamic
system we are going to characterize that particular system by means of some particular
properties. Now there can be a huge amount rather there can be a large extent of properties by
which we can characterize a particular system.

Now as far as thermodynamics properties are concerned we are we usually refer to those
particular characteristics which are associated with energy and its transformation firstly they can
be quantitatively evaluated and mostly we find that since we are dealing with macro systems
which are made up of a large number of ultimate particles I am not going to refer to them as
atoms, molecules, ions nothing because they can be anything.

I would be referring to the smallest portion of my system as ultimate particles, which is the
smallest subdivision of my system which retains the characteristics of the system. So therefore
the properties which are manifested of the system they are actually the result of a collective
behavior of a very large number of ultimate particles you can very well understand that when

6
since we are dealing with phase equilibrium thermodynamics which is the part of classical
thermodynamics which deals with macro sized systems.

We are not going to relate or rather we are not going to deal with thermodynamics at its
molecular level or the statistical thermodynamics as it is called, since we are dealing with
classical thermodynamics, so usually we are going to deal with those particular properties which
are associated with energy and its transformation. We are comparatively fewer in number and
there are some characteristics which a thermodynamic property possesses.

What are these? The first property is it should have a definite value when the system is in a
particular state. Quite naturally when I see that the system existing in a particular state it means
that it has a definite set of its different properties. It has a definite temperature, it has a definite
pressure, it has a fixed volume and so on and so forth we will be coming across other properties
in our due course. And remember one thing when we define the properties of a system in a
particular state it is irrespective of how that system has come to that particular state?

(Refer Slide Time: 12:24)

It can take up any path it can start from any other initial state but when it has come to some
particular state say I or some particular state F, no matter what was the initial state? No matter
how it has come from the initial state it is going to have the same properties. So therefore what I
mean to say is the properties of a system they are state functions they define the state of a system
they are determinable irrespective of how the system was brought to that particular state and

7
when any particular system is undergoing a change of state from say one particular state i to
another state f then the change in the property dP is equal to Pf minus Pi. It does not matter in
what path the system had used in traversing from state i to state f as a result we say that the
properties are usually exact differentials.

(Refer Slide Time: 12:47)

Well, it is very good that you have identified it. Let me tell you some other things that although I
have told you that the thermodynamic properties they are the collective behavior of a large
number of ultimate particles which are characterizing that or rather which are making up that
particular system we need to remember that while thermofile for determining thermodynamic
properties it is not important to understand or consider the structure of matter and usually there
are very few variables which can be used to describe the state of matter and while on the other
hand the ultimate particles in order to characterize them we need to know the structure of matter
and therefore for each and every particle we need to know its position its momentum its velocity,
so a large number of variables are required for complete specification of the state of the ultimate
particles but when the ultimate particles are completely random and they are manifesting their
average characteristics which are manifested as thermodynamic properties, so for finding out this
thermodynamic properties it is not important to understand this structure of matter or we need
not go to the molecular level.

8
Now when I am mentioning that well, we have a particular system it should be characterized by
some particular properties when it is in a definite state it automatically implies that the whole
system should possess one set of properties or in other words, when a system is in some
particular state and I want to measure the temperature by inserting a thermometer I should be
getting the same temperature no matter where I insert the thermometer in the or if I insert the
thermometer in different parts of the system.

(Refer Slide Time: 15:01)

So therefore what I mean to say is that a particular state of a system is meaningful only when it is
an equilibrium state and then the next question comes is what is an equilibrium state? In other
words how do I define an equilibrium state? We should remember one thing, if we leave a
system to itself if we do not disturb a system it will finally attain a some particular state where it
does not change on itself, more importantly it does not have any tendency to change by itself.

This tendency to change or not having the tendency to change is much more important than the
fact that it does not change, what do I mean by this? I mean that suppose you have a wall, a wall
does not move you are pushing it very hard, right? It resists the movement but there is some
particular unbalanced force which is acting on the wall, fine. So therefore there are 2 things
which bring about the change in property of this system.

First, is there has to be some particular unbalanced force which will bring about this change and
the system it resists that particular unbalanced forces when it cannot resist then it undergoes
some
9
particular change. The important thing about equilibrium is that the system or the body does not
have any unbalanced force acting on it as a result of which there is not only that there is absence
of any change there is absence of any tendency of change at the macroscopic level.

So therefore under such a condition there is no exchange of energy and there is no chemical
change between the that different bodies which are interacting with one another and as a result of
which we find that the whole body, it is in some particular state where it exhibits the same
properties throughout thus the properties neither change with position nor they change with time.
Now you have to remember, as I was telling you when left to itself all systems they come to this
state of equilibrium. Suppose you throw a stone, as small children you must have thrown stones
in the river and you must have found ripples.

The ripples continue for some time and finally they die out and the pond comes or returns to its
quiescent state. In the same way whenever you are performing any particular change you find
that there is some disturbance in the system the disturbance continues and then after a time the
system returns to its quiescent state under which there is no unbalanced forces neither there are
any unbalanced mechanical forces nor there are any unbalanced chemical forces nor there are
any unbalanced thermal forces. So therefore the body is completely at peace with itself and with
the surroundings, this particular state is known as an equilibrium state.

Now when we talk of thermodynamic equilibrium it’s very important for you to remember that
when a body simultaneously satisfies thermal, mechanical and chemical equilibrium only under
that condition a body is said to exist under thermodynamic equilibrium. Mechanical equilibrium
means equality of pressure at all positions or points of the body as well as with the surroundings.
Thermal equilibrium means equality of temperature at all positions of the body and with the
surroundings.

Chemical equilibrium means equality of chemical potential, possibly you are not aware of this
word, we are going to deal in greater details with this particular word and you will find that after
a few lectures chemical potential is going to be our companion throughout the entire course on
this particular subject of phase equilibrium thermodynamics. Any how it is sufficient for you to
remember at this moment that equality of chemical equilibrium between the system and the
surrounding it guarantees chemical equilibrium.

10
After a few lectures I will be taking up one particular class where I am going to prove that
equality of temperature, equality of pressure and equality of chemical potential they are the
primary constants the primary requirements for attainment of thermodynamic equilibrium. Now
before I proceed I have one question for you as I was telling you that the thermodynamic
equilibrium is a state where the system does not change its property with position or time.

(Refer Slide Time: 20:35)

This particular aspect distinguishes an equilibrium state from a steady-state; can you just identify
what is a, what differentiates a steady-state and an equilibrium state? I will just give an example
which will help you to understand the difference between the 2. Say we have got a copper rod,
right and that particular copper rod we arrange it in such a way that one part of the copper rod it
is immersed in say steam at 100 degrees centigrade.

And the other part of the copper rod you have immersed in ice at 0 degrees centigrade will keep
the copper rod in this particular manner for quite a sufficient time what happens, we find that the
end of the copper rod which is immersed in steam, that particular end it has a temperature of 100
degrees centigrade. The temperature as you proceed it keeps on decreasing, finally the
temperature at the other end is 0 degrees centigrade.

So therefore you find that after sometime you find that when you put a thermometer at different
locations you record the same temperature but this temperature is not same for different locations
it varies with position. The rod under this condition is said to be at steady-state. Now what you
do?
11
You simply remove the steam jacket as well as the ice jacket and you keep the rod as it is in the
air after sometime you find that the whole rod has assumed some particular uniform temperature
and this particular temperature does not change, it’s the same throughout the rod and it does not
change with time. Under this condition the rod is in an equilibrium state.

Now when some particular or other let it be here, so therefore when we have any particular
system in an equilibrium state it has a fixed set of properties when it goes to another equilibrium
state it is again characterized by another set of properties. Now there are different equations of
state which relate the different properties of the system and from these equations of state we can
predict some properties in terms of the other properties. For example the most common equation
of state which you know by this time is the ideal gas equation. So from that equation you know
that if suppose you know the pressure and temperature of the gas in a container and if you know
that the gas is in equilibrium state then without measuring the volume just by using ideal gas
equation you can find out its volume you can find out its mass and so on and so forth. You can
find out the number of moles of the gas etc.

Now it is important to remember that some system is in equilibrium state there are a minimum
number of properties which can be varied without changing the state of the system or in other
words, there are a minimum number of properties which must be specified in order to completely
define the state of the system. And these numbers of properties are related to the number of
phases present in then system and the number of components present in that system by a very
famous rule which is known as the Gibbs or rather Gibbs phase rule.

12
(Refer Slide Time: 24:21)

The rule as it is as stated here the rule states that the number of intensive properties which can be
specified that is equal to the number of components which are present minus the number of
phases plus 2 this 2 usually refers to the temperature and pressure conditions. So therefore from
here what do we know? Suppose we have water in a container it is a single phase system, single
component system so therefore for this water there are 2 properties which can be independently
varied other 2 properties which if you mention the state of the water is completely specified they
are its temperature and its pressure. This same water if it is kept in equilibrium with its vapor.

Suppose there is a container we have water and water vapor above it and the 2 phases are in
equilibrium with each other, so in this case you tell me what are the minimum number of interns
or rather what are the minimum number of properties which must be specified. C is 1 here P is 2
here, when there was just water in the container we could independently vary its temperature and
pressure. Now if this water is kept in equilibrium with steam then we find that temperature and
pressure they are no longer independent of each other they become interdependent.

13
(Refer Slide Time: 26:12)

So either you specify its temperature or you specify its pressure the other gets fixed
automatically, with this water you mix up alcohol ethanol say, okay. This ethanol water mixture
now it has 2 components it has one phase so therefore in this particular case we find that what is
going to be F? For this particular case you tell me when we have an ethanol water mixture we
have 2 components, so therefore in this particular case what do we have we know that we can
vary 3 properties independently.

What are the 3 properties? The temperature of the mixture, pressure of the mixture, composition
of the mixture or in other words the mole fraction of at least one component of the mixture.
Same ethanol water mixture when it is kept in equilibrium with its vapor, right? So therefore in a
container we have an ethanol water mixture in equilibrium with its vapor, now under this
condition what do we have? Can we vary 3 properties at will or in this case we find for the 2
phase 2 components systems we can just vary 2 properties independently.

14
(Refer Slide Time: 27:28)

So we can either vary the temperature and composition of one of the phases automatically the
pressure and the composition in the other phase gets fixed or in other words we can vary
temperature and pressure the compositions gets fixed etc. So therefore the phase rule is
something very important it is a very powerful tool in order to deal with phase equilibrium
thermodynamics. There are just 2 things which you need to remember when we are dealing with
this phase rule the first thing is F refers to the number of intensive parameters as is mentioned in
this particular slide, so therefore it has nothing to do with the total amount of mass or the total
amount of volume it just deals with the intensive properties. Well, I forgot to mention when I was
mentioning properties that properties they can be intensive they can be extensive all extensive
properties when divided by the total number of mass or the moles they result in intensive
properties.

So therefore the phase rule it states that if we have a ethanol water mixture then we can vary 3
properties and that is true if we have 99 percent ethanol 1 percent water or 1 percent ethanol 99
percent water, so therefore the important part to remember about phase rule is that it is definitely
a very powerful tool when we are dealing with phase equilibrium thermodynamics but it has it
deals with intensive variables number 1 and it has nothing to do regarding the proportion of the 2
phases.

15
Well, so we stop here at the moment until the next class we continue our discussions on the
introductory aspects of phase equilibrium thermodynamics.

16
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 02
Introduction (contd.)

A very good day to all of you, so we have come to the second class of phase equilibrium
thermodynamics, in the last class I had given you a very brief introduction regarding systems,
surroundings, boundaries and equilibrium state of a system etc. Now when does a system deviate
from equilibrium? As I told you in the last class under normal circumstances if any particular
system is left to itself it will not deviate from the equilibrium state because it is most comfortable
when it is in the equilibrium state there are no driving forces and there is no incentive there is no
need for the system to change.

A system changes only when it interacts with the surroundings by virtue of some sort of an
energy interaction or a mass interaction and when it is interacting it undergoes a process by
virtue of this particular process, it shifts from one particular equilibrium state it continues this
particular path till it reaches a second equilibrium state again when there are no unbalanced
forces and it has no tendency or there is no tendency of the system to deviate from the second
equilibrium state unless again it is forced to do so.

17
(Refer Slide Time: 1:37)

So therefore the process which a system undergoes it is defined as a method of operation in

which specific quantities of heat and rather it is better now we do not discuss about heat, work
etc in which a system interacts with the surrounding by virtue of some sort of energy interaction
as well as may be there can be some mass interaction also and during the process the system
moves from one equilibrium stable state to another equilibrium state and the succession of states
which the system passes during this particular change that is known as the path of the process.

Now let me tell you that usually whenever we undergo any particular process it is quite fast,
right? And it is very difficult for a system to remain in equilibrium during all the states it is
undergoing when it is going from one equilibrium state to another equilibrium state. An idealized
situation is a situation where a system while undergoing a process passes through a succession of
equilibrium states but this is an idealization and often we assume that a system is actually
passing through a succession of equilibrium states during the process and we call such a process
a reversible process.

Now let me remind you again a reversible process is an idealization it is never possible to, we
can try to attain a reversible process but it is never possible to realize a reversible process in
practice, most processes are irreversible. If so why are we so interested about reversible
processes? We always try to rather simulate or model or rather analyze any particular process

18
assuming it to be reversible and then we find that or rather we incorporate an efficiency factor to
come to the actual energy interactions in the irreversible process, why so?

(Refer Slide Time: 4:20)

Why are reversible processes which can never be realized in practice so very important? The
reason is very simple, see suppose we take some particular amount of gas in a pistol cylinder
arrangement a very conventional system which will be dealing now and again in our course of
thermodynamics, again possibly due to the history of thermodynamics originating from heat
engines and heat cycles. So say for this particular system in this case the gas which is contained
in this particular cylinder this is my system, right?

And what I do? We need this system it is in contact with the thermal reservoir at a higher
temperature, fine? So under this condition what happens? Some amount of energy flows from
this particular reservoir into the system, when some amount of energy flows suppose we can
maintain the condition such that, as the energy flows we do not allow temperature rise of the
system on the contrary the system keeps on expanding in volume and the piston keeps on going
up very slowly such that with entry of heat by an infinitesimal amount which occurs when there
is a dT difference of temperature between the thermal reservoir and the system then the heat flow
from the reservoir to the system occurs very slowly.

And under that condition the system, it expands under a very slow rate and we find that with
time the system gradually increases in volume by dV amount and its pressure reduces by minus
dP
19
amount, right? So therefore we can assume that at each and every state the system was in an
equilibrium state. What does it imply? It implies that at each and every instant of time the system
had one particular pressure value, one particular volume value, the volume value went on
increasing very slowly, the pressure value went on decreasing very slowly, it is true but for each
particular state for each particular state the system was undergoing from its initial position of the
piston here to say the final position of the piston here, the each and every state which the system
was undergoing it could be characterized by a unique set of properties or in other words the
system was in an equilibrium state at each and every position.

So therefore if we can plot the state of the system see on a PV plot maybe this is my initial state
and say this is my final state, so I know in this particular case it was I can actually plot the path
of the process. So therefore in this particular case when I can actually trace the path of the
process as we shall see after sometime it is very easy for us to identify the energy interactions
which the system had with the surroundings during its entire path.

It is given by the area under the curve that will be and you also know possibly and I will just
discuss shortly. This area under the curve gives the energy interaction between the system and
the surroundings, so therefore in order to calculate the energy interaction it is very important to
trace the path of the process. The path of the process can be traced only for a reversible process.

20
(Refer Slide Time: 8:15)

So therefore a reversible process although an idealization it is very important we try to simulate

or we try to idealize each particular process as a reversible process where it comprises of a
succession of equilibrium states with driving forces between the system and the surroundings
being differential in magnitude and this occurs for some too idealized situations like motion
without a friction, no pressure difference between system and surrounding and infinitesimal
similar pressure difference between systems and surroundings.

Energy being transferred just like the piston cylinder arrangement which I have shown. Energy is
being transferred with almost no temperature difference between the interacting systems and so
on and so forth. Now well, there is one thing which I would like to mention, right? Now what is
the basic difference between a reversible and an irreversible process? When I ask this question in
the class the answer which I get is that on reversing the process the system comes back to its
original state but this is not the correct answer.

It can always happen even in a cyclic process also that when we reverse the process the system
comes back to its original system the important part is that a difference between reversible and
irreversible processes are with the changes in the state of the surroundings. For an irreversible
process even when we reverse the process the system might come back to its original state but
the surrounding is not going to come back and for a reversible process both the system as well as
the surrounding comes back to its original state.

21
Now well, the most important part of my introductory lecture, why just as I have said a system if
left to itself it remains in an equilibrium state. It changes from 1 equilibrium state to another
through a Series of steps which comprise a processes etc. And the process occurs only when
there is some energy interaction between the system and the surrounding. Now this word energy
you have been hearing about this word since your school days and possibly in your physics book
you have learned whenever I asked in the class what is energy? The answer which I received
almost every year is energy is the capacity to do work then the next question is what is work?

So therefore these terms are very common you have been using them since your childhood
energy, work, process, system all these words are very common in fact the entire concept of
thermodynamics, it is based on common everyday experiences we are dealing with very common
words which you use in your everyday but sadly that does not make the study of
thermodynamics easier, on the other hand it is always almost the contrary thing.

(Refer Slide Time: 11:38)

So therefore in order to understand energy the definitions which I would prefer to give is energy
is the capacity to induce a change in that which inherently resists a change. If you go through this
particular definition carefully, what do you find? You find that the word energy comprises of 2
parts what are the 2 parts? First thing is that there is some particular force which is trying to
induce a change in the system.

22
The system is inherently resisting that particular change and this particular force which is being
applied that particular force it overcomes the resistance to change of the system and then it
performs the change in the direction of the applied force the same thing that you have learned in
physics. In physics also we have learned that energy or the work done that was equal to or say
the work done that is equal to F into dl, it should be dW which is equal to F into dl, right?

It is force into displacement. Just to keep a link or keep a paratue with that particular definition
which you had some across in this case also I define energy in this particular way that it is a
capacity to induce a change in a system which inherently resists the change, so therefore from
this what do we see? We see in that case energy, what does it become? In that case energy
becomes a product of 2 terms.

(Refer Slide Time: 13:48)

The first term is a thermodynamic driving force which inherently overcomes the change which a
system provides to resist a change and once that change is overcome then this particular driving
force it brings about a displacement of the system in the direction of the driving force. So
therefore we find that just like physics we find that we define energy as the product of two
quantities one is the thermodynamic driving force and the other is the thermodynamic
displacement.

The important part here is the thermodynamic driving force and the thermodynamic
displacement they are not defined in the same way as they are defined in physics, for our case

23
defining the thermodynamic both these properties the thermodynamic driving force and the
thermodynamic displacement both of them are thermodynamic properties of the system. Usually
what do we find? A thermodynamic driving force is an intensive property whose difference in
magnitude between any 2 parts brings about a change in some other extensive property of the
system which is known as thermodynamic displacement.

So basically energy in thermodynamics it is a product of one intensive property whose difference

in magnitude brings about a change in an extensive property of the system and the driving force
it controls the extent and the direction of change in the thermodynamic displacement. For
example in the piston cylinder arrangement that we were discussing just now.

24
(Refer Slide Time: 15:49)

In this particular case we found that some amount of energy was flowing and due to this energy
flow it was expanding and why was it expanding? Due to a pressure difference between the
system and the surrounding.

So therefore the pressure difference P that was the thermodynamic driving force which was
bringing about change in the volume, which is the thermodynamic displacement. So therefore for
the pressure volume type of work, we find that work done or the energy transfer in this particular
case it is equal to integral of PdV, right? Now one thing we need to remember in
thermodynamics is that we can define the thermodynamic driving force and the thermodynamic
displacement in any particular way but when multiplied they should give you the unit of energy.

For example we can define the thermodynamic driving force, yeah. So therefore the thing is it is
very important for us to remember that whatever we are doing the thermodynamic driving force
and the thermodynamic displacements are simply properties of that particular system one of
them is an intensive properties which is the thermodynamic driving force.

25
(Refer Slide Time: 18:12)

The other one is an extensive property which changes and that is a thermodynamic displacement
and the choice of these two they are completely arbitrary for example in this piston cylinder
arrangement we could have very well assumed instead of assuming this to be the driving force
we could have assumed the product of the pressure and the piston area which is F as the driving
force and the displacement of the piston as the displacement.

It hardly matters how we are going to define or rather how we are going to divide the energy into
a thermodynamic driving force and the thermodynamic displacement it is important to remember
that no matter how we are dividing it when we multiply the two it should have a unit of energy in
this case there is just one thing which I would like you to remember that since thermodynamics
driving force and thermodynamic displacements they are scalar quantities.

So therefore they are different from the displacement and driving force which we define in
mechanics for example they do not obey the Newton’s third law of mechanics and it is important
to remember that although these two are scalar quantities but they are the collective property the
motion of the large number of ultimate particles which exhibit Newtonian mechanics, right? But
since they are completely random they are average of all the displacements and driving forces of
a large number of ultimate properties, finally sum up to manifest as the pressure change which
brings about the volume change.

26
So therefore it is important for us to remember that in thermodynamics we define energy as a
product of the thermodynamic driving force which is nothing but an intensive property change
and which brings about a change in any extensive property of a system which is the
thermodynamic displacement. Usually since we deal with pressure volumes type of work so
usually the redefined work done as integral PdV.

(Refer Slide Time: 20:33)

27
(Refer Slide Time: 20:37)

Now it is important for us to remember that the energy interactions between the system and the
surroundings are usually in the form of work or in the form of heat, right? So here I am not going
to go into details assume that you already know it or you can take it up as a home assignment I
have just found out what are the different expressions of the work for different processes for
different reversible processes Isobaric, Isochoric, Isothermal, Adiabatic processes. You can very
well take these derivations as your home assignment in case you have any doubts we can have
the interactive session during the course and the TAs’ would be very glad to solve any doubt that
you have.

28
(Refer Slide Time: 20:55)

There are other types of works also which it is not only that PV type of work we can have other
type of systems other than piston gas in the piston cylinder type of arrangement, but we will find
that for each particular system that you consider, we can break up the work done into a
thermodynamic driving force and thermodynamic displacement and when we multiply the two
we get the unit of an energy, right?

29
(Refer Slide Time: 21:26)

Now as I have already mentioned there are two types of ways by which a system and the
surrounding can interact or can exchange energy, it can be either in the form of a work or in the
form of the heat. For both of them we need a thermodynamic driving force and a thermodynamic
displacement. When the driving force is the temperature difference than we call that particular
energy interaction as heat, when the driving force is anything other than the temperature
difference we call it as the work.

30
(Refer Slide Time: 22:38)

Now here I would like to mention one thing, since we are dealing with properties, so therefore
this driving force and the displacement both of them can be the property of a system both of
them can be the property of the surrounding or else one of them can be the property of the
system the other of the surroundings. For example suppose I take up this particular arrangement I
arrange my system such that my system comprises of not only the air in the piston cylinder
arrangement but also some portion of the surroundings as well.

Now in this particular case we find that the pressure difference which was bringing about the
volume difference everything was a property of this system. So therefore the energy change
which took place here, is also a property of this system. Similarly, suppose I define my system
somewhere outside such that this whole thing is a property or is or it lies in the surrounding then
in that case we find that the whole energy interaction was a property of the surrounding. But
suppose I have my system as the gas contained in the piston cylinder arrangement then we find
that maybe not this case a better case is when work is done.

In this case what happens suppose I apply or I gradually increase the pressure here I find that the
piston it keeps on coming down and it compresses the gas, so in this particular case what
happens the pressure which is applied that is a property of the surrounding and the contraction of
the compression of the gas or the volume decrease of the gas this particular volume is a property
of the system.

31
So therefore for this case the energy interaction which occurs whose property should it be?
Because in this case actually the energy interaction is occurring across the boundary which is a 0
thickness interface between the system and the surroundings and energy interaction occurs
because there is the pressure of the surrounding forces of volume change inside the system under
this condition we say that it is energy in transit, it is neither a property of the system nor a
property of the surrounding.

(Refer Slide Time: 25:36)

It is a path function it is energy in transit and as a result of which we consider this type of energy
in transit as path functions whose differentials are inexact due to which we are going to represent
heat and work interaction between the system and the surroundings by means of a different sort
of a differential as compared to the exact differential by which we had depicted a property. This
is something very important which I will be following in the class I would want you people also
to follow this that we would like to differentiate between inexact differentials and exact
differentials. The path functions will be shown by inexact differentials and the state functions
will be shown by exact differentials.

32
(Refer Slide Time: 26:01)

Now in this case so therefore I would like to mention that heat and work they are quite unique,
why? Because they are the energy interactions between a closed system and their surroundings
and we can have the same effect both by heat as well as by work. For example in this particular
case the expansion of the system could have been brought about by pressure changes in the
surrounding in which case we say that work is being done, in this case we say that work is being
done on the system we can also bring about this same change by a temperature differential with
the surrounding in which case we say that heat is being transferred to the system.

If you observe the initial and final state in this case and this case we find that they are the same
but the change in this case was brought about due to work done on this rather work done by the
system from the surrounding and then this case it was due to the heat transfer between the system
and the surroundings, right? Now you have to remember that before the change and after the
change is over we just see the change of state, we see the initial volume of the system, the initial
temperature and pressure of the system, we see the final volume and the final temperature and
pressure of the system.

We are in no position to understand how this change has been brought about once the change is
completed, so therefore we can just see the change which has brought about. So therefore it is
very important to remember that heat and work have meaning as long as the change is going on

33
once the change is over we find a change in the property of the system we do not know how that
change has been brought about.

(Refer Slide Time: 27:45)

So therefore please remember there is no total heat or work content of the system there is a total
energy content of the system but there can not be any heat or work content of the system they are
just manifested as long as the process is going on and then they are manifested once the process
is over by a change in property of the system which shows that there was an energy interaction
between the system and the surroundings.

It does in no way show that there has been or in what particular form that energy interaction had
taken place. Now before I end I will give you a very interesting example of how the choice of
system dictates whether it is work done or it is heat transferred. During winter every day we heat
up our water in our hostels by using an immersion heater most of us we have immersion heaters
which we dip in the bucket of water and we heat the water.

Now suppose we assume that our system is just water in the bucket and nothing else then in that
case how the water gets heated up? There is a temperature difference between the rod of the
immersion heater and the water due to which energy gets transferred and it is heat transferred
which heats up the water. Now suppose you think that well, no the water is not going to be my
system, my system is the entire thing inside the bucket that means the immersion heater the

34
water everything contained in the bucket. In this case how does the water get heated up? It gets
heated up because of the electrical work which is being done on the rod by the battery.

So therefore in this case it is work done on the system it is important to remember why I was
emphasizing so much on the choice of system, the choice of system is very important because
that decides whether we are having heat interaction or work interaction as we proceed you will
find that the choice of system may make your calculations difficult or it may make your
calculations easier as well.

(Refer Slide Time: 29:55)

So therefore choice of the system is very important and before I end I would like to have like to
discuss the sign convention which we are going to use for my work and heat in the entire course
of phase equilibrium thermodynamics. The sign convention is that when heat enters the system it
is positive and when system does work on the surrounding it is positive. So therefore we
followed this sign convention and proceed for our next class, thank you very much.

35
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 03
First Law of Thermodynamics

Well good day to all of you, so in the last 2 classes I have already completed the introduction and
by this time you know that what is a phase? Of course I had completed to rather I forgotten to
mention what a phase is? A phase is that portion of matter which is uniform throughout in
physical structure and chemical composition or in other words it is a homogeneous piece of
matter. So therefore any system that we select that may comprise of single phase or that may
comprise of more than one phases, okay.

Now when we have a system comprising of more than one phases then naturally within that
particular system the 2 phases are going to interact they both the phases they settle down to an
equilibrium state. And in thermodynamics what are you going to study particularly? We are
going to study the properties or the relationships between properties of the 2 phases of any
particular system and when the system has a whole interacts with the surroundings by virtue of
either heat or work, how the states of the 2 phases they change? How they are connected? And
what are the relationships governing phase equilibrium when 2 or more phases are in equilibrium
within any particular system?

Now before we go to 2 phase systems, the first thing which you are going to deal with is the
simplest thing we are going to deal with homogeneous systems or single phase systems. We are
going to find out the relationships between the properties of the different single phase or
homogeneous systems and the relationships of these properties with the energy interactions
which the system has with the surroundings. We are going to take up close systems first and then
we are going to deal with open systems or rather we are going to extend the analysis for open
systems at one and an isolated systems at the other end.

So therefore after we study the energy interactions, next we need to know that whether any
process is possible in a system or not or whether there are some processes which are possible
some processes which occur spontaneously to bring about the change in a system and some

36
processes they are not possible and we need to expend energy for them or rather to make them
happen.

Or in other words we shall be discussing in brief about the feasibility of the different processes
which a system may undergo. And after that once we have understood more or less the heat work
or rather the energy interactions between a system and surroundings for the time being, if
initially we are going to take single phase systems we understand what are the feasible processes
in the system, then the next thing which we do is before going into the details of phase
equilibrium we are going to deal with single component 2 phase systems and then we go for 2
components or multi-component multiphase systems.

So gradually we proceed and as we proceed the complexity increases and as I have already
mentioned in my introductory lecture that since many of you will be interested in petroleum
refineries etc, so I will be touching upon some portions of hydrocarbon thermodynamics the
remaining can be clarified from the book of Edmister or any other book which you find is
rudiment.

(Refer Slide Time: 4:22)

Now the entire we the paradigm of thermodynamics it is governed by primarily the 2 laws along
with zeroth law and third law of thermodynamics which is also there. Now these 2 laws they had
originated from a series of experiments which had been performed by Joule and Kelvin between
1843 to 1848 they are known as the famous paddle wheel experiments what did they do? The

37
experiments were very simple at that time they were quite crude but they came up with some
very important observations which formed the basis of the 2 important laws of thermodynamics.
What was the experiment that they did?

The experiment was they had contained water in a insulated vessel and in that particular vessel it
was connected with a paddle wheel and this particular paddle wheel it was rotated by the
lowering of the mass which was attached to the pulley arrangement on the top. So therefore what
these 2 scientists they had tried to do? They had rotated this particular paddlewheel with quite
sufficient speed and they had heated the water contained in the vessel, okay.

And how much amount of work was done or how much amount of heat was rather how much
amount of work was done on this particular water that could be manifested only by a change of
state of the system from an initial temperature T1 to a final temperature T2 which was noted by
means of the thermometer. Now after the paddle wheel they also use other types of devices to
perform work on this same volume of water kept in an insulated rigid container.

They found out that no matter whatever device they used in order to perform this amount of
work it was found that the amount of work performed was proportional to the amount of
temperature increased. For each case the temperature increased they have maintained the
temperature increase as constant and they found out that for each case the amount of work
performed it was proportional to the temperature increase which occurred.

And after each experiment after the temperature increased what they did? They broke the
insulation and immersed this volume of water inside the water bath to bring it back to the
original temperature and that they could find out what was the heat given up to the amount of
water contained in the water bath. Now this particular experiment which they did they found 2
very important observations which formed the basis of the first law and the second law of
thermodynamics.

The first thing as I have already mentioned they found out that the amount of work done was
proportional to the temperature increase in each case, no matter how the work was performed
and they observed that whenever the system it undergoed the cyclic change, what was the cyclic
change? When the initial and final states of the system are the same mind it, cyclic changes can
be reversible they can be irreversible it is important to remember that a reversible cyclic changes

38
both the system and the surroundings come back to their original state and for a irreversible
cyclic change it is the system which is brought back to its original state the surroundings does
not come.

When they performed a series of irreversible experiments because as we have already discussed
it is not possible to obtain reversible processes or reversible cyclic processes in practice. So what
did they find? They found out that whenever a system was undergoing a cyclic change however
complex the cycle maybe it was found that the cyclic integral of dQ was always proportional to
the cyclic integral of dW where the constant of proportionality was 1 in SI units.

As a result of which the first law can be mathematically expressed as given as this particular
expression. Now from this particular expression they found out some very important
observations which led to the law of conservation of energy as we all know. It is important is to
remember that first law does not directly give you the law of conservation of energy it is a
consequence of first law. In this particular experiment they also had another very important
observation, they found out that when they were performing of work the temperature was
increasing and then when after the temperature increased they broke down the insulation and
immersed this particular volume of water in a water bath when it gave up heat and came back to
its original state.

The same experiment was performed as I have already mentioned by different ways of work the
same process was followed and the people found out that it was possible for going from state 1
to state 2 during that experiments that is they could heat up by performing work for all the
experiments but the reverse process could not be performed, they could not use the heat content
of this particular system in order to lower this particular weight.

So this was another very important observation they found out or in other words they could not
first use the heat in order to perform work although the law of conservation of energy was
obeyed in the process. So therefore these are the 2 observations first thing was that as it is shown
by this mathematical statement from where we are going to see shortly we arrived at the law of
conservation of energy and the second observation was while doing work and increasing the
temperature of the system is the feasible one the reverse process could not be performed.

And if we have to use heat transfer in order to perform some amount of work they found out that
some amount of heat has to be lost to a surrounding at lower temperature and only the remaining

39
amount of heat some amount of useful work can be performed. Thus while the entire amount of
work can be converted to heat the entire amount of heat cannot be converted to work as a result
of which now we know that heat is a lower grade of energy as compared to mechanical energies.

(Refer Slide Time: 11:26)

So these were the 2 observations based on which the 2 laws of thermodynamics were formulated.
Let us now see the formulation of the first law of thermodynamics, as I have told you it gives
you the equation as dQ. Well, I will just mention that it is not the, these here are not cut so
therefore they do not show an exact differential this is a typographical error which has happened
but you please remember during your studies it’s very important that you show the inexact
differentials by either this thing or you can also write it down as Delta Q minus Delta W. In the
class I will be cutting the Ds to denote inexact differentials.

So therefore from this particular process suppose again I distance all of these suppose I assume
reversible processes and I plot the reversible processes not on a PV plot but on a generalize F-r
plot where F is a generalized thermodynamic driving force, r is generalized displacement. Now
here suppose my initial state was 1 from this initial state I go to the state 2 by a reversible
process which can be plotted reversible process 1a2 which can be shown by a solid line here and
I come back from here by another reversible process 2b1. Now my entire process 1a2b1 is a
reversible cyclic process.

40
So for this entire process I can write down that integral of 1a2b1 dQ minus integral 1a2b1 dW
this is equal to 0, so therefore I can also break it down and write it down as dQ 1a2 plus integral
dQ 2b1 minus integral dW 1a2 minus integral dW 2b1 this is equal to 0. I can rearrange it and
write it down as integral dQ minus dW 1a2 plus integral dQ minus dW 2b1 this is equal to 0, I
can write it down in this particular way.

Now instead of 2b1, I can also come back by another process and again a reversible process by a
different path which is 2c1. So for this particular 1a2c1 for this particular cyclic integral also I
can write it down as 1a2c1dQ minus integral dW1a2c1 this is equal to 0. In the same way I can
break it down and I can write it down as integral 1a2dQ plus integral 2c1dQ minus integral
1a2dW minus integral 2c1dW that is equal to 0. Now if I subtract equation 2 from equation 1
what am I going to get? I will just write down the equations once more for your convenience and
then I perform the subtraction.

(Refer Slide Time: 16:41)

First equation is 1a2 plus dQ 2b1 minus dW1a2 minus integral dW 2b1 equals to 0 and the
second equation is dQ1a2 plus dQ2c1 again dW 1a2 dW 2c1 equals to 0. Now if I subtract one
from the other what do I get? I find that these dQs and these dWs they cancel out and what do I
get? I get integral dQ2b1 minus integral dQ2c1 minus integral dW2b1 minus integral dW2c1 this
whole thing equal to 0 or in other words can I write it down as integral dQ minus dW 2b1 this is

41
equal to integral dQ minus dW 2c1 can I write it down in this particular way? And moment I
have written it down in this particular way can you tell me what does it imply?

It implies that dQ it depends upon the path dW depends upon the path but the difference between
dQ and dW that is equal for both the paths 2b1 and 2c1. When the initial state and the final state
initial state is 2 in both cases the final state is 1 in both cases when the initial and final states are
the same then the difference between 2 dQ and dW this particular quantity remains the same,
what does it mean?

It means that dQ although dQ is a path function dW is a path function the difference between the
two must be a state function and that should depend upon the initial state and the final state. And
this particular difference the difference between dQ and dW this is stored as internal energy or as
energy of the system and this energy of the system it depends upon the initial state and the final
state it does not matter whether we are coming from the initial 2 the final state via path c or via
path b the difference is the same this is nothing but E2 minus E1, right?

(Refer Slide Time: 19:01)

So therefore from here what do we get? We get rather the first thing which we get is that the
difference between these 2 is equal to the energy stored in the system and this particular energy
stored this Delta E this is a property of the system, this Delta E this comprises of the Kinetic
energy of the system, the potential energy of the system and also the internal energy of the
system

42
plus and this internal energy comprises of the energy of the ultimate particles which comprises of
of a the different motions of the molecules etc etc.

(Refer Slide Time: 19:07)

And this is the internal energy which was manifested by the increase in temperature in this
paddle wheel experiment. Usually in thermodynamics we deal with we deal with stationary
systems which do not have much displacement.

(Refer Slide Time: 19:19)

43
So therefore usually these 2 terms are 0 and therefore Delta E reduces to Delta U. And therefore
from here we get that for a closed system dQ minus dW equals to dU this occurs for a closed
system. So therefore for a closed system we can always say that the net energy change in the
system is equal to the net energy transfer across the system boundary.

Now what happens for an isolated system, for an isolated system we know that it does not allow
any sort of work either or rather any sort of interaction between the system and the surroundings
so therefore for an isolated system what do we get? dQ equals to 0, dW equals to 0 which
naturally brings you to the dE for an isolated system equal to 0 which is nothing but the principle
of conservation of energy.

So therefore we find that from the first Law what do we get? We find that for closed systems the
net energy change of the system is equal to the net energy interaction between the system and the
surroundings and energy is conserved for an isolated system. Within an isolated system suppose
there are differences in driving force between 2 parts of the system then definitely the 2 parts
will interact but the total energy of the system remains constant which is known as the principle
of conservation of energy and this particular principle of conservation of energy it includes all
types of energy including the Kinetic energy or the mechanical energy the internal energy and so
on and so forth.

(Refer Slide Time: 21:56)

44
Now from here what do we get for an open system, for an open system what do we have let us
see? Suppose we and let me remind you for most of the things that we work out a chemical
reactor, a nozzle, a pump, a compressor most of these they deal with open systems. Now for
open system suppose I take up an open system in this particular manner it is just a black box
because it can be anything. So therefore I take it up as a black box.

So say for example it can be a reactor it can be anything so from here we find that at pressure P1
some amount of volume is entering here the pressure is P2 and some amount of volume V2 goes
out here we find that shaft work possibly this particular device it performs some amount of shaft
work may or may not perform, some amount of heat enters in this particular system and then we
find that here some additional work is required other than the shaft work to push this volume of
fluid inside and this work is done on the system. So therefore it is negative work and some
amount of work the system does to push out some volume of liquid or some volume of this
system out through it.

(Refer Slide Time: 23:04)

So therefore this was my reactor and what do I do? In order to use the first Law for closed
systems I need to keep the mass of the system constant.

45
(Refer Slide Time: 23:10)

So therefore I devise my system such that its mass remains constant while some particular fluid it
keeps on flowing through it and the device interacts with the surrounding by virtue of work and
heat. So what do we do? We assume that the system comprises of the amount of mass which is
contained in this particular control volume and the amount of mass which is going to enter in this
control volume during time dt. So therefore at time t this is my system, this whole thing is my
system. And at time t plus dt I assume that my system has some amount of mass has gone out
through this, so now my system changes into this particular configuration.

So what do I find? I find that my system comprises of a fixed mass of gas which is enclosed by a
flexible boundary and I break my entire process into small timescales dt during this time scale dt
say some amount of mass Delta m is entering the system some amount of mass Delta m is going
out of the system. So therefore at time t my particular system comprises of the entire mass here
plus the Delta m mass which is going to enter here and at time t plus dt my system comprises of
the time the mass which has already and the mass which has gone out during this time t plus dt.

Now in this particular case what is my dW? dW definitely will comprise of the shaft work and
along with that it is going to comprise of the work which was done on the fluid so that it can
enter and the work which the system did on the fluid such that it can go out. So therefore the
total work in this case comprises of the shaft work and the work which was done on the system
in pushing this fluid plus the work done by this system in pushing out the fluid.

46
So therefore in this particular case my dE equals to dQ minus dW where this particular dW can
be written down as minus P2V2 Delta m minus dW shaft, right? So therefore for stationary
systems where there is no potential energy change also for those particular cases E can be written
down as U. So therefore in this particular case I get U2 minus U1 into Delta m this is nothing but
equal to dQ minus dW shaft plus P1V1 Delta m minus P2V2 Delta m or in other words if I bring
it here to this side I get U2 plus P2V2 Delta m minus U1 plus P1V1 Delta m this is equal to dQ
minus dW shaft. This is my first Law for open systems.

(Refer Slide Time: 27:20)

I am sorry this Delta m should have been here this Delta m, sorry for making this mistake or in
other words what I get is, I get that what this particular for open systems what I get that while in
closed systems my dQ minus dW was equal to dU.

47
(Refer Slide Time: 27:27)

In this case I find that dQ minus dW shaft this is equal to some particular term which also has
which is a property of the system and has the units of energy that is U2 plus P2V2 here U1 plus
PIV1 and as possibly you are aware of this fact that U plus PV is nothing but the we have
defined it with another particular property group which is known as the Enthalpy of the system.

(Refer Slide Time: 28:04)

So therefore for open systems we can write it down as dH which equal to dq minus dW shaft. For
closed systems we had dU was equal to dq minus dW. Now suppose we take a close system at

48
constant volume, what happens? And suppose we assume that the only work which can be
performed is PV work or in other words dW equals to PdV. So in this case for constant volume
work done is equal to 0 and dU equals to dq.

Suppose I take up a constant pressure process and or other a Isobaric process again assume only
PV work is done. So in this case dW equals to PdV which is nothing but equal to d pV as well
isn’t it? So therefore for this particular case what do I find? We find for this particular case d of u
plus PV equals to dq.

(Refer Slide Time: 29:16)

49
(Refer Slide Time: 31:00)

The same u plus PV which I had got in for a close system, I find that we have come across the
same u plus pV for an Isobaric close system as well. So we find that this particular term u plus
pV which also has a property of energy, so this we encounter very frequently so instead of
writing u plus pV every time we thought of defining it as one particular term and we define this
particular term as the Enthalpy of the system.

So therefore from the first Law what did we do? We found out that the first Law deduces the law
of conservation of energy for an isolated system and it gave the law as dU equals to dq minus
dW for a closed system and particularly since as we know that we are primarily very frequently
we encounter isobaric processes or isobaric processes this reduces to dq equals to dH and also for
open systems we find that law reduces to dq minus dW shaft equals to dH. So at the end of the
first Law we have defined 2 new properties one is the internal energy of the system and the other
is the enthalpy of the system.

Next class I would like to elaborate a little more on the enthalpy of the system and I would like
also to talk about how much amount of temperature rise we can obtain by some particular heat
interaction between the system and the surroundings or I would like to discuss a little more about
the heat capacities and the specific heat capacities of the systems before I close the first Law and
go for the second law of thermodynamics, thank you very much.

50
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 04
Second Law Of Thermodynamics

(Refer Slide Time: 0:33)

Well, good day to all of you to continue with our discussions on the first law of thermodynamics
what did we do? In the last class we have ended with the law of or the statement of the first law
for different systems and we found out that usually we find out that even for close systems since
it is very easy to maintain a constant pressure and we always prefer to perform processes under
atmospheric pressure. So therefore we encounter constant pressure processes very frequently.

51
(Refer Slide Time: 1:00)

So for such processes we find that the term u plus pV usually dominates and the same thing
happens for the open systems as well as a result of which we had in or rather we have included
the term enthalpy in our energy balance or in the first law system. Now before since enthalpy so
very important and we will be dealing with it very frequently, so before I proceed further I would
be I would like to spend a very short while discussing the important or the some interesting
aspects of the enthalpy of a system.

(Refer Slide Time: 1:36)

52
Now what is enthalpy basically it is a convenient property as you know which combines U and
PV or in other words it combines internal energy and the flow work and we have seen that it has
energy units but the important question is can we conserve enthalpy just like we have conserved
energy, for isolated systems the first law gave us the law of conservation of energy in isolated
systems, right? So in the same way do we have something like the conservation of enthalpy?

(Refer Slide Time: 2:41)

So let us see that, now if you that find suppose we have a rigid vessel where maybe it is rigid it is
immovable and maybe an isolated system so therefore there is no amount of internal energy
change in it. Say for example we have got some particular system which is enclosed by a
adiabatic boundary condition. So therefore this vessel it is rigid and enclosed by an adiabatic
wall and just here we have a piston cylinder sort of an arrangement, okay.

It is fully enclosed but we have a piston cylinder kind of arrangement from where we can impose
a pressure P on the system since it is rigid it cannot move, so therefore my dV equals to 0, right?
And since it is enclosed by an adiabatic wall so therefore I know that my dQ equals to 0, fine.
And since there is no other work being performed on it, so therefore naturally my PdV equals to
0 or my dW equals to 0 and since it is a closed system, so naturally for this particular system I
know my dU is also equal to 0.

Now I keep on increasing the pressure on the system, right? So therefore when I keep on
increasing the pressure on the system but since it the system inside it cannot move and there is no

53
heat transfer, so therefore the dU remains constant in the system since dQ equal to 0 dW equal to
0 holds even when I am applying pressure, so therefore what happens to the enthalpy change of
the system? The enthalpy change is in this particular case my d or rather delta H that is equal to
U2 minus U1 plus P2V2 minus P1V1. If I write it in differential form this is dH equals to dU
plus d of PV my volume remains constant, so therefore this becomes dU plus VdP.

I know my dU equals to 0 but dH is not equal to 0 because dP is not equal to 0, so therefore what
do I find? I find that even when dU equals to 0 even when we find that there is no energy change
of this particular system the enthalpy of the system can increase since the pressure of the system
is increasing.

So therefore in reality we cannot have something like the conservation of enthalpy just like we
have conservation of energy or in other words enthalpy is not a true energy like enthalpy energy.
And enthalpy can be included in the energy balance equation for 2 instances which we have
already shown, the first instance was for the open system which I have shown where we were
performing reversible flow processes and the second case is non-flow processes at constant
pressure where only expansion work or volume change work is involved.

So therefore we find that enthalpy can be included in energy balance equation for under only 2
circumstances the one for which I have derived the equations and incorporated enthalpy in my
first law those 2 processes are reversible flow processes and non-flow isobaric processes
provided they are undergoing only PV type of work and fortunately I should tell you that
majority of the industrial important processes they fall in one or other of these 2 categories.So
therefore we include enthalpy in our energy balance equation and we usually forget that enthalpy
is not a true energy in its strictest terms.

Now the other thing which I was thinking of discussing before closing the chapter on first law is
that we can apply first law for different close systems, right? I give it to you as an exercise for
applying first law and finding out the work interaction for isobaric I have already done isochoric
also I have already done but adiabatic, poly-tropic and isothermal processes you can take ideal
gas as the working fluid and you can do it even for a generalized system as well.

Now whenever for these particular systems we find that there is some amount of heat interaction
or there is some amount of work interaction and that brings about a change in property of the

54
System, the only property change which occurs due to heat interaction is usually the temperature
change, temperature change can occur both by heat interaction as well as work interaction but
how much amount of the temperature change will occur due to some amount of heat input on the
system that depends upon the heat capacity of the system, with the same amount of heat input
different systems will not exhibit the same amount of temperature change.

(Refer Slide Time: 8:21)

You are already aware of the heat capacity but again since it is quite a unique property so I would
like to spend some time on the heat capacity of the system. As this is very evident you have been
studying it from your school days, so you know that the heat capacity of the system is defined as
the amount of heat which it absorbs and the temperature rise which occurs as a result of this heat
change for the instantaneous heat capacity it is usually defined by this particular form.

55
(Refer Slide Time: 9:10)

Now there are some important things which I would like to mention the first important thing is
that heat capacity has got a meaning only when the heat input brings about a change in the
temperature of the system. So therefore it is meaningless when phase change occurs because
under that condition becomes infinitely large since the heat transfer in this case does not bring
about any temperature change.

The other important thing a very interesting thing is that supposed due to this heat input when
there is a chemical change occurring then we find that may be due to the chemical change there
is some amount of heat which is evolved or some amount of heat which is absorbed in, so
therefore the heat can be absorbed for the chemical change along with increasing the temperature
of the system.

Now under such conditions we will find that if we try to evaluate C and if the reaction is
occurring at a very slow rate and internal equilibrium is not established then the C which we are
going to measure and the C which we are going to measure when internal equilibrium is
established are going to be different. So therefore it is very important to note the condition under
which we are measuring C where a chemical reaction is taking place within the system and
usually it is specified that the heating here is slow enough for equilibrium to be maintained.

The C which is measured under that condition is the C which we should be taking because if the
measurements are taken very fast such that before internal equilibrium is attained we try to

56
measure the temperature rise amount and the amount of heat which is going on then they the
value of C which we will be getting will be different from the value of C which we will be
getting when internal equilibrium is established and the measurement is rather the heating is very
slow and internal equilibrium is established.

(Refer Slide Time: 11:25)

For the same arguments C usually refers to a closed system and the most important fact about the
heat capacity is although it is a property of the system it is a path dependent and it has got no
meaning until the path of the heating is specified. So therefore for each path of heating we can
have a different value of C and the 2 most commonly used are particularly important kinds of
heat capacity are the heat capacity at constant volume and the heat capacity at constant pressure.

We find that the heat capacity at constant volume is just the rate of increase of the internal energy
of the system with temperature and the heat capacity at constant pressure which we use much
more frequently is the rate of increase of the enthalpy with temperature it is important to
remember that this includes the flow work along with the increase in internal energy of the
system provided PdV is the only form of work. So therefore although it is a property of the
system it is path dependent, the heat capacities are extensive properties while the molar or the
mass heat capacities they are intensive properties.

Well, the next thing which I would like to mention is that so long we have been discussing about
the law of conservation of energy. Now we find that whenever a process takes place then for that

57
particular process the total amount of energy interaction with the surrounding is equal to the total
amount of energy increase or rather total amount of energy change within the system but even
while obeying this particular concept also we find that some processes can be possible some
processes can not be possible.

(Refer Slide Time: 13:19)

For example I take up few processes which are shown here, suppose for each case I would like
to go from state A to State B what is it? It is some very simple I have 2 equal blocks of copper at
20 and 30 degrees centigrade I bring the 2 blocks together and then each block is at 25 degrees
centigrade I can spontaneously come from state A to State B but the reverse I have 2 blocks at 25
degrees centigrade I bring them together and then I find one block goes to 30 and the other block
comes to 20 degrees centigrade that particular process is never possible.

58
(Refer Slide Time: 14:07)

Again in the similar way suppose we have some particular gas at one half of an insulated
container the phenomena of free expansion as it is called. We, maybe this is an insulated
container in this particular container this portion is evacuated and here we have some particular
gas at say some particular temperature T and somebody particular pressure P I remove this
particular partition the gas automatically it expands and fills the entire space here but when the
gas has filled and I try to push some particular partition the gas will automatically never come
back and fill one half of the container.

So while this free expansion, this phenomenon of free expansion as it is called this is a
spontaneous process the reverse never occurs. In the same way suppose I have one other
container again say I assume an insulated container and I have gas A here and I have gas B here,
I remove this I find that gas A and B mix together to form a homogeneous mixture but I have to
use tremendous amount of energy if I want to separate this homogenous mixture into gas A and
B and I try to confine gas A at one portion of the partition gas B at the other portion of the
partition it is never possible in this particular way. I can keep on citing several such examples
where we find that while the process in one particular direction is possible the process in the our
other direction is never possible even when it does not flout the the law of conservation of
energy.

59
In all the systems you find that 2 blocks at 20 and 30 degrees centigrade and each coming to 25
degrees centigrade the amount of heat transfer which would have occurred with the same amount
of heat transfer we can always postulate the 2 blocks at 25 degrees centigrade can be made to
contact and one comes at 20 the other goes at close to 30 degrees centigrade the same amount of
energy interactions in pen and paper will show you that same the amount of energy interactions
can take place when we go from state A to B and when we come from state B to A but all of us
know that while state A to B is highly possible state B to A is never possible.

Now why does it happen? Why do some particular processes are feasible while the reverse
process is not feasible? Why some processes occur spontaneously and others do not occur
spontaneously.

At one point of time it was thought that whenever any process it gives out heat it is spontaneous
whenever it has to absorb heat it is not spontaneous but then people found out there are several
endothermic reactions as well which are spontaneous. So therefore they felt the same way the
Joule’s experiment while we could use the way to heat the water we could not use the thermal
energy of the water in order to perform work unless we make some other arrangements as I have
discussed. So why are all these things happening? Then people found out that well the primary
reason is that some processes are feasible and others are not feasible.

For example heat always flows from a higher temperature to a low lower temperature never the
reverse thing happens, so why are these things happening? So therefore in order to define these
things the second law of thermodynamics has come up and the second law of thermodynamics
that postulates that heat can flow from, rather that postulates that while some processes are
feasible while the reverse processes are not feasible.

Well, there are 2 statements of the second law of thermodynamics which I will just be displaying
tomorrow but what I am going to discuss right at the moment is that while the people found out
that some processes are feasible they wanted to develop or they wanted to propose a quantitative
measure of why some processes are feasible why some others are not?

Whether it can be quantitatively predicted so that suppose they want to perform any particular
process they would find that particular quantity they would be evaluating that particular quantity

60
and based on the results they get they would be able to decide in which direction they should
proceed.

(Refer Slide Time: 19:07)

So the second law was in attempt in this particular direction. The second law has a large number
of statements one statement is by Kelvin Planck the other statement is by Clausius. Now let me
tell you that all these statements they are with reference to the heat engines and the heat pumps
because mostly as I have told you that the oldest and endeavors of man were in terms of heat
engines and heat pumps. So Kelvin Planck statement in a nutshell it says does the same thing
which Joule had observed that it is impossible to continuously absorb heat from or rather to
convert heat into work by working with a single reservoir that would lead to a perpetual motion
machine of the second kind and if that were possible then Joule could have performed the
experiments in both the directions.

And Clausius said that if heat has to be transferred from a body at lower temperature to a body at
higher temperature then work has to be done to make this particular transfer possible, right? So
these were just qualitative statements, now based on this further developments were done the
first development was a Carnot cycle and the Carnot’s theorem which followed from them. What
was the Carnot cycle? The Carnot cycle was a reversible cycle or the reversible heat engine cycle
where Carnot had devised it in terms of a PV work of a gas contained in a cylinder.

61
Let me tell you Carnot cycle is just a reversible cycle comprising of 2 adiabats these are the 2
adiabats and the 2 isotherms 2 reversible adiabats and 2 reversible isotherms it can be made to
apply on any particular system performing any type of work, so the correct representation is
instead of representing it in terms of a PV plot it is better to represent in a plot of F and r where F
is the thermodynamic driving force, r is the thermodynamic displacement F and r can be P and V
it can be Sigma and A it can be anything depending upon the system.

Now in this particular Carnot cycle there were some very important theorems which he had
proposed. Now among this theorems the first thing which we proposed is that any irreversible
cycle has a efficiency much lower than a reversible cycle. So therefore the efficiency of a
reversible cycle will be the maximum. The next thing is for all reversible cycles operating
between the same 2 high-temperature and low-temperature reservoirs no matter whatever path
they take the efficiency of all reversible cycles will be the same irrespective of the working
medium and irrespective of the processes if they are reversible cycles they will have the same
efficiency for each and every cycle and the efficiency will be higher as compared to a
irreversible cycle.

Now based on this they proposed that a ideal gas Carnot cycle that means when ideal gas is used
as a working substance in a Carnot cycle then for that particular situation he found out that Q1 by
Q2 equals to T1 by T2 this has been developed from the concept of efficiency of the cycle. Now
we know how to define the efficiency of a heat engine Carnot cycle was devised for a heat
engine where in very brief let me tell you a heat engine is a device which takes up heat from a
high-temperature reservoir and this heat engine it transfers heat to a low-temperature reservoir
and thus work it takes up heat Q1 rejects heat Q2 and thus work with the remaining amount of
energy Q1 minus Q2, right?

Now suppose we use ideal gas as the working fluid in this heat engine then Carnot found out that
Q1 by Q2 is equals to T1 by T2. Later on Kelvin he thought or rather it was his great intuition he
thought that why only ideal gas has the working fluid if suppose we can devise a temperature
scale where the heat has the thermometric property then definitely we should be able to propose
an absolute temperature scale completely independent of the working substance where the
temperatures of the 2 reservoirs will be in proportion to the amount of heats taken up and

62
rejected in them and therefore we should be able to produce or rather we should be able to devise
an absolute temperature scale and this gave rise to the concept of the absolute temperature scale.

Since it is outside the domain of phase equilibrium thermodynamics and I have got a limited time
I have just 20 hours so therefore I will not be going into the details of this but you can take up
any textbook and go through the definition and the development of the absolute temperature
scale, right?

(Refer Slide Time: 26:02)

But what is of importance to us is, that based on this we find that the very important proposition
was suggested it was that for any particular working substance when it is performing a reversible
cycle as a Carnot cycle that means the reversible cycle comprises of 2 adiabats and 2 isotherms
then in that case we can always write down that Q1 by T1 equals to Q2 by T2 or in other words
Q1 by T1 minus Q2 by T2 equals to 0.

And if you remember that Q1 has been taken up or rather heat provided to the system so
therefore it is positive Q2 is heat rejected from the system that is why it is negative than taking
this sign into account we can always write it down in this particular form which gives us the
equation as Sigma Qi by T equals to 0 and if we have an irreversible Carnot cycle then in that
case what do we know?

63
We know that the efficiency of an irreversible Carnot cycle is one minus Q2 by Q1 and the
efficiency of a reversible Carnot cycle is one minus T2 by T1. So therefore in this particular case
we know this is less than this or in other words we can always write it down as T2 by T1 it is less
than Q2 by Q1.

(Refer Slide Time: 27:20)

Or in other words in the same way we can write it down as Qi by T Sigma this is less than 0, so
therefore in any particular working substance operating in a Carnot cycle we find that Sigma Qi
by T is less than equal to 0 for a Carnot cycle where the equality sign is applicable for the
reversible process and the inequality sign is applicable for a irreversible process. Now next we
also know that any particular cyclic process plotted on a F-r plot that can be broken down into a
series of Carnot cycle.

64
(Refer Slide Time: 28:54)

I am not going into the details of all this because I assume that you know the basics of
thermodynamics or you can find it out in anybody particular textbook of thermodynamics. For
any particular process we can break it down into an adiabatic process, an isothermal process,
adiabatic process and another isothermal process and we can bring the adiabats as close as
possible because they are never going to intersect with one another and therefore any reversible
cycle can be broken down into a series of Carnot cycles and for each of the Carnot cycle this
particular equation holds if it is a reversible cycle.

65
(Refer Slide Time: 28:59)

And this particular equation holds if it is an irreversible cycle.

(Refer Slide Time: 29:52)

So for any particular reversible cyclic process we can always write down this is equal to 0 and
for any particular irreversible process we can always write down this is less than 0. So
combining the 2 we can always write down dq by T for any particular cyclic process it is less
than equal to 0 which is another consequence of the Carnot’s theorem and this was first proposed
by Clausius and it is known as the Clausius inequality. And from here stems out the criteria for
feasibility of different processes which we will be discussing in the next class where I will be

66
introducing the concept of entropy and we will be discussing the feasibility of the different
processes just like I had introduced internal energy and enthalpy from the first law I will be
introducing entropy from the second law and once we have introduced this we have understood
the energy interactions and the energy change of different systems we can understand when
some processes are feasible and why some processes are not feasible on a quantitative manner.

After that we go into the depths of phase equilibrium thermodynamics as I have already
mentioned earlier first we will be dealing with single phase single component the simplest type
of systems. We can also deal with single phase multi-component systems where the compositions
does not change as the process proceeds.

So both of them will have the same treatment and we are going to deal with them then we are
going to deal with single component 2 phase systems then multi-component 2 phase or
multiphase systems. So with this I end up this particular class, thank you very much.

67
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 05
Second Law Of Thermodynamics (Contd.)

Well, good day to all of you again, so to continue with your discussions where we had left it in
the last class I had started discussing the second law of thermodynamics and before that I had
very briefly or quite rapidly glanced through the introductory concepts and the first law of
thermodynamics you are requested to refer to any particular standard textbook go through the
things which I have discussed in the class and in case if you have got any further doubts, me and
my TAs’ will be glad to clarify those particular doubts in our interactive and or question-and-
answer session.

(Refer Slide Time: 1:09)

Yesterday I had started the second law of thermodynamics and I had given you the 2 statements
the Kelvin plank statement and the Clausius statement and if you go through these 2 statements
carefully just the way I was mentioning in the last class that most of the statements they have
been developed with respect to heat engines and heat pumps.

68
(Refer Slide Time: 1:48)

Now very briefly what is the heat engine? It is a cyclically operating device which converts heat
to work, for this particular purpose what it has to do? It has to take up heat from a high-
temperature reservoir, now what is a reservoir etc? Those things will get in any standard
textbook and you are supposed to refer them. So therefore what a heat engine does is, it takes up
heat from a high-temperature reservoir say Q1 amount of heat and it rejects heat to a low-
temperature reservoir some amount of the heat and it does work with the remaining portion and
so therefore this is what a heat engine is?

What does Kelvin Plank statement says? It says that it is impossible for a heat engine to operate
without this particular portion or in other words we cannot convert the entire Q1 into W that we
have already discussed in the last class that due to this reason heat is referred to as a low-grade
energy while the reverse the entire W can be converted to Q1 but the entire Q1 cannot be
converted to W there has to be a heat interaction with a low-temperature reservoir where some
amount of heat has to be rejected and the remaining portion can be converted to W.

So therefore Kelvin plank statement says that this particular leg of the heat engine should be
there and what does Clausius statement say? It suggests the feasibility of operation of a heat
pump, heat pump is nothing but it is just the reverse of a heat engine what it does? Its basic
purpose is to take up heat from a low-temperature reservoir and to pump that heat to a high-

69
temperature reservoir and for this some amount of work has to be done on it without this
particular amount of work it cannot take up the heat and reject it.

Just to be very brief heat pumps can be either heat pumps or a refrigerator the difference between
the 2 being in the purpose fits their going to serve. The purpose of the heat pump is to keep the
high-temperature reservoir hot and purpose of the refrigerator is to keep the low-temperature
reservoir cold, so as a result in the case of refrigerator the high-temperature reservoir is the
ambient and for the heat pump the low-temperature reservoir is the ambient, this is the basic
difference between the 2.

What does Clausius statement say? Clausius statement says that for a heat pump to operate this
much amount of work has to be done without this particular portion a heat pump cannot operate
its quite evident because suppose a heat pump could have operated without this amount of work
it automatically implies that heat can flow from a low-temperature to a high-temperature which
we know it is not possible.

Now I leave it to you as an exercise to show that the Kelvin plank statement and the Clausius
statement they are equivalent to one another and the violation of one automatically implies the
violation of the other and after that what we did? We found out that the most efficient way of
operating any particular process or any particular cycle is by a reversible process.

Well, so therefore a reversible process or a reversible heat engine cycle will definitely ensure that
the heats taken up from the high-temperature reservoir and rejected to the low-temperature
reservoir occurs under differential temperature difference or they occur under isothermal
conditions.

So therefore these 2 processes should comprise of reversible isotherms in order to get the
maximum efficiency. What we were discussing was that the Carnot cycle is the heat engine cycle
which comprises or rather it is a reversible heat engine cycle which comprises of 2 reversible
isotherms where the heat is being taken up from the high-temperature reservoir and heat is
rejected to the low-temperature reservoir and the 2 isotherms are connected by 2 adiabats where
the net work is performed such that the net work performed is equal to Q1 minus Q2, if anyone
of these processes comprising the reversible cycle is irreversible then it becomes an irreversible
cycle, okay.

70
(Refer Slide Time: 6:52)

Now after this based on this cycle Carnot has proposed some very important propositions which
we have already discussed in the last class and based on those particular propositions we had
derived the Clausius inequality which states that in any particular cyclic or rather in any
particular reversible cycle the quantity dq by T it disappears along a reversible cycle and that is
less than 0 along an irreversible cycle, so therefore the 2 equations combined give us the
Clausius inequality which we have derived in the last class.

Now I would just like to remind you that in order to derive this, what we did? We first based on
the efficiency of a heat engine cycle we first derived this particular equation for a reversible and
an irreversible Carnot cycle and then I had proposed that any particular reversible cycle can be
broken into a series of Carnot cycle this was based on the supposition that any reversible process
is equivalent to a reversible isotherm, a reversible adiabatic and another reversible isotherm.
While this can be done again I leave it to you as an exercise for you to prove that any reversible
process is equivalent to 2 isotherms and a reversible adiabat.

Now based on this any irreversible cycle can be broken down into a large number of adiabats and
since we know that 2 adiabats can never coincide or can never intersect again why they cannot
intersect? Is for you to find out and in case you have doubts we are going to clear it in the
interactive session. So therefore for any cycle it can be divided into a large number of closely
spaced adiabats with reversible isotherms connecting the different adiabats such that the entire

71
reversible cycle can be broken into a number of Carnot cycle, for each Carnot cycle this
particular equation or this particular proposition is applicable which gives for any reversible
cycle dq by T is equal to 0 and in any portion of the cycle is irreversible then the term dq by T is
less than 0.

(Refer Slide Time: 9:58)

Now let us see what this particular Clausius inequality implies this gives rise to a very important
relationship or it gives rise to a very important property which decides when a process is going to
be irreversible or when a process is feasible and when it is not? Now let us see how it is done?
Again based on a F-r plot maybe from state one to state 2 we devise a reversible process, so the
process proceeds along 1a2 and the system comes back to its original state along another
reversible process say 2b1.

As a result the system has undergone a cyclic reversible cycle 1a2b1. Just note that since it is a
reversible cycle, so therefore all processes in the cycle are reversible so therefore each state the
system has passed through in the course of this process is an equilibrium state therefore we can
trace the path of the process by a solid line on the F-r plot. Now therefore for this reversible
cycle from Clausius inequality what can we write? We can write 1a2b1 dq by T this is equal to 0.

What does it imply? It implies that 1a2dq by T plus 2b1 dq by T this is equal to 0 or in other
words we can write it down as minus dq by T 2b1, now it is important to note that 2b1 is a
reversible cycle, so therefore if we can reverse the heat interactions then the heat interactions

72
along 2b1 is just the opposite of the heat interactions along 1b2, so therefore this is equivalent to
1b2 sorry dq by T.

So therefore we find that along this particular process 1a2 and 1b2 we find that the quantity dq
by T reversible along path 1a2 is equal to the quantity dq by T along the reversible path 1b2, so
therefore what does it imply? It implies that provided the path is reversible the quantity dq by T
it is independent of the path followed it just depends upon the initial state and the final state.

So naturally when any quantity its change is independent of the path follows and depends only
on the initial and final state. So naturally that particular quantity as we know, it has to be a
property of the system. So therefore we find that although dq is a path function the term the
quantity dq by T is a state function and this state function as we know this is defined by an
entropy and therefore this particular quantity is nothing but equal to S2 minus S1 the change in
entropy of the system in going from state 1 to state 2, change in entropy of system in going from
state one to state to state two.

So therefore we have found this out. So therefore we find that for a reversible cycle we find that
the quantity dq by T is equal to the change in entropy of the system in going from the same
initial to the same final state now this was for a reversible cycle.

73
(Refer Slide Time: 13:44)

Now suppose I say that the system returns from state 1 to state 2 by an irreversible process say
2c1 since in this case the path is undefined there are several non-equilibrium states in the
process. So therefore the curve joining it is depicted by a dotted line and not a solid line by the
irreversible process.

Now for this particular case for the cycle say integral 1a2c1 for this case the quantity dq by T is
less than 0 from Clausius inequality, isn’t it?

(Refer Slide Time: 14:25)

74
So elaborating this, what do we get? We get integral 1 a2 dq by T plus integral 2c1 dq by T this
is less than 0 and we know that this particular quantity dq by T this is nothing but equal to
integral 1 to 2 ds and we also know that if we just reverse the process or rather if we just reverse
the heat interaction then the reverse process or in the reverse process the heat interaction just gets
reversed. So therefore this is nothing but equal to integral 2 to 1 minus dS, isn’t it?

So there because dS depends just on initial and final state and nothing else, so therefore instead
of this particular term dq by T, can we not write it down as 2 to 1dS plus dq by T along 2c1 that
is an irreversible process it is less than 0 or in other words dq by T irreversible is less than dS for
that same change in states of the system say from an initial state to the same final state, so
therefore in this particular case what do we get? We find out that the amount of or in other words
dS is greater than equal to dq by T where the equal to sign it holds for a reversible process and
greater than sign holds for an irreversible process.

In order to convert this particular inequality to equality we can also write it down as dq by T plus
the amount of entropy which is generated as a result of irreversibility of the process. So therefore
we find that for reversible process entropy generated due to the irreversibility equals to 0 and dS
equals to dq by T and for an irreversible process dSG is greater than 0 and dS it is greater than dq
by T.

So therefore this is what we get and we find that for a process to be feasible there has to be some
amount of entropy generation which automatically implies that entropy of any particular system
it keeps on increasing there is nothing like the conservation of entropy just as we have a
conservation of energy there is nothing like the conservation of entropy and entropy is always
generated during any particular irreversible process.

75
(Refer Slide Time: 18:09)

Now very interestingly suppose we take up an adiabatic process. For adiabatic process what do
we have? We have dq equals to 0, so therefore dS which should be equal to dSG or in other
words dS should be equal to dSG. When the process is reversible adiabatic then naturally we
know that dSG equals to 0 and dS equals to 0 as a result of which we had already stated possibly
you are aware of the fact that any irreversible adiabatic process is termed as an isentropic process
because all the reversible adiabatic processes have to be accompanied by or rather can be
depicted by dS equal to 0 when the process is not adiabatic what happens?

Sorry, when the process is irreversible, for an irreversible adiabatic process what do we get? For
an irreversible adiabatic process definitely since it is adiabatic dq equals to 0, right?

76
(Refer Slide Time: 19:10)

And from this particular equation then what do we get?

(Refer Slide Time: 20:11)

From this particular equation we get dS is greater than equal to 0, right? So therefore we find that
when the process is irreversible adiabatic then in that case it is not an isentropic process although
dq equals to 0 there is some amount of entropy generation which results in an increase in entropy
of the process even when there is no heat transfer in this particular process.

77
So therefore from this what do we deduce? We deduce that all reversible adiabatic processes are
isentropic. Now if I say that well, if you are convinced about this statement, is the reverse
statement true? What do you think should all isentropic processes be reversible and adiabatic?
Just let us give a thought at it, let us see that if this is definitely undisputed.

(Refer Slide Time: 20:36)

Let us see what happens for an isentropic process we know that for an isentropic process dS is
equal to 0, now what does this imply? This implies that dq by T plus dSG equal to 0. This
implies either dq by T equal to 0 and dSG equals to 0 or this particular thing when dS equal to 0
implies this then this suggests a reversible adiabatic or isentropic process this is an reversible
isentropic process.

But we can also have an isentropic process with this being obeyed when this is obeyed under that
condition what do we find? We find out that for an irreversible isentropic process is also possible
where dSG is equals to minus dq by T what does it imply? This implies that we can have an
isentropic process even when it is not reversible and adiabatic but under that condition the
process has to be irreversible and there has to be entropy generation and this particular entropy
generation has to be equal to the amount of heat which is lost by the system to the surroundings.
So it implies 2 things, first thing is an irreversible isentropic process is possible only when heat
is lost from the system to the surroundings, your reverse direction of heat transfer is not possible
number 1.

78
Number 2 we find that only when the process occurs in such a way that the system loses heat at
the same rate at which entropy is being generated then we can definitely have an isentropic
process which is neither reversible nor adiabatic but it is isentropic. A classical example is
maybe a liquid flowing through a pipe under where there is a lot of friction produced and the
flow is such that the fluid loses heat to the surroundings at the rate at which the heat is produced
due to friction.

(Refer Slide Time: 23:26)

So therefore this is something very important for you to note and it is very important for us to
remember that all adiabatic reversible processes are isentropic but all isentropic processes are not
reversible, we can have isentropic processes but these have to be irreversible number 1 and
number 2 for such processes to occur the system has to lose heat to the surrounding, heat transfer
in the other direction is not possible. So this is one aspect which I wanted to discuss.

79
(Refer Slide Time: 24:01)

Now let us see what happens for an isolated system, let us take up an isolated system because
again as I was telling that for most of the cases we would always like to refer to some sort of an
idealization in order to find out how the system behaves under idealized condition? For an
isolated system what do we have dq equals to 0 we know. So therefore for this particular case
your dS has to be greater than equal to 0, agree?

So therefore for an isolated system we can have dS equals to dSG, where dSG equals to 0 for
reversible processes, dSG greater than 0 for irreversible processes. So for this, what do we finally
deduced? We deduce that in an isolated system entropy is always generated if any particular
irreversible or feasible process occurs. So therefore from here what is the condition of feasibility
of a process in isolated system? We find that if dS is equal to 0, we know that the process is
reversible it is feasible in both directions that means both the forward and the backward process
can occur.

If we note if we find that we calculate the entropy change of the isolated system when it is
performing a process and we find that dS is greater than 0, then we know that the process is
feasible in the forward direction but it is not feasible in the backward direction because in the
backward direction dS will be less than 0 and suppose on calculation of dS you find dS is less
than 0, what do you deduce? You deduced process not spontaneous or not feasible. It is
spontaneous in the backward direction.

80
So from here we have deduced the criteria of spontaneity of any particular process in an isolated
system the criteria of spontaneity is dS greater than equal to 0 where this equality sign is
applicable when the process is feasible in both the forward and the backward direction and the
greater sign it shows that the process is feasible in the forward but not in the backward direction
and dS less than 0 it signifies that if the process has to proceed in the forward direction energy
has to be applied to it but in the backward direction or in the reverse direction the process is
feasible.

So we continue our discussions a little more discussions on entropy in the next class, good day.

81
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 06
Second Law of Thermodynamics (Contd.)
Entropy Change During Spontaneous Processes

(Refer Slide Time: 0:27)

Well, so in the last class we had concluded with this particular slide where we had shown you the
criteria of spontaneity of a process. So therefore in an isolated system how you going to find out
whether a process is going to be possible you are going to calculate the entropy of the process.
Now for calculation of the entropy we need to remember that the entropy is equal to the heat
interaction divided by the temperature at which the interaction occurs only when the interaction
occurs along a reversible path.

82
(Refer Slide Time: 1:01)

Or in other words we know that dS can be evaluated from dq by T provided the path is a
reversible path it can be any Dem path but it has to be a reversible path but once the entropy
change has been calculated along a reversible path even between the same 2 initial and final state
if the system undergoes an irreversible path also the entropy change will be the same because it
depends just on the initial and final states.

What I mean to say is that along an F-r plot suppose the system moves from state one to state 2, I
would like to know or rather I would like to take a system from state 1 to state 2, I would like to
know whether this particular process is possible or not or am I going to find out whether this
process is possible.

What I will do? I will try to find out the entropy change which the system undergoes when it
moves from state 1 to 2. Now in order to find out the entropy change between state 1 and 2 or in
order to find out s2 minus s1 what do I need to do? I need to connect these 2 states by a
reversible path, a reversible path may be fictitious it may not have any particular connection
where the actual irreversible path which the system had followed in moving from state one to
state 2 but we calculate the entropy change along this reversible path and it is important to
remember that the entropy change along the irreversible path is also the same.

83
For example let us take up a very few common example which I had shown to you in the last to
last class that these processes are feasible and the other reverse process is not feasible and I had
taken up 3 examples, what were the 3 examples?

The first example was free expansion, okay. In this particular case what did I do? There was a
cylinder which is broken by or rather which is divided by a partition and it is surrounded by an
adiabat an insulated wall such that no heat transfer from the surrounding occurs and I had some
amount of gas which was concentrated here at within the volume V1 and then this particular
partition was removed and the entire gas expanded and occupied the entire, sorry this volume is
V2 the gas occupied the entire volume V1 plus V2.

We know that this is an spontaneous process but the reverse is not spontaneous, so therefore if
this process has to occur then definitely this must be accompanied by an increase in the entropy.
Let us see whether it happens or not or in other words how to calculate the entropy change of this
particular process of the gas expanding from a volume V1 to a volume V1 plus V2 when it is
kept enclosed in a adiabatic container.

What is, we have to first consider the initial and final states of the system what was the initial
state of the system? Initial state of the system was gas at V1 and say some particular temperature
T, what was the final state of the system? It was the same gas which has occupied the volume V1
plus V2 and it was under the same temperature T and the process was adiabatic. I want to
calculate the entropy change of the process what should I be doing?

I should devise a reversible path connecting the initial state and the final state now they are
connected by an adiabatic process, can you tell me what should be the equivalent reversible path
connecting these 2? We find that from this state to this state the temperature remains constant and
there is a volume expansion.

Can we assume that the equivalent reversible path can be a reversible isothermal process? During
which gas undergoes expansion from V1 to V1 plus V2, please note that this reversible
isothermal process is not adiabatic but this is an equivalent reversible process connecting the
same 2 initial and final states which the gas had during the actual adiabatic free expansion
process. Now if we try to calculate the entropy along this reversible isothermal process.

84
What will be dS equals to? This is nothing but equal to dq by T, what is dq by T? This is nothing
but dU plus PdV by T. Now suppose we assume an ideal gas, right? For an ideal gas will be
proving later but for the time being it is sufficient for you to know that for an ideal gas U is a
function of T only, right? And this is given by CvdT, so therefore since the temperature remains
constant dU for this case should be equal to 0 and therefore dS should be equal to PdV by T.

Suppose it is an ideal gas then for this particular case what do we have? We have R by V dV, so
therefore what dS equals to in this particular case or integral dS from say state 1 to state 2 this
should be equal to R by V dV from state 1, so the volume is V1 to V1 plus V2 or in other word
this is equals to R ln V1 plus V2 by V1 and as we know this is nothing but greater than 0. So
now do you understand that why free expansion is a spontaneous process?

Because the entropy generated in the process is greater than 0 in order to calculate the entropy
generated we had devised a reversible isothermal process connecting the initial and final states
which had no connection with the actual process which took place but it was an equivalent
reversible process which was connecting the same initial and the same final states we connected
the entropy change via the reversible process related the entropy change to the actual irreversible
process which had occurred and then found out that the process occurred because the entropy
change was greater than 0.

Now I have one question for you suppose the gas would have actually undergone a reversible
adiabatic process then in that case we find that entropy change is greater than 0, where is the
anomaly? I have told you that in this case the gas undergoes a reversible process, if the gas
undergoes a reversible process than dS should be equal to 0 but I find that for a reversible
adiabatic process dS is greater than 0, how do you explain this anomaly?

85
(Refer Slide Time: 8:51)

In this case try to consider what happens this is the gas and this particular ideal gas it is
occupying a volume V1 and it is undergoing a reversible expansion and it has finally the final
volume is V1 plus V2, right? From this particular initial state to this particular final state it had
undergone a reversible process, now during this reversible isothermal process in order to
maintain a constant temperature what was necessary?

It was necessary that some amount of heat was introduced or rather some amount of heat was
flowing into the process due to which the gas expanded and from initial volume V1 it went to
initial volume V1 plus V2. So therefore since heat was flowing inside this along with the change
in the entropy of the system which in this case is the ideal gas confined within this cylinder there
must be some entropy change in the surrounding as well from which heat was flowing into this
particular gas.

So therefore when I had said that dS equal to 0 this was for an isolated system. So therefore in
order to calculate the entropy change for this particular system I need to find out the dS of this
system and I need to find out the dS of the surroundings and then the total dS the sign of the total
dS is going to decide whether the process is going to be reversible or not.

Now we know what is dS of this system? We have already calculate this is Rln V1 plus V2 by
V1. Let us see what is dS surrounding? Why is the entropy changing in the surrounding?
Because heat is flowing from the surrounding into the system, what is amount of heat which is

86
flowing? This amount of heat is leading to this particular amount of change. So therefore from
first law we know that the amount of heat which has entered minus the amount of work which
has been done is equal to the change in internal energy of the gas.

In this case since the gas is at constant pressure dU equals to 0, as a result of which dq equals to
dW and since there is only Pv work associated here, dW is PdV since only Pv work is associated
here, so therefore this is RT by V dV and therefore in this particular case we find that the amount
of heat which has flown from the surrounding is equal to RT by v dV and what is the total
amount of heat which has flown from the surrounding into the system? This is equal to V1 to V1
plus V2 or in other words this is equal to RT ln V1 plus V2 by V1, agreed?

Now what is the entropy change of the surroundings then? We need to remember that this
amount of heat has flown into the system, so since it has flown into the system it has a positive
sign but if we consider the surrounding as a system then the surrounding is losing this amount of
heat, so therefore for this particular case dS surrounding is minus dQ by T where this dQ is equal
to RT by v dV, 1 by T and there is a minus sign here.

(Refer Slide Time: 12:36)

So therefore if we need to calculate dS surrounding then what do we need to do? The dS

surrounding again let me repeat it and let me write it down, dS surrounding is nothing but equal
to minus Rdv by V and from state 1 to state 2 the change is from V1 to V1 plus V2 as a

87
result of which this is minus R ln V1 plus V2 by V1. So therefore this is dS this is nothing but
Delta S surroundings and we have already calculated Delta S system earlier.

(Refer Slide Time: 13:12)

(Refer Slide Time: 13:20)

If you remember Delta S system was equal to R ln this particular thing, this actually should be
delta S system.

88
(Refer Slide Time: 13:27)

So therefore Delta S system this is equals to R ln V1 plus V2 by V1. So therefore if we add the 2
what do we get? We get Delta S total we assume that the system plus surrounding comprises of
an isolated system. So this is equal to Delta S system plus Delta S surrounding which is nothing
but equal to 0 which shows that a reversible isothermal process has 0 entropy change but a free
expansion it has a positive energy change.

(Refer Slide Time: 14:08)

89
We need to remember that since the container was insulated in this particular case, so therefore
for this particular case dS surrounding was equal to 0. So therefore dS total was equal to dS
system which is this.

(Refer Slide Time: 14:33)

This explains why free expansion as well as a reversible isothermal process both have the same
entropy change of the system but one of them is feasible the other in one direction the reverse is
not possible and the reversible isothermal process is possible in both the directions. So by this
example I tried to show a number of things, first thing which I tried to show was that how to
devise a reversible path and how to calculate the entropy change for any particular irreversible
process.

The second thing which I showed was the entropy change I proved that the entropy change is
greater than 0 for a very well known irreversible process and the entropy change is equal to 0 for
a reversible process but we had to remember in this process that the entropy change had to be
calculated for both the system and the surroundings because the feasibility that we had proposed
in terms of entropy change was for an isolated system.

90
(Refer Slide Time: 15:39)

Now there is also one more very common example which I was discussing in the last class that
we had two copper balls one was at 20 degrees, 30 degrees we brought them together and the
temperature was 25 degrees I said that the this particular process is possible the reverse one is
not possible. So let us see why this happens two identical balls one is at Th one is at Tc, right?
Now when both of them are brought in contact with one another Q lost by the hot body that is
mC Th minus T, Q gained by the cold body naturally this is plus and this is minus we have mC T
minus Tc, we know that under thermal equilibrium conditions Qh equals to Qc as a result of
which T equals to Th plus Tc by 2.

Now let us see that in this spontaneous process what is the entropy change? Now we can assume
that this particular composite system comprising of the 2 bodies is an isolated system because
whatever heat it is giving the same heat is being taken up by the cold body and we assume that
there are no heat losses to the surroundings. So therefore this particular unit can be assumed as
an isolated system. Now let us take up the entropy change of the hot body this is nothing but
equal to dQh by T what is Qh? This is mCdT by T where T is varying from Th to T.

What is dS for the cold body? This should be minus this should be plus. So therefore this is dQc
by T which is nothing but equal to integral of Tc to T, dT by T mC, what is the total entropy
change? dSh plus dSc this will be equal to mC ln T by Tc plus ln Th by T or in other words this
is going to be ln Th plus Tc whole square by 4 Tc, isn’t it? So therefore from this what do we
get?

91
From this we get that this is the total entropy change which has occurred in this particular
process?

Now let us see the sign of this how can we see it? We know Th is greater than Tc, so therefore Th
minus Tc has to be greater than 0. So therefore Th minus Tc whole square has to be greater than
0.

(Refer Slide Time: 18:41)

Or in other words Th square plus Tc square minus 2ThTc has to be greater than 0. I add a 4 Th
Tc and I add 4 Th Tc this side what do I get? I get Th plus Tc whole square is greater than 4 Th
Tc or in other words what do I get? I get the denominator is greater than the numerator as a result
of which the total dS total for this system is greater than 0.

What did I assume in this process? I assumed that there are no heat losses to the surroundings
and the heat transfer which was occurring that occurred so very slowly that we can assume a we
can assume that the heat transfer was occurring irreversibly, right?

92
(Refer Slide Time: 20:00)

So therefore from these 2 processes what I had deduced is, there was another process which I
leave you as an exercise the mixing of 2 ideal gases. In this particular case also yesterday I had
discussed, so this particular example is present in a large number of textbooks so I will not go
into details of this suppose we have N1 moles of a gas, right?

And I have N2 moles of another gas, okay. I simply mix them up and we will find they
spontaneously mix up to give me N1 plus N2 moles of the gas but if I have N1 plus N2 moles of
gas then they do not spontaneously unmix to form this, so therefore this also should be a
spontaneous process in order to find whether it is spontaneous or not what do we need to do? We
need to find out the entropy changes which occurred during the process.

In order to find entropy change we need to devise a reversible path, what should be the reversible
path in this particular case? Just see the process for the gas 1, what happens? There were N1
moles of gas at a temperature T, pressure P and then it gets mixed up here, right? For this final
state also the temperature was T the pressure was P.

But remember one thing this pressure of P here was for N1 plus N2 moles of gas it was but in
this case the pressure P was just for N1 moles of gas, so therefore actually what did this gas do in
the process of mixing?

93
It expanded from the total pressure P to a partial pressure P1 in this particular case and the gas 2
it expanded from a total pressure P to the partial pressure P2, such that P was equal to P1 plus P2
this was the exact process which takes place during the mixing of the 2 gases both the gases they
expand from the total pressure to the partial pressures such that the partial pressures when added
up gives the total pressure.

So in this particular case also what can be the reversible path? The reversible path in this
particular case is the reversible again the temperatures are all constant, so in this particular path
this particular process we can assume that there are 2 reversible processes comprising the whole
system of the mixing of gases the first process is reversible isothermal expansion of gas 1 from P
to P1 and then we assume that there is a semi permeable membrane here through which the gas 1
after expanding it just diffuses into this particular container, right?

So therefore diffusion of gas 1 at P1 through a semi permeable membrane, right? Now since in
this particular case there is no amount of heat interaction so for this particular case we can
assume dq equals to 0 and dS equals to 0 and for reversible isothermal expansion we have
already deduced the entropy change the same thing happens for gas 2 as well so therefore in this
particular case we will find that the dS for the gas 1 this can be given by N1 R ln P1 by P and dS
for the gas 2 is N2 R ln P2 by P.

(Refer Slide Time: 24:19)

94
As a result of which we find that for this particular case the dS total of the gas integrating from
state 1 to state 2 which gives you Delta S total is integral 1 to 2 dS for gas 1 plus integral 1 to 2
dS for gas 2 and we know that this is nothing but equal to N1 R ln X1 plus N2 R ln X2 where X1
is nothing but equal to P1 by P, X2 is nothing but equal to P2 by P, right? And therefore if we
would like to calculate the molar enthalpy the total molar change in the molar entropy of the
system then we get this is nothing but equal to Sigma Xi R Sigma Xi ln Xi.

(Refer Slide Time: 25:22)

So therefore the molar entropy change of mixing of a large number of gases is nothing but R
Sigma Xi ln Xi and since we know that sorry minus there is going to be a minus sign here very
sorry because it is V1by V2 so I missed out a minus sign in everything so therefore this has to be
a minus here.

So we find that since ln Xi is less than one so therefore this particular term for mixing of several
non-identical gases that is greater than 0 and that is why mixing of gases it is a spontaneous
process unlike the un-mixing of gases which is not a spontaneous process. So with this I think I
have completed my discussions on entropy. There are several other aspects of entropy which you
need to consider for an example how entropy is a measure of the randomness of the, it gives the
measure of the extent of randomness of a particular system.

How it is related to the thermodynamic probability of the system? How entropy is related to the
available energy of the system but I am not going to go into the details of all those things
because
95
they do not pertain to phase equilibrium thermodynamics you can go through it through any
particular textbook and in case you have doubts on those particular topics we can take them up in
the interactive courses but as far as the assessment of this course or the assignments of this
course are concerned they will just be concentrated on the topics which you are covering in this
particular class.

(Refer Slide Time: 27:02)

So with this we complete the first and second law and what we have found out rather where we
have come up is from the first law of thermodynamics we have got dQ equals to dU plus dW and
the other equation is dS equals to dQ by T for a reversible process. We have also found dS for an
isolated system is greater than equal to 0 which gives the condition of feasibility of a process. So
with this I end this class and we are going to discuss much more interesting results arising from
the equations or arising from the combined first and second law which we have discussed so
long, thank you very much.

96
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 07
Criteria Of Spontaneity

(Refer Slide Time: 0:35)

Well to continue with our discussion regarding the entropy the maximization principle, so in this
particular slide I have noted down the entropy maximum postulates which states that entropy is
one particular function which is a function of different extensive parameters of the system we
can calculate the entropy just like we did in the last class.

You can calculate the entropy as a function of the different extensive parameters and then we
find out the entropy we know that for different values of the extensive parameters we get
different values of entropy and out of the values of entropy that we have got the state which
corresponds to the maximum entropy is the stable state or it is the equilibrium state.

97
(Refer Slide Time: 1:21)

Now if we take a concrete example say for example suppose I have a very rigid insulated
container and in this particular rigid insulated container there are 2 compartments which are
again which are set here and there is a Gas1 here and there is a Gas2 here and this has a internal
energy U1 say V1 and it contains a large number of moles of say N1, N2 and so on and here and
in this particular case what we have is that we have U2, this container 2 it has a volume V2 and
again there are number of moles of the gas say N1 component into N2 component into, here also
it was N1 1 and N21, right?

So here we have N1, this is N2. So in this particular way for both these compartments we find
that there is a Gas1 here there is a Gas2 here and the Gas1 it possesses a total internal energy of
U1 and the total volume of V1 and it has got a number of components say 123 and the total
number of moles are N1 of component 1, N2 of component 2 etc and in the compartment 2 we
have the Gas2 whose total internal energy is U2 and total volume is V2 and it has the
composition or rather the mole numbers of the different components as N21, N22 and so on and
so forth.

Now in this particular situation we find that this whole thing is contained in a rigid impermeable
an adiabatic container, right? And the 2 compartments they are also separated by a partition
which is again rigid, impermeable and adiabatic partition, right? Now this particular composite
system this is an isolated system and each particular compartment that is also an isolated

98
compartment, so now at this particular moment the gas in compartment1 is in a state of internal
equilibrium the gas in compartment two is in a state of internal equilibrium and the entire
composite system is also in a state of internal equilibrium.

Now suppose what I do is, I just try to make this particular partition slightly movable means I
remove the attachments here such that the partition can move freely within the container what is
going to happen? We find that the partition will move some way this way that way and finally
settled down into a new position which is a new equilibrium position of the gases in
compartment 1 and 2. So I have removed one internal constant and allow the partition not to be
rigid but to be movable and we find that both the gases have settled down they can interact with
one another and they have settled down to a new equilibrium state.

Next what I do? I remove the adiabatic covering from the partition such that each can flow from
one compartment to the other. Again I find that in the 2 compartments a good amount of heat
exchange takes place finally thermal equilibrium within the composite system is established and
the gases in the 2 compartments settle down to a new particular temperature.

So one another internal constant got and we attain another internal another equilibrium state.
Next what I do? I make a number of holes in this partition such that gases can diffuse from
compartment 1 and 2 or vice versa. I find that the gases keep on diffusing between the 2
compartments and finally they settle down into a new equilibrium state where more or less there
is uniform composition of the 2 gases in the 2 compartments, So in this process what I have
done?

I have removed internal constants one by one and I have the gases have settled down to a new
equilibrium states, right? And the point is just to find out which particular equilibrium state is
going to be most stable we need to calculate the entropy change of the gases as each of the
internal constants are removed and then we need to find out that particular value of entropy
which is the maximum for any particular combination of the extensive variables U1, V1, N1, N2
etc.

So therefore we find that finally this particular isolated system is going to settle down in a new
internal equilibrium state which can be decided by the entropy maximization principle where the

99
entropy can be calculated as an objective function of the different extensive variables possible, so
this is known as the entropy maximum postulate.

(Refer Slide Time: 7:19)

And in order to calculate entropy I would like to remind you certain properties which entropy has
first thing is entropy is an extensive quantity or it is additive.

(Refer Slide Time: 7:37)

So therefore the total entropy of this particular system S total it is simply a summation of the
entropies of the subsystems. The next important thing is entropy it is a continuous differentiable

100
and monotonically increasing function of energy or in other words entropy it is a first order
function of energy. So therefore what does it imply? It implies the first order function implies
that S, suppose I increase each V extensive parameter by an equal amount of lambda then in that
case my entropy increases by the same amount lambda. So this is first thing we find that for a
system entropy it is a homogeneous first-order function of extensive variables these facts help us
to calculate entropy of the different systems.

(Refer Slide Time: 9:07)

101
(Refer Slide Time: 9:27)

What is the second thing? The second thing is that entropy of a system that entropy is continuous
it is differentiable and it is a monotonically increasing function of energy, what does this imply?
This monotonically increasing function of energy that implies that del S del U for constant V,
N1,N2 etc is greater than 0 and you we will be seeing later that actually del U del S at constant
V, N1, N2 etc that is equal to T which we will be seeing later.

Now from this particular equation we get something very important. We find out that since
entropy it is a monotonically increasing function of energy and since we can express del S del U
in this particular terms and we know that S it is a function of U, V, N1, N2 so on. So similarly we
can write in terms of energy that U is also a function of S, V, N1, N2 etc which implies that
energy is a single valued continuous differentiable function of S, V, N1, N2 and so on, what does
it mean? It means that the entropy maximum principle also reduces to the energy minimum
principle so therefore we find out that there can be systems where the entropy remains constant
but the energy of the new state is the minimum.

And you will be recalling that in mechanics we find that the stable states correspond to the
energy minimization principle I just wanted to show that the entropy maximum principle
corresponds to the energy minimum principle and if we keep the energy to be constant than its
entropy maximum which governs the stable equilibrium state and if we maintain the entropy of

102
the system constant which we usually do in mechanical systems then it is the energy minimum
principle which governs the stable equilibrium of the system.

So these 2 statements they are analogues to one another or in another words the stable state can
be given by either the entropy maximum principle a postulate or it can be given by the energy
minimum postulate and we will be seeing later that there are several other terms also or several
other energy terms whose minimization also can give us the stable equilibrium states.

(Refer Slide Time: 12:48)

And well the last part is, it just tells us that at the 0 of temperature the entropy of the system
vanishes or in other words this gives rise to the third law of thermodynamics. So therefore with
this what I meant to say is that we had derived the equation this was the equation which I had
derived, this works from the combine first and second law we had derived this particular
equation, right?

103
(Refer Slide Time: 13:01)

So therefore the equation was dU was equal to dQ minus dW, right? This was the equation when
only pV work is concerned then dW can be replaced by PdV which has been done in this
particular equation.

From second law what did I get? I got this particular equation but we need to remember that this
equation is applicable only for a reversible process, isn’t it? So therefore let us see that what
happens in this particular case is, so in this particular case what do we see? Well, in this
particular equation then what do we see? We find that when there is this from the first law this is
applicable for closed systems the closed system we had derived this equation for both reversible
as well as irreversible processes, isn’t it?

104
(Refer Slide Time: 14:54)

Next from second law we have derived dQ was equal to TdS this was for a reversible process
only, right? And then if we can replace this then we get dU equals to TdS minus PdV the
equation which I have shown here. Now let us see for which condition this equation is
applicable, this was applicable for a reversible process and a closed system this was applicable
for a closed system and for any particular process.

Now if you observe this particular equation what do we find? We find that this equation gives us
a relationship between internal energy, temperature, entropy pressure, volume all of these are
properties of the system. So if all of these are properties of the system then they all of them
should depend upon the initial and final states they should have nothing to do with the path
which we follow, so therefore finally from these 2 equations we come across a equation which
should be applicable for any process reversible or irreversible provided initial and final states or
conditions are same.

We need to remember that for reversible processes TdS will be equal to dQ for irreversible
processes TdS will not be equal to dQ in the same way PdV will not be equal to dW, so the
difference between dQ and TdS will be equal to the difference between dW and PdV and
accordingly we could derive this equation from this particular equation there are certain things
which I need to mention about this particular equation. The first thing is this is applicable for any
process but we have to remember that, two it is for any process but this is applicable when

105
volume change is the only form of work that is only Pv is only form of work or in other words
when the work done is just of the PdV type of work.

Next is it is applicable for a closed system in a state of internal equilibrium. So therefore this
equation is applicable for a closed system in a state of internal equilibrium when volume change
is the only form of work and more importantly we need to remember this can also be applicable
when any particular system undergoes a internal chemical reaction as a result of which its
composition changes but whatever be the case the composition change or the chemical reaction
just affects the change in volume there is no other change within the system.

So therefore this can be applicable even when the system inside the close system that has some
reaction going through into it but we need to remember that the affect of the reaction will be a
change in volume and there will be no other work other than volume change occurring within
this particular close system which is in a state of internal equilibrium. So under these conditions
we find that the equation is applicable and if we replace this PdV with dpV minus VdP then we
can just write down the internal energy equation in the form of enthalpy. So from the combined
first and second law we have obtained these 2 generalized equations.

(Refer Slide Time: 18:56)

106
And the other thing which we have obtained from the combined first and second law is the
criteria of spontaneity of a process but we need to remember that we could obtain the criteria of
spontaneity just for an isolated system, what does this demand? This demands that in order to
find out whether a process is feasible or not we need to find out the entropy change of the system
and also the entropy change of the surroundings if the system is not isolated.

Here we need to remember that although finding the entropy change of a system is not very
difficult finally the entropy change of the surroundings can often not be very easy and we would
always like to concentrate our attention on the system we would like to calculate any change in
state of the system in terms of the change of its properties this is much more straightforward than
calculating the entropy change of the surroundings.

So therefore what we would like to do is, we would like to concentrate on this system and we
would like to find out whether any criteria of spontaneity can be obtained by just considering the
system without bothering about the surroundings for a closed system. For the time being we
concentrate on systems where only Pv work is being performed, later on we are going to see
what happens when even non-Pv work is involved in this particular case, right?

So before I start I would just like to you remind you something which I said a little time back
what I meant to say is that when I say dS isolated equals to 0 and when I consider this particular
equation what do I get? I get dU equals to TdS minus PdV or in other words what I get is dS
equals to 1 by T dU plus P by T dV, right?

And from isolated system what does it imply? Isolated system implies that system has a constant
internal energy and the system has a constant volume such that the system which means that the
system does not perform any Pv type of work there is no change in internal energy of the system
automatically which implies that the system has no interaction with the surroundings in the form
of heat and work and for such a system we know that dS is greater than equal to 0.

Can this not just be reverted and written? If you observe this equation can I not write down that
dU at constant S, V should be less than equal to 0, does this equation not imply this? It is almost
equivalent to say that a circle is an enclosed surface which has a minimum area for a given
parameter or a maximum perimeter for a given area, both the statements are the same. In the
same way this statement and this statement they are equivalent to one another and we find that

107
for constant entropy constant volume condition dU less than equal to 0 should be the criteria of
spontaneity, right? Where U is the internal energy change of the system, mind it.

Similarly if we take up the other equation which is dH equals to TdS plus VdP from this
equation we know that dS is greater than equal to 0 is the criteria of spontaneity does it not imply
dH at constant S, P is less than equal to 0. So therefore from these 4 equations or rather from the
last 3 equations what do we get?

We find out that for an isolated system dS U,V is greater than equal to 0, from here we
automatically see that if the entropy of the system is kept constant then for this isentropic process
the system has to interact with the surrounding such that the volume is also kept constant for
such a system we can have dU is less than equal to 0 in the similar way that if this entropy and
the pressure is constant then it implies that dH S,P should be less than equal to 0.

What I mean to say is that again I will repeat the same thing entropy maximum principle also
corresponds to the energy minimization principle which you are so familiar with in your
mechanics class. Now as I repeat that we would be more interested to find out the condition of
equilibrium or the condition of spontaneity by just observing the system when the system
undergoes some particular interactions with the surroundings.

(Refer Slide Time: 24:43)

108
The 2 most common type of interactions which we are mostly interested or which we
encountered in our practical applications are firstly system surrounding interacting under
constant temperature volume conditions and second thing the more common work the more
common phenomena is under constant temperature and pressure, now when such a thing happens
let us see that if by looking at the system only we can find out the criteria of spontaneity.

Again we go back to the same equation and let us see dU equals to TdS minus PdV using the
Legendre transform can we write down this equation as, in this particular form? I believe we can
do this, so therefore what does it give you? Does it give you d of U minus TS. Well, I would just
change this a little and I will write that will be better.

(Refer Slide Time: 25:46)

Sorry, if it’s a condition of spontaneity that I want to find out, the condition of spontaneity which
gives you dS is greater than equal to dq by T or TdS is greater than equal to dq which is greater
than equal to dU plus PdV provided only Pv work happens and the system is a closed system.

So therefore for this particular thing using the Legendre transform I can write it down as or in
other words can I not write it down as d of U minus TS is less than equal to minus pdv minus
SdT and what is this U minus TS? This is nothing but the Helmholtz free energy it is nothing but
the Helmholtz free energy which we find that under conditions of constant temperature and
constant volume this is less than equal to 0.

109
So what I have deduced? From the condition of spontaneity for any system we know that the
condition of spontaneity is dS greater than equal to this from here I have deduced that for any
system when the temperature and the volume remains constant the Helmholtz free energy is less
than equal to 0. In the same way suppose we would like to find out the condition of spontaneity
under constant temperature pressure conditions.

(Refer Slide Time: 27:43)

We start from the same thing TdS is greater than equals to dq which is greater than equal to dU
plus pdV we need to remember that we are just working again with closed system only PV work
So therefore this will be greater than equal to dU plus d Pv minus vdp this can be written down
as TdS minus SdT, so therefore what we get? We get U minus TS plus pv this is less than equals
to minus SdT and plus Vdp or in other words as we all know this is nothing but the Gibbs free
energy and this equation implies dG at constant temperature pressure conditions is less than
equal to 0.

110
(Refer Slide Time: 28:46)

(Refer Slide Time: 29:05)

So therefore from the condition of spontaneity we found out that if you really want to find out
the condition of spontaneity when the system and the surroundings are interacting under constant
temperature volume conditions than the actual conditions rather whether the process is
spontaneous or not can be found out by calculating the Helmholtz free energy if it is less than
equal to 0 the process is spontaneous and if the system and the surrounding interacting under
isobaric isothermal conditions we need to calculate the Gibbs free energy. If the total Gibbs free
energy of the system is less than or equal to 0 the process is spontaneous.

111
(Refer Slide Time: 30:08)

So therefore we have once more deduced 2 other important properties the Helmholtz free energy
and the Gibbs free energy and I would like to remind you that since we are mostly dealing with
constant temperature constant pressure conditions this Gibbs free energy is going to be very
important for all phase equilibrium and chemical equilibrium problem and this is going to stay
with us for the remaining portion of our phase equilibrium thermodynamics class. So we
continue with these discussions on Gibbs free energy Helmholtz free energy in our next class,
thank you very much.

112
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 08
Criteria Of Spontaneity (Contd.)

(Refer Slide Time: 0:25)

Hello everybody, so till now we have gone through or rather we have discussed several experts
of the first and second law and based on these 2 laws we found out we have developed these
particular equations in this particular case. So the things which we have the different equations
that we have defined the first equation was this and based on the first law this was the equation,
based on second law this was the equation.

113
(Refer Slide Time: 1:37)

Now if you observe this equations we will found that as a consequence of the first and second
law we have defined 2 new properties from first law defines dU and the second law defines dS
and if you observe closely you will find that both these properties have been defined as
differences and not as absolute values which implies or rather from this we find that all the other
equations that we have suggested for example the different criteria of spontaneity that we have
selected you find that all of these they are defined in the terms of difference quantities.

And let me tell you this is the greatest or rather this makes phase equilibrium thermodynamics
much more difficult, why? Because since we defined all the properties as difference properties,
so as a result of which the specification or the definition of standard states become very difficult
and it is also very important that when you have a big or rather when you have a equation
comprising of different properties it is important to ensure that the standard states for all the
properties are taken or they are the same. So these particular small small aspects as we shall see
while proceeding towards in the succeeding classes we will see, these are the things which made
this subject slightly more difficult and it is very important that a student concentrates on this
aspects when he is dealing with this particular subject.

114
(Refer Slide Time: 2:41)

(Refer Slide Time: 3:16)

So based on the equations we find that these are different criteria of spontaneity for different
interactions between a system and the surrounding, now let me remind you once more that we
have done this only for assuming that we are dealing with just Pv type of work. We all of us we
know for isolated system it's the entropy maximization principle and which gives us the energy
minimization principle and the enthalpy minimization principle for constant S and V and S and
P.

115
(Refer Slide Time: 3:23)

Now before I proceed I would just like to ask you one thing if you look at this particular
equation, what does it show? It shows that for a spontaneous process the energy has to keep on
decreasing, does it mean that the spontaneous process will only proceed when the energy of a
system keeps on decreasing or does it mean that energy of the system has to keep on decreasing
for a spontaneous process and finally the energy will or the system will sync into your 0 energy
sort of a sync?

It is not that, this particular statement is basically a disguised statement about entropy. This
basically means that for a constant entropy process, if a process has to be spontaneous and it is
occurring under constant entropy conditions that the internal energy has to decrease at the
expense of energy moving from the system to the surroundings or in other words it means that if
the entropy of a system is maintained constant then in that case heat has to flow or rather energy
has to flow from the system to the surroundings for a spontaneous process to occur.

So this is in reality a disguised statement about entropy and from these 2 in the last class we had
defined that for constant volume and temperature conditions it is the Helmholtz free energy
which is important for constant pressure and temperature conditions it is the Gibbs free energy
which is important and it is important for you to know that usually in your practical applications
we deal with constant temperature and pressure conditions and therefore Gibbs free energy you

116
will find that is going to be our companion throughout this particular course on phase
equilibrium thermodynamics it is also important for reaction equilibrium.

(Refer Slide Time: 6:04)

Now after this, what do we get? So from there what does it imply? It implies from this particular
equation it implies that more or less we find out that from this particular equation we find out
that suppose we are plotting say G, suppose spontaneous reaction occurs and in that particular
spontaneous reaction we are plotting the Gibbs free energy versus time what do we observe? We
observe that initially the Gibbs free energy it keeps on decreasing, right?

This is the time where the spontaneous reaction is occurring and after a while when equilibrium
has achieved then we find the dG becomes equal to 0 and this signifies the equilibrium condition
for this particular case it was dG less than 0, for this particular case it is dG equal to 0. Same way
if you would like to plot the Helmholtz free energy as a function of time for a constant volume
and constant temperature conditions we find that while this spontaneous reaction is occurring
that Helmholtz free energy keeps on decreasing and finally when equilibrium has achieved dA
becomes equal to 0, right?

So therefore if we observe suppose at constant temperature and pressure Gibbs free energy I
have said this is equal to G minus TS, so at constant temperature suppose we would like to plot
or rather we would like to find the change in Gibbs free energy when the process occurs this can
very well be written down the way I have written it down, Delta G equals to Delta S minus T

117
Delta S. From here what is the first thing that you observe? You observe that enthalpy dominates
at low-temperature and entropy dominates at high-temperature, right?

So therefore when we know that a spontaneous process has to occur for that particular condition
Delta G is less than 0, how can Delta G be less than 0? When then Delta H is less than 0 and
Delta S is greater than 0. We know that for such a situation the process it is a feasible process or
in other words the process is spontaneous.

Again suppose we find that for any particular equation Delta H is greater than 0 and Delta S
greater than 0. So for this particular situation we definitely know that for this situation Delta G
will be greater than 0 as a result this process is not feasible but it will be feasible in the reverse
direction, right?

The problem starts when Delta H and then Delta S they have the same sign which means that
both of them are positive or both of them are negative under that condition the change in Gibbs
free energy represents a compromise of the system’s tendency to decrease enthalpy and increase
entropy under constant temperature and pressure conditions.

(Refer Slide Time: 9:42)

Let us state a definite example in this particular aspect. Suppose we take a definite example here
what do we find? We find that the very typical example of say ammonia and HCl reacting to find
NH4Cl. We know that for this particular reaction the heat of reaction is known to us as well

118
as the Delta sR is known to us both of them are negative, just if you observe the entropy change
of this particular reaction what will you get?

You are going to find that the entropy change it is negative. Just if you are not aware enough it
will immediately appear to you that this reaction should not occur in the forward direction but
the reverse reaction is possible but in order to find this out since this particular reaction occurs
under constant temperature and constant pressure conditions it is important to find out Delta gR
at the same conditions of 1 bar pressure and 298.15 Kelvin it is not difficult you can just find it
out by plotting or rather by calculating Delta hR minus298.15 Calvin into Delta sR.

What is the value that that you get under this condition? We find that the value is minus 91.2kilo
joules per mole under this condition. Immediately from the value of delta gR that this reaction is
spontaneous which is the actual case under normal circumstances. Same way suppose I take
another very common example vaporization of water although this is a physical process I have
denoted it in the form of a in the form of a reaction.

We know the standard enthalpy of vaporization and the entropy of vaporization of water at 1
atmospheric pressure around 100 degrees centigrade, let us calculate Delta g vaporization at 1
atmosphere pressure and exactly at 100 degrees centigrade that is 373 degree Kelvin. Just
calculate it and you are going to find that this will almost be equal to 0, now repeat the
calculation for 1 atmosphere pressure and say 363 degree Kelvin or 90 degrees centigrade, just
calculate it and you are going to find that it is going to be plus 1.10 kilo Joules per mole.

And suppose you calculate this for 383 degree Kelvin what do you find? You find this is about
1.0 this is around minus 1.08 kilo joules per mole. So immediately from these particular values
of delta g vaporization, immediately you have got the answer that why water vaporizes at 100
degrees centigrade and steam condenses at 100 degrees centigrade since Delta g vaporization is
almost 0 at that temperature therefore this process at 100 degrees centigrade is reversible and it is
feasible in both directions.

And this also gives you the answer that why your condensation of steam to water is spontaneous
at 363 degree Kelvin because Delta g vaporization is positive, so naturally the reverse reaction is
spontaneous and at 383 degree Kelvin we find that the reaction it is spontaneous in the forward
direction.

119
So therefore just by calculating the delta g vaporization values we know that the reaction can
occur in both ways for 100 degrees centigrade and this particular direction it is spontaneous for
383 degree Kelvin and this particular reaction is spontaneous for 363 degree Kelvin. We can take
up several other examples as well but anyhow these are the 2 typical examples which I have
shown so that you can appreciate the importance of calculating Delta g vaporization.

(Refer Slide Time: 14:46)

Well, so far as I have already mentioned that we have just considered a closed system with just
Pv type of work this I have been repeatedly mentioning you. What happens if suppose there is
some amount of non-pv work also performed by this particular system any particular closed
system? Let the system be closed but it performs some non-pv type of work. Naturally along
with pdV then what do we have? Along with pdV then we have 1 Fdr term as well, isn’t it?

Where F is the thermodynamic driving force and r is the thermodynamic displacement, under
this condition do the same criteria of spontaneity hold? For constant S and V what was the
criteria of spontaneity? The criteria of spontaneity was dv less than equal to 0, do you expect the
same criteria to hold in this particular case? No, if we just substitute the equations at constant S
and V you find that the criteria dU S,V less than equal to 0 does not hold for constant S and V
conditions.

On the contrary what should happen under this particular condition? Under this particular
condition we find that dU plus dW non-pv this particular term should be less than 0. In the same

120
way for constant S and P what should be the conditions of spontaneity? It should be dH plus dW
non-pv is less than equal to 0 same way if we continue we will find that for constant volume
temperature conditions this is the equation and for constant pressure temperature conditions this
is going to be the equations.

So therefore from here what are the things that that we have found out? We have found out that
when there is some amount of non-pv type of work under that condition, we find that whenever
there is some non-pv type of work under that condition the amount of useful work or the non-PV
work that we can get from a system or that has to be performed on a system when it moves from
one equilibrium state to another is equal to the decrease in the internal energy of the system at
constant entropy volume conditions if the process occurs under constant entropy pressure
conditions then it is a decrease of the enthalpy and usually what we find?

We find the reactions occurring or the processes occurring for constant temperature volume or
constant temperature pressure conditions and under these conditions we find that the amount of
non-pv work that can be obtained from a system is equal to the decrease in the Helmholtz free
energy or the Gibbs free energy of the system. Again I would like to remind you that since we
are mostly concerned and interested with constant temperature and pressure conditions, so
therefore dG T, P is going to be of interest to us, right?

121
(Refer Slide Time: 18:16)

So therefore for non-pv type of work we will find that more or less. Well, so from the last slide
let me just write down this equation once more, what does this equation show? This is the
equation which I have derived in the last line, right? Now from this particular equation if you
observe this equation we can write it two particular ways, the first thing is we can write it down
as dG T, P is greater than equal to minus dW non-PV.

What is minus dW non-PV? If you remember these the sign convention which we are using this
is Q in is positive and the Q out that is negative, work done by the system is positive, work done
on the system is negative, so therefore minus dW non-pv is a work done on the system. So we
find out that the work which has to be done on the system in order to bring or rather to affect
some amount of change in the Gibbs free energy.

It is for a reversible process we will find that dG T, P equals to minus dW non-pv and you will
find that for an irreversible process we find that the useful work which is done on the system is
minimum for the reversible process. So from these 2 things what do we deduce? We deduce that
useful work done on the system it is minimum for reversible process.

And similarly we can also write it down in this particular case if a system has to perform work
on the surroundings then that has to be accompanied by a decrease in the Gibbs free energy we
find that the decrease in the Gibbs free energy is equal to the non-pv work for reversible process

122
and this is less than the non-pv work for the irreversible process. So from here what do we find?
We find that the useful work obtained from system that is maximum for reversible process, right?

So therefore what can be deduced from here? We can write the equation in 2 particular modes
one shows that the decrease in Gibbs free energy is equal to the useful work done by the system
and we find that the maximum work that can be extracted from the system is obtained for
reversible process and whatever useful work we can obtain for the system for an irreversible
process is less than useful work that can be obtained for a reversible process in the same way if
you have perform work on the system then we find that the minimum work that has to be done
on a system in order to effect some particular change in the Gibbs free energy that is minimum
for a reversible process for any particular irreversible process the work which we have to
perform is will be much greater in that particular case.

(Refer Slide Time: 21:54)

Or in a nutshell suppose we would like to tell this is a nutshell, so therefore we can say it as that
the work done by a system is either less than or equal to the decrease in Gibbs free energy for
constant temperature and pressure conditions for example in an exothermic process. And the
work done on the system is either greater than or equal to increase in Gibbs free energy for
constant temperature and pressure for example in an endothermic process where it is important
to remember that this is applicable only to a closed system.

123
Now when we relate these particular terms with the useful work which can be obtained from a
system of the useful work done on this system does this tell you or does this suggest any
particular similarity with any term which you had studied earlier? If you recall you will find that
in mechanics of potential A(r) was defined as A(r2) minus A(r1) and that was equal to the work
needed to go from r1 to r2. If you remember this thing, so therefore in mechanics a potential was
defined as the work needed to go from r1 to r2 and that was equal to the difference A(r2) minus
A(r1).

The decisive point in this particular case was that Ar it did not depend on the particular way it
was chosen to go from 1 to 2 as long as the acting forces were given as the derivative of the
potential with respect to the coordinates if you compare we will find that in this case also U, H,
A and G their differences give us a measure of the amount of work needed from going in going
from state 1 to 2 and they are independent on the particular way that we have chosen in going
from 1 to 2.

So therefore by analogy with mechanics all these 4 terms can be termed as thermodynamic
potentials and they can be related to the work or rather useful work attained or performed on a on
a system for different coupling of the system and surroundings I repeat dU gives you the non-pv
or the amount of useful work that can be obtained under constant entropy volume conditions, H
for constant entropy pressure conditions, A for constant temperature volumes conditions and G
for constant temperature and pressure conditions.

124
(Refer Slide Time: 24:52)

(Refer Slide Time: 25:04)

From here therefore in a nutshell, we can go to the properties of thermodynamic potentials but
before we do this I would just like to present an alternative derivation of decrease in Gibbs free
energy being equal to the non-pv type of work I have done an alternative derivation for both dG
as well as dA. I would like to remind you that since both of them are thermodynamic properties,
so therefore in whatever way we do the derivation we are supposed to arrive at the same results
provided initial and final states are the same.

125
So in this particular case what I have done? I have started from the basic definitions and I have
defined dU from the first law and for and then at constant temperature and pressure I find that dG
is equals to the reversible work which can be performed by a system and minus the PdV work
which is there because Pv work is anyhow in that particular case it is there.

So from here can you tell me that suppose I give you an statement say that suppose a statement is
given to you say for example I state Delta gf the Gibbs free energy of formation of water at say
some particular condition say delta gf water at say 298.15 Kelvin and 1 bar pressure this is equal
to say I tell you it is equal to minus 237.1 kilo joules per mole, can you tell me what this
particular statement means?

If I tell you Delta gf water is equal to at 298.15 Kelvin and 1 bar pressure equals to minus 237.1
kilojoules per mole. Well, this means that suppose we perform some particular reaction say water
hydrogen at gas plus half O2 gas to give you H2O liquid and for this particular condition
hydrogen is at 1 bar pressure 298.15 Kelvin, oxygen is at 1 bar pressure 298.15 Kelvin and water
formed is also at 1 bar pressure and 298.15 Kelvin. So therefore from this particular forward
reaction we can actually get 237.1 kilo joules of energy we can get from here.

It also means that if we have to decompose water into its constituents we have to provide 237.1
kilojoules per mole of energy provided the reaction takes place reversibly, right? So therefore the
non-pv type of work in this particular case was the electrical work which has to be provided or
rather which is obtained in a galvanic cell where hydrogen and oxygen react to form water or
minus or this much amount of energy is necessary to decompose water to obtain its constituent
elements.

126
(Refer Slide Time: 28:42)

So therefore after this in a nutshell we would like to discuss the properties of the thermodynamic
potentials, what are the properties of the thermodynamic potentials? We all of us know that they
all of them are the thermodynamic potentials are thermodynamic properties and all of them are
extensive properties, what does it mean? It means that U should be equal it should be number of
moles into the molar internal energy, same way the total enthalpy of the system is the number of
moles into the molar enthalpy of the system, your total Helmholtz free energy should be equal to
the number of moles into the molar Helmholtz free energy of the system and Gibbs free energy
should be the number of moles times the molar Gibbs free energy of the system and just like the
total properties here also we know that the molar Gibbs free energy is nothing but H minus TS
the same way a would be equal to U minus TS.

Now here I would very categorically like to mention something which you should be
remembering throughout the entire course of phase equilibrium thermodynamics. Since we are
dealing with a large number of properties here, so therefore all total properties will be defined in
terms of capital letters and all molar properties will be defined in terms of small case letters, we
would be maintaining this particular nomenclature throughout our class or rather throughout this
particular course on phase equilibrium thermodynamics.

So therefore coming back to the properties of thermodynamic potentials all of them are extensive
thermodynamic properties they decide the criteria of spontaneity of a process for different

127
coupling between system and surroundings I need not repeat, you decide the criteria for constant
S and V, H for constant S and P, A for constant T and V, A for constant T and V, G for constant T
and P.

The other thing is it gives a measure of the useful work of the non-PV work to be specific
because we have found out that it is W non-PV minus PdV. So therefore it gives the measure of
useful work a system can perform under given conditions of coupling with the surrounding
medium.

(Refer Slide Time: 31:34)

Well, the most important property of the thermodynamic potentials with which we will be
primarily interested is that if you observe each and every of this particular properties if you
observe this property these the different particular equations we will find this was the combined
first and second law that we have already discussed. From here we have come here, just
substituting A equals to U minus TS we get this equation substituting G equals to H minus T is
you get this equation, so for each thermodynamic potential you have got one set of differential
equations number 1.

Number 2 is, we already know that all of them are properties of the system, so therefore they
have exact or rather they are exact differentials, what are the properties of exact differentials? If
you remember suppose d is an exact differential, so therefore in that particular case suppose P is
an exact differential and it is a function of say x, y then in that case I can write down dP is equal

128
to del P del x at constant y dx plus del P del y at constant x dy this can also be written down as
Mdx plus Ndy, isn’t it?

And under that particular condition I can also write down delM del y at constant x is equal to del
N del x at constant y because both of them are equal to del2p del x del y, since P is a property so
therefore it hardly matters which differentiation I perform first and based on this where all these
terms have been differentiated I have simply written down these properties in this particular form
and then I have performed this particular equality and from here we get something which is very
important and we get a set of equations which are extremely important and these equations are
known as the Maxwell’s equation.

Just observe the Maxwell’s equations what do you get? You find that the Maxwell’s equation
they give us the variation of entropy with pressure the variation of entropy with volume with
pressure and these are related to the PVT properties of the different substances. So therefore we
find from the differential equations of the different thermodynamic potentials we come across a
set of Maxwell’s equations and these Maxwell’s equations are very important because they relate
the non-measurable properties with the measurable properties.

So therefore in the next class we are going to continue and find out how Maxwell’s equation can
be used to relate measurable with non-measurable properties? Thank you very much.

129
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 09
Thermodynamic Network

(Refer Slide Time: 0:31)

Hello everybody, so in the last class while discussing the properties of the thermodynamics
potentials we came across or rather we developed a set of Maxwell’s equations which I have
written down on or which have been written down here for your convenience. Now if you
observe this particular set of Maxwell’s equations you will find that these equations they actually
relate the change in entropy with the change in pressure change in volume.

So in other words this basically relates the changes in entropy with changes in pressure volume
and we will also see how they are related to the changes in temperature? And from here we are
going to derive the changes of enthalpy, internal energy etc with changes of pressure volume etc
and this is as an endeavor to relate the measurable and the non-measurable properties. So that we
can estimate the non-measurable properties simply by measurement of the measurable properties.

130
(Refer Slide Time: 1:37)

Now among the measurable properties the common ones as we all know they are pressure,
volume, temperature out of which pressure and temperature are most easily measured and most
easily controlled along with that we also know that beta and Kappa these 2 the coefficient of
isothermal compressibility and the coefficient of thermal expansivity these 2 can also be
measured. Now let us see how we can play with these set of Maxwell’s equation in order to find
out say the change of entropy the isochoric change of entropy with pressure.

What does isochoric change in entropy means? It basically means del s del p at constant v, right?
Now if you observe then this is related to, if you see the first Maxwell’s equation this will be
equal to del v del t at constant s, so therefore we can write it down as say for instance s from
some particular initial state say P2v1 to say v1P1 this will be equal to del s del p at constant v
integral of say from P1 to P2 dP which can also be written down as minus integral P1 to P2 del v
del T at constant s dp, right?

In the same way suppose I tell you to find out the isobaric change in entropy with volume, what
does this signify? This basically refers to del s del v at constant p, right? So therefore we would
like to find out integral of this as a function of volume that means it is from v1 to v2 this is dv,
okay. So therefore this is equal to s say P1v2 minus s P1v1 which will be equal to if we relate it
we find that in this particular case their relevant Maxwell’s equation is the second Maxwell’s
equation, isn’t it?

131
We find out that the second Maxwell’s equation which del s del v at constant P with del P del T
at constant S. So therefore this will be equal to integral v1 to v2 del P del T at constant s dV,
right? In the same way suppose I would like to find out the isochoric change in temperature,
what does that mean? That means del s del T at constant v. Now can you tell me in what way can
I find this particular quantity out?

Can I not write this quantity as del u del T at constant v divided by del u del s at constant v?
What is del u del T at constant v? We have already discussed the specific heat capacity as
constant volume and constant temperature, what is del u del s at constant v? remember the
differential equation du equals to Tds minus PdV from there we find del u del s at constant v is
nothing but equal to the absolute temperature, right?

Same way if you want to find out del s del T at constant p we can simply write it down as del h
del T at constant P divided by del h del s at constant P all these things they come from the cyclic
rules again we know del h del T at constant P is nothing but equal to del q del T at constant P
which was equal to Cp and this term again from the original dh equals to Tds plus VdP we get
this is equals to T. So therefore we find that in this particular way we have been able to relate the
change in entropy with change in pressure with change in volume with change in temperature at
constant volume the change in temperature at constant pressure.

Now before I proceed I would like to remind you that when we were discussing the Maxwell’s
equation or the evolution of Maxwell’s equation from the thermodynamic potentials we had
derived the equation in terms of total quantities and I had used all sorts of capital alphabets in
order to denote vt sorry v, s, g, a, u and h. Now as I told you that all of these are extensive
properties and the same relationships which hold for total properties are also applicable for molar
properties or for specific properties at this moment we are going to deal with molar properties
and therefore since we are dealing with molar properties I have written all this Maxwell’s
equations in terms of lower case alphabets meaning that I have been relating the change in molar
entropy with the change in molar volume, pressure and temperature.

It’s important for you to remember that the same equations that I have derived here can be
derived for total quantities as well the important thing to remember is if you are dealing with
total quantities all the quantities here should be total, if you are dealing with molar quantities all

132
the quantities written down is molar that is the only thing that we need to remember. Well, now
suppose we go for the?

(Refer Slide Time: 7:59)

Suppose in this particular case, now let us see once we have derived or rather once we have
discussed how entropy changes with pressure, volume, temperature etc then we should be in a
position to derive that Tds equation which are known or in other words we should be able to
derive entropy as a function of volume and temperature and entropy as a function of pressure and
temperature.

Let us see how we go about doing this, for in order to derive this in terms of volume and
temperature we just need to write this as ds equals to as a function of temperature and volume, so
this is del s del T at constant vdT plus del s del v at constant T dv in the last slide I have already
derived this as nothing but equal to Cv by T dT plus del p del T at constant vdv the Maxwell’s
equation have been written on the side so that you can refer to them and find out the way I have
derived.

This can also be written down in terms of beta and Kappa in that particular case this becomes
CvdT plus beta by Kappa dv, right? And using the and if in case you would like to find entropy
in terms of temperature and volume then in this particular case again we write it down in terms
of temperature and pressure, sorry I think I have mentioned volume it was pressure. So I can
write it down in terms of temperature and pressure and in this particular case this becomes CP by

133
T dT minus del v del T at constant pdp which can also be written down as Cp by T dT minus v
beta dp, right?

Now usually what we do is, these particular equations they are written in the form of Tds for our
convenience, right? And we usually write them down as Tds equals to CpdT minus T of this or
in other words here also we would like to write it down as CpdT T of this, right? Here also
suppose I would like to express it in terms of your Tds equations then in that case the equation
can be written down as it can be written down in the form of Tds equals to CvdT plus T of beta
by k Kappa, now these 2 equation they not only show us how s varies with temperature pressure
and volume temperature but they can also be used for finding out a large number of other things
as well.

(Refer Slide Time: 11:11)

For example suppose we would like to find out u as a function of T and v, how to proceed in this
particular case? We already know the basic equation, what is the basic equation that we know?
We know that du I would just like to make the pen a little thicker, so that is it is easier for you to
see. We already know du that is equal to Tds minus pdv, this we know. And from the first Tds
equation which I have written down, from this particular Tds equation can I not substitute this
particular Tds and write it down as CvdT plus T del P del T at constant v minus p dp, right?

Or in other words I can also write it down in the terms of beta kappa etc. I can also write it down
as CvdT plus T beta by Kappa minus P sorry this is dv I made a mistake this is going to be dv,

134
right? So therefore by using the 2 Tds equations the first Tds equation I could actually express u
as a function of CvdT in this particular equation, okay.

Now from here let us see some very interesting result that we get, say for example for an ideal
gas what do we get? For an ideal gas, what is del h? Suppose from here sorry not del h what is
del u del v at constant p? We know for an ideal gas Pv equals to RT for one mole since we are
dealing with molar properties, so therefore in this particular case if you substitute del p del T at
constant v then for an ideal gas this particular term it becomes R by v. So RT by v is nothing but
equal to P. So therefore what do we find? The internal energy dependence on volume for an ideal
gas it disappears off and we find that del u del v at constant T that is that becomes equal to 0,
right?

And I would like to remind you that when we were dealing with the first and second laws under
that condition I had told you that for an ideal gas du equals to CvdT. You had assumed that what
I was telling was correct I could not prove that at that time. Now I show you from Maxwell’s
equation why for an ideal gas del u del v at constant T equals to 0. Same in the similar way
suppose I would like to find del u del P at constant T.

How to find this particular term out? We know from the cyclic relationships del u del p at
constant T del P del v at constant T that will be equal to del u del v at constant T isn’t it? From
the cyclic relationships we have already proved that this is equal to 0. We know that this term is
not equal to 0. So therefore what does it imply? It implies del u del p at constant T will be equal
to 0, right? So therefore we have not only deduced du in terms of CvdT but we have also shown
that for an ideal gas u is independent of pressure and volume it is just a function of temperature
and nothing else which is a very well-known equation.

135
(Refer Slide Time: 15:51)

In the same way suppose I would like to derive the equation for, in the same way if I would like
to derive the equation for h(T,P) right? In that particular case what do we get for h T,P suppose I
in the same way in this particular case I would again like to remember that sorry dh equals to
again from the Tds equation, Tds minus vdp, right? Again I substitute the second Tds equation
here the same way that I had done and then from there once I substitute what do I get?

I get dh is nothing but equal to CpdT plus v minus T, delv del T at constant P dp or in other
words this can also be in terms of beta kappa etc. We can write it down as CpdT plus V into 1
minus beta T into dp, right? So therefore and this particular equation is particularly important
because for most of the practical applications we come across enthalpy changes of physical as
well as chemical processes. In this case also suppose I try to find out del h del p at constant T for
an ideal gas what do I get?

Suppose I do it for an ideal gas, so the for an ideal gas let me find out del v del T at constant p
what is del v del T at constant p for an ideal gas? It is nothing but equal to R by P. So therefore
RT by P for one mole of the gas is nothing but the molar value therefore this whole term cancels
out if you are dealing with ideal gas which shows us that del h del P at constant T for an ideal gas
is 0. Just the way we have proceeded for internal energy we can show in this particular case del h
del v at constant T will also be equal to 0.

136
I will just write it down properly, so this is also equal to 0. So therefore for an ideal gas what do
we know? h it is a function of temperature only h ideal we denote ideal gases with a subscript 0.
So therefore from here we can write h0 is nothing but equal to T1 to T2 Cp 0 dT. At this juncture
I would just like to remind you once more something which I had mentioned in the last class
something about nomenclatures. By this time you must already have been aware of the fact that
we are dealing with a large number of properties and if we do not denote the properties
consistently we might get lost in it.

So therefore again I would like to remind the consistent nomenclatures that we are going to
follow in this class all the total properties the total energy, the total entropy, the total enthalpy etc
all of them will be denoted by capital letters and all molar quantities will be denoted by
lowercase letters and you must have noticed from the beginning of the class I am just using
lowercase letters just to emphasize upon you that I am dealing with molar properties only but at
the same time it is important for you to remember that the same equations that I am deriving will
be applicable for total properties as well.

137
(Refer Slide Time: 20:02)

Now I have introduced one more additional nomenclature, what is it? Whenever I am dealing
with ideal gases I am just putting superscript 0 just to differentiate that these properties refer to
ideal gases in contrast to any particular real gas or any particular real substance, right? Now so
by this particular time we have already come across we have deduced the 2 Tds equations from
the 2 Tds equations we have also derived the equations for dh and du.

And we have also derived or rather we have also shown that both enthalpy and internal energy
for an ideal gas both of them they are unique functions of temperature and both of them are
independent of pressure as well as of volume, right? Now before I proceed further I would like to
ask one particular question to you, we have this particular equation, at this juncture I would like
to remind you that for most of the situations we will be dealing with changes in enthalpy and
entropy as we come across changes of these quantities in practical applications.

So primarily I will be dealing with dh, ds quantities but it is important for you to keep in mind
that you can proceed similarly for du as well, right? Suppose I have some particular gas at a
pressure a P1 and a temperature T1. I would like to take it to another state with pressure P2
temperature T2, now how do I propose to proceed? Suppose I would like to plot the changes on a
P-T curve, right? So here what do I have? I have one particular gas at P1 T1, right?

I would like to take it to P2 T2 state, what are the different ways by which I can perform this
particular process? One is I can perform an isobaric process initially, okay where I go from P1

138
T1 to P1 T2 and then I follow an isothermal process. This I can indefinitely do for this particular
process initially I have performed, the initial process was isobaric, so for this particular process I
can write this down as del h del P dp this has to evaluated at T equals to T1 this integration will
be from P1 to P2 plus del h del T at dT this is at constant pressure, this is at P equals to P2 and
this integration is going to be from T1 to T2.

This is one particular thing that I can do, what have I done for this particular part? Initially I have
performed an isothermal compression from P1 to P2 at temperature T1 and then I have
performed an isobaric heating from T1 to T2 at pressure P2, right? On the contrary there is one
other thing that I can in this particular case. I can first perform the isothermal process and then I
can perform the isobaric process very well. Since at both these points I am dealing with
properties which are dependent on the initial and final states as you already know so hardly it
should matter in this case.

For this particular case what is my h2 minus h1 equals to? In this particular case initially I am
going to write T1 to T2 del h del T at constant P equals to P1 into dT plus integral P1 to P2 del h
del P at constant T equals to T2 dp, right? So therefore I can either perform it by this process or I
perform it by this process, normally it should hardly matter, okay but let us see from the practical
point of view which particular process we should be or rather we should be opting for.

Now tell me what is del h del T at P equals to P1? This should be something like this should
refer to Cp at P1 which should be equal to del h del T at P equals to P1 and for this particular
case what do I need? For this particular case I need Cp, I need information on Cp at P2 and in
order to find these 2 properties, what do I need? I just need if you would recollect I simply need
the PVT behavior of the gases and this I need for sorry I have made a mistake in this particular
case this is not here, for this case I will be needing this one is equal to Cp at P2 sorry.

So therefore we find, in order to find this particular integration and this particular integration for
these 2 integrations I just need the PVT behavior of gases, right? I need to find out how h varies
with P at T1, for this case I need to find out how h varies with P at T2, right? But from the
practical point of view we know that Cp data are available for ideal gases only, Cp data are not
available for or rather they are much rarer and much less accurate for higher pressures, right?

139
So therefore usually at low pressure Cp data are available. So therefore from practical
considerations we would neither prefer this particular process nor we would prefer this particular
process, from practical considerations what are we going to prefer? We are going to prefer I will
select a separate colour for your convenience. We shall be preferring, we would like to come
down to pressure P equals to 0 when at P equals to 0 we would like to perform the change in
temperature and then from here we would like to go to P2 T2.

So the process which we would prefer in this particular case, that particular process is we would
prefer to undergo an isothermal expansion of the gas from P1 T1 to P equals to 0. At this P
equals to 0 where the gas behaves as an ideal gas we would like to bring about the change in
temperature and then once the change in temperature is affected at the low-pressure then from
this P equals to 0 we would like to go to P2 T2, right?

So this is the practical path which we are going to adopt. We neither adopt this path nor we adopt
this path also normally just we would have felt that it hardly matters we can take up any path but
normally but under practical considerations we would prefer this path. So for this path suppose I
would like to write h2 minus h1 for this particular path equals to, let us see what should be the h2
minus h1 for this particular path?

140
(Refer Slide Time: 28:11)

For this particular path it is going to be the h2 minus h1 expression that should be equal to it
should comprise of an isothermal expansion. So it is P equals to P1 to P equals to 0, this del h del
p at constant T dp plus at this particular pressure at this p equals to 0 we would like to perform
del h del T at constant P from T1 to T2 at P equals to 0 dT plus from P equals to 0 to P equals to
P2 we would like to perform del h del P at constant T dp, isn’t it?

So therefore this would be at T2, this would be at T1 and we know the value of this now, this is
nothing but equal to integral T1 T2 Cp 0 dT and we all of us know that Cp is a unit function of
dT, normally Cp 0 is given by a plus bT plus c(t) square plus so on and so forth. So therefore we
can find out Cp 0 as a function of T and you can use that data for due to which we had devised
the roundabout path in bringing about the change of this state from P1 T1 to P2 T2.

So now if I complete this particular equation, so what do I get? I find that this will be equal to
this amount this is nothing but equals to v minus T del v del T at constant P this whole thing
integral P equals to P1 to P equals to P almost 0 dp and this particular path that also comprises to
the same term del v del T a constant p dp but in this case this is from p equals to 0 to p equals to
Pt this here we have supposed to evaluate it at T2, here we are supposed to evaluate it at T1,
right?

So therefore for this particular case what we have done actually? This particular case we have
devised a path in order to use Cp0 and other than the data on Cp0 we find that we just need

141
data on the PVT Behavior of gases which tell us how volume is going to vary with temperature
at the constant pressure. Once we have an idea regarding the PVT Behavior of the gases we are
able to calculate these 2 terms and we can find out h2 minus h1 for any change of state from any
initial state T1 P1 to any final state T2 P2. We will continue with our discussions in the next
class on this particular thermodynamic estimation of properties, have a good day.

142
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 10
Thermodynamic network (Contd.)
(Refer Slide Time: 0:39)

Well, so we continue with our discussions on finding out the enthalpy of any particular real gas
when it is going from pressure T rather from condition P1T1 to a condition P2T2. Now I had
already written down the equations in the last class where I had shown that I may need data of
Cp0 at the ideal gas this particular data this is for P equals to 0 and for other than that what I
need is? I need the 2 particular relationships of del h del P at constant T, this particular
relationship I need for 2 temperature conditions, one is at the lower temperature the other one is
at the higher-temperature.

And we found out that if we can just evaluate these 2 and find out Cp0 then we shall be in a
position to find out h2 minus h1. The same is applicable for s2 minus s1 as well. Now just look
at the equations very carefully, when you look at the equations what do you find? What is this
term? This particular term is nothing but the change of temperature of an ideal gas at from T1 to
T2. What is this? This term basically arises if you think logically had the gas been ideal would
we have this particular term or this particular term?

143
This term and this term would not have been there if the gas would have been ideal, suppose for
an ideal gas say for instance for an ideal gas if I write h20 minus h10 then only this term would
have been there Cp0 dT from T1 to T2 where h10 is at T1 P1 h20 was at T2 P2, right? So
therefore we find that these 2 terms they arise just because the gas is not ideal the gas is real. So
therefore can we not see we try to find out h of T2P2 the change of state or the change of h when
a gas travels from T1P1 to T2P2, now we can take it by several processes. We have already
discussed 3 processes and selected one among the 3 just because it was convenient or the data
was available on by that particular process.

(Refer Slide Time: 3:04)

Now let us see if we can go for a Fourth process as well. Well, what is the fourth process if we
see? It is the fictitious path let me tell you but it hardly matters because the initial state and the
final states are actual and we would like to go from the initial state to the final state.

From this initial to this final we had already devised 3 parts earlier, one was an isobaric process
followed by an isothermal the other one was an isothermal followed by an isobaric and the third
one which you had adopted was an isothermal expansion of the gas to P equals to 0 than i it was
an isobaric heating of the ideal gas and then it was an isothermal compression to the real gas
state to T2P2.

Let us see this fourth particular state what is this? You have a real gas at T1P1, say for example
by some means we can just convert this real gas to the ideal gas at the same conditions of

144
temperature and pressure. Once we have done this then after this we take this ideal gas from
T1P1 to T2P2. Once we are taking this we know what should be the enthalpy changes for each of
these and at T2P2 I change the ideal gas to the real gas condition, right?

Now in the entire process we find that the only hitch in the process is to find out the enthalpy
change involved when the real gas is changed to the ideal gas conditions at T1P1 and when the
ideal gas is changed to the real gas conditions at T2P2. Now this particular enthalpy change is
nothing but the enthalpy change associated with the gas because it is real.

So therefore this can be written down as a different function between the actual enthalpy of the
gas and the enthalpy of the gas had it been ideal. These particular terms which I have used here
and here these are basically they are known as the residual properties this is known as the
residual enthalpy or they are also known as the departure functions, why? Because they show the
departure of the property from its ideal gas value and if you have to define the residual properties
or the departure functions we need to specify one or the value at one particular limiting
condition.

The limiting condition for each case is that, for example for Delta h star it was that Delta h tends
to 0 as P tends to 0. So therefore when we try to find out Delta h star at any condition other than
P tends to 0 then in that case it is quite natural that del of delta h star versus delta P at constant T
this is nothing but del of del h del P at constant T minus del h 0 del P at constant T and we know
that this term is equal to 0. So therefore what do we find? We find that the variation of the
departure function with pressure is nothing but the variation of the actual function with pressure
and we already know what this particular function is.

145
(Refer Slide Time: 6:51)

This particular function as we have already mentioned this is nothing but, so therefore del delta h
star delta P at constant T since it is nothing but equal to del h del P at constant T. So therefore
this is nothing but equal to V minus T del v del T at constant P and nothing else, isn’t it? In the
similar way we can also write down suppose we would like to find the departure function for
entropy in this case also it is nothing but the difference between this delta s star is nothing but the
difference between the actual value and the ideal value in this case we need to remember that the
entropy of an ideal gas it is not 0.

It has a nonzero value if you remember this nonzero value this is nothing but equal to R by P dp,
right? So therefore for this particular case also we can write it down in the form as del s or in
other words I will write the final expression I leave it to you for that deriving, this will be equal
to P equals to 0 to P del v del T at constant P minus R by P dp this is equal to delta s star at T and
P, right?

Therefore delta s star at T and P, this is nothing but equal to this particular expression, so
therefore from here what do we find? We find that the entire botheration of finding out the
properties they simply boil down to finding out the departure functions. If the departure
functions can be found out then and the departure functions can be found out once we know how
those particular properties vary with pressure at a constant temperature?

146
In this particular way we can define all sorts of departure properties, for example we can define
delta g star as g minus g0. We can define the residual volume as v minus v0 and what is v0? All
of us know it is RT by P, so therefore if we substitute this as v minus RT by P we find that the
residual volume is nothing it can be expressed in terms of compressibility factor where this
particular z it is nothing but the deviation of the actual volume from the ideal gas volume.

So therefore this is nothing but Pv by RT which is equal to 1 for an ideal gas and it has got some
value greater than 1 for a real gas. So in this particular way what have we done?

(Refer Slide Time: 10:21)

We have first defined the total properties then we have defined the molar properties than we have
define the residual properties and again I would like to remind you, we can have two sets of
residual properties, total properties and molar properties and whatever relationships that we have
defined for the total properties are also applicable for residual properties, for example we can
very well write down that Delta u star equals to T Delta s star, sorry let me write down in terms
of du star is equal to d T delta the ds star minus Pdv star.

Just I will write it down properly once more d of delta u star equals to T of delta s start minus Pd
of delta v star the same way we can write down delta g star equals to Delta h star minus T Delta
s star we can write down Delta a star and equals nothing but and we can continue like this it was
just to show that whatever equations we have derived till now they are applicable for total

147
properties, they are applicable for molar properties, they are applicable for total residual
properties, they are also applicable for molar residual properties as well.

(Refer Slide Time: 11:31)

Well, after this the next thing which I would like to, so from here there is one particular thing that
we could very much appreciate what was the thing? The thing is from here we could appreciate
the true worth of ideal gas equations because we found out that from here that the ideal gas
equations they can be used as a convenient base for calculation of real gas properties that is quite
evident from here

The next thing is residual properties you must remember that they are particularly advantageous
for the gases although they can be valid for condensed phases as well as for the gases but they
are particularly advantageous for gases, why? Because for gases we find that the corrections are
usually small to all the major terms, right? So the corrections to the ideal gas value since they are
small for real gases so therefore residual properties are particularly advantageous for gases. For
liquids we find that the residual properties they must include large enthalpy and entropy changes
of vaporization, so therefore they are not very much useful for those particular conditions.

148
(Refer Slide Time: 12:54)

Now let us try to derive some other types of equations we have already derived some equations
for example we have shown you how u varies with T and v, we have also shown you how h
varies with T and P and we have also found out some very important rather we have also found
out some very important things which you had already known we have found out the, that for an
ideal gas h and u both of them are unique functions of temperature pressure etc.

(Refer Slide Time: 13:29)

149
Now I would like to derive certain other things again from the basic Tds equations that we have
found out, now from the basic Tds equations which are already written down here if you observe
the 2 basic Tds equations we find we can find out some very important things, for example
suppose I would like to find out Cp minus Cv equals to from this particular equation, what do I
get in this particular case?

(Refer Slide Time: 13:45)

I get Cp minus Cv dT this is equal to T del p del T at constant v dv, just see that, no matter how
much I am writing? I am not going to use uppercase alphabets showing that every time I am
using or rather every time I am dealing with molar properties. In the same particular way can I
not write it down as dT equals to T del P del T at constant v dp by Cp minus Cv plus again this
particular term by Cp minus Cv.

Just see what I have done in this equation I have expressed T as a function of V and p that is only
thing that I have done and I know that all of these are properties, so all of these are exact
differentials, right? So therefore by this equation should be equivalent to the equation del T del v
at constant pdv plus del T del p at constant vdp, isn’t it?

So therefore this particular term and this particular term the coefficient of dv these 2 should be
equal again this particular term and this particular term these 2 should be equal, isn’t it? So in
that particular case I should be able to write del T del v at constant P this is equal to T del p del T
at constant v by Cp minus Cv or in other words from here I should be in a position to write down

150
Cp minus Cv, what is Cp minus Cv in this particular case you tell me? This is nothing but equal
to T del P del T at constant v, del v del T at constant P.

(Refer Slide Time: 15:47)

Now let us adopt the cyclic rule, what was the cyclic rule if you remember? Cyclic rule was del P
del T at constant v. Just that the normal cyclic tool del v del p at constant T equals to minus 1.

(Refer Slide Time: 16:03)

So therefore using this particular equation we should be in a position of substituting this

particular equation or rather this particular term, if you substitute this term then what do we get?

151
(Refer Slide Time: 16:33)

We get Cp minus Cv this will be equal to minus T del P del v at constant T, del v del T at
constant P whole square. Just look at this particular equation, this equation I should say this
equation is extremely important.

(Refer Slide Time: 16:52)

Let us see what is the importance of this particular equation? If you observe this equation what
are the important points that we see? The first thing which we see is, that if you observe this term
there is a minus here, right? And this term always has to be positive this particular term, okay.

152
(Refer Slide Time: 17:44)

We will continue it here, so therefore if you observe here what do we find? We find that this term
is always positive and this term it will be always negative, isn’t it? So therefore this whole
particular expression on the right-hand side this will always be positive as a result of which Cp is
always greater than Cv which you have been knowing so long. So this is the first observation that
we get and this is evident because Cp includes the flow work or the work of expansion on or
compression which Cv does not include.

What other useful things does it show? Let us see what are the other useful things that we can see
from here? The other useful things are suppose at T equals to 0 degrees Kelvin, Cp becomes
equal to Cv. Can you tell me under what other conditions Cp can become equal to Cv? Definitely
at T equals to 0 Kelvin Cp can become equal to Cv there is one other condition when del v del T
at constant P this becomes equal to 0.

When does this happen? This happens for a condition where v equals to v min which
automatically implies rho equals to rho max. Or in other words can this not happen for water at 4
degrees centigrade where we know that water has got the maximum density, under that particular
condition Cp equals to Cv and therefore we take a constant value of the specific heat capacity of
water usually assuming it to be at 4 degrees centigrade and again if we substitute Pv equals to RT
and we try to find out del p del v at constant T and del v del T at constant P then for an ideal gas
what do we get?

153
You can substitute it and you can find it out, for an ideal gas this reduces to Cp minus Cv equals
to R again something which you have already known so long, right? And I leave it as an exercise
to you to express Cp minus Cv in terms of beta and Kappa. So this was one very interesting thing
that we have observed.

(Refer Slide Time: 20:04)

Let us see there is certain other interesting things also which we can find out by using the Tds
equations in this particular case. What can we do? Initially we have done Cp minus Cv, let us see
what is Cp by Cv in this particular case? Just divide one equation by the other under constant
entropic conditions; what do you get for that that particular condition?

154
(Refer Slide Time: 20:34)

Just divide and tell me what you are going to get in this condition? So at constant entropy
conditions we can write down CpdT at constant entropy this is nothing but equal to minus T del
P del T at constant v dv at constant s, sorry this was Cv I wrote. CpdT again at constant S, can
we not write it down as T del v del T at constant p dp s. Now if I divide 1 by the other under
constant entropic conditions, do I get something like minus del v del T at constant P, del P del v
at constant s considering these 2 terms divided by del P del T at constant v, right?

So switch again if we use the cyclic rule we can derive it and we find that this is nothing but
equal to del v del p at constant T divided by del v del p at constant s. So what is it? It is basically
the isothermal compressibility divided by the adiabatic compressibility and we know that this,
what is this adiabatic compressibility? This is nothing but the exponent of an isentropic process.
Let us see what is the exponent of an isentropic process if we take down the basic equation dh
equals to this again we take down the basic equation du which is equal to Tds plus Pdv again for
constant entropic conditions we can write down del h del u at constant s this is nothing but equal
to minus v by P, del P del V at constant s this is the term which we need here.

155
(Refer Slide Time: 22:58)

And we know what is del u del h at constant s, this particular term is nothing but the exponent of
an isentropic process, so therefore from this what do I get? From this I get gamma is nothing but
equal to minus v by p, del P del v at constant s, right? Now this has got some 2 very important
implications the first implication is that for an ideal gas suppose we substitute it in this particular
form.

(Refer Slide Time: 23:12)

This can be written down suppose from this particular equation Cp by Cv.

156
(Refer Slide Time: 23:23)

So therefore for an ideal gas I can write it down as del p del v at constant T which reduces to Cp
by Cv for an ideal gas and you already know that for an ideal gas Cp by Cv equals to gamma this
is one important implication of this equation.What is the other thing? Can we not write it down
as del ln P del ln V at constant s which tells you the equation of an isentrope which is Pv to the
power gamma equals to constant, right? So therefore there are certain other things also that we
can derive from this, for example say the thermodynamic velocity of sound, how do we define
the thermodynamic velocity of sound let us see? This is equal to del p del rho at constant s,
right?

Now how in terms of specific volume we can write this down? In terms of specific volume what
do we get? This is del P del v at constant s into v square, right? Or in other words can we not
write this del p del v at constant s.

157
(Refer Slide Time: 24:41)

We have already found out its expression del p del v at constant s can be expressed in terms of
gamma and can be expressed in these terms.

(Refer Slide Time: 24:50)

So therefore what is this term equals to? This term is nothing but equal to P gamma minus v
gamma by v. If we substitute instead of this minus P gamma by v then what is the
thermodynamic velocity of sound equal to? This is nothing but equal to root Pv gamma for an

158
ideal gas we can write it down as, a equals to root gamma RT another equation which you
already know.

So therefore we find that we can come across a large number of interesting derivations by using
the Maxwell’s equations. I will be not going into great details through all the derivations what I
would just suggest is, you can find out the effect of heating or the effect of cooling when you
would like to cool or rather by expansion by using an isentropic process, isenthalpic process
keeping constant internal energy etc.

(Refer Slide Time: 26:00)

What I mean to say is, it would be interesting if you can find out del T del P at constant s, del T
del P at constant h, del T del P at constant u and you can compare which particular process is
more effective in affecting cooling by expansion. So therefore there are lot of other things that
you can do as well but I would not be going into the details there is just one thing which I would
like to discuss before I end this class on the relationships between measurable and non-
measurable properties.

Initially I had already told you that usually Cp values they are available at low pressures then the
gas behaves ideally. Suppose we would really need Cp values at higher pressures, would we then
stop the computer shell or stop the process or in other words we should have some particular
equation to deduce Cp at higher values from the ideal gas values. we are going to get that

159
equation rather you are going to get some sort of that relationship if you can predict the property
del Cp del P at constant T, what is this particular property let us see?

Can we not write it down in this particular way? Now we know that h it is a, this is at constant T
we know h again it is a property. So therefore we can also write it down as del del T of del h del
P, we can just invert the order of differentiation and we know what is del h del p at constant T,
what is it? So this is nothing but equal to del del T of v minus T del v del T at constant P this is
also at constant P and then if you expand this particular differential what do we get?

We get this del v del T, del v del T cancels out we get minus T del 2 v del T square at constant P.
So therefore Cp at any particular pressure temperature this is nothing but Cp0 at P equals to 0 T
minus T integral P equals to 0 to that particular pressure P, del 2 v del T square at constant P dp.

(Refer Slide Time: 28:35)

In a similar way we can also find out del Cv del v and try to find out Cv at conditions other than
the ideal gas condition.

160
(Refer Slide Time: 28:41)

I would not be elaborating further here I stop here on these particular derivations but I would just
like to impress upon you that if you are aware of the PvT behavior of the gases than using the
PvT behavior of the gases we will be in a position to find out all particular non-measurable
properties in terms of the measurable properties and we can quantify them and we can find out
what are the energy changes associated with the process? Whether the process will be feasible or
not by calculating the entropy change and so on and so forth.

So therefore the next class will be on how to or rather we will be on the discussion of the PvT
behavior of gases, there is nothing to discuss about the PvT behavior of ideal gases we will be
taking up the PvT behavior of real gases, we will be seeing the most of equations that are
available. We will be discussing the equations in brief so that once you are aware of those
equations we will be in a position to use those equations and find out changes in enthalpy
entropy etc by using the PvT behavior of the gases, have a good day

161
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 11
Tutorial
Hello everybody, so by this time what are the things that we have covered? We have covered the
introduction part the first Law and the second law. Before we proceed further I would like to do a
few problems on these particular topics and then proceed and the problems which I will be
dealing with today, they will be primarily confined to the things which we have already covered
and I would also like to mention that the assignments along with these problems will be
consulted while preparing the question papers for your final examination.

(Refer Slide Time: 1:03)

162
(Refer Slide Time: 1:12)

(Refer Slide Time: 1:27)

Now let us go to the first problem here, let us see what the first problem is? The first problem is
that, well, there are 3 properties of the system, right? One property is, it can be PVT or anything
first property say is z and we have expressed z as a function of 2 other properties x and y just to
keep matters general I have mentioned these at z, x and y they can be PVT they can be
something, now the thing is that we know that x and y are properties of the system, right?

163
(Refer Slide Time: 2:08)

And this is the equation of state of the system, the equation of state which relates z with x and y
and there is we have mentioned one particular equation of state and what is it given? It is given
that whether this particular function form is fine if all the 3 parameters x, y and z have to be
properties of the system, how do you go about it? We know that an equation of state relates the
different properties of the system.

(Refer Slide Time: 2:34)

164
How do we define properties? If you recall one of my slides you will see that we had defined
properties of the system as the characteristic of a system which is associated with energy and its
transformation and we had mentioned a few features of the system, what is it? Firstly it should
have a definite value when the system is in a particular state it should we determinable
irrespective of how the system is brought to that particular state.

(Refer Slide Time: 3:11)

Or in other words it is independent of the path followed in moving from one state to the other, it
depends upon the initial state and the final state. So therefore suppose I write down dz equals to
say it is a function of x and y. So it should be del z del x at constant y dx plus del x del y at
constant x dy, right? Now where this should be an exact, since it is an exact differential we can
always write it down as M dx plus N dy.

Where M is del z del x at constant y, N is del z del y at constant x. Now suppose we double
differentiate this equation, so in that case suppose we are differentiating it in the form del del y
of del z del x at constant y this should be equal to del del x of del z del y at constant x this is at
constant y this is at constant x. It should hardly matter whether we are differentiating z with x
first or with y first or in other words both the things del 2 z del y del x should be equal to del 2 z
del x del y, isn’t it?

If this has to hold true only then we can say that z is a property of the system, now for this to
hold to, what does it imply? It implies that del M del y at constant x should be equal to del N del

165
x at constant y. So what do we need to know? The equations which have been proposed in this
particular case, the equation is xydx plus y square dy.

So therefore this is nothing but M we have denoted this as M this as N. So what do we require to
do in order to find out whether z is a property or z to be a property dz has to be an exact
differential, if dz is an exact differential then the order of differentiating z with respect to x and y
is not going to matter del 2 z del y del x should be equal to del 2 z del x del y and this
automatically implies that del M del y at constant x should be equal to del N del x at constant y.

So for the equation of state which is provided, we just need to verify this particular thing. Once
we verify this we find that this holds true then we can always say that this particular equation of
state is thermodynamically consistent. So let us see what is del z del N del x at constant y under
this particular condition? Here we find that del M del y at constant x this is nothing but equal to
X, right? And what is Del N del x at constant y this is equal to y square.

So therefore we find that this is not equal to this and therefore this is not a proper equation of
state or this is not a proper functional form between x, y and z in order to ensure that z is a
property. So therefore what is the question? What will be a proper functional form? Now in order
to get a proper functional form we need to suggest corrections either in M or in N or in both such
that this equation is valid.

Now this can be done in a number of ways for example suppose I suggest or suppose we propose
that M is equal to say xy cube, N is equal to x square y square I have changed both of them. Now
for this particular situation what is del M del x del y at constant X let us see? This is nothing but
equals to 2xysquare, let me write it in a different pen, if we propose M equals to xy cube N
equals to x square y square.

For that case we find del M del y at constant x is going to be 2xy square similarly del N del x at
constant y is also going to be 2xysquare. So therefore this can be one particular equation of state
relating x, y and z we have obtained this by modifying both the terms.

166
(Refer Slide Time: 8:10)

So therefore a correct equation of state in this case can be xy cube dx plus x square y square dy
but please remember this is not the only equation of state that we can use. We can use some other
equation of state as well, for example suppose we suggest that dz equals to y cube by 3 plus y
square dy, what do we get in this particular case? We get del M del Y at constant x is equals to y
square del N del x at constant y equals to y square. So therefore this can also be a equation of
state where we have just modified one term and kept the other term, this term is kept constant
only this term has been changed.

In the same way, we can also write it down as say xy dx plus x square by 2 dy. So this can also
be an equation of state, right? So therefore we will find that for any particular property it is
important to remember that the differential has to be an exact differential and if you recall I had
repeatedly requested you in order to differentiate between exact and inexact differentials. Exact
differentials I had shown by d, inexact differentials by a cut d dQ, dW in this way, du in this way
you can very well understand the difference while this depends upon the path this is just a state
function. So therefore it’s immutual how we perform this u, it is just important that initial and
final states are the same.

167
(Refer Slide Time: 9:58)

And the equation of state that cannot be anyone equation of state connecting the 3 variables this
can be one equation of state, this can be one equation of state where I have just changed one
parameter and along with that this can also be an equation of state, this can also be in equation of
state. So therefore this is one thing which I wanted to emphasize upon that how you are going to
find out whether any equation of state is thermodynamically consistent or not?

(Refer Slide Time: 10:20)

168
Now let us go to the second problem, what is the second problem? We had expressed internal
energy or rather to be more specific it’s wrongly written it is the specific internal energy, I am
sorry about it. It should be the specific internal energy. So therefore we have a equation relating
the specific internal energy with the temperature which automatically suggests us number of
things. The first thing is internal energy is a function of temperature only in this particular case
and the equation of state for that particular fluid.

(Refer Slide Time: 11:36)

Mind it, it’s not specified that it is a gas or a liquid, it is just a fluid, the equation of the state for
that particular fluid is also suggested. If you find equation of state it is something of Pv equals to
a constant into T, right? But the constant is not equal to R in this case. So therefore it possibly is
not an ideal gas but it can be represented by an equation of state for a gas and we also observed
that as it seems it is Pv equals to say C (t) or C1T, so therefore since it resembles the equation of
state of an gas that is why u has been expressed in this case as a function of temperature only.

Now what are you required to find out here? You are first required to find out Cp and Cv for this
particular gas or for this particular fluid and then you are supposed to compute some change of
state for this particular fluid from some initial state to some final state, what is the initial state?
The pressure of the initial state is given it is 1MPa and the temperature of the initial state is given
the final pressure P2 is also given and what are you required to find out?

169
In going from this state to this state you are required to find out the work done and the energy
change in this particular process it's delta E but for this case delta E will be equal to delta u. So
therefore and since it's given that it is an adiabatic process we know for this particular process Q
equals to 0 and therefore we know from first law W will be equal to delta U or delta E in this
particular case. So therefore we find that the problem reduces to finding either W or delta U,
right?

Now there is a small thing which you need to consider Delta U equals to Q minus W from our
sign convention, if you follow our sign convention then in that case your W will be equal to
minus of delta U or Delta U will be equal to minus of W, this you need to keep into
consideration. So for this particular change in state the T2 is not given, so you can compute
either the Delta U or you can compute the work done if you find one, you need to find the other.

(Refer Slide Time: 13:57)

In order to compute this particular change in state what is given? You have the equation of state
and you have a relationship relating the specific internal energy with the temperature, you have
these 2 things in mind. If you observe the equation of state you find that t is in degree plus
centigrade which automatically implies that this is nothing but the absolute temperature of the
system. Now we can do the first part first or even in the first part is not giving you will find that
unless you find out Cp or Cv one of them, it is difficult for you to compute the second part.

(Refer Slide Time: 14:47)

170
So let us first let us concentrate on the first part and then we will see how the information we
have gathered from the first part helps us to find out or solve the second part of the problem,
what is given here? Let me write it down it’s given u equals to 178 plus 0 .718t degree centigrade
we need to find out Cv in this particular case. By definition what is the specific heat at constant
volume?

We know that for most of the cases Cp will not be equal to Cv, if we knew that it was an ideal
gas we could have found out Cv and we could have computed Cp as Cv plus R, in this case it is
difficult to do because we do not know or rather this equation is not an ideal gas equation as is
quite evident. So therefore in this particular case we cannot use this particular, in this case this
equation is not going to be applicable. How to find out Cv?

Cv by definition is del u del t at constant v, isnt it? So therefore if we differentiate this equation
with respect to t, what do we get? We get 0.718 kilojoules per kg degree Kelvin whether it is
going to be kg or moles or whether it is going to be joules or calories that all depends upon the
units in which u and t has been specified.

171
(Refer Slide Time: 16:20)

In this particular case if you observe we find that the specific internal energy has been specified in
terms of this kilo joules per kg t is in degree centigrade, so the unit of Cv is as given.

(Refer Slide Time: 16:30)

How to find out Cp? What is Cp by definition? Del h del T at constant P, what is this h? This is
nothing but equals to u plus pv, right? What is, if from the problem what is u? u is nothing but
equals to 178 plus 0.718t. What is pv equals to? This is going to be 0.267 into t plus 273, so
therefore from this equation can you not express h in terms of temperature? What is the
equation?

172
I will just write down here in this particular equation we will find h is nothing but equals to
250.891 plus 0.985t, isn’t it?

So therefore from here we can find out that h if you differentiate h with respect to t, what do you
get? You get the Cp value.

(Refer Slide Time: 17:51)

So in this particular case what is the Cp value that you are getting? Cp which is nothing but equal
to del h del t at constant P, please remember small t is for degree centigrade capital T is for
degree Kelvin, from here we get this to be 0.985 kilo joules per kg Kelvin there is another
interesting thing that I would like to mention in this particular case.

We notice that in this particular case what is Cp minus Cv equals to? Cp minus Cv if you subtract
one from the other what do you get? You get this is equal to, sorry your Cv was equal to 0.718,
so this is 7, this is 6, this is 2 it is 0.267, what is this 0.267?

173
(Refer Slide Time: 18:54)

Just refer to this problem you will find this is the constant C1T which would have corresponded
to R for an ideal gas equation. So therefore what does it imply?

(Refer Slide Time: 19:06)

174
(Refer Slide Time: 19:37)

It implies that if you have a equation of Pv equals to C1T, then in this particular case Cp minus
Cv will always be equal to the constant which is C1, for an ideal gas this constant C1 reduces to
R, right? But here the very common tendency which I find among students is they just see you
have to find out Cp, Cv. u is given, so they find out Cv and then they simply add it up with R and
try to find out Cp but did not notice that it is applicable only for an ideal gas and for no other
situation.

(Refer Slide Time: 20:04)

175
Now let us come up to the next part of the problem, now for this part of the problem there are
initial conditions given final condition is given the process is specified, what is the process? The
process occurs according to this particular, this particular equation, right? And what is this
equation? This equation as you know this equation is for a polytropic process, isn’t it? It is
already mentioned it is adiabatic, so it is an adiabatic polytropic process that we are going to deal
with this, fine.

So for this process if you know you know P1v1 to the power 1.2 should be equals to P2v2 to the
power 1.2 the initial conditions are given you need to find out the final conditions but if you wish
to use this equation you will find the equation is in terms of Pv whereas initial conditions are
given in terms of P and T which is quite understandable as I have already mentioned that
pressure and temperature are the easily measurable and more accurately measurable variables.

So for that what do you need to do? You simply need to convert or express v in terms of t and for
that you need to use this particular equation.

(Refer Slide Time: 23:07)

So therefore how do we go about this then? We know that Pv equals to 0.267 into capital T say,
okay. I just make it first for convenience. So P1 0.267T1 by P1 to the power 1.2 should be equals
to P2 0.267 by P2 into T2 whole to the power 1.2, so by this equation I could actually relate the
pressure temperature of state one to the pressure temperature of state two. I can solve this out

176
and I can find out the temperature t it comes out to be 49.25 the temperature in degree centigrade
comes out to be 49.25 centigrade.

So once I know and I can always compute delta u for this particular case, what is it? It is mCvdT,
right? Integral of this or in other words since Cv remains constant. So it is T2 minus T1, m you
already know it's given 2kgs, Cv you have already found out from the previous problem it is
0.718, you know T2 by this time you know T1 by this time, so therefore it is 49.25 minus 200
which gives you minus 216.48 kilojoules.

As a result you know W it is nothing but equals to minus of delta u which is nothing but equals
to 216.48 kilojoules. Now this is one particular process of doing it, you can also do it by first
computing W and then computing u that can also be done.

(Refer Slide Time: 23:17)

177
(Refer Slide Time: 23:30)

You know that the process is an adiabatic polytropic process, now if you recall that when we
were discussing these particular things I had already discussed the PvT work in various quasi-
static processes. I have discussed PvT Work for an isobaric, isochoric, isothermal and then
adiabatic process, in fact this is specifically not an adiabatic process this is a polytropic process it
becomes adiabatic when this n becomes equal to gamma, right?

178
(Refer Slide Time: 24:12)

So can we not use this particular equation here in order to find out, if we can use this particular
equation here, this particular equation, this is the equation so can we not use this particular
equation in order to find out W. We can do it, right? But again in order to use this equation you
know P1, you do not know V1.

(Refer Slide Time: 24:39)

So you again have to express V1 in terms of temperature in terms of temperature, where V1 in

this particular case I have already written down your V1 is nothing but this part, isn’t it? So

179
therefore you can substitute V1, n is nothing but 1.2, P2 and P1 both are known to you so
therefore you can find out W in this case and equate this W with Delta u.

(Refer Slide Time: 25:01)

I give it as an exercise to find out W independently from this equation and compare the value
with the W which you have obtained by equating this with Delta u. Just to show that once the
path is specified no matter how you do the computation? The work done will be the same for
both the cases.

180
(Refer Slide Time: 25:25)

Now we move on to the third problem, let us see what the third problem is? This is also another
problem in order where we perform some particular change of state and from that particular
change of state we try to find out the heat interactions and with that one additional thing also we
would like to find out in this particular case the entropy change, right? What is the process that
we have devised? This system is an ideal gas and for an ideal gas what is the process? It is a
reversible isothermal expansion.

So if it is a reversible isothermal expansion, what does it imply? It implies that the change of
state it will be from P1V1 temp T to say P2V2 T, isn’t it? And for this particular change of state
since it is an ideal gas we know that since T remains constant for an isothermal process du is
equal to 0. So therefore for this process we know dQ is going to be equal to dW. So therefore to
find heat interactions we can find out the work done, from the work done can find out the heat
interactions.

181
(Refer Slide Time: 26:57)

For an isothermal process again if we go back to the equations which I had shown for an
isothermal process Pv equals to constant and this is the expression for the work done, isn’t it? So
therefore in this particular case P1V1 is nothing but equals to RT1. So therefore this can also be
written down as RT1 ln P1 by P2, agreed? So therefore we can find out the work done and this
will be equal to the heat interactions under this particular process.

(Refer Slide Time: 27:37)

182
So accordingly in order to find out the heat interactions for reversible isothermal expansion ideal
gas you have already got the expressions we had already derived the expressions.

(Refer Slide Time: 27:51)

So for this particular case your dW reversible was equal to as I have shown you it was nRT1 ln
P1 by P2 or else you can also write it down as n RT1 ln V2 by V1, right? We know that ideal gas
dU equals to 0. So therefore dQ reversible will be equal to dW reversible will be equal to n RT1
ln V2 by V1, right?

But here we are required to find out the entropy change as well, if you recollect what is dS? It is
nothing but equal to dQ reversible by T in this particular case T remains constant. So therefore in
this case if you substitute it, we will find out that this is n RT1 by T1V dv which is nothing but
equal to, if you integrate this particular equation say from state one to state 2 say S1 to S2 or in
other words just from state 1 to state 2.We need to integrate this also from state 1 to state 2, so
therefore this gives us n R ln V2 by V1 this is equals to delta s for this particular process, right?

183
(Refer Slide Time: 29:39)

e
But this was pretty simple but the problem does not end here we are required to find out the heat
interactions then the entropy changes for 2 other processes which provided the initial and final
states are the same. What are the 2 other processes? The 2 other processes are isobaric expansion
followed by isochoric cooling, adiabatic expansion followed by isochoric heating and all the
processes have to be reversible.

(Refer Slide Time: 30:02)

184
(Refer Slide Time: 30:08)

So if we just plot it, simply for ease of your visualization this was state 1, this was state 2 this
was the process which we performed this was P1V1, this was P2V2 say T1 here also it was T,
this process we have already performed and we have found out the heat interactions and the
entropy changes in for this particular case this is say process A, what is process B? Isobaric
expansion so that means pressure P is constant

This is isobaric expansion followed by isochoric cooling, right? So the process B is, you proceed
in this way and you proceed in this way. Now for this particular process it comprises of 2
processes say B1 and B2, for this particular B1 plus B2 we can very well write down dq
reversible along B1, what is this? This is du plus dW reversible along B1, what is dU in this
particular case? This is an isobaric, this is dU is going to be CvdT plus dQ reversible under this
case. So therefore this is going to be P1dv, isn’t it?

So therefore q reversible along B1 this is going to be your integral and here if I specify the states
it is going to be P1, all of these are small they are molar P1 v2 and say T3. So therefore this has
to be integrated from T1 to T3 CvdT plus P1, V2 minus V1, agreed? So this is a q reversible for
this particular path.

What is dq reversible for B2? In this particular case we have integral T3 to T1 CvdT, since this is
an isochoric process so naturally dv equals to 0, so naturally dW reversible which is equals to
Pdv will be equals to 0, right? So therefore in this particular case what do we have? We have dq

185
reversible equals to dU. So therefore automatically we find that when such a situation arises the
heat interaction behaves or can be evaluated from this as a state function as it is.

So now what is dq reversible for the entire process B1 plus B2? You add this and this with this,
when you add you will find that in this case the integration was performed from T1 to T3, in this
case it is performed from T3 to T1. So therefore since we have just the initial and final states are
the same we are just reversing it. So therefore these 2 cancel out and we find this is equals to P1
into V2 minus V1, I am sorry I am repeatedly writing it s capital. So therefore this is dq
reversible.

What is the entropy change now let us see? Entropy change ds along B1 it is going to be integral
of this integral this is dq reversible by T along B1, right? So therefore dq reversible T along B1, I
need to integrate this I need to integrate this, so therefore what does it amount to T1 to T3 CvdT
by T plus nR integral V1 to V2 dv by v, agreed which is nothing but equals to T1 to T3 CvdT by
T plus n R ln V2 by V1.

What is ds along B2 tell me? This is nothing but T3 to T1 CvdT by T, right? Now add this
particular (ds) and this particular (ds) to get the delta s along path B1 plus B2, what is this? This
is equals to n R ln V2 by V1.

186
(Refer Slide Time: 34:51)

Compared this with the del delta s that you had obtained for an isothermal process, what do you
find? You find that the delta s for both the cases are the same but we find that the dq for this
process and the dq for this process they are not the same.

(Refer Slide Time: 35:08)

Again you can repeat it for the next process which is given, the next process is nothing but your
adiabatic expansion, say for the next process what is it? This is basically the adiabatic expansion
followed by the isochoric heating. So therefore here we start from P1V1T1 along a process say

187
C1 to arrive to P3V2T2 and then from there we come to this process. For this process also, we
can proceed I believe you could have done it on your own.

(Refer Slide Time: 35:40)

But suppose we work it out say dq reversible along C1 that is equal to 0 because it is adiabatic,
so therefore dS along this process it is going to be 0, dQ reversible along C2 is nothing but du
which is equal to CvdT again since for an isochoric process dW equals to 0 I have already told
you. So therefore what is q reversible along C1 plus C2? It is just CvdT, right? What is dS for
this process let us see? This is going to be integral T2 to T1 CvdT by T, right?

188
(Refer Slide Time: 36:22)

And we can also for this particular path C1, say for this particular path C1 we can also write that
CvdT equals to minus nRt dv by v, why? Because du equals to minus w since dq equals to 0,
right? So therefore instead of the CvdT by T, can we not substitute it with minus V1 BY V2 nR
dv by v, which gives us minus nR ln v2 by, in this particular case there was a minus, very sorry
so it becomes a plus so therefore this is ds in this particular case, right?

(Refer Slide Time: 37:20)

189
(Refer Slide Time: 37:28)

Again if we compare the q reversible and the ds’s what do we find? I can just superimpose it on
this and if you observe all the 3 what do you find? For this particular path this nR ln v2 by v1,
for this path it is nR ln v2 by v1, in this case also it is the same which automatically implies delta
s is a property of the system and no matter what path we follow, provided the paths are reversible
we know that delta s it is going to be the same provided the paths are reversible.

(Refer Slide Time: 38:25)

On the contrary we find delta s from this particular formula that ds equals to d, we find it from
this particular formula but here even if you are finding it from a path function it is a property but

190
the heat computed for all the 3 cases this is for case 1, this is for case 2 and this is for case 3 this
is for case 3 we find that the heat interactions along the three-paths they are different. Although
the entropy change is the same for all the 3, thank you very much and I look forward to further
interactions while we proceed in this particular course, thank you.

191
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 12
Gibbs free energy as a function of Temperature and Pressure

Hello everybody, so if you recollect in the last class what we have done we are trying to relate
the non-measurable properties with the measurable ones the importance of the non-measurable
properties being that they were important to find out the heats of different type of reactions to
find out whether the reactions can be spontaneous or not to relate the path functions with the
state functions and so on and so forth.

(Refer Slide Time: 1:12)

Now doing that during the course of the discussion I had mentioned that the 2 most frequently
used measurable properties are those of temperature and pressure because they are easily
measurable and they can be easily controlled and regulated, accordingly if you remember what
we have done in the last class, we have tried to express enthalpy change and entropy change in
terms of changes in pressures and changes in temperature.

And of course we have done a lot of other things also and in each of the cases we are trying to
relate different types of properties different properties in terms of the PVT behavior of the gases
and where we came to realize the importance of the PVT behavior of the gases, now before we

192
proceed I would like to spend some more time in order to understand the variation of Gibbs free
energy with temperature and pressure.

Now if you look at this particular equation the variation we find is quite straightforward, okay.
Since Gibbs free energy it is related with changes in temperature and pressure, we are having
straightforward manner, so immediately we understand that the variation of Gibbs free energy
with temperature is given by means of the entropy of the process and the variation of Gibbs free
energy with pressure is given by the volume of the process.

Now just as I have mentioned earlier for Gibbs free energy also we will be primarily dealing
with the molar Gibbs free energy variation with temperature and pressure. So let us see first at
constant pressure what do we get? At constant pressure we will find del g del T at constant P that
is nothing but equal to minus of the molar entropy of change but we need to remember that molar
entropy itself it is a non-measurable property. So therefore we find that the more convenient
equation or more convenient relationship will be to express Gibbs free energy in terms of some
measurable parameters for that what do we do?

193
(Refer Slide Time: 3:05)

We write down or rather we use the energy, g equals to h minus Ts which can otherwise be
written down as g by T equals to h by T minus s. Now suppose we differentiate del g del T with
P at constant pressure, what do we get? We get del del T of h by T at constant pressure minus del
s del T at constant P which is nothing but equals to 1 by T del h del T at constant P minus h by T
square minus del s del T at constant P, now what is del h del T at constant P?

If you recollect this is nothing but equal to Cp, so this becomes Cp by T minus h by T square. If
you recollect we had already derived what is del s del T at constant P that too is equal to minus
Cp by T which gives you or rather which relates the variation of Gibbs free energy as a function
of temperature this is related in terms of the enthalpy of the process.

Or in other words if we try to relate del delta g del T with Delta T at constant pressure this is
nothing but this can be related to the enthalpy change of the process

194
(Refer Slide Time: 4:42)

And this particular equation I would like to mention this Particular equation it is very important
it is known as the Gibbs Helmholtz equation and it is very important to derive the temperature
dependence of equilibrium constants in chemical reactions.

(Refer Slide Time: 5:02)

Now suppose and we would like to find the, for a pure substance how does Gibbs free energy
change? As a pure substance it is heated from say 0 degree Kelvin to some particular temperature
very high temperature T degree Kelvin.

195
(Refer Slide Time: 5:25)

Here also we use the same equation in order to find it out we find that g as a function of T, it is h
as a function of T minus TsT. Now suppose we start from a very low temperature say 0 degree
Kelvin, now the substances they remain solid under that particular condition, right?

(Refer Slide Time: 6:09)

So therefore if we try to evaluate h T minus h 0, what do we find? We find that initially it is a

solid and the solid it is heated from 0 to some particular melting point of this solid, so it is given
by this particular expression as I have already written down here, so therefore sorry this is the

196
expression in heating when the solid is being heated from 0 Kelvin to the melting point I
continue the heat further what happens the solid starts melting, when the solid starts melting then
we have to get the latent heat of fusion.

Once the entire solid has melted away then the liquid starts heating and this is the expression or
this is the enthalpy change for the liquid heating from its melting point to its boiling point then
the vaporization process starts and the substance it starts absorbing the latent heat of
vaporization, finally the entire substance has vaporized it is in a vapor state and then the heat the
vapor to the high-temperature some high-temperature say T, right?

(Refer Slide Time: 7:16)

So therefore this whole expression gives the enthalpy change of the pure substance when it is
heated from a temperature of 0 degrees centigrade to a higher temperature T Kelvin such that at
T Kelvin the substance remains in the gaseous state. Now for this particular substance if we try
to plot h T minus h 0 as a function of temperature, what do we find?

We find that for this particular case initially what happens? Initially the solid gets heated since I
plotted h T minus h 0 it starts from the origin h 0 is nothing but it is this molar standard enthalpy
of the substance at 0 Kelvin at an 1bar pressure. So therefore usually it is the ideal gas enthalpy
at 1 bar pressure that the gas behaves or it remains or behaves as an ideal gas. So therefore this
continues till the melting point of the solid.

197
After that what happens it absorbs delta h fusion but does not change in temperature, next again
there is a liquid heating which continues till the boiling point and after that again there is a latent
it absorbs the latent heat of vaporization in this particular case this is the latent heat of
vaporization delta h vaporization and in this particular case this is the Delta h fusion and then
finally after that it’s going inside the writing anyhow the vapor phase it is being heated.

So therefore if we plot h T minus h 0 the enthalpy change of a pure substance when it is being
heated from 0 Kelvin to some particular higher temperature the curve it has something of this
sort.

(Refer Slide Time: 8:57)

Now if you plot the entropy change for the same substance when it is being heated? It is given by
this particular expression.

198
(Refer Slide Time: 9:09)

Now for this case also if I plot s as a function of T or s as a function of temperature, now here I
would like to mention actually this particular expression it should have been g T, suppose I
express g T as a function of h 0 since I would like to keep the same standard state for enthalpy as
well as the Gibbs free energy, the actual expression is h T minus h 0 minus T at s T minus it's
just 0 at sT which is nothing but 0.

So therefore I plot sT as a function of temperature and this is the exact expression that we are
going to give see and with this case also we will find that the nature of graph is the same.
Initially it’s heated and the solid increases in temperature its entropy also increases then we have
for one particular, during the melting process it absorbs Delta h fusion at T fusion in this
particular case it is absorbing delta h fusion at T fusion and after that again the liquid gets heated
then it absorbs delta h vaporization at the boiling point and then the heating it continues.

So therefore in this case also we will find the nature of the curve is just the same while it is in the
single phase the entropy rises the temperature rises and when the change of phase occurs that
entropy rises at a constant temperature and so on and so forth. So this is nothing but the melting
point of the solid and this is nothing but the boiling point of the solid.

Now in order to obtain gT minus h 0, so what do we need to? For each particular point we need
locate the point here and we need to subtract T sT from here from this particular expression to

199
get the resultant gT minus g0. Now let us see that while the enthalpy curve and the entropy curve
they were more or less similar in nature with just differences in the magnitude of the different
quantitative measures.

(Refer Slide Time: 11:54)

Let us see the nature of the gT versus h h0 curve, in this particular case what should be the nature
of the curve let us see? The nature of the curve in this case, Suppose we plot this, the x the y axis
is gt minus h o and the x axis is T degree Kelvin, right? So what is the nature of the curve that
you would expect when we subtract T of sT from hT minus h o, we find that the nature of the
curve is completely different, it is something of this sort where we find that this corresponds to
the melting point of the substance this corresponds to the boiling point of the substance.

200
(Refer Slide Time: 13:07)

(Refer Slide Time: 13:25)

So in this particular case what do we find? We find that there are some unique features of the
Gibbs free energy variation with temperature curve as compared to the enthalpy variation and the
entropy variation. The first thing we find that if we see del g del T at constant P is nothing but
equal to minus s. So therefore we find that since s is an infinitesimally positive quantity. So
therefore the slope is always going to be negative which is shown here it decreases with
increasing temperature number 1.

201
(Refer Slide Time: 13:57)

Number 2 is we find that it is a continuous function of temperature even at phase transition, why
does this happen? In a nut phase transition, what is delta g transition equals to? This is equal to
delta h transition minus T transition into delta s transition, right? And what is this delta h
transition equals to? It is nothing but T transition into delta s transition because at a transition
temperature the entropy increases by absorbing the heat and the heat is only used up as the latent
heat for that particular transition.

202
(Refer Slide Time: 14:29)

So therefore we find that at the generation temperature this term and this term are equal as a
result of which delta g transition equals to 0 and this explains why even if it increases with
increasing temperature and we find that even at phase transition we find that the Gibbs free
energy it is a continuous function for function of temperature and this signifies that at phase
transition when the 2 phases are in equilibrium under that condition delta g transition equals to 0.

But even if it is a continuous function of temperature we need to remember that the slope
becomes discontinuous at each transition, why? But what is the slope the slope is nothing but
minus s. So therefore if we are we are concerned with this particular portion then in that
particular portion we will find delta g delta T at constant P this is going to be minus s entropy of
the solid.

On the other hand if we are concentrating on this particular portion we will find that for sorry
this is delta g. If delta g delta T at constant P this will correspond to the entropy of the liquid. We
very well-know that entropy of the liquid is greater than entropy of the solid. So this explains
why even if Gibbs free energy is a continuous function of temperature even at phase transition
the slope becomes discontinuous at each transition the slope corresponding to the changes in the
entropy before and after the transition.

203
(Refer Slide Time: 15:59)

And we very well-known that s of solid it is less than s of liquid which is less than s of the vapor
phase, so naturally what does it suggest? It suggest that the negative slope it is steepest for the
vapor phase which I have shown here this particular case corresponds to the solid this
corresponds to the liquid state and this corresponds to the vapor phase. So we find that the slope
is steepest for the vapor phase and it is least steep for the solid phase.

Now there is another interesting thing that we can observe, if we observe the points near the
phase transition very closely we find suppose at the phase transitions we try to extend the 2
curves say we try to extend the solid curve beyond the melting point, we try to extend the liquid
curve again below the melting point.

Now what do we find? We find that even if the liquid were to exist below the melting point its
Gibbs free energy would have been higher as compared to the solid phase and we find that even
in the solid would have existed beyond the melting point its Gibbs free energy would have been
higher as compared to the liquid phase. So therefore since a system or a or any particular
substance is stable only for a minimum Gibbs free energy condition at conditions of constant
temperature and pressure it automatically suggests why a solid is favored at low-temperature
below the melting point? Why the liquid is favored at high-temperature beyond the melting
point?

204
(Refer Slide Time: 18:19)

And from the same logic we can also understand why the vapor phase exists beyond the boiling
point? And why the liquid phase exists below the boiling point? And if we recall once more,
again I would like to say that g is nothing but equals to h minus Ts as I had already discussed
earlier since s is associated with T, so therefore this particular term this becomes important at
high-temperature this we had already discussed earlier and naturally this particular term this is
important at low-temperature.

205
(Refer Slide Time: 18:24)

So therefore what do we conclude? We conclude that a solid is solid phase is favored at low-
temperature, why? Because it has the lowest enthalpy and the vapor is favored at high-
temperature, why? Because it has the highest entropy, so the existence of the gas at high-
temperature is due to the entropic contribution to Gibbs free energy while the existence of solid
at low-temperature is due to the enthalpic contribution at the low-temperature phase.

206
(Refer Slide Time: 18:59)

Now here it will not be out of the place to just mention something which I have forgotten to
mention while I was discussing entropy. How did we find out or what is the expression to find
entropy? It is nothing but integral dq by T, right? Or in other words we will find that if we give
the same amount of heat suppose this is q this is a reversible, so this is q reversible by T.

Suppose I give the same amount of q reversible to a substance at different temperatures, we find
that the change in entropy it will be proportional to the or rather inversely proportional to the
absolute temperature which signifies that if the same amount of heat is applied reversibly to a
substance at different temperatures then the entropic change will be much more for lower
temperature as compared to the entropic change at higher temperature this is almost similar to
understanding that if you sneeze in a very quiet library the disturbance of the people around will
be much more as compared to if you sneeze in a very busy shopping mall. So therefore this is
something which I had forgotten to mention, so I thought it will not be out of the place if I
mention it here.

207
(Refer Slide Time: 20:31)

Now let us discuss the variation of Gibbs free energy with pressure. Now what do we know
again if I recollect the equation dg equals to minus sdT plus vdp, right? So therefore del g del p
at constant T it is nothing but equal to the molar volume of that particular solid.

(Refer Slide Time: 20:52)

So therefore we find that at constant temperature if we are talking then this should be del P, so
there it gives us del g del P at constant T equals to volume from where we can find out the and

208
the Gibbs free energy change for any process when a system goes from P1 to P2 is nothing but
equal to vdP, right?

(Refer Slide Time: 21:22)

So therefore from here for solids and liquids what can we say? For solids and liquids we can say
that your delta g in this particular case, this is equal to say suppose g at say P2 this is equal to g
at P1 plus v into P2 minus P1, right? And since this or it can also be written as v delta P which is
almost equal to 0 for liquids and solids which tells us that g P2 is almost equal to g P1 for liquids
and solids or in other words Gibbs free energy it is more or less insensitive to pressure for liquids
and solids.

Now what happens for an ideal gas suppose we say, say we take up one mole of an ideal gas,
right? For one mole of an ideal gas this is going to be integral vdp integral and we know this is
nothing but RT by Pdp which gives us the expression as RT ln integral P1 to P2 this is also P1 to
P2, this is nothing but RT ln P2 by P1, so therefore what type of expression is this? We find that
for an ideal gas the variation of Gibbs free energy with pressure exhibits an logarithmic effect
which is nothing but due to the entropic effect because if you remember the enthalpic change of
an ideal gas with pressure equal to 0 which we have already derived.

This particular equation it is particularly very important in chemical reaction equilibrium

involving gas phase reactions, right? So therefore we had discussed how Gibbs free energy
varies

209
with pressure for liquids and solids, we found out how Gibbs free energy varies for ideal gas and
it is very evident that if you are dealing with real gases then naturally we have to know how the
molar volume of the real gas varies with pressure or which again we have to refer to the PvT
behavior of the gases.

And we will be discussing after one or 2 classes that while trying to find out the variation of
Gibbs free energy with pressure for a real gas. We come across a concept of fugacity and we
define another important property which is the fugacity we will be discussing it and we will be
discussing in details the variation of Gibbs free energy with pressure for real gases shortly.

(Refer Slide Time: 24:19)

Well but before that, before I end the class what I would like to do in this particular case is that I
would like to discuss just like the way I have found out how Gibbs free energy varies with
temperature in the same way I would like to find out how Gibbs free energy it is a function of
pressure, right? So what type of graph would you expect in this particular case?

We know again let me write down the equation del g del P at constant T this is equals to v, now
we take up the same substance that we had done we start heating it or rather we start
compressing it from a very low pressure where the substance naturally at that low-pressure it
usually exists in the gaseous phase as we start compressing it, so therefore in the gaseous phase
what do we find as we start compressing it?

210
The del g del P at constant T that will be corresponding to the molar volume in the vapor phase,
so this is del gv del P at constant T that will correspond to the molar volume. Now this will
continue till the vapor does not condensed to the liquid phase, once it has condensed to the liquid
phase we will find that del gL del P at constant T this is going to correspond to the molar volume
of the liquid phase and then again beyond the melting point we find that del gs del P at constant
T this will be corresponding to the molar volume of the solid phase.

Now what do we know? We know that volume it is also a positive quantity, so from there what
do you understand? We find that the variation or the slope of Gibbs free energy versus pressure
is always positive or in other words Gibbs free energy keeps on increasing as pressure increases
this is the first thing, the other thing what do we find? That at low pressure we always
encountered a vapor phase, right?

So therefore at low pressure what will you expect? We are going to expect something of this sort
where the slope corresponds to the vapor phase, fine. Then beyond the pressure corresponding to
the or the pressure under which condensation occurs we find that the liquid phase exists and the
slope will be corresponding to the molar volume of the liquid phase and beyond the condition
where the liquid is going to form the solid phase we find that it is going to correspond to the
molar volume of the solid phase.

There is one more thing that we know, we know that v of the vapor phase this is greater than the
v of the liquid phase which is greater than v of the solid phase, so what does that explain? It
explains two things, the first thing is Gibbs free energy always increases with pressure it always
has a positive slope, the slope is the steepest in the vapor phase which is less steep in the liquid
phase and it has the minimum steepness in the solid phase.

Again the way I have described or I have discussed while discussing the variation of Gibbs free
energy with temperature in this case also if we extrapolate the curves on either side of the
transition points it automatically becomes clear why we encounter the vapor phase below a
particular pressure below the transition pressure while we encounter a liquid phase above the
transition pressure and again in the same way if you are working with the liquid and the solid
phase it automatically becomes clear why the solid is favored at the highest pressures and the
liquid is favored at intermediate pressures right?

211
(Refer Slide Time: 28:26)

Now just one thing I would like to mention that this particular nature of the curve that I have
drawn this particular nature it is for a normal substance say for carbon dioxide and this particular
curve has been drawn for some particular position near the triple point say for instance it has
been drawn under this particular point and the things which I have drawn is I have taken a
vertical line from very low pressure I keep on following the vertical line and then we find that
initially there is the vapor phase then we have a liquid phase and then we have a solid phase at a
higher pressure.

212
(Refer Slide Time: 28:53)

So therefore this particular the graph has been drawn for a condition where T it is close to the
triple point but it is greater than it is greater than the triple point but it is close to it, it is far
removed from the critical point, do you think that the Gibbs free energy variation with pressure it
will have the same characteristics at constant temperature no matter which temperature I select,

(Refer Slide Time: 29:30)

213
For example suppose instead of drawing it for a condition near the triple point suppose I try to
draw it at the triple point exactly at the triple point suppose I try to draw. In this particular case
do you expect the same nature of the graph to hold?

(Refer Slide Time: 29:45)

In this particular case, what is the nature of the curve that you are going to get? gp as a function
of P, in this particular case you will find that at the triple point this is the vapor phase and here
we find that this is going to be the solid phase, this is the vapor phase, this is the solid phase and
the liquid phase is going to exhibit the nature of the curve something of this sort and this
particular situation is for P equals to Pc and for any other pressure other and this particular curve
has been drawn for T equals triple point, this is also the triple point.

We find that at P equals to P triple point all the 3 phases have got the same Gibbs free energy, so
they are all of them are equally probable of existence but once we start increasing the pressure
from the triple point pressure at the triple point temperature we find that there is greater
likelihood of the solid to exist over the liquid phase.

214
(Refer Slide Time: 30:48)

(Refer Slide Time: 30:50)

In a similar way suppose we would like to plot it at some other temperature where T is less than
the triple point temperature, in this particular case what type of curve did you think you are going
to expect? We have the vapor phase at low pressure we have the solid phase here this is the vapor
phase, this is the solid phase and the liquid phase will be existing we never have a liquid phase
for the temperature conditions less than the triple point.

215
(Refer Slide Time: 31:21)

Now suppose instead of this we try to plot it near the critical point at the critical point as all of
you know more or less the vapor and the liquid phases they become indistinguishable and the
molar volume of the vapor is almost equal to the molar volume of the liquid.

216
(Refer Slide Time: 31:41)

So if we plot g p g versus p at P for a condition say P almost close to T equals to Tc, what do you
expect? You expect that more or less since the slope of the curves are going to be almost equal at
the transition places and you find that the solid exists above the vapor and liquid coexistence
curves for all particular conditions.

And at P equals to TC you have just a continuous curve throughout that is no discontinuity in the
curve and the solid phase this is just a fluid portion and the solid phase curve it remains above
the fluid curve under all particular conditions.

217
(Refer Slide Time: 32:40)

Now before I end I would just like to mention that all these curves that I have drawn here they
are for a pure substance and for example carbon dioxide or benzene or anything, do you expect
the same types of curves to hold when we are going to deal with water?

(Refer Slide Time: 33:10)

If you note the PT curve for water it is very evident that the solid liquid coexistence curve has
got a slight negative slope as compared to the light positive slope for carbon dioxide, benzene or
any other normal substance this is because as we all know that the molar volume of the solid is

218
greater than the molar volume of the liquid in case of water, there are 2 other substances which
also exhibit this particular characteristic the 2 substances being Bismuth and Antimony, for these
thing we have some particular differences when we try to plot the g versus P curves at the triple
point and for conditions less than the triple point.

I leave this as a home assignment for you, try it out and try to find out the g versus P curves for
different conditions or rather for water at the triple point and less than triple point and in case
you have doubts you can get back to us and we can clarify the doubts.

(Refer Slide Time: 34:01)

219
(Refer Slide Time: 34:05)

But as I hint it is very evident more or less the curves get inverted, the situations which you had
got for g equal this particular situation is going to exist for T less than triple point and we find
that this particular situation is going to exist for T equal to triple point or slightly higher than the
triple point.

So with this more or less we have come to an end to our discussions on property estimations and
the properties of homogeneous systems which primarily comprise of one phase I would just like
to mention that these components which we have dealt with they are homogeneous systems they
can be single component they can be a mixture of components but due to the processes which are
undergoing there is no change in composition there is just change in volume.

Or in other words to be more specific so long our discussions were confined to closed systems,
in the next class we are going to deal what happens when the changes that we have discussed are
also accompanied with changes in composition as well, we are first going to deal for single
component systems and then we will be dealing with multi-component systems, have a good day.

220
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 13
PvT Behaviour of gases

(Refer Slide Time: 0:32)

Hello everybody, to recall what we have done till now, we were using the Maxwell’s equations
to find out your entropy as a function of temperature and pressure the Tds equation is written
down here. We had similar equations in terms of your T and v as well then we had also expressed
enthalpy in terms of temperature pressure and we had also done this for several other properties
if you recollect.

We had tried to predict Cp minus Cv, we had tried to predict Cp by Cv, we had tried to predict u
as a function of T and v etc. Now in all these equations if you recollect, what do you find? We
have used thermal data and volumetric data in order to predict these particular thermodynamic
functions thermal data are those data which tell you how this function vary with temperature and
volumetric data are those which tell you how these particular properties they vary with volume
or density and pressure, right?

So as far as this part is concerned this either Cp or Cv etc, there also we have already discussed
that more or less data is available for ideal gases as a result of which we would like to we would

221
prefer to carry out this particular temperature change at very low pressure where we can use the
data of ideal gases in order to predict the enthalpy change due to temperature change of gases the
other part was to find out the dependence on pressure.

We find that this particular dependence it is a function or it depends upon the PVT Behavior of
substances, right? Or in other words this is easy to evaluate provided we have a relationship of
this form for the substance which we are going to consider, fine. And we find that these
relationships they are available, plenty of them are available the reason for this plentiful
availability is just that there is no particular relationship which is applicable over a wide range of
conditions for different pure components as well as the mixtures.

We are going to discuss PVT Behavior of pure components in this class remembering fully well
that mostly in thermodynamics we would be requiring to discuss or rather we will be dealing
with thermodynamic properties of mixtures. So we will be doing PVT Behavior of pure gases a
little amount about condensed phases and then after we complete fugacity we will be predicting
or we will be discussing different mixing rules and combining rules to predict the
thermodynamic properties of gaseous mixtures, right?

Now we know that this type of relationship they are available in different forms, one of them is a
graphical representation, one of them is a mathematical equation we also have a third form which
is as tables tabular data. So these are the 3 forms usually in which PVT Behavior of gases they
are or rather sorry PVT Behavior of any particular substance it is available.

So we are first going to deal with graphical equations and our graphical representations and then
we are going to deal with mathematical equations and as far as tables are concerned you are
already aware of steam tables and several such tables are available we will not be discussing
them in the process or in our course of learning of thermodynamics you may have to refer to
several tables, so just be aware that PVT behavior of pure components in different phases are
also available of tables the steam table is the most common example of it.

222
(Refer Slide Time: 4:32)

Now let us take up the graphical representations, for a pure substance it can exist in 3 forms it
can be a solid, a liquid and a vapor state, right? And what are the properties that we can vary
independently F in this particular case it is C minus P plus 2, C is one, so if P is 2 then in that
case F becomes equal to1 if suppose P is 1, F becomes equal to 2, right? So therefore what can
be and what are the variables that we have at hand? We have temperature of the gas we will
prefer the molar volume because we would like to deal with intensive variables and the pressure
of the gas.

So out of this we can vary 2 and keep the third one constant and try to find out the different
variations. So for example we can generate curves of say T as a function of v, we can generate
curves of say P as a function of v this will be at constant pressure and this will be at constant
temperature. We can also have curves of P as a function of T at constant molar volume. So these
are the 3 ways by which graphically the PvT Behavior of pure substances can be represented.

Let us take up the Tv behavior, so suppose we have taken up a substance say for a very low-
temperature say minus 20 degrees centigrade we take up benzene for instance it remains solid
under that temperature we keep on applying heat to it say the initial temperature or rather initial
condition of existence of benzene is here for temperature is very low say minus 20 degrees
centigrade benzene is in this particular case is solid, we start heating it what happens? Its
temperature, it absorbs energy temperature increases along with that it expands.

223
So therefore even both of them they start increasing after that it is continues till the solid keeps
on getting heated and it expands this keeps on continuing till we reach the melting point of the
solid, at the melting point what happens? It changes to the liquid phase due to this change its
volume increases but this change as all of us know occurs at a constant temperature. So therefore
under this condition we find that they get a horizontal line which marks the fusion the process of
fusion of the pure substance.

After that this fusion process continues till the point entire solid has melted and it forms the
liquid during this entire time we find some amount of solid keeps on melting liquid form till at
this particular point we find entire solid has melted and it is converted to the liquid phase, in this
particular the volume corresponds to the molar volume of the liquid at the melting point and this
particular temperature it is nothing but the melting point of the solid.

Now with the liquid we continue this heating process, again as the liquid gets heated the
temperature keeps on increasing and along with that the volume also keeps on increasing, in this
case possibly it is not clear from my particular figure but in this case the rise will be much
steeper as compared to the rise in the case of the solid. So this continues as long as the liquid
remains a liquid and it has not reached the boiling point then at one point it reaches the boiling
point. So this is the T boiling point of the pure substance at the boiling point again the same
thing occurs as the fusion this is the vaporization curve this curve continues till the entire liquid
has vaporized and the pure component vapor is formed.

Once liquid has completely converted to the vapor again the temperature keeps on rising in this
particular case. So therefore this part it corresponds to the volume of the vapor at the boiling
point it is the T boiling point, right? This corresponds to the specific volume of the liquid at the T
boiling point and we know there is a remarkable increase in the volume and due to this
remarkable increase in the volume naturally it absorbs a good amount of latent heat.

Now this whole thing the entire heating process, what have I done? The entire heating process it
is performed at constant pressure, say the pressure is 1 bar pressure, so what is this? This is
basically an isobar on the Tv plot.

224
(Refer Slide Time: 9:31)

We can construct such isobars at different pressures, right? And this is what precisely I have
done here. We have constructed isobars at different pressures at increasing pressures and we find
that we get such identical curves as I have already shown.

So the only difference between these curves is that we find that as we go for higher and higher
pressure the solid liquid coexistence region that more or less remains the same what happens is
the liquid vapor coexistence region that keeps on becoming lesser and lesser as we perform the
experiment at higher and higher pressure. So therefore in this particular curve after that what we
do? We have performed the experiments at different pressures and we have located the melting
points and the boiling points etc.

And then we have joined the melting points boiling points at different pressures and these
particular curves they are known as the saturated curves these curves which is the locus of the
different saturation points at different pressures are known as the saturation curves. Now what do
the saturation curves do? They join points at different pressure temperature conditions where 2
phases are in equilibrium it can be a solid liquid in equilibrium it can be a liquid vapor in
equilibrium.

So therefore what is this curve? This particular curve this is a saturated curve with respect to the
solid, what is this curve? This is a saturated liquid curve with respect to fusion, remember one
thing we have 2 saturated liquid curves, one is with respect to fusion the other is with respect to

225
boiling, right? So therefore this is a saturated solid curve this particular curve, right? This is a
saturated liquid curve with respect to fusion, fine.

And this is a saturated liquid curve with respect to boiling and we have a saturated vapor curve
here. Now we find that more or less the saturated liquid curve and the saturated solid curve with
these 2 curves they are more or less parallel to each other while we find that the saturated liquid
curve with respect to vaporization and the saturated vapor curve they gradually incline towards
each other and finally they converge at a point.

(Refer Slide Time: 12:42)

So therefore on close observation what do we find? We find that the change in volume which I
have marked here, so therefore in this particular way if we keep on making them at different
pressures and then if we join these particular points then we get, I will do it with a different pen
that will be easier for you. We can have this, we can have this then we can have this, we can have
this, so therefore what do we find? We find that this is the saturated solid curve, saturated liquid
curve with respect to the melting, saturated liquid curve with respect to vaporization.

Saturated vapor curve and we find that while these 2 lines they are more or less parallel to each
other for this particular portion we find that Vv minus vL that keeps on decreasing at higher
pressure , right? This keeps on continuing and as we go for higher and higher temperature also
you will find that this keeps on decreasing till at some particular temperature we will find more

226
or less that Vv minus vL this becomes equal to 0 and this particular point, in fact I have not
drawn it as a point the drawing is not good.

(Refer Slide Time: 13:59)

If you find here you find at this particular point at this point more or less your change in molar
volume becomes equal to 0.

(Refer Slide Time: 14:04)

Now when the change in molar volume becomes 0? What other thing happens? You know due to
this particular change the substance it has to absorb latent heat and due to this change, absorption

227
of latent heat there is volume expansion due to which the substance changes from the liquid state
to the vapor state.

(Refer Slide Time: 14:30)

So therefore isn’t it quite natural that if we consider the portion of the curve here we find that
Delta v the change of volume is much higher, here it is much lower.

So therefore it is quite natural that here the latent heat which the substance has to absorb in order
to change from the liquid to the vapor state should be much higher as compared to the latent heat
it needs to absorb at this pressure and therefore this should continue and here we believe the
latent heat of vaporization that it has to absorb is going to be equal to 0.

228
(Refer Slide Time: 15:09)

Or in other words, suppose I plot delta h vaporization as a function of temperature then what is
the curve which I expect? It should have a high latent heat vaporization to start with it keeps on
decreasing and then at T equals to Tc we find delta h vaporization equals to 0, fine.

So therefore this was the temperature volume curve which we found out and what did we find?
We find that mostly for our cases since we are dealing with solution liquid vapor substances we
would like we would prefer to keep our discussion, since we are not dealing with physical
methodology or solid state physics we will not be dealing with the solid state much.

229
We will be primarily confining our attention to the liquid vapor phase, so therefore we will be
more or less interested in this particular portion of the Tv diagram, fine. And here what do we
find? We find that as long as we are inside or we are at in a region here we find that the isobars
they have free particular regions of different slopes. There is an isobar here then there is a
horizontal portion and then there is another portion. So if we continue this we will find at the
critical point there is going to be a horizontal inflection beyond the critical point the isobars will
be something of this sort.

230
In the same way we can plot our Pv diagram also, we are going to start in the similar fashion that
we have done here we take our substance we start heating it and then as we heat it at constant
pressure initially the substance was here we heated the pressure remains constant it is here and
then it changes to the vapor phase, sorry liquid phase then the liquid gets heated and then it goes
to the vapor phase and then the vapor is heated to a very high temperature.

So this is what this 2 is a saturated state with of saturated solid state, this is saturated liquid state
with respect to fusion, saturated liquid state with respect to vaporization, saturated vapor state
this can be performed at this was possibly say at P equals to 1 atmosphere. We can keep on
performing these experiments for different pressures and then finally we can draw the curves in
this particular way. So in this particular case also we can we can draw the curves in this
particular way we find that we are going to get something identical this is a solid-state, here also
this is solid, this is solid plus liquid liquid, this is liquid plus v this is v, here also this is solid s
plus L, L plus v and this is the vapor state.

Again we would like to be concentrated here if we find that the isotherms which we draw here
then in that case the isotherms in this particular case it is going to be almost vertical in the liquid
region because we know they are all practically incompressible and then during the change of
phase we know it is more or less horizontal it is going to be horizontal curve because remember

231
one thing change of phases occurs under isothermal isobaric conditions therefore they are
horizontal both in the Tv plot as well as in the Pv plot.

And then in the vapor region it exhibits more or less a rectangular hyperbola sort of a thing. So
therefore as long as we are operating at T is less that Tc we find that the saturation curve it
comprises of 3 particular regions with the middle region decreasing as we go for higher and
higher temperatures and finally at T equals to Tc we find that there is a horizontal inflection and
beyond that they are going to have curves something of this sort.

(Refer Slide Time: 19:11)

This is what I have shown in this particular figure if you see I have tried to show this, this is for
carbon dioxide. Now there is one thing I have already told you about the point of termination of
the liquid vapor curve, right? Liquid vapor dome, sorry, this is nothing but the critical point, what
about the initial portion? Where does the liquid vapor dome or in other words where from where
do this saturation curves they originate?

Now all of these curves they originate from a triple point line, if you observe this triple point line
you will find that at this particular line the solid liquid and the vapor phases they coexist in
equilibrium as a result of which F equals to 0 under this condition and therefore this line it is a
very important constant characterizing any particular substance where it is characterized by a
specific temperature, a specific pressure and a specific volume.

232
(Refer Slide Time: 20:23)

For instance suppose we take up 2 particular substances say water and carbon dioxide, what do
we find? We find that for water the triple point pressure it is, for water T triple point this is
273.16 most of you all are aware of it, P triple point this is4.58 millimeters mercury and for
carbon dioxide if you see you will find that the T triple point it is 216.55 Kelvin, P triple point
this is equals to 5.11 atmospheres but does this bring any, does this strike any cord in your mind
the importance of this data?

233
(Refer Slide Time: 21:13)

If you observe this data what do you find? You find that the triple point curve for carbon dioxide
this particular curve here as well as here this occurs at a pressure much higher than the
atmospheric pressure.

234
(Refer Slide Time: 21:51)

(Refer Slide Time: 22:03)

So therefore at atmospheric pressure where are you exactly? When you’re working with carbon
dioxide at atmospheric pressure you are lying in this particular region, what is this region? It is a
solid here, it is a solid plus it is a vapor region here and in between this region it is marked by a
solid plus vapor. So what does it imply? It implies that if you’re working at a pressure below the
triple point pressure then in that case you’re working in this particular zone where the solid will
directly be converted to the vapor phase as a result of which we will find that substances sublime
if their triple point pressure is higher than the atmospheric pressure.

235
(Refer Slide Time: 22:14)

(Refer Slide Time: 22:17)

On the contrary if the triple point pressure is much less than the atmospheric pressure then at
atmospheric pressure what are you doing basically? You are operating in this particular region
where you find that the solid will be converted to the liquid and the liquid then will be converted
to the vapor and this gives you the answer to questions as to why some substances sublime.

In reality all substances sublime if you are operating below the triple point pressure, for example
if you’re working with water and you’re working at a pressure of say 1 millimeters mercury that

236
at such a great amount of vacuum then you’ll find that water will also sublime. So the question
of why some substances sublime is that, it’s triple point pressure is either higher than
atmospheric or lower than atmospheric.

(Refer Slide Time: 23:01)

So therefore we find that the vapor liquid dome that we have drawn here, the vapor liquid dome
it starts from the triple point curve, this is the triple point curve and it extends to the critical
point. Now there is one particular question which I would like to put before you, we have a triple
point curve and a critical point marking the beginning and end of the vapor liquid dome, correct?
The solid liquid coexistence curve also starts from the triple point curve as is very much evident
from this particular figure.

But here we have not marked an end to this solid liquid coexistence region, for example here we
know that both of them they merge at the critical point and beyond the critical point we have
only vapor phase and the characteristics of the critical point is that if you’re working above the
critical point then just by mere completion we cannot reduce the or rather just by mere cooling
we cannot convert a vapor to a liquid phase.

Do you have certain identical termination point for the solid liquid coexistence curve? Or does
the solid liquid coexistence region or the curves they continue indefinitely? This is something
that you should be wondering and here I would like to mention that there is no termination point
for the solid liquid coexistence curves they keep on continuing, what is the reason behind it? The

237
reason is, if you observe the molecule behavior of liquids and vapors you’ll find that the main
difference between them is about the degree of freedom of movement of the molecules in the
liquids they are much more bound, in the gases they are much more free.

So therefore when we are heating a liquid, what we are doing? We are just making the molecules
more free from one another we are decreasing the intermolecular attraction between them as a
result of which they are getting converted to the vapor phase. It is basically a matter of the
degree of the freedom of which is enjoyed by the molecules.

(Refer Slide Time: 25:24)

So therefore for such a case beyond the critical point if you’re going then we find that we can
very well, suppose we are in this particular region we want to come to this particular region. We
can cross either follow this particular isotherm and we can perform this thing or else we can even
ignore this particular discontinuous region and we can make this transition by these particular
process where we find that without this particular change of phase this particular heterogeneous
liquid vapor dome we can also make a gradual change from one to the other.

But when we are considering the solid liquid transition state then it is not just a matter of the
degree of freedom enjoyed by the molecules. In this case actually the solid structure has to break
and the liquid has to form as a result of which under this particular condition there has to be a
zone of discontinuity where the solid structure starts breaking, breaks completely and forms the
liquid

238
state due to which the solid liquid coexistence curve it keeps on continuing indefinitely there is
nothing like a critical point for a solid liquid coexistence curve.

(Refer Slide Time: 26:48)

In fact if you also consider solid-solid transitions there also we will find that there is nothing like
a critical point. Well this was, one more thing before I end I would just like to remind you that
these curves will not be complete unless we introduce one more information into them, what
have we shown here? We have shown the pure component existence zones. We have shown the
zones of existence of the 2 phases.

Now remember one thing, when we are working with 2 phases then it’s not only the temperature
and pressure that is important, in fact when you fix one the other automatically gets fixed. What
is the other thing which is important here? The other thing is the proportion of liquid and the
vapor which is there when you’re operating inside the liquid vapor dome. So in order to
complete this particular figure it is very important that we also plot along with isobars, isotherms
we also plot the constant quality lines.

And remember one thing the first constant quality line is the saturated liquid zone where we
know the x which is the dryness fraction which is the mass of dry vapor divided by the total mass
of the moles of the liquid vapor mixture that is equal to 0 at the saturated liquid curve and it is
equal to one along the saturated vapor curve. So therefore the first quality line is x equals to 0,
the next quality line is x equals to 1, the intermediate quality lines can be drawn.

239
All these quality lines they start from the critical point and end at the triple point curve and once
everything is inserted the saturated or rather the phase diagrams for pure substances are
completed. Here also there should be the constant quality lines and in from these constant quality
lines, suppose we are operating at this particular point we can know that pressure of that
particular point, we can know the mixture volume and we can also know the temperature here
and from the mixture volume we can know what is the proportion of the liquid and was the
proportion of the vapor in this particular case.

So the liquid and the proportion of liquid and the vapor they can be obtained from a lever rule
principle, okay. They are proportional to the distances of the horizontal portion of that point from
the liquid saturated liquid curve and from the saturated vapor curve. So therefore just I would
like to show how we can do this?

240
(Refer Slide Time: 29:08)

We know what is x equals to? x is nothing but the quality or the dryness fraction, right? What is
this quality or the dryness fraction? Suppose they’re working with nv moles of vapor nl moles of
liquid and then nv moles of vapor, okay. So therefore we know that this is equal to 0 as I have
already mentioned because nv equals to 0 on the saturated liquid curve no matter what pressure
or temperature you have operating equal to one over the saturated vapor curve.

241
(Refer Slide Time: 29:46)

(Refer Slide Time: 30:04)

And then suppose we have located the molar volume here in this particular case, suppose say it is
VM, so therefore and we are working with say n moles of the gas, right? So therefore nVM
should be equal to nlVL plus nv Vv, fine. Now dividing throughout by n, what do we get? We
get V is nothing but equals to 1 minus mv by mv plus mL the VL plus mv by mv plus mL Vv or
in other words the V for the mixture VM which is nothing but equals to Xvv plus 1 minus x VL,
right? And this has been obtained from the lever rule, so therefore by measuring the horizontal
portion here.

242
(Refer Slide Time: 30:46)

(Refer Slide Time: 30:56)

Suppose we are at this particular position by measuring the horizontal portions, we can find out x
and 1 minus x and we already know these 2 so we can find out VM from here

243
(Refer Slide Time: 31:13)

And here I would like to mention that the similar relationship this additive property in fact this
holds for all extensive properties of pure substances we can have s equals to 1 minus x sL plus
xsv similar way we can have the same thing for enthalpy the same thing for internal energy and
so on and so forth we can keep on having. So therefore from here what do I do? Till now what
have I done?

I have discussed just 2 curves we will be discussing the PT curve in the next class and in all the
curves that I have discussed it is important for you to know this is TV curve in all the curves you
will find that at this particular point more or less we have something of this sort. In fact we can
we can also plot different curves for Ps and Ts also.

Here also in the Ps curves also we find that there is a horizontal inflection at the critical point,
here also we will be having a horizontal inflection at P equals to PC this is at T equals to TC, this
is P equals to PC, this is T equals to TC, for all the curves we find there is a horizontal inflection
and for each of the curve we know that there is going to be a horizontal portion here as a result
of which for each copy can write down del P del V 2 phase equal to 0.

Del T del V 2 phase equal to 0, same applies for these 2 and we can also write down del P del V
at T equals to TC equals to 0 del 2 P del v2 at T equals to Tc equals to 0. Same thing for del T
del V at p equals to Pc equals to 0 del 2P sorry del 2T del V2 at P equals to PC equals to 0. So

244
therefore this particular and similar way we can write for these also, we can write del P del s at
constant at 2 phase this is also equal to 0.

Del T del s 2 phase equals to 0 and therefore, so therefore what I mean is such a relationship can
be written down for any particular closed substance in the 2 phase region and at the critical point.
So we will continue with our discussions regarding the graphical representations in the next
class.

245
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 14
PvT Behaviour (Contd.)

(Refer Slide Time: 0:34)

So therefore we had already discussed if you remember we had discussed the Pv and the Tv
diagrams already the only thing which was left was the Pt diagram. Again we start with the same
thing suppose we have pressure versus temperature, again we will we start with benzene at very
low temperature it was one and then we keep on heating it at constant pressure it comes to some
particular point and then there is change of phase since the change of phase occurs under
isobaric, isothermal conditions, so therefore 2 and 3 they coincide here.

Then again it starts to heat, it heats till a quite till the liquid changes to a vapor phase again
phases 4 and 5, if you recollect their transition from liquid to vapor that occurs at a particular
point in this particular case and then the vapor continues to be heated. In this particular way we
can keep on drawing these particular curves for different substance at for different specific
volumes we can keep on drawing these particular curves.

And as we go on for higher and higher volumes we rather sorry higher and higher pressures, we
find that these particular curves are there just the location of this saturation curves are different,

246
again we again we join the different saturation point to give us saturation curves and under that
case what do we get? We get something of this sort, this is going to be this, I could not throw
them well, this and then this is going to be something of this sort, right?

So we find that we have a solid vapor transition curve which is nothing but the sublimation
curve, okay. Now below this sublimation curve we have a solid here and then we have a liquid
vapor this is known as the vaporization curve, it separates, this is the solid zone, this is the liquid
zone and this is the vapor zone. So it separates the liquid and the vapor and we also have a fusion
curve which marks the transition from the solid and liquid.

Again I would like to remind you that the fusion curve it can continue indefinitely as I have said
but this particular curve it has to terminate at point C which is given by Pc, sorry yes Pc, Tc. So
therefore this terminates here and this does not terminate and the sublimation curve also
terminates all these 3 they terminate at the triple point which reduces to a point here, since this
triple point is marked by a particular T triple point and P, right?

(Refer Slide Time: 3:15)

247
(Refer Slide Time: 3:24)

Where I would also like to mention one more thing which will be evident from this particular
graph we find out that more or less the slopes of sublimation Curves and vaporization curves are
always positive, the slope for sublimation curve is steeper as compared to the vaporization curve
and usually the slope of the fusion curve is also positive, it is negative only for certain
substances.

248
(Refer Slide Time: 3:34)

For example the most common substance that we come across which is water, now why all these
things happen? You just you need to remember that water contracts on freezing but exactly why
it happens and all the characteristics of the PT curves that we’ll be dealing after a few more class
where we will be dealing with equations like Clausius Clapeyron equation, Clapeyron equation
all of these are going to tell you going to predict the slopes of this Pt curve and also the reasons
for number of things that we observe in this particular case.

249
(Refer Slide Time: 4:07)

(Refer Slide Time: 4:11)

Now by this time you must have realized that this particular point the critical point here and the
critical points which I have shown here these critical points they are constant characterizing the
substance just like triple point’s, right? So therefore for every substance if you want some very
reliable property then definitely you would like to locate the triple point. Now as it is quite
evident that it is easier to measure pressure temperature as compared to volume, so mostly what
we do? We measure the critical pressure temperature and then try to predict the critical molar
volume from these things, right?

250
(Refer Slide Time: 4:51)

So therefore in order to determine this there is a very popular experiment which I will be
mentioning in very brief which is known as the disappearing meniscus experiment, in this
experiment it is a very common experiment which are performed in the chemistry laboratory.
Usually this is performed with SF6 since its critical point is near about 45.5 degrees centigrade
or something. So therefore it is easy to perform the experiment, what do you have? You have a
closed sealed tube.

Now when you have a closed sealed tube what does it mean? And then here you introduce some
portion of the liquid that you want whose critical point you want to measure. Usually the
experiment as I have said their performed with SF6, right? Now what happens? In this particular
system more or less you have introduced some amount of liquid, so therefore the mass remains
constant and then since it is sealed from all ends so it's its total volume also remains constant,
right?

So therefore what does it imply? It implies that the molar volume of the liquid that also remains
constant, fine. Now under this condition if you start heating the tube, what you expect is going to
happen? Quite naturally they are suddenly of which we would like to say that the moment you
start heating the tube the liquid starts vaporizing and more and more vapor will be formed and
then at one point the entire tube will be filled with the vapor.

251
(Refer Slide Time: 6:38)

Now this will not always happen it will depend upon the molar volume with which we have
started for example if I refer to the previous figure it is going to be evident to you. Suppose I
refer to the previous figure when in this particular case it appears to you, yeah. If you just
observe this figure it all depends from where you have started the experiments.

Suppose your molar volume was somewhere here. Suppose your molar volume was say
somewhere here from here you start heating it, what do you what are you basically doing? You
have kept the molar volume constant and you are increasing its temperature, fine. So therefore
you keep on heating keep on heating what happens? You find that that you are shifting actually
towards higher quality lines.

252
(Refer Slide Time: 7:29)

If you draw quality lines here, you are shifting towards higher quality lines till you reach the
quality line x equals to one which is nothing but the saturated vapor curve and when you have
reached this curve you find that your meniscus it keeps on decreasing till at one point it
disappears and the whole thing is filled up with the vapor phase at you have entered your
superheated vapor region, fine.

(Refer Slide Time: 7:45)

253
(Refer Slide Time: 7:50)

Suppose you are starting with a very low molar volume, yeah with a very low molar volume you
have started. For example, say for instance you have started from somewhere here, right? Now
you keep on heating it, what are you doing basically?

(Refer Slide Time: 8:01)

Just notice you are following a vertical line, what are you doing here? You are going from a
higher quality line to a lower quality line and then you keep on shifting quality lines till you
reach x equals to 0 the saturated liquid curve and then if you continue the heating you have
actually entered into the sub cool liquid region.

254
(Refer Slide Time: 8:25)

So in this particular case what happens? You find that your liquid meniscus it keeps on going up
and the, this continues till finally the whole thing becomes filled up with the liquid. Now suppose
you start the experiments with some particular volume of the molar volume of the mixture such
that the molar volume it corresponds to the critical point.

(Refer Slide Time: 8:52)

255
(Refer Slide Time: 9:17)

Now if you start heating it what will you expect? You are going vertical in the same way, please
note that you are following a constant quality line in this particular case, you keep on following
this constant quality line and continue the heating you will find that no matter how much you
heat your line is going your meniscus will be remaining fixed at one your particular location it
neither rises nor it falls, it simply remains constant at this particular portion and this thing it
continues till you are reaching the critical point the moment you have reached the critical point
the meniscus suddenly disappears.

And you know that this is the condition of, this is the critical condition at that particular
condition if you measure the temperature and the pressure you are able to locate Tc and Pc, you
can keep on performing the experiments you will find that the meniscus disappears comes back
means you are just operating at this particular point, you just increase the heat again decrease the
heat just decrease the heat again increase the heat. If you keep on continuing this you will find
that the meniscus keeps on coming and going but it remains at one particular location and this
corresponds to your Tc and Pc. So by this a very easy way of finding out the critical temperature
and once you have found out the critical temperature then from the critical temperature it is much
easier to find out the critical volume.

256
(Refer Slide Time: 10:19)

Just I will show you what exactly happens, suppose this is the critical temperature and suppose
this is rho or in other words we can also plot the critical the volume also in this particular case.
Exactly what are we, are we doing here? So therefore or in this plot rho versus T that is going to
be easier. So what do we do? Suppose for the liquid we are plotting the orthobaric density which
means the densities along the liquid vapor coexistence curve, what do we find? As we keep on
heating the liquid density keeps on decreasing, right? The liquid density, this is rho liquid it
keeps on decreasing because good the liquid expands and it also vaporizes at the same time.

What happens to the vapor? Much amount of vapor is being added and also we find that the
vapor it expands, so we find that the liquid density keeps on decreasing the vapor density keeps
on increasing till at T equals to Tc both the densities they become identical, right? So therefore
here also we find that that if you would be plotting the rho bar which is nothing but rho liquid
plus rho vapor by 2.

If you are plotting this versus temperature then we find usually this is the slope of this particular
curve and this particular curve it is usually linear in nature and the slope it can be obtained as rho
bar equals to A plus Bt, right? So from here you can find out A and you can find out B and then
you can find out rho bar at T equals to Tc, so this is a standard way of finding out T equals to Tc
or the critical conditions for pure component substances.

257
(Refer Slide Time: 12:31)

So apart from this I would like to say these are the basic phase diagrams but the phase diagrams
which we use mostly for our purposes as you already know that those phase diagrams are
primarily apart from these phase diagrams they are primarily the temperature entropy and more
common is the enthalpy entropy diagram which is commonly known as the mollier chart.

The importance of mollier chart is the first thing is, it can be plotted on an elongated scale and
here if you observe you find that we have the isobaric lines and we have the these are the
isobaric lines we have the, these are the isothermal lines and we have lines of the constant quality
lines.

So therefore this particular curve it is complete in all respects and if we know the pressure and
temperature of the different points and then we are in a position to go to find out its enthalpy to
find out its entropy and if suppose the conditions are such that it lies within this 2 phase dome
then in that case we can also find out the dryness fraction or the quality of the mixture.

So therefore usually we find that the mollier diagrams are the diagrams which we use most
frequently in order to find out the properties measurable as well as non-measurable of pure
component substances in the liquid phase mostly in the liquid phase, vapor phase as well as in
the liquid vapor region.

258
Now this was all about the graphical representations of PvT behaviors apart from this naturally
as I have already mentioned we need to have some equations when I tell you about the equations
the first equation which comes to your mind naturally for the gas phase is nothing but the ideal
gas equation this is the first thing which comes to your mind, isn’t it?

(Refer Slide Time: 14:22)

(Refer Slide Time: 15:05)

And we know as we all know that the ideal gas equation which is so very known to you for one
mole it is Pv equals to RT we find that this is applicable only for this you all of you know this is
applicable for very low not even low very low pressure and high-temperature and therefore this

259
particular quantity z which is known as the compressibility factor which is nothing but Pv by RT
this is equal to 1 for ideal gases and this is not equal to one for real gases, this you already know
and I have just depicted this in the figures here where we show that that z is not equal to one for
any particular real gas and as the temperature keeps on decreasing the deviation from ideality
increases it is also a function of the measure of the gas.

So naturally since ideal gas equation does not suffice, what is the next thing? We would like to
go for the different equations which would be predicting the PvT Behavior of the gases. Now
what are the different equations that we have? The equations can broadly be classified in 2ways
the first one is known as the PvT Behavior of the gases.

(Refer Slide Time: 15:47)

So therefore they can be broadly classified in 2 ways, one is the cubic equation of state the
immediate example of this, all of you are aware of is van der waals equation and the other one is
the virial equation of state. When we are not sure whether the gas obeys any particular cubic
equation or the virial equation we often refer to the compressibility factor chart which we will be
discussing after we finish discussing these particular equations of states.

260
(Refer Slide Time: 16:34)

(Refer Slide Time: 16:43)

Now why is the cubic equations known as cubic equations or in other words to be more specific,
what is the specialty of cubic equations? Now the general form of cubic equation don’t get
confused these are the whole lot of equations that have been proposed by different researchers
the general forms being these but we are not going to deal with the host of these equations for
your purpose be comfortable with primarily Van der waals equation which you already know and
this equation is already known to you the functional form is already known to you and then there

261
are 2 other equations which we will be using for conditions where the van der waals equations
fails, right?

Now the functional form if you observe you will find for all of these what is the functional form?
There are usually 2 constant equations where the constants are denoted by small a and small b
for Van der waals, capital A capital B for Redlich Kwong and alpha and beta for Peng Robinson,
we find that volume correction term is the same for all of them it's only with the intermolecular
interaction which is represented by more and more complex equation in the different equation
states.

(Refer Slide Time: 17:37)

Now let us see under what conditions each particular equation is applicable? This will be evident
if we observe say the experimental pressure versus density data for ethane, okay. Here what do
we observe? If we observe see this particular line it is the solid line it is the experimental curve
which has been observed. Just note the Van der waals equation of state, we find that beyond 200
atmosphere it fails completely and under this particular circumstance we find more or less both
the Redlick Kwong as well as the Peng Robinson equation of state both of them are applicable
for this particular condition, okay.

And both of them are quantitative and if you observe closely you will find that the Redlich
Kwong equation it appears to be slightly better as compared to the Peng Robison equation of
state. Just to verify this I have given you one particular problem for calculating the molar volume

262
of ethane this is also for ethane at 300 Kelvin 200atmoshperes, please do it with 3 particular
equations of states and compare the answers that you get and compare the values with the
experimental values from where you will be in a position to understand that Redlich Kwong
equation is definitely better than Van der waals equation at higher pressures.

The other thing which I would want you to do is, there is also another thing let me explain
before I go for this particular equation of state, now what is the specialty of cubic equations of
state? The first specialty is, if you write it down in the form of the volume then what do you get?
Then for all the equations you get a type where you find that it is cubic in volume. Now when
this cubic in volume it should give you 3 roots?

(Refer Slide Time: 19:46)

Now usually what do we find? We find that if you are working at conditions greater than T
greater than Tc, P greater than Pc for this particular case you will find that there is just 1
physically realizable root I should write it down as 1 physically realizable root the other roots are
complex just to give you a feel I have already given you 2 problems, do those 2 problems then
you can take a few other problems you will find that when you are operating conditions much
higher than the critical conditions than under those conditions you get one physically realizable
root.

263
(Refer Slide Time: 20:32)

Now when you are working for conditions where T is less than Tc for that particular condition
what do you get? That means what I mean to say you are working in this particular zone, in this
particular zone here this particular zone here, you will find that under this condition you get 3
roots, right? One root the lower root it corresponds to the liquid volume the other root
corresponds to the vapor volume and there will be an intermediate root which is the spurious
root, why?

Because suppose you are trying to predict the behavior of the Pv behavior using any equation of
state, it can be Van der waals, it can be Peng-Robinson it can be Redlich Kwong equation of state
you will find none of the equation of states can predict this discontinuity in behavior, what they
do? They predict a continuous curve, what is a continuous curve? Curve is something of this
sort, this is what they predict, right?

Now we find that suppose for T less than Tc and P equals to the saturated pressure, which
corresponds to this particular pressure under that condition if we work then we get one particular
physically realizable root, what is a physically realizable root? It has to be positive, it has to be
real and very important which it's not very evident it has to be greater than B, right?

The total volume of the gas has to be greater than the volume occupied than the molecules, fine.
So therefore you will get 3 physically realizable roots if you are working within this liquid vapor
dome, you will get one physically realizable root when you are working in this particular

264
domain. So there and then we find that the isotherm or the isobar is predicted in this particular
way where we find that the smaller volume it corresponds to the liquid volume this corresponds
to vL, the large volume corresponds to Vv.

Now this particular region what does this correspond to? This particular region it correspond to
sub cooled vapor which is a meta-stable state. What does this particular region correspond to?
This corresponds to a super heated liquid which is again a meta-stable state, right? So therefore
this region and if you find this region these 2 regions they are meta-stable states they can be
obtained under very special conditions and if you observe this particular portion you find that for
this particular portion del P del v at constant T is greater than 0 the slope is positive which
cannot happen.

So naturally this particular region it is a spurious region, this region cannot exist as a result of
which the intermediate root which you get has got no meaning. So therefore from here what do
we conclude? We conclude 2 things the strength of cubic equations of state is that they can not
only predict the volume or the behavior in the vapor phase they can also do so for the condensed
phase and more importantly for the liquid vapor coexistence region.

(Refer Slide Time: 23:56)

So therefore we can predict the molar volumes for liquid state, vapor state and the liquid-vapor
coexistence curve and just to show you 1 particular example I have given you one particular
problem here, no not this problem sorry, this particular problem where I have the liquid argon

265
and argon is exist in the 2 phases under this particular condition and the experimental density
molar density of liquid argon and vapor argon are provided here. I would like you to verify this
using the Van der waals, Redlich Kwong and the Peng-Robinson equation of state, remember for
all of this you are going to get 3 roots for the volume.

(Refer Slide Time: 24:25)

And the 3 roots as I have mentioned they correspond to 3 points joining the horizontal line such
that the area under and the area above the horizontal line are equal. So from each of this you will
get 3roots out of which the higher root will correspond to the vapor volume, the lower root will
correspond to the liquid volume.

266
(Refer Slide Time: 25:25)

And we find if you compare in this particular case is you are supposed to work out this problem,
if you compare you find that the answer for the Peng-Robinson equation of state is this, sorry.
The answer for the Peng-Robinson equation of state if you compare the different equation of
states then what do you get? You get that the Peng-Robinson, yeah. You get that for the Peng-
robinson equation of states whatever you have got these particular values they are closest to the
experimental values.

Redlich Kwong equation of state they are slightly further off and Van der waals equation of state
completely fails this is what I have tried to show in this particular curve. Van der waals equation
could not be plotted because it was out of the range, if you observe here this is the Redlich
Kwong curve, this is the vapor liquid coexistence portion the Redlich Kwong curve, this is the
Peng-Robinson curve it shows that the Peng Robinson it's slightly better than the Redlich Kwong
equation in the liquid vapor region.

So therefore among the different equations that I have suggested for the different equations what
do we find? We find usually that the Peng-Robinson equation of state it is better for the liquid
vapor coexistence curve, the Redlich Kwong equation of state is better for higher pressure
conditions you are requested to workout examples and see it for yourselves.

267
(Refer Slide Time: 26:28)

And the strength of cubic equation of state is that it can predict the saturated volumes along with
the vapor volumes. The weaknesses is that it fails to predict the condition in the liquid vapor
dome the situation I have shown here and I have also compared it with the experimental
situation, right? Here you get 3 physically realizable roots out of these 2 are meaningful and in
this particular case you get just 1 physically realizable roots.

Now the other things which I would like to mention about the cubic equation of state is as you
have already observed in the cubic equations of states what do we see? All of them comprise of
some constants, right? Now these constants are completely empirical constants. Now each
particular gas is going to have the different constants, it is very difficult to rather to find out or to
or to accumulate or rather to collect the different constants for the different gases, fine.

So what people usually try to do is they try to express the constant of the cubic equations, say for
instance for Van der waals it is a, b, for Redlich Kwong capital A capital B these things are their
they people try to express them in terms of the critical constant, why?

Critical constant are much easily measurable as we have said they are much important much
more constant as compared to other particular physical properties of pure substances, so in the
next class or rather I will just be discussing very briefly this is left as an exercise for you in order
to express the empirical constant in terms of the critical constants and then once we do it, we find
that from the cubic equations of state we arrive to a very interesting equation of state which

268
is known as a generalized equation which leads us to some more interesting results, thank you
very much.

269
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 15
PvT Behaviour (Contd.)

Well to continue with the discussions of the last class, what did we do? We first found out the
graphical representations of the PvT behavior and after the graphical representations of the PvT
behavior then we came to the mathematical representations of PvT behavior and then we
discussed we found out that there were different equations of states, the cubic equation the Virial
equations of state and we started discussing the cubic equations of state and there we found out
that the cubic equations of state they have got some very strong advantages.

(Refer Slide Time: 0:51)

270
(Refer Slide Time: 1:24)

The first advantage is that we can, as we have found out in the last class the first advantage was
it, since it is cubic in volume it is capable of predicting not only the behavior in the gaseous
phase but also in the 2 phase region as well as in the liquid region that was number 1 and then we
also found out the drawback in this particular case that in the 2 phase region we found out that it
was not capable of predicting the isotherm the nature of the isotherm it showed a continuous
curve we had discussed all these things.

271
(Refer Slide Time: 1:42)

Well, the next thing and I had also shown you the list of the different cubic equations which have
come up and for the present class I had already mentioned that these 3 equations they are the
most important equations which we will be dealing with as we have seen in all the 3 equations
there are 2 empirical constants which are characteristics of the particular depth and nature of the
gas and when we are dealing with mixture, naturally it is a function or it is a mixture property
which can be expressed in terms of the pure component properties as we shall see later.

Now for each and every gas if you are taking up say for example we Van der waals equation, we
need to find out a, b for everything, so we are just increasing the number of known parameters
that we need for when we are going to deal with PVT behavior of gases that is not a very
desirable situation.

So therefore in order to reduce the number of unknowns usually what is attempted is to express
the empirical constants in any cubic equation of state in terms of the critical constants because as
we have already seen the critical constants they are true constant characterizing any particular
pure substance.

272
(Refer Slide Time: 3:02)

(Refer Slide Time: 3:09)

So therefore there has always been an attempt to predict the empirical constants in any cubic
equation of state in terms of the critical constants and it is very easy to do it, say for instance if
you take up the Van der Waals equation of state, this is the equation, right? So therefore we have
one equation as P equals to RT by v minus b minus a by v square we can have another equation
if you differentiate it this can be one equation this can be the other equation, so from these 3
equations we should be in a position to find out a and b, right?

273
(Refer Slide Time: 3:46)

And these are at T equals to Tc, sorry. So from this we can find out a, b but a more
straightforward technique of doing this is that if we remember that at v equals to vc T equals to
Tc and P equals to Pc, so what we can do? We can just write down the equation in the cubic state
its v cube minus 3 v square vc plus 3v vc square minus vc cube equals to 0, from where did it
come? We know v minus, v equals to vc v minus vc equals to 0. So v minus vc whole cube will
be equal to 0 I have just expanded this, right?

And then I have compared this with the original the van der waals equation written in the cubic
volume, what is it? This is going to be v cube minus b plus RT by P v square plus a by P v minus
ab by p equals to 0, right? And here we know that at the critical point, what do we have? We
have Tc, we have Pc, we have Pc, we have Vc here, right? So therefore this equation has been
written down as a cubic in volume at the condition of v equals to vc, this has been written down
in the cubic of volume at the condition of v equals to vc.

274
(Refer Slide Time: 5:06)

(Refer Slide Time: 5:09)

So therefore both these equations should be identical or in other words the constants should
correspond to the constants should be equal in both the equations, so if we compare the
constants, what do we get? By comparing the constants we get in fact we get three equations by
comparing the constants, if you see we get 3 particular equations and we will find that 2 of the
equations they are same, so we get 3vc equals to b plus RTc by Pc this is one.

And then the next equation which we get is 3vc square which is equals to a by Pc and Vc cube
this is equals to ab by Pc, from these 2 equations immediately what comes out? From these 2

275
equations immediately we get b equals to vC by 3, right? We get this and if you substitute this
particular equation in this term, then what do we get? We get a equals to 3vc square into Pc,
right?

So we have expressed a and b in terms of critical constants but both of them contain Vc, at again
I repeat the same thing that temperature and pressure they can be much easily measured, more
accurately measured, more easily regulated so naturally we would like and it is much more easily
evaluated.

As we have seen Tc and Pc is much easily evaluated as compared to vc for which usually we
resort to the disappearing meniscus experiment which we have discussed in the last class. So
therefore we would like to express a and b in terms of Tc, Pc instead of Vc. So for that what do
we do? We take help of this particular equation and we know that in this equation we get 9b is
equals to b plus RTc by Pc, right? 3vc from there we get which gives us b equals to RTc by 8Pc,
fine.

And in the same way we can also say that a equals to 3vc square Pc we have found out again we
can substitute vc equals to 3b we can substitute it here and finally from there what do we get?
We get a equals to 27b square Pc which gives us as 27Pc RTc by 8Pc whole square, right? So
therefore a is obtained as 27 RTc whole square by 64Pc, fine. So therefore we could express both
of them in terms of Tc and Pc.

276
(Refer Slide Time: 7:48)

Now this particular exercise I have shown to you for the Van der waals equation of state. You are
advised to take up a similar exercise for both Redlich Kwong equation of state as well as for the
Peng-Robinson equation of state and you will find that for both the cases proceeding in the
similar manner we would be in a position to find out the empirical constants of the equation in
terms of Tc and Pc. So therefore this has enabled us to reduce the number of constants that we
required in order to characterize the properties of any particular gas but more important than this,
this exercise brings up very interesting results let us see what is the result that this exercise
brings up?

277
(Refer Slide Time: 8:32)

(Refer Slide Time: 8:39)

278
(Refer Slide Time: 10:06)

Again we start from the basic equation this is the basic equation, right? If we again start from the
basic equation and we substitute a and b in terms of what we have obtained. We have obtained a
and b if you substitute them in terms of this equation and this equation, what do you expect we
are going to get? From here what do we get?

For the first equation substituting a and b we get P plus 3Pc vc square by v square into v minus
Vc by 3 this is nothing but equal to RT, agreed? Now suppose I divide both the sides with PcVc
or I multiply with 1 by PcVc, suppose I do this then what do I get? Just observe the equation
very carefully I get P by Pc plus 3Vc by v whole square into V by Vc minus one third equals to
RT by PcVc this RT can also be written down as RT by Tc Pc Vc by Tc, can I not write it down
in this particular way?

279
(Refer Slide Time: 10:17)

Just observe this part, what does this show? P by Pc is nothing but it is known as the reduced
pressure, so therefore this can be when I rearrange this can be written down as PR plus 3 by vR
square VR minus one third equals to 8TR by 3. Do you understand the importance of the
equation that we have of obtained? Here what have we done?

280
(Refer Slide Time: 10:34)

We had started just note with the Van der waals equation it had two constants characterizing the
gas, so this equation should be having different forms for each and every gas that we take from
here where have we arrived at?

281
(Refer Slide Time: 10:45)

(Refer Slide Time: 11:10)

This particular equation which has no constants characterizing a gas, it almost appears like the
ideal gas equation Pv equals to RT where it is the same equation for any particular gas that we
use. So therefore we find that starting from the basic Van der waals equation, for starting from
this particular equation we find that we have arrived at a particular equation which is
independent of any particular constant, right?

282
(Refer Slide Time: 11:44)

And then so therefore this particular is expected to be same for all the gases or in other words we
can say that 2 gases when they are having the same Pr and Tr should have the same Vr, number
1. Number 2 suppose we try to calculate the z from here. From here if we calculate z, what is z
equals to? Again starting from the basic Van der waals equation this is nothing but z can be
expressed in this particular form and from here we find out again if you substitute a and b, what
is the final form that we that we get?

From here the final form which we get is Vr by Vr minus one third minus 9 by 8 VR TR, so
therefore what do we find? That z also at the same conditions of see what have they got? VR it is
a function of TRPR, z it is a function of apparently it appears it is a function of TRVR. So
therefore Z is also since VR is a function of TRPR z is also a function of TRPR so therefore we
find that in this particular way if we start substituting different properties which we will be doing
later we find that all of them are unique functions of TR and PR.

So what does this mean? This means that the gases when they are at the same conditions of TR
PR they exhibit the same reduced other properties or rather they have the same reduced values
for the different properties. Now this particular characteristic of the gas which has come out from
the, this can be obtained from any cubic equation of state and this particular property of the gas it
is known as the or it is given by the law of corresponding states this is known as the law of

283
corresponding states that all gases under the same reduced conditions exhibits all same type of
properties.

And from here we can also find out one more thing, what is ZC equals to can you tell me? ZC is
equals to suppose you calculate ZC from here we find that this is equal to PCVC by RTc. You try
to find it out from the basic Van der waals equation and we find that ZC for all gases it reduces to
a unique value which I have mentioned here it's 0.375 as obtained from Van der waals equation it
is sorry.

(Refer Slide Time: 14:24)

284
(Refer Slide Time: 14:36)

We find that the from the Van der waals equation if you are trying to do, we find that more or less
we find that z is obtained as a unique value which is 0.375 for Van der waals equation 0.333 for
Redlich Kwong equation 0.3 from the Peng-Robinson equation of state and the way I have
obtained this particular equation which is written here the way I have obtained this particular
equation and derived this equation it can be done in identical fashion for other cubic equations of
state.

285
(Refer Slide Time: 15:27)

Naturally the more complex the equation is what happens the more complex the reduced version
will also be and the more difficult the deduction is going to be but with any cubic equation of
state we find that we can finally express the behavior of any gas using the reduced cubic
equation of state in terms of the reduced parameters and those equations that do not have any
constants characterizing the gas and for each particular equation of state we find that they predict
a unique value of the critical compressibility factor the values which we have got our different
for different equations but each equation shows that ZC should be having a unique value
provided the gases correspond to the 2 this is known as the 2 parameters law of corresponding
states.

286
(Refer Slide Time: 15:50)

For example if I give you a very specific example say carbon dioxide and nitrogen, right? For
these 2 gases I can very well tell you that just like all the gases for these 2 gases say suppose at a
constant TR and say at VR equals to 20 TR equals to 1.5 both the gases they should have a PR
which is equal to 0.196, right? Because they obey the law of corresponding states for carbon
dioxide I know PCo2 equals to 14.3 atmospheres 1.9 liters per mole and TC, co2 actually this
should be PC VC and TC Co2 this is 456 Kelvin.

For nitrogen also we find the PC is going to be 72.87 atmospheres TC is going to be 304.14
Kelvin, VC is going to be 0.094 liters per mole, right? So therefore from here and we know that
at say VR equals to 20, TR equals to 1.5 they should have the same reduced pressure and this
particular state it corresponds to say if you calculate it out, you will find that this corresponds to
a pressure of nitrogen as 6.58 atmospheres, volume of nitrogen equals to 1.8 liters per mole,
temperature of nitrogen 189 Kelvin.

And in this particular case your, for the PCo2 this corresponds to I’m sorry these were the
corresponding states I am extremely sorry the PC these are the states which correspond to this
and the PC VC for these are 72.877 atmospheres TC equals to 304.14 Kelvin, VC equals to
0.094 liters per mole, right? And for nitrogen the PC these are the 2 corresponding states I am

287
extremely sorry. For nitrogen this is going to be 33.55, for nitrogen this is going to be 33.55
atmospheres VC is going to be 0.09 liters per mole and the TC is going to be 126.2 Kelvin, right?

And so therefore the corresponding states are what? Corresponding states this is PC VC for CO 2,
this is PV VC for nitrogen and from there we find that carbon dioxide at 14.3 atmospheres 1.9
Litres per mole 456 Kelvin that corresponds to nitrogen at 6.58 atmospheres 1.8 liters per mole
and 189 Kelvin. So from here what do we find? We find that for any particular gas if we know
the critical constants then for that gas at any particular condition we should be able to find out
the reduced parameters we know that even an unknown gas will have for the same reduced
parameters, say for the same reduced TR and PR we will have all the other reduced properties
the same.

So therefore what we can do from here we can find out the critical constants for any particular
unknown gas from the known critical constants of a known gas. So therefore this shows that we
have started from an approximate equation the cubic equation but the cubic equation what have
they shown us? They show us that the properties of all gases are same if they are compared at
conditions at same conditions related to the critical point or if they are compared under the same
reduced parameters and this is a very powerful law this is given or this particular law of
corresponding states is exhibited by all the cubic equations of states that we have the functional
forms are of course different.

288
(Refer Slide Time: 20:22)

So from there we got an idea that for all gases there should be a relationship as z is a function of
TR PR this is given by all cubic equations of state but mind it the functional form which is given
that is different but definitely that such a relationship exists this is evident from all cubic
equations of state and the functional form for that we need to resort to experiments. So from
resorting to experiments we found that this is indeed a characteristic which is shown by different
particular gases and for all the gases what do we find?

289
(Refer Slide Time: 21:20)

We find that indeed for all the gases we find that it’s really true that z it has a unique relationship
with PR and TR and for all for wide number of gases this particular graphs have been plotted
these graphs are available in any book of thermodynamics and we find that from these particular
graphs they have been drawn for 2 different conditions of PR one is for a lower PR and the other
is for a higher PR, we find from these 2 equations indeed they have been tried for a number of
gases and indeed z is a unique function of PR with TR as parameters and they hold very good for
non-polar gases and for similar molecules of the gases these are extremely applicable.

Okay, more or less we find that this particular law, this law of corresponding states it appears to
fail near saturation curve this is something important it appears to fail near the saturation curve
and also at the critical point apart from these 2 places more or less we find that the predictions
they are within plus minus 5 percent for non-polar and slightly polar gases. More or less we find
that other than these 2 conditions they can predict the law of corresponding states holds within
plus minus 5 percent for non-polar and slightly polar gases remember deviation is large for
highly polar gases.

So therefore we find that this is a very powerful tool we have and people have often tried to use
this when? In absence of any other additional data and it is also very important to note that cubic
equations they are approximate in nature but the law of corresponding states holds for a large

290
number of gases usually for most of the conditions, now before I end the class I would just like
to place before you why this happens?

Why do gases obey the law of corresponding states the reason is if we take up any temperature
scale even the Kelvin scale that has been made by us the gas does not know or the numerical
value of the temperature which we assign to the gas is not known to the gas but the gas knows
it's critical conditions, it knows how far removed it is from the critical conditions. So related to
the critical condition it behaves identically, all gases behave in the same way at the critical point,
right?

So when it is equally removed from the critical point it behaves identically and that is the reason
why each and every gas it obeys the law of corresponding states now there is one thing I would
like to mention that if suppose we can make a finding which is totally leak proof absolutely
accurate then there is no need of any additional research that never happens whenever we find
something we find that there are some fallacies, some limitations which have to be overcome, so
therefore additional research goes on in this particular way, this case also it was the similar thing.

People found out that law of corresponding states is very powerful it can be applicable and the
most of the conditions except the conditions which I have mentioned and we find that more or
less the deviation it’s less than 5 percent for extremely non-polar molecules etc but it otherwise it
can extend to 10 to 15 percent as well.

So naturally people wanted better predictions, so for better predictions what was needed? They
thought that the deviation from the law of corresponding states it primarily arises because of the
intermolecular interactions between the molecules. So therefore if a third parameter can be
incorporated which would account for these intermolecular attractions or intermolecular
interactions than possibly a better law of corresponding states can be proposed.

291
(Refer Slide Time: 25:50)

Accordingly people started thinking about what can be the third parameter to be incorporated in
the law of corresponding states? And 2 schools have come up with 2 particular third parameter
for one particular school it was z it’s a function of TR PR and ZC. For the other particular school
which is a much more popular thing apart from TR PR they have introduced something known as
the acentric parameter. So in the next class what we do? We go for discussing the 3 parameter
law of corresponding states and we find that the 3 parameter law is more accurate than the 2
parameter law of corresponding states. So we go for the 3 parameter law when we find that the 2
parameter law is not very accurate we continue in the next class.

292
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 16
PvT Behaviour (Contd.)

(Refer Slide Time: 0:33)

Well, to continue with our previous discussion where did we end? I had mentioned that normally
this is very good but under certain circumstances we found that there is good amount of
deviation therefore to improve predictions people started thinking about any particular third
parameter which can be incorporated here which takes into account intermolecular interactions
that is expected to improve the 2 parameter law of corresponding states.

293
(Refer Slide Time: 1:11)

So people started thinking about the third parameter, what did they observe? The first thing
which people observed that if really all gases obey this 2 parameter law of corresponding states
then from here all gases should have the same value of Zc, so what they did? They started
finding out Zc for a large number of values which I have shown here.

294
(Refer Slide Time: 2:03)

If you observe you find that the Zc values for the different gases they are definitely close to one
another but there is a variation if you observe here that too these are mostly for non-polar
molecules the variation is from 0.26 to 0.3 the variation occurs in this particular way. So
therefore people thought that definitely the third parameter should be something or rather the
third parameter should be the critical compressibility factor where in for 82 compounds when ZC
was found out, it was found that ZC was varying from 0.23 to 0.30 with majority of T cases lying
between this particular situation.

So therefore it was thought that when any gas obeys the 2 parameter law of corresponding states
they should have a Zc average of 0.27 and whenever there is a deviation for this then Zc should
be introduced as a third parameter and accordingly the third parameter law of corresponding
states was introduced and lot of graphs of Zc were there and this was Zc equals to 0.27 any
deviation was taken into account by separate graphs, right?

This was one class of thought this particular class of our or rather this particular definition it was
given by the, it was given at the University of Wisconsin but the more popular approach is the
thing which was suggested by the University of California, what they suggested? They thought
that if the gases they correspond to the law of 2 parameter law of corresponding states then all
the gases they should have the same saturated reduced pressure saturated vapor pressure at the
same reduced temperature, right?

295
(Refer Slide Time: 3:36)

So therefore what this particular case again if I bring the 2, if this is applicable then all the gases
they should have the same PR saturated at same TR this should happen but people found out that
this is not the case, right? And this particular deviation was taken up as the third parameter in the
3 parameter law of corresponding states how was the deviation or how was this third parameter
brought about?

296
(Refer Slide Time: 3:56)

The third parameter was brought about by plotting if you find I have made it here by plotting log
of PR saturated with 1 by TR, this is actually PR saturated, right? So therefore they plotted log
10 PR saturated as a function of TR and then they found that for different gases the slopes of
these curves they were different and they assumed that definitely a perfect molecule should be
one with completely spherical molecules that should be a perfect gas they are assumed Argon to
be a perfect gas.

And then therefore they thought that the limiting value of omega should be such that omega
equals to 0 for perfect gases with spherical molecules like Argon. Accordingly they defined
omega in the way that I have mentioned here this was the definition of omega which gives minus
log 10 PR saturated at TR equals to 0.7 minus 1.0, why? Because for argon we find this term
becomes equal to one for argon and therefore omega becomes equal to 0 for argon. So they took
up omega as this is known as an acentric factor and this gives you the deviation of the
intermolecular potential of any molecule from that of a simple molecule or rather simple gas
with spherical molecules, right?

297
(Refer Slide Time: 6:26)

And this was taken up as the third parameter in the 2 parameter law of corresponding states
which gave us z equals to TR, PR and omega and the functional form relating z with TR, PR,
omega was something of this sort z where this is the z 0 predicted from 2 parameter model,
right? And plus omega z1 TR PR this is the functional form relating Z with TR, PR, omega
which gives you that for normal circumstances far removed from the critical conditions and the
saturation curves for more or less not maybe polar molecules z can be obtained from the 2
parameter law of corresponding states, when there is some deviation then we need to find out
omega.

Omega can be found out from the slope of PR, log PR saturated versus TR curve and then we
need another set of equations to find out just like z versus TRPR we need another set of
equations to find outside z1 as a function of PR with TR as parameter. So therefore along with z
0 additional similar curves were generated for z1 from which we could we can find out z0 and z1
for any particular gas evaluated its omega and accordingly calculate the compressibility factor
for that particular gas.

298
Here I have just for your reference I have presented the z1 curves as well along with the z0
curves if you observe we have this z0 curves here and we have the z1 curves here and the omega
curve here with all those things we should be in a position to predict the behavior of any
particular gas even if you do not know which equation it corresponds to simply by assuming the
3 parameter law of corresponding states for greater accuracy and 2 parameter law of
corresponding states for most practical applications.

Now this was all about the law of corresponding states and the use of compressibility factor
charts and the cubic equations of state we need to remember that the cubic equations of state
after all they are approximate in nature, right? Despite all the great advantages they are
approximate in nature number 1 and they have certain empirical constants. Now a better
theoretical basis or based on a better theoretical basis the real equation of state has been
proposed.

299
(Refer Slide Time: 8:22)

What is the virial equation of state? Let us see possibly all of you are aware of this particular
equation of state it can be written in a pressure explicit form as well as in a volume explicit form
as we have noted here and we find that these particular constants this BC etc this particular
constants they give us the interaction between 2 molecules 3 molecules etc. Now if you observe
this equation what do you find?

The first thing is, this equation has got some theoretical basis it is not entirely empirical, the next
thing is you find that more or less for very low pressures high-temperatures you can neglect all
the terms Pv by RT becomes equal to one, so it reduces to the ideal gas behavior. When it starts
deviating from ideality the deviation is not very high we can just include the first parameter here,
when the deviations are still higher we can include the higher-order terms.

So therefore depending upon the situation we can include more number of terms and thus
incorporate the extent of deviation from non-ideality just by using one particular equation. Now
the first thing is and the other important thing I also mentioned that the constants can be
evaluated from intermolecular potential even when we do not have any particular experimental
data. Now usually for the conditions which we come across we find that usually the first constant
the B and the B primes they’re mostly important.

300
(Refer Slide Time: 10:12)

We usually for most practical purposes we can use that virial equation by truncating after the
second term. So therefore this particular portion or this particular portion usually is sufficient for
us and the relationships between the different constants have also been mentioned in this
particular equation, right? So therefore there is not much that I have to say about the virial
equations of state there is just certain thing which I would like to mention regarding the first
constant which is usually much more important and which is of much greater concern for us.

301
(Refer Slide Time: 11:03)

Now if you observe this particular constant, what do we find? There are certain important things
that I would like to mention, first thing is if you take up this particular equation, what do you
find? You find that your B, this is simply nothing but limit rho tends to 0 del z del rho at constant
T equals to 0, right? So therefore what does it imply? And we know this B prime this is the
relationship between B and B prime is that B prime this is equal to B by RT, fine.

So therefore if suppose we are plotting from this particular equation if you are plotting z versus
P, what do you get? You get a linear equation you should be getting a linear equation at low to
moderate pressures where we can just consider this part at low to moderate pressure we find that
if we plot z versus P we should be getting a linear plot whose slope should be given by B prime
or whose slope should be given by B by RT.

302
(Refer Slide Time: 12:04)

So therefore if we are plotting in this particular case then we find that the slope in this particular
for each of the equation, what is the slope equals to? Slope is nothing but equal to B by RT
remember one thing all these Bs and the Cs etc, all of them are functions of temperature and they
are independent of the pressure or the density of the particular gas.

Again we find that if we plot this particular B, this is B basically if we plot this particular B with
temperature we find that for very low temperatures the B is negative for all the gases and then
under one particular temperature we find that the B becomes equal to 0, what does it imply? For
low to moderate pressure it implies that definitely the B 0 means the gas is behaving ideally, so
therefore it appears that at the condition where the temperature where V equals to 0 the gas
behaves ideally it’s actually not so.

At the condition of the temperature where B equals to 0 the attractive and repulsive forces are
equal to one another and this happens at a specific temperature which is again a property of the
gas and this particular temperature is known as the Boyle temperature often denoted as, this
temperature for each particular gas this temperature is denoted by TB.

For this particular case we find that the attractive and the repulsive parts of intermolecular
attraction cancelled and the gas appears to behave ideally, provided we have neglected the effect
of virial coefficient beyond the second. So therefore all these things are applicable for low to
moderate pressure. So naturally finding out the first virial constant it’s quite important and there

303
is a very easy way by which we can find it out again for understanding the very easy way we
refer to the basic equation, what is this?

(Refer Slide Time: 14:04)

This is equal to B prime this is equal to 1 plus B by RT into P, agreed? Now if you multiply both
sides where RT by P, what do we get? We get V equals to V0 plus B as a function of, it’s a
function of T, what is B equals to can you tell me? It’s nothing but equal to V minus V0 which is
nothing but equal to the residual volume for that particular gas at the pressure at any particular
desired pressure.

So therefore we can find out data on the second virial coefficient from the data of residual
volume if it’s available at any particular pressure conditions provided we can neglect the
contribution of third virial coefficient at higher terms. So therefore these are some of the few
things that I wanted to tell you but before I end there is something interesting that I would like to
mention.

I have already told you that when we are using the virial equations of state, when we are using
the virial equations of state I have already told you that usually we are dealing with the first 2
terms or we come across the first virial coefficient which is important and I have also highlighted
just because the first virial coefficient and one more thing for most of the cases this particular
form is much more convenient. So we are mostly dealing with the coefficient B and I have also
shown you the physical significance of B.

304
Now there is something interesting which I would like to tell you, while for normal
circumstances the virial equation with 2 terms is sufficient at the critical condition we need at
least 3 terms for a proper representation of the PvT Behavior, let us see how and why? I would
like to show you that how near the critical condition we use at least the third virial coefficient for
a correct prediction of the Pv behavior near the critical point.

(Refer Slide Time: 16:27)

Let us see what happens if we start with just the 2 equations of state we truncate it after the 2
equations, what is it? This is RT by v plus RT B by V square, right? What is del p del? Sorry this
is P equals to, what is del P del V at constant T? It is minus RT buy V square minus 2 RTB by V
cube. What is del 2 P del v2 at constant T? This is nothing but equal to simply I am doing I am
performing differentiations and nothing else I get the equation as this.

What happens at the critical point? At the critical point my T equals to Tc, this T equals to Tc, so
accordingly here I should be getting I would be putting Tc Vc I will be substituting all the
constants with the critical values, right? So once I do this and we know that at T equals to Tc
both of them are 0. Now just observe the equations closely from this equation, what do you get?
You get B equals to minus Vc by 2.

From this equation what do you get? Simply solve it and you find B equals to minus Vc by 3, so
therefore you are getting inconsistent results if you are just considering 1 particular coefficient at
the critical point. So therefore you get inconsistent results and you need to consider more number

305
of terms for a correct prediction of the PvT Behavior near the critical point, what do we need to
do then? For this particular case we if we consider more than one particular critical constants
then we are dealing with this particular equation, right?

So in this case if we are performing del v del T, we get minus RT by V square minus 2 RT B1 by
v cube minus 3RT B square by V to the power 4, again if we differentiate it once more we find
out this equation becomes 6 RTB1 by v to the power 4 minus 12 RTB square by v to the power
of 5, right? At the critical point again I would like to substitute everything with the critical
constants, okay.

And then what do I have? I have got 2 particular equations B1 and B2 in this particular case
sorry B and C I had written them in the form of B and C, this is B and this is C, I am sorry. This
is B and this is C. So therefore from these 2 equations now we will find that we have from these
2 equations we will find that we have come to the state where B is equals to minus Vc and C is
equals to Vc square by 3. So therefore we find that what the critical isotherm if you substitute
these 2 values we find that Pv by RT which is nothing but Zc this reduces to 0.33 PcVc by R this
reduces 0.33 at the critical point.

So from here what do we deduce? I deduce that it is fine to use the virial equation of state with
just 2 terms for conditions far removed from the saturation curve and at the critical point but for
conditions near the critical point in order to obtain consistent PvT behavior we at least need to
consider 3 terms in the virial equation of state, right? So with this more or less I have completed
the some of the discussions on the PvT.

306
(Refer Slide Time: 20:51)

One more thing suppose I would like to mention is the generalized equations in terms of the PvT
behavior or rather in terms of the virial coefficients. Now here we find that from generalized
compressibility factor chart just the way I have deduced the compressibility factor chart from the
cubic equations of state we should be able to predict the virial coefficients also from the
generalized charts. So we know that z equals to one at P equals to 0.

So therefore we also know that at low pressure z it is a linear function of P at constant T, right?
So for low to moderate pressures, what do we have? We have Pv by RT this is equals to 1 plus B
prime P this is equals to 1 plus BP by RT which can be written as one plus B Pc by RTc PR by
TR. For non-polar molecules we can write it down as BPc by RTc equals to B0 plus omega B1
and when you substitute this particular equation in the equation for z, we get z equals to 1plus
BPc by RTc which is equal to one plus B0 PR by TR plus omega B1 PR by TR, right?

So from Pitzer correlation we know z is nothing but equal to z0 plus omega z1 from where we
can say that z0 is nothing but 1plus B0 PR by TR and z1is nothing but B prime PR by TR and
from here we can also deduce that B0 can be obtained as 0.083 minus 0.422 by TR to the power
1.6 and B1 can be obtained as 0.139 minus 0.172 by TR to the power 4.2, now these are the
generalized virial coefficients called coefficients correlations which are applicable at low to
moderate pressures for non-polar gases only. So therefore this reduces us to the rather so that we

307
can also deduce the law of corresponding states using the virial equation just the way that we
have done from the cubic equations of state.

So with this we would like to do a few problems so that you get much more conversant with PvT
behavior of gases and then we proceed to find out how the PvT behavior of gases can be used in
the equations to predict the enthalpy, entropy, internal energy, Gibbs free energy, Helmholtz
energy and host of other new parameters that we will be introducing from the PvT behavior of
gases? So this is all for the day and we proceed and we do a few problems in the next class.

308
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 17
Tutorial

(Refer Slide Time: 0:25)

Well, let us see the first problem. The first problem is written here we need to estimate the
pressure in a tank of 1 meter cube filled with100 KG methane when left in the surroundings
temperature is 40 degree centigrade, what are the data which are given to you? The data of TC
PC and omega are given to you. What is the gas? It is methane; you know the molar mass of
methane, what is the molar mass of methane?

309
(Refer Slide Time: 0:52)

It is very, you can always find it out this is equals to 16 into 10 to the power minus 3 KG per
mole, right? TC PC and VC are given, so quantity of methane in the gas CH4 this is equal to 100
by 16 into 10 to the power minus 3 which is 6.25 kilo moles you know it, fine. What is the molar
volume of methane? VCH4 this will be equal to 1 by 6.25 into 10 to the power 3, fine. This is
nothing but equal to 1.6 into 10 to the power of minus 4 meter cube per mole.

What is TR in this particular case? The temperature is given as 313.15 Kelvin by 190.7 Kelvin
which is 1.642. So therefore you know TR and what is z equals to that is equals to Pv by RT your
P is given, v you have already found out this is the V the P is already given it is P is not given
sorry, R is given P is given at 40 degree centigrade. So if you substitute all these terms what do
you get? You get 46.41 into 10 to the power of 5 PR into 1.6 into 10 to power minus 4 divided by
8.314 into 313.15 you get z as 0.2852 PR, right?

See one thing what are the data you have at hand? With the data you have at hand you are
supposed to find out pressure, agreed? You know TR you can find out z as a function of PR,
agreed? But and you also have a compressibility factor chart which gives you z as a function of
PR for particular Tr, so in this particular case if you have a Tr, Tr of 1.642, so in your particular
condition will lie at any particular point on this particular isotherm, agreed?

But where it is going to lie you do not know. You just know it is going to lie on this particular
isotherm you need a trial and error method to find out exactly where it lies? It is a very difficult

310
thing to do without anything you cannot do a trial and error method but at the same time you
know something else, you know that z has a linear relationship with Pr as well with the slope of
0.2852, so what do you do?

(Refer Slide Time: 3:28)

The best way is to do is you draw a straight line from the origin with z equals to 2.582 Pr and
you see where it intersects with the isotherm at Tr equals to 1.642. So naturally this is going to be
the point in question and for that particular point in question you know what is z0? You find that
the z0 is nothing but 0.87.

311
(Refer Slide Time: 3:52)

So naturally you from here you know that at TR equals to 1.642, what is your PR going to be?
Your PR is going to be z by 0.2852. So you know the PR is going to be 3.05, right? So you know
PR you know TR and for better accuracy you would like to refer to the z1 versus Pr Tr chart.
From the z1 versus PRTR chart if you since you know PR and TR you can find out z1, the z1
you will be finding out just check it up, this is data I am giving you this is nothing but equals to
0.23.

Once you know z1 you know z then what is the next thing? This is basically z0; you can find z as
z0 plus omega z1, omega is already given. So from there you can find out the final z using this
final particular z you can find out PR as z by 0.2852 you get PR as equal to 3.06 and from there
you can find out P and this particular iteration possibly it has to be done a number of times. So
from here you get a particular z again you start from this particular with z0 equals to 0.86 z1
equals to 0.23 PR equals to 3.06, TR equals to 1.602. You get a PR as 3.024, right? With this
particular PR as 0.324 there are several iterations may be which you have to do.

312
(Refer Slide Time: 5:33)

Finally you get your z as your z0 as 0.86, z1 equals to 0.23 Pr at Pr equals to 3.204, TR equals
to1.642, from where you get your final PR as 3.024 when the calculations converge from where
you get your P as 140.34 bar. So this was the first problem.

313
(Refer Slide Time: 5:55)

(Refer Slide Time: 6:02)

Let us go to the second problem it is a pretty straightforward problem you have a cylinder of 0.1
meter cube, so your volume is 0.1 meter cube, the pressure is also given 8.25 mega-Pascal,
temperature is given as 25 degrees centigrade you are required to find out the mass of ethylene
and you are supposed to do it by the Van der waals equation of state 2 and 3 parameter
compressibility factor chart, right?

So for the Van der waals equation a is given b is given but it has to be a trial and error technique
you need to find out, see you know the total volume if you can find out the molar volume then

314
you can find out the mass of ethylene inside this. In order to find out the molar volume you need
to write down the Van der waals equation in the cubic form, once you write it down in the cubic
form you need to solve it iteratively.

Now remember one thing when you have to do iterative calculation, first guess which you should
be using is the value of v obtained from the ideal gas state, right? So therefore that will be your
first guess value for solving this particular cubic form of equation. So therefore from the initial
guess value you get v equals to 1.196 into 10 to the power minus 4 meter cube per mole this
initial value, the initial value has been obtained from the ideal gas equation.

Once you know the molar volume so then from that you can find out the quantity of ethylene as
the total volume divided by the molar volume this has been obtained by solving the cubic
equation of state the first guess value for the cubic equation of state was from the ideal gas value.
So therefore from here you get that the total moles of ethylene in the cylinder was this and the
total mass of ethylene was 23.411 kg provided ethylene obeys the Van der waals equation of
state.

As far as finding it out from the 2 and 3 parameter compressibility correlations what you need to
do? You need to find out TR, find out TR it will be equal to this, find out PR this is going to be
1.6123, since you know TR PR in this case there is no need of any particular trial and error you
can very well find out z0 from the graphs which I have given.

You can also find out z1 the graph is also provided here, once you know these 2 you can find out
z omega is also given I believe, so you can find z as z0 plus omega z1, substitute these 2 and
omega is 0.089 from there you get z. Once you get z you can find out V from this particular
equation. The V from the 2 parameter law it comes out as this is the molar volume 9.765 into 10
to the power of minus 5 meter cube per mole, from where you get the mass of ethylene it is
obtained as 28.67 kgs and from this is from the 2 parameter laws.

Similarly if you proceed with the 3 parameter law you get the mass to be 28.48 kgs, it is worth
noting that there is slight quite amount of difference between the Van der waals prediction and
the 2 and 3 parameter law of corresponding states, naturally this particular deviation it occurs
due considering the approximate nature of the Van der waals equation as well as the inaccuracies
inherent in the 2 and 3 parameter law of corresponding states. It is also important to know that

315
the predictions from 2 and 3 parameter models they do not vary much which automatically
implies that we can safely use the 2 parameter law of corresponding states provided we are not
very near the critical conditions and not very near the saturation curves.

(Refer Slide Time: 9:53)

Now let us go to the third problem, a slightly different problem. What do we have here? This is a
problem on the virial equation, what do we have here? We have methyl chloride at 100 degrees
centigrade the first 2 coefficients are provided and we would like to find out the work required
for reversible isothermal compression of one mole of this Methyl chloride from say one bar to 55
bar at 100 degree centigrade and you are required to work it out both in terms of the Berlin form
of equation and the Leiden form of equation or in other words from the volume explicit form and
as well as the pressure explicit form.

316
(Refer Slide Time: 10:41)

This is a pretty simple problem you have got B1v you have got B2v you have got T you have got
P1 you have got P2, right? From here again you start with a v initial value remember one thing v
initial value will be nothing but RT by P1, from there you are required to find out V final, what is
W equals to? Minus of integral Pdv, you know Pv by RT is nothing but 1 plus B by v plus c by v
square, agreed? So we can easily substitute this in terms of P, in terms of P this equation
becomes RT into 1 plus B by v plus c by v square into 1 by v, right?

So therefore we are going to substitute it here, we are going to get this as RT integral v1 to v2 dv
by v plus B dv by v square plus C dv by v cube into this equation from where we get the work
done as 12.62 kilo joules per mole, right when we are using this particular equation, fine. Now
suppose we use the other equation then in that case we need to find out B prime and we need to
find out C Prime, we know the equations it is B by RT in this particular case also we know the
equation it is C minus B square by RT whole square.

We can find these 2 values, this value is going to be minus 7.817 into 10 to the power minus 3,1
by bar this is going to be minus 3.492 into 10 to the power minus 5, 1 by bar square, once this B
prime and C prime are found the rest is pretty simple we know Pv by RT equals to 1 plus B
prime P plus C prime P square, we know what is v equals to, again we can substitute it in terms
of RT by P one plus B prime P plus C prime P square, fine.

317
(Refer Slide Time: 13:00)

Or else we can write we can just expand and write it down, so therefore from this particular
equation we know what is dv dp equals to? It is equals to RT into 1 by P square plus C prime,
fine. And we know what is W? This is equals to RT integral P1 to P2 minus 1 by P plus C prime
P into dP which gives you the value of 12.596 kilo joules per mole. So therefore this is one
particular way that we can do it there is just one thing which I would like to point out.

318
(Refer Slide Time: 13:33)

(Refer Slide Time: 13:42)

If you if you observe the work done which you have obtained by using the same equation in 2
particular forms in one case you get the result as 12.62 in the other case you get the result as
12.596 they are not completely identical this suggests or this happened, why? Just because we
had taken the virial equation truncated after the third term, this shows that if you would have
taken additional terms the predictions would have been better in this particular case, right?

So in this particular way you can try several other equations for several other problems there are
a lot of problems which are available in a large number of textbooks you can try this out you can

319
try problems involving cubic equations of state, virial equations of states and we will also be
providing assignments to you for the same.

Now before I end there is one more problem also which I would like to do, it is not exactly on
the PvT Behavior of the gases but on phase equilibrium that you have already covered, what is
the problem?

(Refer Slide Time: 14:37)

320
(Refer Slide Time: 14:51)

A very interesting problem it tells you about the transition from the vitreous to the gray form of
selenium, right? What does it tell you? It tells you that suppose we have selenium in the vitreous
state we would like to convert it to selenium in the grey state. It is given that you are supposed to
find out Delta G for this at 25 degrees centigrade, fine. And what is the data given to you for
finding this out? The data given to you is that at 125 degrees centigrade they are at equilibrium,
what does it mean? It means that at 125 degrees centigrade Delta G equals to 0, fine. Using this
data you are supposed to find out this. How to do this? You want to go from this state to this
state.

Remember here it has say some particular value say some say Gse here it has got something say
Gse grey and here it has Gse v suppose, right? From here to here we wish to go we can devise
any path for it. So what do I do? Instead of this path I devise a path this, then this, then this. In
this path what are things that I know? Here the change is Delta G,1 here the changes Delta G2
and here in this case the changes Delta G equals to 0. So therefore can I write Delta G at 25 it is
equal to Delta G1 plus Delta G at 125 plus Delta G2?

Delta G1 is the Gibbs free energy change of selenium vitreous when it goes from 25 to 125
degree centigrade and Delta G2 is the change in the Gibbs free energy when the grey selenium
goes from 125 to 25 degrees centigrade, right? Now if I can find out Delta G1 I can find out

321
Delta G2 I can find out the Delta G transition at 25 degrees centigrade. What are the data I have
for finding these 2 out? The data which I have is the entropy of transitions data, right?

So from the entropy of transition data what do I know? I know Delta G1, this is equals to minus
SdT from 25 to125 this is for the vitreous case and since it occurs under a constant pressure so
therefore this just entropy change which happens and this is this Delta G is going to Delta G 25
this is 125 to 25 again minus S the grey variety dT.

You know the changes in dT, you know Sv, Sg which is independent of temperature. So from
here if you substitute the values what do you get? You get this is nothing but minus 7.4, 125
minus just I am substituting the values minus 10.04, 25 minus 125 which gives you the value as
264 calories. Transition for this transition Delta G is positive which automatically states that this
transition from vitreous selenium to grey variety of selenium it cannot be a spontaneous process
but the reverse case has to be spontaneous process.

So therefore it is obvious that at 25 degrees centigrade the vitreous form is going to be much
more stable and the reverse reaction is going to be the spontaneous reaction and at 125 degree
centigrade either of the states can be stable and if we know the condition at 1 particular state we
can find out the condition at the other state.

So this completes or rather I have tried to do a few tutorials on PvT behavior to give you a better
feel and since I had some time I thought I would just show you the importance of phase
equilibrium conditions to find out any particular transition or any particular state from some
known particular state at some other conditions we will be continuing our discussions regarding
the estimation of properties and then we will be going for mixtures from the next class onwards,
have a good day.

322
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 18
Property Estimation from PVT Behavior

(Refer Slide Time: 0:39)

323
(Refer Slide Time: 1:07)

Well, so today we are going to discuss the property estimation of real gases and other substances
of course but they will be primarily confined to real gases. Now to a quick recap before we go
for the property estimation part if you remember in the last class or rather few classes before this,
what did we do? We used the Maxwell’s equation in order to express certain non-measurable
properties in terms of measurable ones.

What were the different non-measurable properties? We tried to express the change in entropy
with change in measurable properties primarily T and P or T and V they are given by the first 2
equations here, then from there what we tried to do? We tried to find out the rather we tried to
express the change in enthalpy and the change in internal energy again in terms of the PVT
behavior of the gases and also the Cp Cv the specific heat property data and after that we also did
a few other things for example we tried to express Cp minus Cv in this term we tried to express
Cp by Cv and so on and so forth.

We had tried to derive several such properties in terms of the PvT behavior of gases and then we
found that unless we know the PvT behavior we will not be in a position to exploit these
equations further, so therefore we had discussed the PVT behavior of gases we had discussed
both the mathematical equations as well as the graphical representations and of course I have
mentioned that there are some tables as well which can be used for this purpose.

324
Now once we are aware of the PvT Behavior of gases now we can use whatever information we
have at hand in order to find out or rather in order to express your enthalpy, entropy, internal
energy etc in Cp Cv etc in terms of the PVT behavior of the gases.

(Refer Slide Time: 2:34)

Now I would also like to remind you that while we were discussing this we found out that the
behavior of real gases can very well be described with respect to the ideal gas behavior with the
definition of residual properties or departure functions.

325
(Refer Slide Time: 3:20)

We had discussed how the evolution or the definition of these departure functions if you recollect
what I had done was, I had initially proposed that on a Pv plot I have some particular or on a PT
plot if you take it was a TP plot as far as I remember. We have some particular gas at 1 and we
need to take it to 2.

So therefore I have proposed which path are we going to adopt, is it going to be this path or it is
going to be this path and then finally we found out that it does not matter whichever paths we
take but if we have to consider the availability of data then we would not accept or rather we
would not like to adopt either path AB or path CD rather we would like to go to the low-pressure
range, sorry.

We would like to go to the low-pressure range and at this low-pressure range we would like to
find out the enthalpy change or the entropy change due to temperature change and then from here
we would like to reach state 2. So this was the path which we had preferred the only reason being
that we knew the Cp or rather the Cp values are available at low pressure and the gas behaves
ideally and then I had also proposed another path for performing this particular computation.

326
(Refer Slide Time: 4:54)

And this path is already provided here, in this path most of the states that we have intermediate
except the first in the last state they are fictitious ones but it hardly matters because if we start
from one state and end in another state, if these 2 states are the same then it hardly matters
whether we are going from state one to state 2 through a fictitious part or through a practical path
it hardly matters. The property changes that we are going to compute they will remain the same
no matter what path we adopt if the initial and the final states are the same.

Of course if you ask me what is going to happen regarding the energy interactions? Definitely
they are going to be different but at the moment we are just dealing with property estimation. So
therefore the path of execution is not important for us, the initial and final states are important.

327
(Refer Slide Time: 5:41)

(Refer Slide Time: 5:49)

So instead of adopting this particular path and going from state one to state 2, I had proposed an
alternative path where we are converting the real gas at T1P1 an ideal gas at T1P1 then we
perform the entire computation that we desire over the ideal gas we take it from T1P1 to T2P2
and after that we convert this ideal gas to the real gas at T2P2.

So therefore all these intermediate conversion say from this state to this state and this to this path
2 and 3 they are very well-known and they are very easily computed the only hitch for the entire
problem was to execute path 1 and to execute path 4 both of these are fictitious paths they cannot

328
be performed in practice, what are these? We are converting a real gas to the ideal gas and we
would like to know the property change for this case.

Here we are converting an ideal gas to a real gas, we would like to know the property change for
this case. Here we would like to compute the property change at T1P1 here we would like to
compute the same property change at T2P2, right? So therefore if you observe what are we
basically trying to find out in path 1 and path 4? We’re trying to describe the property of a real
gas with respect to an ideal gas under the same conditions of temperature and pressure.

Definitely at high-pressure low temperature conditions we cannot have an ideal gas state. So
therefore the ideal gas state here and here they may be fictitious but it hardly matters because
what we are interested to find out is, if the gas could have existed as an ideal gas under any
particular temperature and pressure condition then what would have been the property
difference?

Now every time this particular thing it was motivated from the fact that if this property change
can be calculated then the rest of the computations can be done under the ideal gas conditions
and which simplifies the computation to a large extent. Second thing is suppose we can define
this particular property at any temperature and pressure we can use this definition and find out
this property for other conditions of temperature and pressure.

329
(Refer Slide Time: 8:28)

And therefore we had defined this particular property change here or this particular change we
had defined this as a departure function or a residual property and we had denoted it with Delta
M star where M is any extensive property we have already discussed these things. So if M is an
extensive property then Delta M star Delta means a difference, M star is the deviation from the
ideal gas value.

So this is nothing but M minus M0, where M is an extensive property and suppose this has to be
done at some temperature and pressure. So this and this are at the same temperatures and
pressures this we can define for any extensive property. In this particular lecture we will be
primarily be confining our attention to the estimation of enthalpy and entropy keeping in mind
that from this we can also compute other properties at our wish.

In this particular lecture we will primarily be confining our interest to the estimation of Delta h
star at any Tp and Delta s star at any Tp these are the departure functions for molar enthalpy and
molar entropy this is defined by h at Tp minus h0 at the same Tp this is defined as s at Tp minus
s0 at the same Tp. So once we are in a position to define these properties then definitely the rest
of the computations become quite easy.

So naturally this class will be confined to finding out or to predicting the departure functions
primarily of enthalpy and entropy. So that we can use this in order to compute the change of state

330
or rather to compute the entropy and enthalpy changes when a gas undergoes a change of state
from any initial to any final state. Now let us see how we are going to do it.

(Refer Slide Time: 10:35)

I think I had already discussed some part of it earlier and I’ll just be proceeding in finding out the
discussion once I have the expressions for Delta h star and Delta s star and before that I would
just like to remind you this also we have discussed earlier that this concepts of residual properties
or the equations which define residual properties they just show the true worth of ideal gas
equations and they just show that the ideal gas equation can serve as a convenient base to
estimate real gas properties and it is also important for you to remember that we can have
residual properties for gases for condensed phases as well as for gases but they are particularly
advantageous for gases, why?

Because if you’re trying to find out the residual properties of condensed phases with respect to
an ideal gas at the same temperature then pressure, the value of the properties will become very
large. So our idea was to find out the properties changes from ideal gas equation and then
incorporate small corrections to account for the deviation of the behavior of the substance from
ideal gas behavior.

Usually any such corrections they are desirable when the deviations from the ideal gas behavior
are small. So naturally if you’re computing these residual properties for condensed phases the
corrections will be very large and it’s not very worth in that particular case, they are primarily

331
useful for finding out the property changes in real gases. Along with that I would also like to
remind you that all these residual properties they arise just because the real properties are
functions of pressure and for most of the cases the ideal properties are not functions of pressure.

So therefore this difference between the real and the ideal gas property primarily for both the
cases the variation of temperature are more or less the same it’s the variation of pressure which is
important and for condensed phases they hardly varies the property hardly vary with pressure. So
with everything put together residual properties or departure functions you may use any term at
your convenience they are primarily applicable to real gases and real gas mixtures.

(Refer Slide Time: 13:18)

We will be dealing with property estimation for real gases in this class, real gas mixtures will be
dealt later on, right. And after this if you recollect we had also defined h minus h0, how had
defined h minus h0? We found that in variation with the temperature that remains the same for
both the cases, right. So therefore the residual property it primarily arises because h is a function
of P I believe I had already done the derivation earlier and therefore we know this expression this
expression is obtained from this particular equation that we had already discussed earlier.

So this is the expression for the departure function for enthalpy and this is the expression for
departure function for entropy, once we can find these 2 we can use any particular equation or
rather all the other properties can be found out from these 2 departure functions and maybe the

332
departure function for volume etc which can be easily obtained we can find out all other non-
measurable properties.

So therefore the main challenge of finding out Delta h star and Delta s star is to compute this
particular integral or to compute this particular integral, how to do this? One thing is we find that
this integral it is a function of let me see if I can write down this integral we find that this integral
it is going to be a function of the PVT behavior of gases, isn’t it?

(Refer Slide Time: 14:58)

I will just write it down for solving it, this is given as integral P equals to 0 to P, v minus T del v
del T at constant P dp or in other words very sorry this is Delta h star equals to, right? Or in other
words we can also write it down as integral P equals to 0 P vdp minus T integral P equals to 0 to
P del v del T at constant P dp.

In the same way we can also write down Delta s star as minus integral P equals to 0 to P, del v
del T at constant P dp minus integral P equals to 0 to P, R dp by P and R R ln p, right? So
therefore apparently if you look at these expressions what do you find? If you have the PVT
behavior of the gases it becomes very easy, we have already discussed the PvT Behavior of the
gases, right?

So therefore from those particular expressions if you can substitute rather if you can express v in
terms of p then you should be able to find out this integral and if you can compute del v del T at

333
constant P and then you can express it as a function of P you should be in a position to find out
this particular integral this is quite easy, right? But if you go back and if you observe the cubic
equations of state that we have discussed say for example the Van der waals equation of state, in
what way have did we express the Van der waals equation of state?

The Van der waals equation of state usually we write it down as RT by v minus b minus a by v
square. Same way we usually for the Peng-Robinson and the Redlich Kwong equation of state
also if you go back to our discussion you will find that all these equations are written in terms of
P equals to some function of v and T or in other words for all these equations we find they are
for these equations temperature and volume are taken as the independent variables and for all
these equations they express P in terms of T and v.

So therefore it is not very straightforward to find out the dependence of v on P from these
particular equations it will be much more useful if we can write down this equations in terms of
not dp but dv, let us see how to do it?

334
(Refer Slide Time: 18:06)

It is not very difficult, what do we know? We already know that from these equations well, I will
write down Delta h star once more so that it becomes easy for me to follow and I also have this
in front of me, right? What do we know already? We know that dPv this can be written down as
Pdv plus vdp, right? So therefore instead of vdp we can write it down as dPv minus Pdv, agreed?
Now also we know from the cyclic rule that del v del T at constant P, del T del P at constant v,
del P del v at constant T this is equals to minus 1, agreed?

So therefore we can express it as del v del T at constant P dp this is equal to minus del P del T at
constant v dv. So therefore we can substitute this equation this expression and this expression in
the expression of Delta h star, what does the expression become? Delta h star therefore it
becomes instead of vdp we can write it down as dPv we will put the limits later on minus pdv,
right? Instead of this we have these 2 terms minus (minus) becomes plus in this case plus T we
have del P del T at constant v dv.

335
(Refer Slide Time: 19:45)

(Refer Slide Time: 19:52)

So therefore if you have an equation of this particular form P equals to so and so you can very
easily find out del P del T at constant V you can find out pdv etc and you can compute this out,
only part remaining are the fitting of the limits of the integral. Initial limits were from P equals to
0 at very low pressure to a pressure of interest.

Now what is Pv at P equals to 0? At P equals to 0 the gas behaves as an ideal gas, so for that case
Pv becomes equal to RT you need to remember that we are all dealing with only molar

336
quantities. So therefore this becomes RT to Pv this becomes at very low volume it's v infinity to
the volume of interest this is also from very low volume to the volume of interest.

So if you solve it what do you get? It becomes Pv minus RT plus it is going to become plus v
equals to v infinity to v, T del P del T at constant v minus P dv, agreed? And this particular term
Pv minus RT, Pv is nothing but equal to, for ordinary gases this is going to be z RT minus RT. so
this can also be written down as RT into z minus 1 plus the term which we have written down,
isn’t it? So therefore from this particular equation we find in a very simple form once if you
know the equation of state we are in a position to find out Delta h star from this equation.

(Refer Slide Time: 21:52)

So therefore we find that delta h star it can be expressed as I just write it down once more for
your convenience integral v equals to v infinity that is a volume of the ideal gas to v T del p del
T at constant v minus P dv. Similarly suppose we start working with delta s star at any particular
temperature and pressure it is nothing but equal to s minus s0 at that temperature and pressure,
what is the expression?

337
(Refer Slide Time: 22:26)

(Refer Slide Time: 22:47)

That I have already shown the expression here, so therefore using that particular expression this
is minus P equals to 0 to P del v del T at constant P minus R by P dp we will just replace del v
del T just as the way we have done with del P del v, so therefore what does it become? It
becomes integral v equals to v infinity to v, del P del T at constant v dv plus integral P equals to
0 to p R dp by P, what is R dp by p let us see?

338
I mean let us find out we have already said dPv this is equals to Pdv plus vdp, right? Now dPv by
Pv this is equal to dv by v plus dp by p, isn’t it? We are dividing everything with Pv, so therefore
from here what do we get? What is dp by P equals to? dPv by Pv minus dv by v, isn’t it? So
therefore instead of dp by P we can substitute in this particular way.

Therefore this becomes del P del T dv from v equals at constant v definitely, v equals to v
infinity to v plus R dPv by Pv minus R dv by v, integral of this whole thing. So therefore this
becomes integral from v infinity to v del P del T at constant v dv plus R, here the integration
should range from RT to Pv d ln Pv minus integral v equals to v infinity to v, R dv by v.

So from there what do we get? This is basically this becomes R ln Pv by RT plus integral v
equals to v infinity to v del P del T at constant v minus R by V whole thing over dv. so therefore
using this equation we can find out delta s star at any temperature pressure conditions using this
equation you can find out delta h star at any temperature pressure conditions. So therefore if we
know which equation it is going to be valid you can very well do it and you can find out these 2
things?

(Refer Slide Time: 25:47)

339
(Refer Slide Time: 26:00)

For the Van der waals equation of state as I have shown in this particular presentation the Van der
waals equation of state if you calculate you are going to find out the values as shown, please do
the calculation they are pretty simple you just need to evaluate the differentials and perform the
integrations correctly. If you do it you are going to find out that for the Van der waals equation of
state we are going to find out this.

340
(Refer Slide Time: 26:05)

Usually we find in the hydrocarbon industries people use the Peng-Robinson equation of state.
So if you perform the calculations with the Peng-Robinson equation of state, the equation is
already provided for you then in that case you are going to arrive at the expressions which have
been provided here.

341
(Refer Slide Time: 26:40)

I will just do it for you in very briefly, so that you can just check up whether you have got the
same thing or not. Say for any particular property h2 minus h1 this is v1 minus v2, T del P del T
at constant v minus P dv I have also already written it down for you plus delta Pv, right? Now
from the Redlich Kwong equation of state if you perform these particular integrations you will
find this is equal to 1.5 a by T to the power of 0.5 integral v1 to v2 dv by v into v plus v plus
delta Pv, okay.

If you integrate it correctly this equation then we find this gives you as 1.5a by b T to the power
0.5 ln v2 by v2 plus b plus ln v1 plus b by v1 plus delta Pv. Now suppose the state 2 is the state
of interest then state one is the ideal gas state. So therefore in the ideal gas state we know v1
tends to 0 ln v1 plus b by v1 this is naturally b equal to 0.

So with this we find out h minus h0 by RT this is given as the expression which I have already
written down 1.5a by b RT to the power 1.5 ln v by v plus b plus z minus 1 and if you make
certain substitutions then more or less I think this is the equation that you are going to get.

342
(Refer Slide Time: 28:35)

In a similar form if you proceed with the calculations of delta s star then also you are going to
find out that s2 minus s1, this is s2 minus s1 this is nothing but R integral v1 to v2, dv by v
minus b plus 0.5 a by T to the power 1.5 integral v1 to v2 dv by v plus b. If you integrate it and
then you find s2 minus s1 is R ln v2 by v, I am calculating this out because this is quite
commonly used in hydrocarbon industries and since this is slightly involved, so therefore this is
T to the power 1.5 ln v2 into v1 plus b by v1 into v2 plus b, right?

If you rearrange it then we find that once you rearrange it and you arrange it properly then finally
you get s minus s0 this will be equal to ln v minus b by RT minus 0.5a by b RT to the power of
1.5 ln v plus b by v, right? So therefore in this particular way you can also try out the Redlich
Kwong equation of state you find that from these equation from the PvT behavior of the gases if
you know that they are obeying any particular equation of state, all the equations are pressure
explicit. So using these you will be in a position to find this out.

What are the other ways of estimating enthalpy, entropy departure functions or other departure
functions? Because these equations that we have derived were although very accurate an exact
but they are very tedious and time-consuming as engineers we would like to have certain faster
methods which may not be so accurate but they would help us a quick to obtain a quick
estimation of the property. So we are going to discuss that in the next class.

343
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 19
Property Estimation(Contd.)

Well, so apart from what we have already done we were trying to find out the departure functions
from the PVT behavior of gases. Now apart from PVT Behavior we can also use the
compressibility factor chart for finding out the departure functions for that we need to just
reorient the equation which we have derived we just need to write them in terms z and Tr Pr etc,
how to do this?

(Refer Slide Time: 0:49)

Let us see v is nothing but equal to zRT by P. Del v del T at constant P then we can write it down
as zR by P plus RT by P del z del T at constant P or in other words we can write it down as R by
P z plus T del z del T at constant P, isn’t it? So in that case delta of delta h star delta P at constant
T, what is this? This is nothing but if we simply substitute this equation here then what do we
get? Instead of v we are going to write zRT by P and then minus T the expression for del v del T
at constant P.

What does it reduce to? zRT by P minus T R by P z plus T del z del T at constant P, agreed? Or
in other words this reduces to zRT by P minus this term is also zRt by P minus RT square by p
del z

344
del T at constant P or in other words this can be also written down as delta h star on integration
you will get this is nothing but equal to minus R integral minus RT square integral 0 to P del z
del T at constant P dp by P, isn’t it?

And so therefore this equation we can also write it down in terms of reduced parameters like
minus R Tr square Tc square for T integral 0 to Pr del z del Tr Tc dPr Pc by Pr Pc, isn’t it? And
then what do we find finally? We find that we have expressed delta h star in terms of minus R Tr
square Tc integral 0 to Pr del z del Tr at constant Pr d Pr by Pr, agreed? So therefore from this we
find we are we could express delta h star in terms of TR and PR.

And so therefore if we know z in terms of TrPr we can substitute it and we can find out Delta h
star, for finding or using z we recollect that they have already shown you the use of
compressibility factor charts which express z as a function of Tr and Pr. Just in the way they have
expressed delta h star we should be in a position to express delta s star as well, how to express?

(Refer Slide Time: 3:59)

345
(Refer Slide Time: 4:15)

In the same way we are we are going to substitute this expression here and we are going to work
it out, this gives minus R integral 0 to P, z minus 1 by P plus T by P del z del T at constant P dp,
isn’t it? And this also we can write it down as R integral 0 to Pr z minus 1 plus Tr del z Del TR
constant Pr whole thing dPr by Pr.

346
(Refer Slide Time: 4:39)

(Refer Slide Time: 4:47)

So we can also find out delta s star and delta h star in terms of z Tr Pr we can use this
compressibility factor charts to find z then we can integrate z as a function of Tr and Pr. Z is the
function of Tr at constant Pr and we can very well do this but there is something more interesting
if you observe the equations that I have written down.

347
(Refer Slide Time: 5:06)

The equation for Delta s star and the equation of delta h star I will rewrite it just for your
convenience. This is integral 0 to Pr del z del Tr at constant Pr dPr by Pr observe this equation
carefully, what do you find? You find that suppose you can express this equation as delta s star or
else delta s star by R then what do you find?

You will find the right-hand side it just express in terms of reduced volumes. So what does it
show? It shows that this parameter Delta s star by R this definitely obeys the law of
corresponding states. In the same way if you can express this equation of delta h star in this way
you find again the right-hand side it contains only the reduced parameters and z and nothing else,
what does it imply?

348
(Refer Slide Time: 6:27)

It implies that these 2 terms delta s star by R and delta h star by RTc they definitely obey the law
of corresponding states. So therefore we need not refer to the compressibility factor charts that I
have shown in here. We should be able to generate identical curves using Delta h star by RTc as a
function of Pr and Tr.

349
(Refer Slide Time: 6:54)

We should be able to generate Delta s star by R curve as a function of unique functions of Tr and
Pr and this has been done by a large number of researchers so long and the curves they are of this
particular form. These curves they have been generated by Edmister and they are used very
frequently in the hydrocarbon industries in order to estimate the reduce enthalpy departure
function as a function of reduced pressure and a function of reduced parameter, okay.

And in the same way graphs have also been generated for the entropy departure function. Now it
is important for us to remember that just like the compressibility factor chart which we have
discussed we had found out that the 2 parameter law of corresponding states it is not very
accurate and we often resort to the 3 parameter law of corresponding states. For the 3 parameter
law what do we need to do?

350
(Refer Slide Time: 7:49)

We simply express z as a function of z0, this is for simple compound for simple molecules and
then we incorporate a correction using a third parameter which is the acentric factor. So therefore
if we wish to incorporate this and express delta h star by RTc in terms of the 3 parameter law of
corresponding states that also we can do in this particular case, for that what we need to know?

We just need to substitute instead of z we need to substitute z0 plus Omega z1 and find out the
relevant equation, what is the equation let us see? The equation becomes minus this becomes
minus RT square by Tc integral 0 to Pr del z0 del Tr at constant Pr dPr by Pr we are simply
substituting that and elaborating the equation by Tc integral 0 to Pr del z1 del Tr at constant Pr
dPr by Pr or in other words we find that we are in a position to write the enthalpy departure
function in terms of something like delta h star by RTc as delta h star 0 by RTc plus omega, plus
this was already minus omega.

So therefore plus omega into this particular term which is h0 minus h of 1 by RTc where this
term it can be written down as omega Tr square integral 0 to Pr del z1 del Tr at constant Pr, dPr
by Pr, right? So therefore we find from here the 2 parameter expression it is nothing but Tr, so I
will just write down delta h star by RTc 0 this is nothing but TR square integral 0 to PR del z0
del Tr at constant Pr dPr by Pr and delta h star byRTC1 this is equal to TR square integral 0 to

351
PR del z1 del Tr at constant Pr d equal, I am confusing the small and all of these are small
alphabets dPr by Pr.

So therefore just like we have discussed for the case of your the compressibility factor in this
case also we find that for greater accuracy we need to incorporate not only the 2 parameter but
also that 3 parameter law of corresponding states. Now similar exercise can also be performed
for the entropy correction or the entropy departure function as well.

(Refer Slide Time: 11:17)

In that case also if we proceed in this particular way, we will find that Delta s star by R it can be
expressed as 0 to Pr z0 minus1 plus Tr del z0 del Tr at constant Pr this whole thing dPr by Pr
plus omega integral 0 to Pr z1 plus Tr del z1 del Tr at constant Pr whole thing dPr by Pr where
this particular path this whole part is nothing it is nothing but delta s star1 by R and this part it
can be written down at delta s star0 by R.

352
(Refer Slide Time: 12:14)

(Refer Slide Time: 12:16)

Now several charts have been proposed as I have already discussed there are charts by Edmister
and one can refer to those charts by going to the book of Edmister we find that there are enthalpy
departure functions, there are entropy departure functions for 2 parameter as well as the 3
parameter law and the 3 parameter corresponding states and usually we prefer the 3 parameter
because 2 parameter module it was anyhow not very accurate for predicting compressibility
factor. So naturally it fails much more for the case of enthalpy and entropy departure functions as
well, right?

353
(Refer Slide Time: 12:46)

(Refer Slide Time: 12:55)

354
(Refer Slide Time: 13:19)

(Refer Slide Time: 13:21)

So these are the 2 types of charts which are available, the initial charts which I had shown they
are for Edmister they are primarily used for the hydrocarbon industries there are also certain
other charts which have been proposed by Lee and Kesler these charts can also be used, you can
very well see that they have the 2 parameter and the 3 parameter enthalpy as well as entropy
departure functions you can see they have been expressed for lower values of reduced pressure
and higher values of reduced pressure one can very well use them and they can find it out, okay.

355
(Refer Slide Time: 13:46)

(Refer Slide Time: 14:15)

356
(Refer Slide Time: 14:17)

We can do a one or 2 problems regarding this so that you can get a better feel of it, for example
suppose I tell you that we would we are interested to find out delta h star and delta s star for
methane say, okay. For methane at 32 degrees centigrade and say 9.28 bar and I tell you to use
Edmister chart for this, so what are you going to do for this? You are required to find out the
enthalpy and entropy departure functions for methane at 32 degrees centigrade and 9.28 bar,
right?

357
(Refer Slide Time: 14:20)

(Refer Slide Time: 14:23)

So let us go to the Edmister’s chart for this, the Edmister’s chart are this, right? Now for methane
the first thing that you need to do, you need to find out Tc and Pc and omega, right? What is Tc
in this case? You can find it from any standard table or etc, Pc is going to be 46.41 bar and
omega this is going to be 0.011, once you know this you are in a position to find out Tr,
remember this Tr and Pr they should be these are in terms of Kelvin. So therefore this also has to
be converted to Kelvin.

358
Now there is one thing which I would like to tell the problems you often tend to make mistakes
just because you do not consider the right units, it is very important when you’re working with
pressure, suppose you are using Antione’s equation very frequently they are given in torr so
therefore the pressure which you’re going to use has to be in torr. When you’re working with
temperature see at what in which unit they are given, so these things are very important if you
want to arrive at correct answers for not only for thermodynamics for anything for that matter.

(Refer Slide Time: 16:04)

So therefore this becomes 305.15 by 190.7 which is nothing but1.6 in the same way you can find
out Pr it is going to be 9.28 by 46.41 which is equals to 0.2. So therefore you need to find at Pr
equals to 0.2 and Tr equals to 1.6 let us go to the enthalpy charts you find out for this particular
situation you are going to get from here if you can locate your Pr equals to 0.2 it's here, here
somewhere and Pr is equals to 0.2 and Tr equals to 1.6 this is the line.

359
(Refer Slide Time: 17:07)

(Refer Slide Time: 17:10)

So if you follow this line and to come to 0.2 you find out that your Delta h star0 by RTc this
whole thing becomes 0.085, check it up for yourself and find out then if you go to the next curve
then you’re in a position to find out Delta h star 1 by RTc find it out you are going to get this is
0.01 similar fashion if you find out Delta s star0 by R, read it out it’s going to be 0.032 Delta s
star 1 by R this is going to be equal to 0. You can use these particular graphs and you can find
them out.

360
Now once this is available you can find out delta h star by RTc it is just going to be this plus
omega of this, so if you calculate it as 0.085 plus 0.01 into 0.011 then from there you find the
value is going to be 8.511 into 10 to the power minus 2 and from here what is the value of Delta
h star then which you are required to find out? This is nothing but minus 0.135 kilo joules per
mole, right? So therefore in the same way if you proceed with Delta s star by R you have got. So
therefore we find out that Delta s star by R this is just this value plus omega of 0. So naturally
this reduces to 0.032 only which gives you delta s star as minus 0.266 used per mole Kelvin,
right?

(Refer Slide Time: 18:55)

So therefore in this particular way you can take up a few other problems and you can find it out
for example suppose take up the problem for propane, instead of methane suppose I tell you to
take up the problem on, find out Delta h star and delta h star for propane at 60 degrees centigrade
and 17.03 bar you can use the Edmister’s chart for doing this I will just give you the answers you
compare and you just see whether you could do it or not.

Delta s star will be minus 4.511 joules per mole Kelvin and the value of Delta h star is going to
be minus 2.163 kilo joules per mole, just try it out and see what you’re getting. In the same way
for using Lee Kesler chart also, suppose I tell you that for the same propane you do it using the
Lee Kesler chart. For the using the Lee Kesler data naturally we have to go to the correct graph

361
for this. These are the graphs for Lee Kasler data, now here also you need to know the value of
TC and PC.

So Tc in this particular case for propane it is 369.9 Kelvin, PC this is equal to 42.57 bar and
omega it is 0.153 provided these are given then you’re in a position to find out your 60 yeah,
you’re in a position to find out Tr here, Tr is 0.9, Pr this is equals to 0.4.

(Refer Slide Time: 20:39)

So from these particular graphs you will find that for Pr equals to 0.4 and Tr equals to 0.9 you
cannot use this graph we have to use this particular graph, right? No, you can use this graph,
sorry. So for this Tr equals to 0.9 and P Pr equals to 0.4 if you read it out you find that that the
value in this particular case it’s almost equal to 0.6 it more accurately 0.59 you can take it as 0.6.

362
(Refer Slide Time: 21:05)

So therefore you will find that Delta h star 0 by RT see in this particular case it 0.6. Again if you
go to the next curve you will find that Delta h star 1 by RTc if you read it here again for 0.4 and
for 0.9 you find it is about 0.75, right?

So therefore from here you will be in a position to calculate Delta h star by RTc it is just this plus
omega of this. So therefore this becomes 0.7109 or 071 you can take, from where we can
calculate Delta h star to be minus 2.186 kilo joules per mole, R in this particular case you said
dealing with joules this is going to be 8.314 and the value of TC it is 369.9 which I have already
mentioned same way if you would wish to calculate Delta s star we can do it as well.

Here also you need to go from 0.4 to 0.9 and you can see the value is about 0.463, isn’t it? So
therefore Delta s star by R0 this is 0.463 and if you’re going to calculate Delta s star1 again from
0.4 to 0.9 it is about 0, his particular value it is about 0.744 something yeah,0.744. So therefore
0.463 plus your 0.153 into 0.744 that gives you Delta s star by R and which is nothing but 0.5768
multiply it with a value of R which is 8.314 and finally you get the value at Delta s star equals to
minus 4.796 joules per mole Kelvin. So in this particular way you you can try a large number of
problems and you can get yourself conversant with the usage of these charts.

363
(Refer Slide Time: 23:41)

Now before I end it is also important to just have a brief discussion on the prediction of departure
functions will be virial equations of state, what is the virial equation of state we know? This is z
equals to Pv by RT for low to moderate pressures we truncate it after the first term this can also
be written down as BPc by RTc Pr by Tr, remember one thing this is applicable for low to
moderate pressure this is something important.

So then what is v equals to? This is RT by P into 1 plus BP by RT which is nothing but RT by P
plus B, accordingly what is del v del T at constant P this is nothing but del B del T plus R by P
accordingly what is delta h star that we get from here? The value of delta h star and the value of
Delta s star what are these values that we get from here? These particular values are this is, just
substitute and find out B minus T, dB dT into P and Delta s star this is equals to minus P dB dT,
right?

So therefore and here also you can express B in terms of B0 and omega B1, you can also express
B in terms of RTc by Pc B0 plus Omega B1 and you can use the 2 and the 3 parameter law what
this class it is just sufficient to know that Delta h star and delta s star they can be defined in this
particular way. Now therefore these are the ways by which you can find out the properties using
the PvT Behavior of gases or the cubic equations you can use the compressibility factor chart
you can use the virial equation of state these were the ways by which you can do it using the PvT
Behaviors.

364
Apart from the generalized equations of state or the generalized compressibility factor charts we
also have some particular figures some particular tables in order to find out or predict the
thermodynamic properties one such common table is the steam table which you are already
aware of. There are certain other charts also available for finding out the properties particularly
enthalpy and entropy instead of 2 going through the details calculations that we have said.

(Refer Slide Time: 26:16)

A few of these are the pressure enthalpy diagram which is shown here, these Particular diagram
is shown for nitrogen and here if you observe we will find that the entropy curves are given the
pressure and enthalpy I have given and we also have the specific volume data and the
temperature data. So therefore it contains large number of data and we have been in a position to
find out the enthalpy and entropy we can have such particular charts for other gases as well.

365
(Refer Slide Time: 27:03)

Well, there are also the common the typical temperature entropy diagram and also the enthalpy
entropy diagram which is very easy and regarding this enthalpy entropy diagram this is known as
the mollier chart this is very convenient because from here we can find out enthalpy we can find
out entropy and since it contains constant pressure lines, constant temperature lines as well as
constant quality lines in the 2 phase region. So therefore this gives us a large amount of
information and this is a very popular chart which is used for predicting the properties of gases.
So with this we end the estimation of the properties of pure component gases or mixture of gases
where the mixture behaves as a pure component, right?

After this we are going to study the properties of mixtures and we are also going to extend our
discussions to homogeneous open systems and heterogeneous closed systems. So with this we
close our discussion on homogeneous closed systems we have discussed the estimation of
properties and the concept of thermodynamic potential and the concept of chemical potential we
are going to use this further in our discussion on phase equilibrium thermodynamics, thank you
very much.

366
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 20
Concept of Chemical Potential.

(Refer Slide Time: 0:45)

Hello everybody, so till now our discussion they were primarily confined to close systems, right?
Close systems means those systems which have been interacting with the surroundings under
different conditions of interest maybe say for example they were interacting with the
surroundings under constant volume entropy they were also interacting under constant P, S, they
are interacting under constant T,V and they were interacting under constant T, P but in all the
cases we had kept in mind that we are dealing with closed systems where all the interactions did
not concern mass changes of the system due to interaction with the surroundings even if there
was a chemical reaction inside the system we ensure that the affect was primarily that of volume
change while the composition and the total mass of the system were kept constant.

367
(Refer Slide Time: 1:25)

(Refer Slide Time: 2:07)

Now under that condition what did we observe? We found out for that particular condition we
had defined different thermodynamic potentials or rather the changes of the different
thermodynamics potentials under different conditions of interactions between the system and the
surroundings and the changes between the different thermodynamic potentials they were related
to changes in other properties when a system moves from one equilibrium state to the other
equilibrium state while interacting with the surroundings under different particular conditions
and these were the 4 conditions of interest which we were dealing with, right?

368
(Refer Slide Time: 2:17)

Now suppose let me and one more thing I would like to remember that all the equations which I
have written down here all this set of equations they are primarily concerned with only PV type
of work of expansion and no other non-PV work involved. If there is a non-PV work then
definitely we have a plus Fdr bar here that we have already discussed. We have discussed both
the criteria of equilibrium both for PV type of work as well as for non-PV type of work. Now
this entire thing this was for, the whole thing was suggested for close systems with only PV
work.

Now suppose I try to deal with open systems when along with energy there is mass transfer
between the system and the surroundings, what do you expect under this particular condition
when we deal with open systems? Keeping in mind that even for open systems we shall be
dealing with only PV type of work, now how have we defined the change of say if it is the first
equation? How have we defined the change of du with change of S and V.

Remember another thing the equations which I have written down here, they are all for total
properties or total extensive properties of the system and the importance of differentiating
between the total and the molar or the mass specific properties will be very clear today itself in
this class as I proceed further. Now in what way have I define the change of u? I have what
changes in say total entropy and total volume. First I have defined del U del S at constant V
times the dv and then I have defined del U del v at constant S.

369
Now suppose there is change in the number of moles, now for the time being again I mention
that we are considering only PV type of work and single component systems these are the 2
things that we are going to consider for the time being. Now for these cases what do you expect
which term is going to be added? Here we are going to have one term del U del n under
conditions of constant s and v, right into dn and moment we add up one more variable, what are
we required to do?

We are required to write Sn here we are required to write vn here. So therefore when dU changes
as a result of change in total entropy, total volume and total number of moles of the system then
the differential equation for dU comprises of 3 terms, the one term denoting the change of U with
s when volume and the total number of moles are constant, the second term dealing with change
of U with volume when entropy and total number of moles are constant and the third term deals
with the change of U with n when total entropy and total volume of the system remains constant.

The same thing is applicable for each and every equation that we have written down for example
here we are going to have something like del H del n at constant s and P and immediately we are
supposed to put one n here and one n here. Same thing is going to apply to this case as well
where we are going to write this down constant T and V both of them are capital and we have to
write it down in this particular way. Here also we are we are supposed to write del G del n
constant T, P here we are going to have a constant n, here we are going to have a constant n,
right?So moment we shift from closed to open systems we find that initially we had 2 terms now
we are having 3 terms in the entire differential equations.

370
(Refer Slide Time: 6:57)

Well, so the next question which comes up is what are these particular terms that we have written
down? What is this particular term? What is this particular term? What is this term and what is
this term? Now just if you look closely at these terms, what is the term let me write it down here
it is del U del n at constant s, v.

371
(Refer Slide Time: 7:35)

So therefore it is the change of internal energy with change in the number of moles, so at the first
intuition it appears that this should be the molar internal energy of the system. Let us see if our
initial intuition is correct or not and the same thing should apply to each and every case for
example del H del n at constant SP this possibly then it should correspond to small h this should
correspond to small a and this should correspond to small g.

Now let us see if our supposition is correct or not? What is Del U del n at constant capital s
capital v? This is equal to u, now what is capital U? Capital U is nothing but n time’s u. So
therefore what is this particular term? This term is nothing but u plus n del u del n at constant
capital S capital V, right? Now try to understand it very well, since this is the molar property so
whatever equation I have written down for the total property the same equation should be
applicable for the molar property when this equation is written in terms of molar quantities,
right?

So therefore we know that du this is equal to Tds minus pdv, what does it imply? It implies that
du at constant molar entropy molar volume is equal to 0, you get my point. So therefore if du at
molar entropy, molar volume is equal to 0 so therefore and we know capital S is nothing but
equal to ns we know capital V is equal to nv. So therefore if we are keeping this constant and we
are varying this it’s quite natural that this also has to be varied this cannot be kept constant.

372
The same way if we are keeping this constant and we want to vary this then this also has to vary
this cannot be kept constant. So what I mean to say is you cannot keep total entropy and molar
entropy constant at the same time when the total number of moles of a system is varied. The
same thing is true for volume you cannot keep total volume and molar volume constant at the
same time when the number of moles are varied.

(Refer Slide Time: 9:54)

So therefore what I mean is that if your du at small s, v equals to 0 than definitely we can write
this down as del capital U del n at small s, v this is equal to u, right? And we also can write this
down as that del capital N del u for small s, v this cannot be equal to del U del n at capital S,
capital V, right? Because you know that this and this cannot be kept constant at the same time
when n is varying.

373
(Refer Slide Time: 10:32)

(Refer Slide Time: 10:38)

Now if this equation has to be true then in that case this term has to vanish but we find that if this
term vanishes then this term cannot vanish. So therefore it automatically implies that this is
nonzero term as a result of which your del U del n at constant capital S, capital V is not equal to
the molar internal energy of the system.

374
(Refer Slide Time: 11:21)

So therefore the first thing which we realized is that this particular term which has been written
down this particular term is not equal to the molar internal energy. So the next question arises
what is it? Then do we have to define it in terms of some other molar properties. Is that molar
property different for each and every differential equation? Then we land up into a set of 4
additional molar properties which are not the true molar properties in that particular sense.

375
(Refer Slide Time: 11:52)

So the next, once it was realized that this quantities they do not correspond to the molar
quantities at least for the first 3 cases. Let us see what the del G case what we get? For the del G
case again we can write down dG it is equal to del G del T, I have already written the equation
on the overhead slides yet I would like to repeat it once more.

Del G del P at constant T, n dP plus del G del n at constant T, P. I repeat again and again the
same thing this equation is applicable only for a single component open system. Now in this case
what is G equals to? Again proceeding in the similar way it is equal to ng. So therefore del G del
n at constant T, P it is nothing but equals to ng del n at constant T, P which is nothing but equal
to g plus n del g del n at constant T, P.

376
(Refer Slide Time: 13:17)

(Refer Slide Time: 13:35)

Here also I write down the same thing, I can write it down as dg equals to minus sdT plus vdp,
right? So therefore dg at constant T, P equals to 0. So therefore del G del n at constant T, P this is
equal to g. So therefore we find that in this particular case this part dg at constant T, P equals to
0. So therefore this part cancels out and we know del G del n at constant T, P is nothing but equal
to the molar Gibbs free energy of the system.

377
(Refer Slide Time: 14:24)

So therefore what we deduce is that although the previous molar quantities did not or rather the
previous quantities the third terms which I have written do not correspond to the molar properties
but del G del n at constant T, P corresponds to the molar Gibbs free energy. Now let us see what
these terms corresponding to? I will be doing the derivations just with del U del n at constant s, v
and please note that the same derivation is going to be applicable for this term same derivation is
going to be applicable for this term I leave it as an home assignment to repeat the derivations
which I have done and to find out the corresponding expression for del H del n and del A del n
subject to the respective constraints.

378
(Refer Slide Time: 14:35)

Let us see what it is? Now again I repeat U is nothing but equals to nu, right? This I know so
therefore I can write it down as dU is nothing but equals to ndu plus udn I can definitely write it
in this particular way. Now what is du equals to? du several times you must have learned by
heart by this time write it down in this particular form therefore this can be written down as Tds
minus Pdv since their properties I can play with them in any particular way it hardly matters
provided initial and final states are constant and I can write it down in this particular way. So
therefore this can be written down as nTds minus npdv plus udn, right?

379
(Refer Slide Time: 15:48)

Now can I write this particular equation as I will just write it down once more to maintain
continuity nTds minus npdv plus udn, right? Can I write it down as T dns minus sdn from
Legendre transform I can very well do this. Similar way I can also write this down as P dnv
minus vdn plus udn. Now you look very carefully what I have got?

I have got dU equals to dn of u minus Ts plus pv, spu all of them are small case alphabets minus
Td ns minus pd nv, what is this term? This is capital S this is capital V, so what do I get? I get
and what is this term? U plus pv minus Ts, it is nothing but the molar Gibbs free energy. So what
do I get? dU is nothing but equals to gdn minus Tds, sorry this is plus Tds minus pdv.

380
(Refer Slide Time: 18:01)

So therefore what do I find? That my initial equation which I had written down for closed
systems along with that I need one more term gdn here in order to extend the closed system
equation to open systems. If we repeat the derivations for dH dA etc we find that for each case
we can incorporate a term gdn and the equation becomes applicable for open systems and even
this gd in term is applicable for the fourth equation as well. So therefore for all the equations I
find that for open systems we find that this particular term it corresponds to the molar Gibbs free
energy.

381
(Refer Slide Time: 18:45)

So therefore we can write it down as del U del n at constant S,v this is equal to del H del n
constant S, P, del A del n at constant T, V this is equal to del G del n at constant T, P which is
nothing but equal to g.

382
(Refer Slide Time: 19:42)

Now therefore from here it is quite evident that this particular term the molar Gibbs free energy
that is quite important and since it is quite important, so therefore this assumes a very important
role in phase equilibrium thermodynamics and it is often called chemical potential of a substance
and where chemical potential of a substance is nothing but it is molar isobaric-isothermal
potential or it can be if mass specific isobaric isothermal potential it all depends upon the way we
have defined the mole numbers or the mass numbers, right? Now this was for a single
component system and from these particular equations we can arrive at another very important
relationship.

383
(Refer Slide Time: 20:07)

Let us see this particular relationship which I have written down. This particular relationship
from there we can write it down as dU equals to Tds minus pdv plus since molar Gibbs free
energy is often called chemical potential I can define it in terms of chemical potential. Now
again applying Legendre transform we can write this equation as equals to dU plus d pv minus
vdp minus of d of uN minus ndu, right?

So again from here we can write d of U plus Pv minus Ts this will be equal to plus SdT minus
vdp plus ndu minus dnu equals to 0 where this we know is nothing but G, this part we know is
nothing but G. So therefore this part and this part cancels out giving us a very important equation
which is very useful for phase equilibrium thermodynamics and this particular equation is known
as the Gibbs Duhem equation.

What does this particular equation suggest? It tells us that for a single component open system
we cannot vary T, p and u at the same time. If we fix up mui and p, T automatically gets fixed up
and it is true for all the other systems. So therefore all of them cannot be varied at the same time
if one of them or rather both of them are varied the third one automatically gets fixed. Now this
particular equation has got a greater applicability when we go for multi-component open systems
which we will be discussing in the next lecture, thank you very much.

384
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 21
Concept of Chemical Potential(Contd.)

(Refer Slide Time: 0:30)

Now let us see what happens for multi-component systems, for multi-component systems if we
take it up thus same equation with little with some more terms again I repeat we are going to deal
with only PV type of work and I will be doing the entire derivation considering internal energy
keeping in mind that the same derivation can be applied for other systems as well.

385
(Refer Slide Time: 0:46)

So in this particular case what did I write down initially? This was equal to del U del s at
constant v dv for closed systems del U del V at constant s sorry this was dS this was dV. Now if
if I have more than one component then what do I do? This is equals to ni not equal to j this is nj
not equal to i, nj not equal to i where I varied the number of moles of the neth component. So
therefore this is going to be del U say del n1 at constant s, v n2 to the final component nM of
whatever number or say nC is the total number of components, okay.

So therefore it goes on continuously into dn1 plus del U del n2 dn2 again this is at constant s, v,
n1, n3 till n and maybe a number of components N plus and so on we can keep on writing. So
therefore this can be summarized and it can be written down as this is equal to del U del s
constant nj not equal to i d s plus del U del V constant s nj not equal to i d v plus summation of i
equals to 1 to n where n is the total number of components del U del ni constant s, v nj not equal
to i.

I am sorry in this particular case I am extremely sorry it should have been total N. I am very
sorry for this mistake it should have been total N, right? So therefore this is the final equation
which we are going to get for multi-component systems and in this particular equation we find
that each and every variation of U with the mole numbers each of them they correspond to the
chemical potential of that particular component mu2.

386
(Refer Slide Time: 4:19)

So therefore this can also be written down as del U del S at constant v, n ds plus del U del V at
constant s, n dv plus Sigma i equals to 1 to n mui dni, right? Where each mui it corresponds to
del U del ni of the ith component where the number of total number of moles of all other
components except the ith component is kept constant, I have written down the equations for the
multi-component open systems in each case.

You can see and in each case again it can be proved that the property change due to the change in
the number of moles of component i when the number of moles of all other components are kept
constant is given by mui dni in each particular case where muis just like mu for single
component systems.

In this particular case also we find that it corresponds to the molar internal energy or the molar
thermodynamic potential subject to the respective constraints and in reality it is the molar
isobaric isothermal potential when the number of moles of any 1 component is changed while the
number of moles of the other components are kept constant and in this case also just like the
previous case we can define the Gibbs Duhem equation in fact the Gibbs Duhem equation is
much more convenient when we are defining it for multi-component systems.

387
(Refer Slide Time: 5:22)

In this particular case the equation becomes sdT minus vdp plus Sigma ni d mui equals to 0 and
at constant T and P this becomes Sigma ni d mui equals to 0 and if we divide it by the total
number of moles it becomes Sigma xi d mui equals to 0. The importance of this equation can be
immediately understood if we know the composition of the mixture and if we know the chemical
potential of n minus one number of components we can use this particular equation to find the
chemical potential of the neth component.

Say for example for a true-component system this just reduces to x1 dmu1 plus x2 dmu2 equals
to 0 at constant T and P. So if we know x1 if we know the change of mu1 then by using this
equation we can find out this particular quantity and there is where the importance of the Gibbs
Duhem equation this is the Gibbs Duhem equation for multi-component system and we will find
later that this is very important to test the thermodynamic consistency of the data that we use
because we need to remember that most of the these data they are obtained from experiments. So
therefore before using them for any particular calculation it is very important to test that this data
are thermodynamically consistent and this particular equation is usually used to test the
thermodynamic consistency of this data.

Now before we go further it is very important to understand that why is this so very important
term? The molar Gibbs free energy given a special name and that took the name as chemical
potential. Is it really a potential in the true sense just like u, h, a and g. Now in order to

388
understand this we have to recall the properties of a thermodynamic potential which I had
already discussed when I was discussing when I was trying to explain why u, h, a and g are
thermodynamic potentials.

(Refer Slide Time: 8:01)

What are the properties of thermodynamic potentials if we recollect? What are the properties of
thermodynamic potentials are? If you remember first is the value has to be independent of the
integration part. What does it signify? It signifies that these particular property or rather these
particular potentials they have to be properties of the system. Definitely chemical potential is a
property of the system we know that. So in this case we do not have any problems, the other
thing which we found out was that all these potentials they must be units of energy, right? Here
also definitely chemical potential it is measured in the unit of energy but that is primarily
accidental, why?

389
(Refer Slide Time: 8:35)

Because the chemical potential mui it is nothing but equal to del G del ni at constant T, P nj not
equal to i and this particular number of moles it is basically a number, have I expressed in terms
of concentration then this would have been a energy concentration instead of an energy unit. So
therefore since mui is defined in terms of ni so therefore it has the units of energy but that is
primarily accidental and what does the third condition of potential imply?

390
(Refer Slide Time: 9:12)

It implies that the minima denote stable equilibrium which we already know while we were
discussing equilibrium under different conditions what did we find?

(Refer Slide Time: 9:28)

We found that dU s, v was equal to 0 the conditions of equilibrium work then dH S, P was equal
to 0 then dA T, V was equal to 0 and dG T,P was equal to 0, right? So therefore we find that for
each and every case the differential quantities are equal to 0.

391
(Refer Slide Time: 9:49)

(Refer Slide Time: 10:32)

But in the case of chemical potential if you observe this particular equation and this particular
equation it immediately appears to you it is not that the dmu is equal to 0, this does not happen
on the contrary Sigma mui ni this is equal to 0. So therefore if you observe this that the
summation of chemical potential for any particular reaction or for any particular process equals
to 0 than it appears that it has been better if it would have been called by the name of the force
because we find that this particular characteristic combines with or rather this particular
characteristic it corresponds to a fourth quantity.

392
But again then we found out that if we name it by the force then we find that we have a force
which has the unit of energy that is not something very convenient and the next thing which we
observed was that very frequently we find that for non-equilibrium condition the driving force
for current they can be identified with the gradient of chemical potential because very frequently
we find that when we are dealing with non-PV type of work then we are dealing with electrical
work and under that condition the current it is usually defined as the gradient of potential energy.

So therefore under that condition we find that it is a force being a derivative of a force that is also
not a very comfortable feeling. So therefore what I mean to say is chemical potential is
something very unique which combines some of the characteristics of potential but it cannot be
consider a thermodynamic potential in the true sense.

It has some characteristic of a thermodynamic force but it cannot be considered a

thermodynamic force in the true sense it has the unit of energy it is an intensive property of the
system but the summation it corresponds to equilibrium and it’s Minima does not correspond to a
stable equilibrium.

So all these things suggest that chemical potential has a unique place of its own and since it
combines a large number of characteristics just because the way it has been defined, so therefore
it has got a multifaceted aspect and since in the absence of any better name we would like to
prefer to follow conventions and we would like to continue calling it as a potential but not a
thermodynamic potential but a chemical potential keeping in mind that it is not a true
thermodynamic potential.

393
(Refer Slide Time: 12:45)

The next thing again which I would again I would like to say that what have we done in this
particular case? For this particular case we have found out or rather we have related each and
every thermodynamic potential with the chemical potential term. Now it would be interesting for
us to find out that if we notice that for an isolated system what is the criterion of equilibrium?

For this particular case it is dS isolated equals to 0 where we know that S is not a thermodynamic
potential. So therefore in this particular place it will be interesting to find out how this particular
term del S del n at constant u, v in what way is it related to the chemical potential? Now this
particle derivation it is slightly involved but we can go through it and we can try to appreciate
keeping in mind that although it is not a thermodynamic potential it is often used in order to find
out the criteria of equilibrium for an isolated system and it has got a very interesting relationship
with chemical potential where the relationship is not of this particular form.

394
(Refer Slide Time: 14:14)

Let us see this particular relationship before we proceed further, now this particular case can we
not write it down in this particular form dS equals to we can write it down as say del S del n at
capital UV, right? We would like to find how this is related to mu? Now this particular equation
you can definitely write it down as ns at constant UV which is nothing but equal to just like we
are being doing for all cases del s del n at constant capital UV.

Now I would like to remind you of a very important rule of differentiation which is given as del
z del x at constant n is nothing but del z del x at constant y plus del z del y at constant x into del
y del x at constant M. So if we apply this particular equation here, what do we get? We get del s
del n at constant capital U capital V is nothing but del s del n at u v plus del s del u at n, v into
del v del n at u, v we can write it down in this particular terms.

Then this particular term can also be expanded in the similar way. We can again write this term
as del s del n at small u small v plus del s del v, u n into del v del n, u capital V, right? And then
we can substitute this equation there. Once you substitute what do we get? We get del s del n at
capital U capital V this is nothing but equal to del s del n small u small v plus del s del v small u
small n into del v del n uv plus del s del u nv, del u del n capital U capital V, we can write it
down in this particular way.

Or if you observe this equation what do you find? The first thing is this particular term del s del
n at small u, v this term this term this becomes equal to 0, what about this term? Here I would
like

395
to remind you that Tds equals to du plus pdv, right? So from there what is this term equal to can
you tell me? This term then becomes equal to P by T, what about this particular term? This term
becomes equal to 1 by T.

(Refer Slide Time: 17:39)

So therefore and we have to keep in mind that capital U is nothing but equal to nu capital V is
equals to nv. So therefore from there what do we get? We get del u del n capital U, v this is
nothing but equals to del del n of u by n U, v which is nothing but equals to minus u by n or in
other words this is equal to minus capital U by n square. Same way we can write it down as del v
del n capital V, u is nothing but equals to minus v by n. So therefore from there what do we get?
del s del n at capital UV it is nothing but equals to minus of u plus Pv by Tn from there what do
we get?

396
(Refer Slide Time: 18:45)

We get that then this becomes equal to and if we substitute this particular term in the first
equation that we have, then finally what do we arrive at?

397
(Refer Slide Time: 18:58)

We arrive at a condition that del S del n capital U, V is nothing but equal to s plus nu minus pvn
by Tn from there what do we get? We get minus of, so we get minus of u plus pv minus Ts by T
which is nothing but equals to minus mu by T. So therefore what do we get? We get that del S
del n at capital U, V is nothing but equals to minus mu by T.

(Refer Slide Time: 19:47)

398
(Refer Slide Time: 19:49)

399
(Refer Slide Time: 20:47)

So therefore we were trying to find this relationship we know now that this is equals to mu by T.
So therefore we combine first and second law for open systems for only Pv work from combined
first and second law and single component system for all these conditions, what do we get? We
get dS is nothing but equals to 1 by T dU plus P by T dv minus mu by T dn or in other words
Tds equals to dU plus pdv minus mu dn. So this is the equation which we derive derivation was
slightly involved with large number of capital and small case alphabets involved.

But by this particular discussion what I wanted to impress upon you is the importance of
identifying the total and molar properties and I would like you to remember that we are not
dealing with mass specific properties at the moment, same equations will be applicable for mass
specific properties as well.

And from here what we came to know is when we have to deal with open systems when the
system interacts with the surroundings not only by energy interaction but also by changes in the
number of moles or by mass interaction then we find that we had to introduce a new chemical
property which was actually not new it is basically the molar Gibbs free energy but we had
called it by a new name the chemical potential.

400
(Refer Slide Time: 22:02)

Why did we have to put a new name to it? Because we found out that for single property system
it was the molar Gibbs free energy but moment we are dealing with multi-component systems we
find that this particular property it becomes equal to mui in this particular case, right?

So therefore just in order to combine both single component and multi-component systems we
had defined or rather we have defined the Gibbs molar Gibbs free energy for single component
systems as chemical potential.

401
And we keep in mind for single component systems chemical potential is equal to this particular
expression which for a multi-component system it becomes del U del ni s, v, nj not equals to i
and so on and so forth and both these terms they are referred by the same name it is known as the
chemical potential and we will remember that the chemical potential is not a true thermodynamic
potential but it has a number of characteristics of a thermodynamic potential.

So in the next class we shall be going to deal with open systems, first we are going to deal with
single component open systems and then we are going to extend it for multi-component open
systems.

402
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 22
Homogenous Open Systems
(Refer Slide Time: 0:51)

Well, hello everybody we continue our discussions on homogeneous open systems that we have
started in the last class if you see in the last class we have derived these 4equations relating a
thermodynamic potentials with the respective parameters for under different conditions of
coupling between the system and the surroundings and we have defined a new term which we
found for pure substances it is nothing but the molar Gibbs free energy and it was defined in this
particular form.

403
(Refer Slide Time: 1:06)

(Refer Slide Time: 1:30)

Now we would like to extend the discussions that we had yesterday slightly further and we
would like to discuss certain other aspects of homogenous open systems. Now since as I have
already mentioned and you also know that the temperature pressure conditions or keeping
constant temperature and pressure it's the most common thing that we encounter in practice. So
therefore we are going to deal primarily with the equation which relates the Gibbs free energy
with the other things.

404
(Refer Slide Time: 1:36)

Now for the time being let me write it down for the time being I would like to deal with the
simplest type of Homogenous open systems, what is it? It comprises of say one component it's a
single component system which exists in 2 phases and let me tell you this is the most common
type of phase transition that we encounter and the phase transitions are usually of the form solid
to liquid or maybe liquid to vapor or maybe solid to vapor and so on, these are the most common
transitions that we encounter that are also be some other type of transition for example an alloy
of say a lead and tin which forms a solder that can also, here also we can have a similar type of
phase transition.

So for the time being, what we do? We concentrate on one component 2 phase systems and we
consider the simple or the simplest type of open systems that we can think of once we are
conversant we are quite familiar with this then we go for heterogeneous closed systems or in
other words we go for a system where a large of homogeneous open systems which constitutes a
heterogeneous closed system.

Now let us see in this particular case we can write down dG as del G del T at constant P, n dT
plus del G del P I have written it down several times in the last class plus del G del n 1 where I
would like to remind you that n1 is the number of moles of phase1 say, n2 is moles of component
it should be of component in phase 1 and this is moles of component in phase 2, so this is del G
del n1 under constant T, P dn1 plus del G del n2 constant T,P dn2, right?

405
Now at constant pressure temperature conditions because usually we know that these types of
phase transitions they occur under constant temperature pressure conditions. So under these
conditions what do we get? We get del G del n1 T, P dn1 plus del G del n2 T, P dn2 this should
be equal to 0. Now in the beginning I have told you that we are dealing with a single component
in 2 phases and we assume that while, for example suppose say it is a liquid vapor mixture,
right? Kept in a piston cylinder arrangement where the pressure is kept constant the temperature
is kept constant we have got a liquid here we have got a vapor here.

Now if you consider the liquid only to be a system then it is an open system because there is
mass and energy interaction between the liquid and vapor phase if we consider the vapor to be
the system then again it is an open system but when we consider the liquid vapor combination to
be a system then it is a closed system. Now what do we have here? We have say n1 moles here
we have n2 moles here and we know that since the composite system is a close system.

So n1 plus n2 equals to n this should be something constant or in other words what I mean to say
is dn1 plus dn2 equals to dn should be equal to 0 which automatically implies that dn1 should be
equal to minus dn2. So therefore if you substitute in this particular equation then what do we get?

406
(Refer Slide Time: 6:12)

(Refer Slide Time: 6:35)

We get in this particular equation then in that condition we get del G del n1 T, P n2 minus del G
del n2 T,P n1 into dn1 is equal to 0. What does this imply? Since we know that dn1 is not equal
to 0 so it automatically implies del G del n1 at T P n2 must be equal to del G del n2 at T, P n2.
From yesterday we already know what is this? It is in fact it is already written here this is
nothing but mu. So therefore from here we can write this down as dmu1 or rather mu1 is equal to
mu2 under this condition.

407
So therefore if suppose transfer occurs under condition such that mui1 is not equal to mui2 then
in that case there is going to be mass transfer till the time when mu1 becomes equal to mu2 and
in the process of this mass transfer what do we find? We find that the system with the larger
chemical potentials shrinks and the system with the smaller chemical potential it increases in size
till the chemical potential of both the open systems

(Refer Slide Time: 7:19)

408
(Refer Slide Time: 8:08)

Say the liquid and the vapor under this condition they become equal moment they become equal
we find that there is no other further interaction between the 2 systems they are under
equilibrium condition. So to summarize if mu1 is greater than mu2 we know that the dn1 is
negative and the mass transfer occurs from phase 1 to phase2 and vice versa if dn2 is positive
then the mass transfer occurs from 2 to 1.

So just the way I have done I will repeat phase with the larger chemical potential shrinks and
phase with the smaller chemical potential it goes till equilibrium is reached or in other words still
mu1 becomes equal to mu2. Now this particular equation we can extend it further to find certain
other characteristics of the transition boundaries, how?

409
(Refer Slide Time: 8:29)

All of us know that what is d mu equals to? D mu it is nothing but equal to dG and we already
know what is dg equals to again referring to this particular equation. Since dG has been written
in terms of molar quantities, so therefore this equation with the number, with the different molar
quantities. So therefore this becomes minus sdT plus vdp, right? So dmu1 it should be equal to
minus s1dT plus v1dp, in the same way d mu2 this should be minus s2dT plus v2dp.

Now what do we know? We know that dmu1 should be equal to dmu2. So therefore for the 2
phases this should be equal to this and from there what do we get?

410
(Refer Slide Time: 9:34)

(Refer Slide Time: 9:58)

From there we get the famous Clapeyron equation which can be derived as dp dT equals to s1
minus s2 by v1 minus v2 or in other words it is delta s by delta v which is nothing but the
entropy of transition or the ratio of the entropy of transition with the volume change of this, so
therefore this gives us the Clapeyron’s equation you can see where we find that the slope of the
pressure versus volume curve it is obtained from the slope of the pressure versus temperature
curve it is obtained from the entropy of transition divided by the volume change on transition.

411
So therefore from this we find that the slope of the of the PT curve if we observe the slope of the
PT curve which I have shown in the last class if you recollect the slope of the PT curves can be
obtained from that particular equation, the equation which we say it is commonly known as the
Clapeyron equation. So from this Clapeyron equation what are the things that we get? First thing
is suppose we apply it there are number of things.

(Refer Slide Time: 11:39)

First thing is this can be applicable to any phase equilibrium problem say any phase equilibrium
of any particular substance provided it is a pure substance we can apply it and if you observe the
important part is we find that we can use the thermodynamic data, the data on delta s del delta v
to predict the appearance of the phase diagrams and their forms, so therefore what I mean is we
can predict the slopes of this PT curves for the different phases and we can also predict which
graph is going to be steeper which is going to be less steep and we can do all these things using
the thermodynamic data.

412
(Refer Slide Time: 11:49)

Let us see what this particular equation reduces to? For example if we consider the solid liquid
boundary when we consider the solid liquid boundary what is Delta s? Delta s is nothing but
delta s fusion which is nothing but delta H fusion by the melting point of the solid so from there
what do we get? We get dp dT is nothing but equal to Delta h fusion by T delta v fusion, right?
Now usually what we find, usually this Delta v fusion is nothing but v liquid minus v solid.

413
(Refer Slide Time: 12:39)

For most of the substances this is positive and therefore it is quite evident if you observe here the
slope of this particular curve it is, we see that it that it is more or less positive. Now very
interestingly if you find if you observe water then in that case we know that delta v for water that
is going to be negative, why? Because yesterday also we discussed and you already know that
the water it contracts on liquefaction.

So therefore if you observe the curve for water what you find? You find that there is a slight
negative slope and that arises because of the negative value of Delta v fusion. We also see that in
this particular case Delta v fusion is usually much smaller and that explains for the very steep
curve the solid liquid transition curve for both the cases that I have depicted here and also for
several other cases that you can you can see or you can take up in different textbooks.

414
(Refer Slide Time: 13:52)

So therefore the conclusion is that for the solid liquid boundary we find that the slope is usually
positive except some examples like water, antimony, bismuth and other thing is slope is usually
very steep and so therefore from here we can we can easily integrate this equation as well and on
integrating this equations say from a pressure P1 to P2 and assuming that Delta h fusion and
Delta v fusion they are more or less constant with pressure.

We can get an expression like the final pressure can be obtained as the initial pressure plus Delta
v fusion ln T2 by T1. So in this particular equation if we know the total number of parameters
are 1, 2, 3, 4, 5, 6 out of these if we know 5 of the parameter we can find out the sixth one. Well,
this was for the solid liquid boundary.

415
(Refer Slide Time: 14:50)

(Refer Slide Time: 16:47)

Now let us apply the same equation to the liquid vapor boundary as well. So if we apply it to the
liquid vapor boundary starting from the same Clausius Clapeyron equation we know dp dT this
is equal to delta s by delta v. So therefore when it is liquid vapor then this is going to be delta s
vaporization and the delta v vaporization.

416
So this is nothing but delta h vaporization by T the boiling point temperature into v vapor minus
v liquid, now in this particular case usually unlike the solid liquid transition we know that
usually the v vapor is much greater as compared to the v liquid and we also know that for low to
moderate pressures this can be equal to, this obeys the ideal gas equation.

So from here we can write this down as dp dT is nothing but equal to delta h vaporization by RT
square by P or we can also rearrange to write to denote dln P dT is nothing but equal to Delta h
vaporization by RT square and this particular equation is known as it is a very famous equation it
has got a lot of applicabilities and this equation is known as the Clausius Clapeyron equation, on
integrating this equation what do we get?

(Refer Slide Time: 17:05)

On integrating this equation we get ln P2 by P1 this is equal to delta h vaporization by R into 1

by T1 minus 1 by T2. So therefore what does this imply? This implies, so therefore I have just
written down the equations which we were discussing for solid liquid transition this is the
equation for liquid vapor transition the Clausius Clapeyron equation has been derived and from
this equation it is quite evident that ln P versus 1 by T should be a straight line.

And here I have depicted for a large number of liquids the log P saturated versus the 1 by T and
we find the linear nature of the curve for all the substances despite the large number of
assumptions which are inherent in this particular derivation. We will discuss the assumptions
shortly after we deal with the solid vapor boundary.

417
(Refer Slide Time: 17:46)

Now in the solid vapor boundary also it is almost the same thing and the only thing that we do is
we replace Delta h vaporization with Delta h sublimation.

As a result of which the equation which we obtained in these particular cases ln P or else I will
write the better form dp dT is nothing but Delta h sublimation by RT square by P and keeping in
mind that Delta h sublimation is nothing but delta h vaporization plus Delta h fusion this
automatically explains that if you observe the PT the curve you will see that the solid vapor
transition line it is much steeper as compared to the liquid vapor transition line and this can be
explained from the fact that it arises since delta h sublimation is involved in the solid vapor the
PT curve of the solid vapor transition curve and Delta h sublimation is much greater as compared
to Delta h vaporization.

So therefore from here we find that generally for transition for the condensed phase to the vapor
phase we find that the equation can more or less be written in the form as log P or log 10 P it is
something of A minus B by T the equation is something of this form and from this log P or ln P
in whatever way you wish you can write it down and you see the utility of this particular
equation.

418
(Refer Slide Time: 19:34)

The utility of this particular equation or even the equation written in this form is first thing we
can predict the P versus T curve from thermodynamic data on latent heat of vaporization. So
therefore from here the first thing that we can find out is that how vapor pressure varies with
temperature this is the first thing that we can find.

The other thing that we can find is how boiling temperature or boiling point varies with pressure.
And the other interesting thing which we can find is we can find out Delta h vaporization from
PT data this is the other important thing we can find data on Delta h vaporization as well as Delta
h sublimation both of them and from the data.

419
(Refer Slide Time: 20:40)

From here if we try to find out Delta h vaporization it is nothing but equal to RT square by P dp
dT. Similarly Delta h sublimation it is nothing but equals to we can write it down in this
particular form. So therefore we find that from these equations which we have derived from the
basic equation of homogeneous open systems the simplest type comprising of single component
and 2 phases, what do we find?

420
(Refer Slide Time: 21:16)

We find that we are in a position to predict the slopes of the PT curve here and also if we know
the slope we can predict the heats or the enthalpies involved in the transition criteria. Now we
need to remember despite the success of the Clausius Clapeyron equation it has got a large
number of assumptions. Now what are the assumptions?

421
(Refer Slide Time: 21:42)

The first and the most serious assumption is for all these integrations that we have done here I
have assumed the basic thing that delta h vaporization is constant only under that condition I
have come from this step to this step and this is the most drastic assumption.

422
(Refer Slide Time: 21:58)

Because if you observe the PV curves we can very well observe for the liquid vapor zone, we
will be discussing this in the next class and then it will be clear to you how the most drastic
assumption is regarding assuming the consistency of the of delta h vaporization the latent heat of
vaporization. So we continue with this discussion in the next class.

423
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 23
Homogenous Open Systems(Contd.)
(Refer Slide Time: 0:36)

Well to continue with our discussions where we have ended in the last class was I had derived
the Clausius clapeyron equation, from the Clausius equation and the final derivation where I
ended was the Clausius Clapeyron equation expressed in this particular form and we find that a
more popular equation which relates vapor pressure with temperature is just a empirical
modification of the Clausius Clapeyron equation, in fact that equation you must have heard and
that equation is much more popular and that is used much more frequently, it is an empirical
modification of the Clausius Clapeyron equation which is expressed as log 10 P equals to A
minus B by t plus c.

In this equation if you, this equation is known as Antoine equation and for most of the purpose
we will find that if you have to find out vapor pressure at some temperature or if you want to find
out the boiling point at some particular pressure usually we use this particular equation if you
compare there 2, we find that there are basically 2 differences one is, it is expressed in log10
whereas this is the 10 base and this is with the ln and the other thing we find that the parameter C
is anything any value other than 273.15 Kelvin, so this equation is much more used.

424
Now coming back to the assumptions which were involved in Clausius Clapeyron equation
despite the fact that it can be so successful and it is used so very frequently, possibly it’s thought
that a fortune cancellation of the errors inherent due to the different assumptions gives or it's
responsible for the success of the Clausius Clapeyron equation. Now what is the most drastic
assumption? I have just mentioned it in the last class.

(Refer Slide Time: 2:23)

425
(Refer Slide Time: 2:28)

The most drastic assumption is that I have assumed as I have assumed delta h vaporization to be
constant based on which performs this particular integration in coming from this equation to this
equation.

(Refer Slide Time: 2:43)

In reality we know that Delta h vaporization it is a significant function of temperature and this
will be evident if you observe the Pv curves which I have shown in this particular diagram. We
find that the liquid region exists at low in this particular region the vapor region exists under this
particular region these are the isotherms which have been drawn the constant temperature
isotherms and
426
we will find that there is a dome where the liquid vapor both of them coexist when any particular
system within this particular dome then it keeps on changing its composition, sorry.

It keeps on changing its phase from the pure liquid states to the pure vapor phase it keeps on
changing its composition from the pure liquid state to the pure vapor phase and when it is inside
the dome we find that depending upon its location it has different proportions of liquid and
vapor. We find that as we go towards higher and higher pressures we find that this liquid vapor
dome it keeps on becoming smaller and smaller.

And the horizontal portion which marks the phase change from the liquid to the vapor region that
keeps on getting shorter and shorter till it reduces to a point at the critical point. So therefore at
the critical point what we observe is the partial the molar volume of the liquid phase and the
vapor phase they are the same and therefore none of the phases they can be distinguished from
one another and under this particular condition we will find that the transition is not abrupt it is
just continuous.

(Refer Slide Time: 4:21)

So therefore accordingly based on this if we plot delta h vaporization versus temperature we find
that that the curve is something of this sort where delta h vaporization becomes 0 at T equals to
Tc. So therefore from here it is quite evident that in no way delta h vaporization is constant
unless we assume the 2 temperatures T1 and T2 over which the transition or over which the
phase transition takes place are very close to one another.

427
So therefore my next attempt will be to find out the dependence of the latent heat of vaporization
on temperature and this particular derivation this has been derived from the or rather this is
known as Kirchhoff’s equation we know that the latent heat of vaporization this is nothing but
equal to h vapor minus h liquid, so therefore d of this is going to give me this, right?

(Refer Slide Time: 5:28)

Now this is the latent heat of vaporization it is dhv minus dhL. If you recall the property
estimations for single phase substances that is in closed systems we had related h with
measurable parameters T and P. If you recall those equations that we had derived dh was equal
to CpdT plus v into 1 minus beta T dp, right when we were trying to express h as a function of T
and P, fine. So therefore we can substitute this equation for the vapor phase and for the liquid
phase in this particular equation once we substitute what do we get?

We get d delta hv this is nothing but equal to Cp vapor minus Cp liquid dT plus v vapor minus v
liquid minus v vapor beta vapor minus v liquid beta liquid T into dp, right? We get this equation
and from this equation suppose I would like to find out the variation of the molar latent heat of
vaporization with temperature, what do we get? The equation is something of this sort vg minus
vL minus vg sorry it's I missed it should be vv minus beta v minus vL beta L T, delta hv by T
delta v, isn’t it?

What I have done? I have substituted dh from the basic equation that we have derived and then
have substituted the Clausius clapeyron equation for dp dT in this particular case. So from there

428
what do I get? I get d delta hv dT in this particular case it is nothing but equal to Cp vapor minus
Cp liquid plus delta h v by T this and this cancels out minus Vv beta v minus vL beta L by Vv
minus vL into delta hv and this particular equation this is known as the Kirchhoff’s equation to
predict the latent heat of vaporization in terms of temperature and pressure and this can find, so
therefore this is the equation this is primarily for the liquid vapor transition.

Now what happens for the, and in this particular case it is important for us to remember that if
we assume that the vapor phase is an ideal gas then for that case beta v is nothing but equals to 1
by T and again if we assume that beta v is much greater than vL as a result beta v beta v will also
much greater than vL beta L and substituting all these things here we finally get d delta hv dT
that is nothing but equal to Cp vapor minus Cp liquid.

The other terms we find they are going to cancel out, so therefore this particular equation gives
us the variation of latent heat of vaporization with temperature.

(Refer Slide Time: 9:18)

429
(Refer Slide Time: 9:36)

Now suppose I would like to find an identical thing for the solid liquid transition. In that
particular case suppose I would like to find the delta h fusion then in that case here also I repeat
the same thing I do the same thing I tried to substitute dhL minus dhs I write this in terms of Cps
CpL etc and then the equation that I get is CpL minus Cps plus repeating VL minus Vs minus vL
betaL minus vs beta s T into delta hf by T delta v, right?

From where again if I start substituting I get CpL minus Cps plus delta hf by T minus vL beta L
minus Vs beta s by vL minus vs delta hf by T and if we assume that vL, vs your CpL, Cps all of
these are constant and I also assume that quite naturally in both the solid and for the liquid they
are incompressible liquids, so beta is negligible as a result of that in this particular case the
equation reduces to delta hf by T which is nothing but CpL minus Cps divided by T.

430
(Refer Slide Time: 11:01)

So therefore from here what did I get? If you observe that we could actually predict the variation
of the latent heat of vaporization and the variation of the latent heat of fusion with temperature in
terms of the specific heats of the 2 phases in fact it is important for you to remember that when I
assumed that the latent heat is constant it means that I assume that this specific heats of both the
phases in or in the respective phases remain constant with temperature.

And so therefore from these equations we can also find that we can predict delta hv at
temperatures other than the normal the boiling point, we can predict delta hv at temperatures
other than normal boiling point using the Kirchhoff’s equation, okay and I would also at the end
like to mention a few other equations which can be used for these purposes those equations are
primarily empirical in nature.

One of the equations is due to Watson it is known as the Watson correlation which gives delta
hv2 equals to delta hv1 1minus Tr2 by 1 minus Tr1 to the power 0.38 where 1 and 2 these two
refer to the temperature of States one and two and Tr is nothing but the reduced temperature
which is nothing but T by Tc.

431
(Refer Slide Time: 12:50)

And for finding out the latent heat of vaporization at the normal boiling point this is easily
estimated from another correlation which is known as the Riedel correlation. From here we can
find out delta hvn it is nothing but 1.093 RTc Tbr ln Pc minus 1.013 by 0.930 minus Tbr. So
therefore what I mean to say is and in this particular equation we will find delta hvn it is nothing
but the enthalpy of vaporization at normal boiling point Tc is nothing but the critical temperature
as we know Pc is nothing but the critical pressure and this equation Tbr is nothing but the
reduced boiling point and reduced normal boiling point of the particular substance with which
we are dealing with.

So therefore we can use the Riedel correlation to find out the molar latent heat of vaporization at
normal boiling point.

432
(Refer Slide Time: 14:07)

And then we can either use the Kirchhoff’s equation or we can use the Watson correlation to find
out the molar latent heat of vaporization at any temperature other than the normal boiling point.

So with this I would like to conclude this particular session on homogeneous on homogeneous
open system, the simplest type of open system which comprises of just one component and it
exists in 2 phases and we found that by using rather by using the basic equations of
thermodynamics we could actually predict either the pressure temperature data from the
thermodynamic data like molar enthalpy of vaporization or molar enthalpy of transition or the
vice versa.

From here we could also find we could also deduce 2 equations namely the Clapeyron and the
Clausius Clapeyron equation and I would like to remind you once more that the Clausius
Clapeyron equation despite its large number of assumptions serves quite well for practical
purposes but usually we use the Antione’s equation to find out the vapor pressure, in the next
class what we are going to do? We are going to deal with heterogeneous closed systems.

So initially what I did? We started our discussions with homogeneous closed system, in
homogeneous closed systems we assume that it was made up of a single component or even if it
is a mixture the composition of the mixture did not change and we tried to define a large number
of properties for those particular systems and we tried to relate the measurable and non-
measurable properties and the basis for that was Maxwell’s equation.

433
After that we just relaxed the assumption further we went for homogeneous open systems and we
deduced the Clausius Clapeyron equation, in the next class we are going to deal with the
heterogeneous closed systems which comprises of a large number of homogeneous open systems
which have been enclosed or rather which are enclosed by a boundary and the entire composite
system does not interact with the surroundings with respect to mass transfer or with respect to
volume transfer it can interact with respect to energy interaction. So we are going to deal with
heterogeneous closed systems in the next class.

434
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 24
Heterogeneous Closed Systems

Well, hello everybody so as I had discussed in the last class in this particular class we are going
to deal with heterogeneous closed systems, right? So therefore what do we assume? We assume
that there are large number of homogeneous systems. Just in the last class I had taken up to
homogeneous open systems inside a closed system. It was one particular pure substance into 2
different phases.

(Refer Slide Time: 0:56)

So therefore in this case I would like to extend it slightly further and I would have say I have pi
number of phases. Since the derivation is slightly involved it is very important that we keep all
our nomenclatures in a proper form, okay. So therefore it has say pi phases and the phases has
are numbered with alpha such that alpha varies from 1 to pi. And if we assume that there are N
numbers of components where each or rather we can take it as C number of components where
the components are numbered as n this is one to C and so long what we have done?

435
(Refer Slide Time: 1:59)

For these particular type of heterogeneous closed system we have already derived the situation
we have already derived the condition of equilibrium for different type of interactions between
the close system and the surroundings and we have derived the following equations that has been
noted down here. As I have already mentioned the first equation is for constant S and V and then
it is S and P then V and T and this is for constant P and T conditions.

So therefore in this case we already know the condition of equilibrium for heterogeneous closed
systems but we know these in terms of thermodynamic potentials or we know these in terms of
extensive parameters. Very frequently we would like to express them in terms of intensive
parameters which can be easily measured or they can be easily estimated.

Now we already know the conditions we know that I had mentioned in the first-class that
thermodynamic equilibrium is the simultaneous maintenance of thermal, mechanical and
chemical equilibrium and that time I had also mentioned that thermal equilibrium implies
equality of temperature, mechanical equilibrium implies equality of pressure and chemical
equilibrium implies equality of chemical potential but just by the statement of someone it is not
always good to get convinced.

So I would prove it from you I would start with the condition of equilibrium which we had
already derived so long from one of these equations and I will arrive at the condition of
equilibrium expressed in terms of intensive variants.

436
(Refer Slide Time: 3:44)

Now here I would like to mention that is particular derivation it was for or this particular prove
was first given by Gibbs and what he did? He used the first equation say dU equals to Tds minus
Pdv plus Sigma I, mui, dni he used the equation for dU in order to do this derivation. The
primary reason for using U as a starting point was possibly because of the symmetry of the
equation if you observe.

We find that in the expression of U each differential on the right-hand side is a differential of an
extensive property and the coefficient that we have that is an intensive property. Usually if you
consider the other equations we find that the differential and the coefficients there are mixtures
of intensive and extensive properties and just because this symmetry is there only in U.

So therefore Gibbs also started proving the condition of thermodynamic equilibrium from dU
from the expression of dU and we will also be doing the same thing and this is and from here we
are going to start and we are going to arrive at the uniformity of all intensive variables at
equilibrium by considering just one function U.

So therefore while we are considering this particular function the generalized expression this was
written for just one phase. Now if we have pi’s number of phases so naturally then in that case
this generalized expression should be written in this particular form it should be T alpha dS alpha
where Alpha equals to 1 to pi. Now here I would again like to mention I have taken up pi number
of phases I have taken up C number of components and the phases are numbered with alpha the

437
components are numbered with n and I denote the phase by means of a superscript I denote
component by means of a subscript.

Since the derivation is slightly involved it is very important for us to remember the conventions
that we are using. So therefore and if we consider this equation again we find that since there are
more than one phase, so therefore this should be P alpha this should be dv alpha this also should
be from Alpha equals to 1 to Pi. What about the last equation? In the last equation we find that
here I should be equal to 1 to C and before that there should be alpha equals to 1 to pi, in that
case as I have already mentioned my subscript refers to the component and my superscript will
be referring to the phases.

So from here what do we get? From this particular situation, so we find that for all the phases I
can write down this equation. Now if I slightly elaborate it and write, what do I get? If I just
breakdown the sigma values it is basically T for phase 1 dS1 minus P1dV1 plus mu1 phase1 dn1
plus mu2 phase1 dn2 plus so on to muC dnC phase1 plus again for the second phase it is T2 dS2
minus P2 dV2 plus mu1 in phase 2 dn1.

Sorry I need to refer to the components here dn1 in phase 2 plus mu2 phase 2, dn2 phase 2 plus
so on till we get muC2 dnC2 I can keep on continuing this for phase 3 phase 4 and finally for the
Pith phase it is going to be P pi dV pi plus mu1 pi dn1pi, mu2 pi dn2 pi plus so on till mu Cpi dn
Cpi, right? And we have to remember that we are dealing with a heterogeneous closed system as
I have mentioned in here.

438
(Refer Slide Time: 8:43)

So when it is a heterogeneous closed system and I am dealing with dU, so therefore this equation
should be supposed to hold, so therefore what do I mean to say is this whole thing should be
equal to 0, right?

Now apart from this we also need to remember that there are also some additional constraints
here, what are the additional constraints? The first thing is the individual variations of entropy
the individual variations of volume and the individual variations of the total number of moles of
each chemical species they can vary but subject to the condition that the total entropy, the total
volume and the total number of moles remains constant or in other words ds which is equal to
ds1 plus ds2 plus so on till ds pi this should be equal to 0 because after all it is a close
heterogeneous closed system.

Same way dv1 plus dv2 plus so on till dv pi this should be equal to 0. In the same one for each
and every individual component we can write down dni alpha this should be dni1 plus dni2 plus
so on till dni pi this should be equal to 0. So therefore what do I get? I find that I have written
down one particular equation in terms of how many variables?

439
(Refer Slide Time: 11:00)

(Refer Slide Time: 10:43)

I have written it down if there are pi numbers of phases then I have written down the whole
equation in terms of pi into m plus 2 numbers of variables and out of that, how many constraints
do I have? I have 2 constraints for sv and since there are C numbers of components I have
constraints for all these C number of components.

So therefore many number of constants, sorry it is not m it is C because C is my number of

components and how many constraints do I have? My numbers of constraints are C plus 2, so
therefore it is always possible that I can eliminate a few of these variables.

440
(Refer Slide Time: 11:10)

And I should be able to write it down in terms of pi into C plus 2 minus C plus 2 number of
variables or in other words this particular equation the equation for dU can be expressed in terms
of pi minus1 into C plus 2 number of variables. Or in other words these number are the truly
independent variables, since we have eliminated other variables using the C plus 2 number of
constraints these are the number of constraints that we have, these are the number of variables
that my system has.

441
(Refer Slide Time: 12:09)

So therefore now I should be in a position to express the entire equation of dU in terms of lesser
number of variables, here the number of variables as I have mentioned is pi into C plus 2. I
should be in a position to eliminate C plus 2 variables from here. How can I do this? I know this
particular equation, so instead of ds1 can I not substitute it with minus of ds2 plus so on to ds pi?
Same thing can I not do it with dv and same thing can I not do it with dni alpha?

442
(Refer Slide Time: 12:55)

So what do I mean is instead of ds1 I can substitute it as minus ds2 plus ds3 plus so on instead of
dv1 can I not substitute it with dv2 plus dv3 plus so on.

443
(Refer Slide Time: 13:04)

And in the similar way can I not substitute each and every ni1 with minus of ni2 plus ni3 etc. So
once I make the substitution and I put it here, so instead of dS1 what am I going to have? Instead
of this dS1 I am going to have minus of ds2 plus so on and so forth, isn’t it? Instead of this dV1
also I should be having minus of dv2 minus of dv3 plus so on and so forth I can write it in this
particular term.

444
(Refer Slide Time: 13:50)

So once I write it down in this particular term I find that there is a ds2 and there is a ds2 here,
there is a dv2 there is a dv2 here as well. Same way there is a ds3 and there is a ds3 here.

445
(Refer Slide Time: 14:04)

So once I start arranging it in this particular form what is the resulting expression that I get? The
resulting expression which I get is T2 minus T1 into dS2 minus P2 minus P1 dV2 plus mu12
minus mu11 dn1 of 1 plus so on and so forth till muC2 minus muC1 dnC2 and in the same way I
can write T3 minus T1 dS3 minus P3 minus P1 dv3 plus mu13 minus mu11 dn13 this was dn12
dn13 plus so on and so forth till muC3 minus muC1 into dnC3 and this we can keep on
continuing till for the pith phase we can write T pi 1 minus T1 dS pi P pi minus P1 dV pi plus
same way mu1 minus mu11 dn1pi continuing till mu Cpi minus muC1 dnC pi, right?

446
(Refer Slide Time: 15:52)

(Refer Slide Time: 16:13)

So therefore I had started with this particular equation I had minimized I had considered the
constraints and then I could express this equation in lesser number of variables I could express it
I had started with pi into C plus 2 variables I could express it in terms of pi minus 1 into C plus 2
variables and this is the final expression that I obtained and I know that for the closed
heterogeneous closed systems this whole thing should be equal to 0.

447
(Refer Slide Time: 16:37)

Now if this has to hold if you observe the individual terms we find that in order for the entire
expression to be 0 the individual variations or the individual coefficients of ds2 dv2 dn12 dn22
etc all the individual coefficients of each and every differential they should be individually be
equal to 0 if the entire composite dU has to be equal to 0. And what does it imply? It
automatically implies that each of these terms they will be individually 0 provided T2 minus T1
equals to 0 again T3 minus T1 equals to 0 and so on. The same way I can write V2 minus V1
equals to 0 V3 minus V1 equals to 0. So in general sorry not v very sorry P2 T3 minus P1 very
sorry.

448
(Refer Slide Time: 18:55)

So in general from here what do I deduce? I deduce that for thermodynamic equilibrium in a
heterogeneous closed system starting from dU s,v equals to 0 I have arrived at a condition that
T1 equals to T2 equals to Tpi, P1 equals to P2 equals to P pi, Mu1 of component1 equals to mu1
component 2 to mu1pi. In the same way we can continue to use C1 equals to mu C2 equals to mu
Cpi and therefore what have I arrived at the end?

At the end I find that the thermodynamic equilibrium express in terms of the thermodynamic
potential reduces to the condition of equality of temperature, equality of pressure and equality of
all the chemical potentials of each and every component present in the mixture and therefore we
can express thermodynamic equilibrium either in terms of the extensive properties or we can
express it in terms of the intensive properties it is all the same.

449
(Refer Slide Time: 19:13)

And very importantly from here I have come across another very important observation, what
did I observe? We observe that I had started initially with pi into C plus 2 numbers of variables.
Now this entire pi into C plus 2 numbers of variables are not independent, why? Yesterday we
had derived Gibbs Duhem equation and while deriving the Gibbs Duhem equation we found out
that for each and every phase the S,T and all the compositions are not independent if suppose it
is a single component then S,T and Mu cannot be varied independently if 2 are fixed the other
automatically gets fixed.

So therefore we find out of this pi into C plus 2 numbers of variables that we have the actual
number of variables are pi into C plus one because there is a Gibbs Duhem equation relating the
different number of variables in each of the phases that we are considering. So therefore from
here what do we find?

450
(Refer Slide Time: 20:37)

We find that in this particular case the actual number of independent variables the number of
independent variable in each phase is m plus 1 since the Gibbs Duhem equation relates or rather
relates the meth variable with the m plus 1 variable. So therefore the number of independent
variables in pi number of phases should be equal to pi into m plus 1, right? And if the system is
in a state of internal equilibrium then in that case we find that as we have just now found that
there are pi minus 1 sorry all of these are C, I am extremely sorry. There are pi minus 1 into C
plus 2 equilibrium relations connecting them this we have just now we have found this out.

So therefore we find that if there are number of variables are pi into C plus one then for a state to
be in a state of internal equilibrium there are pi minus 1 into C plus 2 equilibrium relations. So
from there what do we deduce? We deduce that the number of degrees of freedom which we had
defined earlier as F, which is nothing but the number of intensive variables if you recollect they
were dealing with intensive variables used to characterize a system minus the number of
relations or restrictions connecting them, what is this equal to?

This is going to be equal to Pi into C plus one minus this particular equation, so therefore F is
nothing but equal to pi into C plus one minus pi minus 1 into C plus 2 which is nothing but equal
to C minus pi plus 2 and if you recollect this is the Gibbs phase rule that I have stated without
proving and this is a proof which is available which shows us that for any particular system in a
state of internal equilibrium be it homogeneous be it heterogeneous be it a pure substance be it a

451
mixture of substances whatever it is, the number of intensive properties that are needed to
characterize the system depends upon the number of components and the number of phases and
the relation is expressed by the Gibbs phase rule.

Now well, I end here but before I end I would just like to mention that I have written down the
definitions of the number of phases and the definition of phase and definition of component and
the number of degrees of freedom or variance because very frequently I find that there are quite
number of confusions regarding these things.

(Refer Slide Time: 23:37)

So therefore you can go through these particular definitions which states that the phase is nothing
but a part of the system uniform throughout in chemical composition and physical properties and
each particular phase is separated from the rest of the system by means of boundary surfaces
when we have single phase systems they are homogeneous when we have more than one phases
then they are known as heterogeneous systems.

452
(Refer Slide Time: 24:06)

And the number of components is nothing but the minimum number of species which are
necessary to define the composition of the system. For example if we have an acetic acid water
mixture then in that particular solution there will be acetate ions, hydrogen ions, hydroxyl ions
and so on and so forth but the number of components are 2 its acetic acid and its water since they
are soluble.

So therefore the number of phases is 1 in this case and according with number of degrees of
freedom or the variance that I have already written down. So therefore in this particular class I
have dealt with closed heterogeneous or rather heterogeneous closed systems and I have derived
the criteria of thermodynamic equilibrium with respect to intensive variables and from there I
have also deduced the Gibbs phase rule which you will be using very frequently henceforth.

In the next class we continue with our discussions on the physical interpretations and the means
of determining the chemical potential on mu which we have already come to realize is quite an
important parameter as far as phase equilibrium thermodynamics is concerned, thank you very
much.

453
 Course on Phase Equilibrium Thermodynamics

By Professor Gargi Das

Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 25
Tutorial
(Refer Slide Time: 0:23)

Well, so today we have come to the to the tutorial after the 4th week. So we are already halfway
through the thermodynamics course. I believe you must have faced quite a number of doubts and
we have also got some very interesting clarification and doubts and questions from your side. We
welcome much more questions on this. In fact, we would also welcome suggestions from you
regarding (ex) explaining once more some portions which were not very clear or maybe certain
portions which you think had become more difficult and you would like to have, like me to
explain it once more.

It will be good if you can put up all the suggestions on the portal. We will try to accommodate
them as much as possible. Now coming to the tutorial after the end of the 4th week. Means we
are already halfway through and this is the tutoria particularly pertaining to the next week.

454
(Refer Slide Time: 1:21)

So in this particular, the 1st problem is that let us say this this problem how how do we handle
this particular problem? What does it say? It says that liquid mercury it boils at about 357 degree
centigrade.

(Refer Slide Time: 1:36)

So therefore the problem 1, the boiling point of mercury that is given as 357 degree centigrade.
Heat of vaporisation is also given in calories per gram, mind it. So therefore it is not it just note
that this particular heat of vaporisation which is given, this is given as the specific latent heat of
vaporisation and not the molar latent heat. So therefore you have to keep it in mind that you are

455
going to convert this to the latent to the molar latent heat of vaporisation by multiplying it with
the molecular weight of mercury which is 201.

And then it is given, it is required that the pressure which is required to distill mercury at 150
degrees centigrade. So therefore we know that at the boiling point of mercury, naturally it is
going to have the saturated vapour pressure, right? So therefore, at the P at this particular boiling
point, this is nothing but equal to 760 millimetres mercury and we want to distill mercury not at
the boiling point but at 150 degree centigrade.

So therefore T2 that is 150 degrees centigrade and we would like to find out what is the
approximate pressure required for this particular distillation. Now you know that from the data
which is given, you are given the boiling point, the corresponding saturation vapour pressure
which is known to you, your latent heat of vaporisation is given. If you assume that the latent
heat of vaporisation does not change from 150 degrees centigrade to 357 degrees centigrade
which is definitely as we have already discussed is quite a drastic assumption but nevertheless,
we make this assumption to for simpler calculations.

Then in that case, we find that the Clausius–Clapeyron equation can be quite suitable for it. What
is the Clausius–Clapeyron equation let us recall. We had, I had already discussed the Clausius–
Clapeyron equation in the class. It was LN P2 by P1 is equal to delta H vaporisation by R 1 by
T1 minus 1 by T2. Is not it? So therefore if we use this, we find that we are required to find out
P2. We know T1 right?

Delta H vaporisation, this has to be 68 into 201 and this is given in calories. So therefore the
corresponding R also has to be taken in calories. So therefore this is 2 and 1 by T1 is 1 by 373
minus 1 by 630 right? So therefore if you solve this out, then what do you find? On solving this
out what do we find? We find that we get P2 as 0.425 millimetres mercury. So therefore we find
that this is the required pressure which is required to distil mercury at 150 degrees centigrade.

This was quite a direct Clausius–Clapeyron equation and let me tell you that that you will be
given some such direct applications of Clausius–Clapeyron equation in assignments and then
possibly will be evaluated also on the basis of these particular problems which are done in the
class. I would like to mention that whatever evaluation process we take up, that is at the final

456
semester exam or the final test that we take, more or less you will be dealing with whatever
problems have been done in the class, the problems or the questions which are asked in the
assignments.

One more thing also I would like to tell you that in the course of phase equilibrium
thermodynamics for the last 4 weeks, you must have faced problems by because you you had to
deal with a large number of formulae for estimation of properties and maybe several gas laws
and several other equations. Now we would be you will be required definitely to remember
certain simple equations. For example, Clausius–Clapeyron equation you are supposed to
remember all or it is not very difficult to derive it.

Ideal gas equation, Vander Wal's equation you have been doing for a very long time. I suppose
you already know these equations by heart. But all the difficult equations, say Maxwell's
equations and several such equations, they they will be posted before your test and you can refer
to those particular equations when during a test season but it is not a bad practice to remember
some of these equations for your future usage of thermodynamics right? So therefore this was
one particular application of Clausius–Clapeyron equation.

(Refer Slide Time: 6:45)

457
Now let us go to the next problem. A slightly more involved problem. What does it tell you? It
tells you that water boils at 95 degrees centigrade in a particular hill station, right? We know that
at the mean sea level, water boils at 100 degrees centigrade. So therefore, definitely the pressure
on the top of the hill station is quite high as compared to the sorry it is quite low as compared to
the pressure at the sea level., right?

So therefore, we know that at the sea level, the pressure the water boiling point, this is 100
degrees centigrade and here, the boiling point is 95 degrees centigrade. Why does this happen?
Why is the boiling point difference? Naturally because the pressure here has to be lower as
compared to the pressure here. What makes the difference in pressure between the 2 places? It is
naturally the difference in height between the 2 places.

So definitely, between P1 and P2, there is a difference in height which brings about the
difference in pressure, isn’t it? So therefore, this arises due to rho gh of the air column between
the 2, is not it? So therefore, from here we know that P2 minus P1 into A will should definitely
be equal to rho gh right? So therefore due to this particular altitude difference, there is a pressure
difference which brings about the difference in the boiling points.

And you are here the question which is there, you are asked to estimate the approximate altitude
and the 2 conditions. What are the 2 conditions? 1st condition is atmosphere is isothermal. What
does it imply? It implies that in the atmosphere, PV equals to constant. Or in other words, that V

458
is this molar volume or in other words, your P2V2 at the altitude is equal to say P1V1 at the sea
level. So this one information is given.

Possibly if we are going to use this information in order to express P2 in terms of P1 and then
relate this particular equation to from the basic force balance equation to find out H. This is
problem 1 and the problem 2 is to repeat the whole thing if the atmosphere is adiabatic. Let us
deal with problem 2 part A first and then we go to part B, right? So here again use the same
equation LN P2 by P1 this is equal to delta H vaporisation by R I had already written it down
earlier 1 by T1 minus 1 by T2, fine?

Now here we know everything except P2. This is the only unknown that we have. You find other
than that, we know everything. What is P1 equals to? P1 is at this P1 it is the atmospheric
pressure. So therefore, from here we know and Delta H vaporisation it is also given. It is given as
2256.94. In this case, just note that it is given in kilos joules per kg right? So it is better you
convert it to joules because we know R in terms of joules.

So therefore, this becomes into 10 to the power 3 joules per kg, right? So therefore this is
2256.96 into 10 to the power 3 joules per kg. Accordingly your R this is going to be 8.314.
Again in this case, you note that the latent heat of vaporisation is the specific latent heat, it is not
the molar latent heat. But in this case, you note that it is the the substance or the fluid in between
it is air.

So therefore what do we do? We simply multiply it with the molecular weight, the average
molecular weight we assume and from here I am just substituting T1 and T2 minus 1 by 368.15.
Just note certain things, when you are doing these problems, firstly T1 and T2 even when they
are given degrees centigrade, they have to be converted to Kelvin, the absolute scale
temperature.

Next thing is please note the unit in which Delta H vaporisation is expressed. If it is expressed as
per kg, we are required to convert it to per mole. If it is and also note the unit in which Delta H
vaporisation is expressed, you are required to express R also in the same units, right? So from
here you know that LN P2 by P1 this is equals to minus 0.1778, right?

459
What are you required to find out? You are required to find out the height difference when you
know the the ratio of the pressures or the pressure difference between the 2 right? So you need
one more equation to relate the pressure difference to the altitude of that particular place. From
physics, we know how we are supposed to these.

(Refer Slide Time: 12:42)

We take an infinitesimal volume element say. From the infinitesimal volume element, this is say
the infinitesimal volume element right? And in this infinitesimal volume element the cross-
sectional area is say A, the small height is say dh, here the the pressure suppose it is say P plus
DP into A. Here the pressure is P into A. And this pressure is acting here. And we find that the
the weight of the fluid which is which is acting downwards what is it? It is Adh, the volume into
divided by the specific volume into g right?

So therefore for for this particular volume we know that since there is no acceleration, so from
force balance what we get? From force balance, we get, this is going to be P plus dP minus P into
A, this is equals to Agdh by V. Or in other words, it is better you write it down in this way. This
is equal to 0. Is not it? So P plus dP into A minus PA plus this, this is the pressure force and this
is the gravitational force, this equals to 0.

From where we we get, dP is nothing but minus gdh by V. Please note that V is nothing but the
specific volume, right? Now for isothermal conditions, what do we know? We know as I have

460
told you, P2V2 equals to P1V1 where P1V1 refers to this to the sealevel, this refers to the sea
level and this refers to the height above the altitude. Agreed? So therefore can we not write down
this particular equation, say equation A if we can substitute dP, we find that this equation dP, this
gives you as this is gdhP by or rather gdhP2 by P1V1 is not it?

From where, we find that sorry this is P L or dP by P this is nothing but equal to minus gdH by
P1V1 integral from 0 to H and this integral extends from P1 at the sea level to P2 at the altitude
h? So from here what do we get? We get LN P2 by P1 this is equals to minus g by P1V1 into h
agreed? And we know that P1V1 since it is since for this particular case, we can we can easily
write express it as g by RT1, the temperature here into M into h where we know what is M.

M is nothing but the molecular mass of air right? It is the molecular weight or the molecular
mass of air. What is this equal to? 28.97 into 10 to the power minus 3 kg per mole. This is the
molecular weight of air right? g also you know it is the acceleration due to gravity. In SI units,
this is going to be 9.81 metre per second square fine? You are supposed to find out h. R
accordingly you have to substitute R in the variant equations.

So therefore in this particular case, R is going to be 8.314 in terms of joules and your molar mass
I have said, so therefore now equating ln P2 by P1 we have already obtained lnP2 by P1 in this
particular case if you see.

461
(Refer Slide Time: 17:12)

462
This is equals to minus 0.1778. So we can relate the 2. Minus 0.1778 this will be equal to this
particular portion here, so we can substitute all the terms here. This is equals to 9.81 into 28.97
into 10 to the power minus 3 into h divided by 8.314 into the temperature 298.15. So from here,
what do you get? You get h equals to 1550.8 metres. Well I will I I will I will I will write it
slightly more.

This is 1550.8 meters, right? So therefore what did you do in this case? In this particular case,
basically using Clausius–Clapeyron equation you found out what should be ln P2 by P1 equals to
right? And then what you did? You perform the force balance here. From the force balance,
equating the forces which is acting on this particular infinitesimal volume of air of cross-
sectional area A and height dh, the forces which are acting on it are the pressure forces which act
due to the pressure difference and this is balanced by the force of gravity on this particular
volume element.

(Refer Slide Time: 18:50)

463
 So accordingly, from this particular
force balance, you get this equation. And then the only thing which is given to you connecting
dP and V is that the air column or the atmosphere behaves as an I be its it is in under isothermal
conditions. So from there, you get this particular equation for relating dP and V. Accordingly
you make the relationships and from here, you get this particular equation.

(Refer Slide Time: 19:21)

464
Now this I have find, you have 2 equations for ln P2 by P1. One is this equation, one is this
equation. Simply equate the 2 equations and you get h. Now this was for the 1st part of the
problem. What about the 2nd part? The 2nd part, it is slightly involved.

(Refer Slide Time: 19:36)

465
It is said that the atmosphere instead of behaving isothermaly, it behaves adiabatically and the
gamma is given. So therefore in this particular case, we find that the relationship between, the
Clausius–Clapeyron equation, that particular equation it is it is applicable in this case, no matter
whether the atmosphere behaves under isothermal or adiabatic conditions. So for part B,
definitely we can take up this particular portion.

(Refer Slide Time: 20:09)

466
And we can write down for part B, ln P2 by P1 equals to minus 0.1778 this we have obtained
from Clausius–Clapeyron equation right? This we already know. Now for force balance, the in
the same way that we had done the force balance here, in the same way we can perform the force
balance, same approach and we get this equation as well. So we write down this equation also,
this equation is dP equals to minus gdh by V.

The only thing comes that when you want to relate P and V, then in the last problem you had
tried to relate them assuming the atmosphere to be isothermal, now you will be trying to relate
them assuming the atmosphere to be under adiabatic condition. So in the previous problem, the
relationship was P2V2 equals to P1V1, in this particular case, the relationship is as you know, it
is P1V1 gamma equals to P2V2 gamma right?

This is the only equation that you are going to use. This is the only difference from the previous
problem but naturally, the calculations become slightly involved in this particular case. So
therefore if we if you are proceeding in the same way, we find that in this particular case, dP
equals to P1V1 the power gamma minus gamma V to the power minus gamma plus 1 into dV,
just the same way that we had we we had proceeded in the previous case, we do it and then we
substitute this particular equation in this equation right?

When we substitute, just substitute it and then you see what you get. The equation is going to be
something of minus gamma P1V1 to the power gamma V to the power minus gamma plus 1 dV

467
will be equal to minus dh by V. Or we can write it down as V the power minus gamma dV equals
to gdh gamma P1V1 to the power gamma, right? So you can you can keep on expanding it and
finally when you expand it, you find that V2 by V1 to the power 1 minus gamma, this is equal to
1 plus 1 minus gamma by gamma gh by P1V1.

You can again elaborate it for further. This V2 by V1 to the power gamma we can always write
it write it down as P1 by P2 to the power gamma minus 1 by gamma, these are all adiabatic
equations. This can be written down then as 1 plus 1 minus gamma by drama, the same thing that
I have written down here gh by P1V1 right? So therefore from here, what do we get?

(Refer Slide Time: 23:31)

We get from here that P2 by P1, this is simply 1 plus the reciprocal of the thing which I had
written down there, gh by P1V1 whole to the power gamma by gamma minus 1 right? And from
there, I get lnP2 by P1. I can I can very well substitute everything here. Now instead of P just
like I had done in the previous case, instead of P1V1 I can substitute it with RT1. For 1 mole,
this is going to be RT1 by M. And therefore I can I can from here I can very well write, this I
will write later. From here, I can very well write this down as P2 by P1 equals to substituting all
the values we are going to get 1 minus gamma is 0.4 because gamma is 1.4 as it is mentioned.

468
g is equal to 9.81 just like the previous case, molecular mass is 28.97 into 10 to the power minus
3 into h this divided by gamma which is 1.4 into R 8.314 into T which is 298.15, the room
temperature. So simply substitute it and this we know, this is ln P2 by this is P2 by P1.

(Refer Slide Time: 25:17)

469
And from Clausius–Clapeyron we had obtained ln P2 by P1 equals to this. From here, we get P2
by P1 this is equals to 0.8371. So I can very well equate this P2 by P1 with the value which I
have got here, I can equate both the values, this is equals to 1 minus 3.2755 into 10 to the power
minus 5 into h whole to the power 3.5 right? So from here, I get h equals to 1512.3 metres.

So you find in order to find the altitude, it depends upon if the atmosphere is behaving
isothermally or if the atmosphere is behaving adiabatically, it is this portion simply matters when
we try to relate the pressure and the specific volume. And for this particular equation, it is
important to consider the atmosphere in one particular state but other than that, the rest of the
problem is same for part 2A and 2B. We need to perform a force balance in order to relate your
height with the pressure and then we need to perform or rather we need to find out the ratio
between the pressures from Clausius–Clapeyron.

And equating the Clausius–Clapeyron equation with the force balance equation where we have
made necessary corrections as by considering the atmosphere 1st as isothermal and then as
adiabatic and after that equating, we find out the altitude. So therefore, this is one particular way
by which you can find out the altitude provided the altitude is high enough to bring about a
reasonable difference in the boiling point of water right?

So now we go to the next problem. We had enough of Clausius–Clapeyron. So at the moment,

we would go for some property estimations. Now in this regard, I would like to tell you that we
have already discussed property estimations and we found out that there are several equations

470
and when we do not know exactly which equation the gas obeys, then under that condition, we
either refer to the compressibility factor chart or we refer to the to the reduced coordinate charts
which expressed Delta H star and Delta S star in terms of a unique function of reduced
temperature and pressure.

Now mostly in industries, we find that the charts, they are very very much used but since it is
difficult for us to provide chart etc, so therefore we will be considering primarily on finding out
these departure functions from equations of state. Now since for example purpose, using Redlich
Kwong and Ping Robinson which is again mostly used in industries. Particularly the Redlich
Kwong equation is very much used in the hydrocarbon industries.

The procedure is the same but since the calculations become slightly involved and we are all
dealing with hand calculations for the purpose of this class, I we would primarily like to be
rather primarily like to keep our problem within simpler equations of state. It can be a Vander
Waal's equation of state, it can be a Berthelot equation of state which is also a very simple
equation of state or some particular equation of state which is much less involved, just to
simplify calculations but it is important for you to remember that usually in industries, we I
either go for the generalised charts, charts which have been provided by by various workers or
we use, usually we use the Redlich Kwong equation of state.

(Refer Slide Time: 29:23)

471
So if you remember, when I was taking the class, I had derived the expressions for enthalpy and
entropy departure functions right? And I had also mentioned if you if you recall, I had also
mentioned that these departure functions, they are expressed in the terms of PVT where you
require V to be a function of P. We had found that most of the equations, the cubic and the real
equations, it is often more useful for us if we can express the functions in terms of P as a
function of V and T.

So therefore we have to make some particular modifications of this equation such that we could
express the enthalpy and entropy departure in terms of dV instead of dP. I do not know whether
you remember this or not but we had in order to change these particular expressions, what did we
do? VdP is nothing but equals to be known that V, PV equals to RT, right? So from there we
found out VdP this is nothing but equals to dPV minus PdV. This is true not only for ideal gas,
this is true for every gas obeying any particular equation of state.

So instead of VDP, we can we can make the substitute as dPV minus PdV. What happens then?
Then what are the limits of dPV? When P equals to 0, we know the gas behaves as an ideal gas.
So PV becomes equal to RT under this particular condition, right? So so therefore we make this
following substitution. If we make this following substitution, then instead of VdP we write it
down as dPV minus PdV, right?

472
Now when we write this as dPV minus integral PdV fine? The limits are from P equals to 0 to P
equals to P. For P equals to 0, we know the gas behaves as an ideal gas. The gas behaves an ideal
gas at P equals to 0, what is the corresponding PV equals to? For ideal gas, we know PV is
nothing but equals to RT for 1 mole of the gas. And for P equals to P, naturally this is going to be
PV, right? And we know that for an ideal gas, the volume can be approximated as V infinity or
V0 fine?

And volume here, it is V. So instead of this 1st term, we can substitute it with the 2 terms. And
then if you recall, we had mentioned that from the cyclic rule, we can write DelV Del T at
constant P, this particular term, we can very well replace it with Del P Del T at constant V. How?

(Refer Slide Time: 32:17)

473
From the cyclic rule, Del V Del T at constant P into dP constant T this can be very well written
down substituted as minus Del P Del T at constant V into dV at constant T. So therefore from the
cycling rule, we have gone there. So therefore instead of this particular term, we can very well
substitute it in this particular fashion fine? By doing this, what do what have we actually
achieved?

By making these particular substitutions, we find that the equations which we had obtained here,
this particular equation, we can write it down, I will just write down the equation once more for
your convenience. V minus T Del V Del T at constant P whole thing as dP which can be written
down as P equals to 0 to PVdP minus T integral P equals to 0 to P Del V Del T at constant PdP
right? So instead of VdP, what do we do?

We substitute it just the way I have written down, we substitute it with PV and then integration,
what does it give? It gives this PV minus RT. So therefore from this particular term, we get PV
minus RT and for the next substitution, what do we get? It is V equals to infinity to V T Del P
Del T at constant V minus PdV this is at constant T, fine? Or else we can write it down as in
terms of Z if you want to write, RT into Z minus 1 plus integral V equals to infinity to V T Del P
Del T at constant V minus P into dV.

474
So if you observe this equation, you find that you can find out the enthalpy departure function for
any particular equation which expresses P as a function of T and V. Now let us see what was the
what was the problem which has been which was given, the previous problems which we were
dealing with.

(Refer Slide Time: 34:44)

This particular problem, this was to calculate the enthalpy and entropy departure function of
what? For an ethylene gas, right? Now for this particular ethylene gas, the gas is at 8.25 MPa
pressure.

(Refer Slide Time: 35:10)

475
So P is given 8.25 MPa and the temperature is given, T small T is given as 25 degrees
centigrade. It is given that the gas obeys Vander Waal's equation of state. The a of the, a and b
are not given, you can note it down. It is 453.046 into 10 to the power minus 3 Pascal metre cube
per mole whole square, b is 0.057 into 10 to the power minus 3 metre cube per mole. This is
given, right? So once these 2 are given, then in that case, what you need to do.

(Refer Slide Time: 36:11)

Just in this particular case you know that here in this equation, you can substitute the the for the
Vander Waal's equation if you see, this is the Vander Waal's equation. This is the Vander Waal's

476
equation and till this particular equation, it was a home assignment. If you if you make this the
the necessary substitutions here, then in that case the final equation which you arrive at is this.
Once you arrive at this equation, find, finding out Delta h star is no problem.

T and V are known, RT are known, A is already given, so you can find out Delta h star. Same
way, you can find out Delta S star as well. Right?

(Refer Slide Time: 36:54)

And if you wish a slightly more elaboration on this on the substitution, then I would like to tell
you that well P is RT by V minus b minus a by V square. Just to tell you how we arrived at this

477
at this particular equation this particular equation from this equation. This was your home
assignment. I believe that most of you have done it already. In case you have got stuck up, I will
just like to do it once more.

P is given, Del P Del T at constant V is very simple. R by V minus b. So therefore Delta h star I
have already found out. This is PV minus RT. So P is already given, so PV minus RT is very
simple for you to find out. This is going to be equal to RT into PV by RT minus 1 right? And
that if you substitute Del P Del T at constant V here then what do you get? You get integral V
equals to infinity to V, this is going to be RT by V minus b from here if you observe, minus
again RT by V minus b plus a by V square dV fine for V equals to V to infinity.

From here, what do we get? We get from here, RT RT cancels out, so you get PV minus RT and
here these 2 cancels out. So therefore minus a by V. You just make the necessary substitutions
and you are going to get the answer for this particular equation.

(Refer Slide Time: 38:48)

The answer for this particular equation is…for this particular equation P, T, everything is given.
So therefore for this particular equation, your Delta H star if you make all these substitutions,
then in that case, I will just make the substitutions for you. 1st term is PV minus RT minus a by
V. What is P equals to it is given? It is given 8.25 mega Pascals. So into 10 to the power 6 into
1.1673 into 10 to the power minus 4, this is V.

478
One thing I would like to remind that this V you have to find out from the basic Vander Waal's
equation you are required to find out V under this pressure temperature conditions. To find it out,
you find V to be equals to 1.673 into 10 to the power minus 4. It might happen that in some
problems just to reduce the calculation time and all, we might be providing you the volume but
normally under conditions, you are supposed to find out the specific volume, right?

Minus RT? What is minus RT? R in this case is 8.314, T is 298.15 minus a by V. a is already
given, 453.046 into 10 to the power minus 3 divided by 1.1673 into 10 to the power minus 4,
right? Make the calculations, you find that the Delta h star which is nothing but the actual value
h minus the ideal gas value, this is going to be minus 5.397 kilojoules per mole. Make the
calculation and find it out for yourself.

(Refer Slide Time: 40:39)

In the same way you are supposed to proceed for s minus s0. This also if you recall, this part also
we had derived. Again here, we have Del V Del T at constant PdP. Substitute it with this. And
we have R by P dP right? So therefore in this case also, you you can substitute it. If you make
the necessary substitutions, then what do you get? Once you make the necessary substitutions, I
will just write it down once more.

(Refer Slide Time: 41:13)

479
Your Delta s star then becomes equal to, in this case it is minus Del V Del T at constant P
integral P equals to 0 to P minus R by P dP at constant T fine? So this if you make make the
substitution, you get V equals to infinity to V Del P Del T at constant VdV plus R dP by P
integral P equals to 0 to P. How to make this? This is already substituted, so you have V equals
to infinity to V Del P Del T at constant VdV.

For this case, what do you write? For this case, you write integral dPV by PV you simply write
down the equation, then you make the you you just write it down and divide by V, you get this as

480
R integral dPV by PV minus dV by V right? Here here again you are going to put the limits,
what are the limits? This is RT to PV. What are the limits for this? V equals infinity to V.

So what do you get finally? Finally the things which you get is RlnPV by RT which is again
nothing but z. This this particular thing is nothing but z. So you can also write it down as Rlnz.
Plus you get V equals to infinity to V integral Del P Del T at constant V minus R by V I will just
write it down. Integral V equals to infinity to V this is dV, this is dV is not it? I have taken this
term out and I have combined these 2 terms because both of them are in terms of dV. So
therefore what do I get here?

This is Rlnz plus integral V equals to infinity to V Del P Del T at constant V minus R by V dV
which gives me on slightly little more substitution, it gives me Rlnz plus Rln if you observe, this
particular case, we get dV by V, so RlnV minus b by V infinity minus V minus RlnV by V
infinity. Agreed? So therefore, from here we can you can make the substitutions. Or in other
words, you can also write it down as Rlnz slightly more substitution, V minus b by V infinity
minus B minus RlnzRT by P into P0 by RT.

Can we now do it? Because V infinity is the ideal gas volume right? V is zRT by P. I have just
made some substitutions. So therefore this gives me as Rln because Rlnz Rlnz they are going to
cancel out and finally what I get, RlnV by B by V minus b plus RlnP by P0. Agreed? Which is
nothing but RlnP into V minus b by P0 into V infinity minus b. Or in other word, this is nothing
but RlnP into V minus b by RT. is not it?

So therefore you find Delta s star from a very simple equation. Again if you if you refer to the
previous problem which was there, if you if you go to the previous problem you find that in this
particular equation you know R, what is R? 8.314 lnP you already know 8.25 into 10 to the
power 6 right V minus b fine? This is 1.1673 into 10 to the power minus 4 minus b 0.057 to 10 to
the power minus 3 divided by R 8.314 into 298.15.

You can find it. This is nothing but minus 13.431 joules per mole Kelvin. Please try it out for
yourself and see if you can get this particular equation or not, right?

So so this is we have found out enthalpy and entropy departure functions, we have assumed that
it obeys the Vander Waal's equation of state. So therefore we have expressed it in terms of a

481
(pre) from a pressure to of volume explicit form that is we have expressed is expressed it as P as
a function of V and T. We have used that equation and we have found out the enthalpy and
entropy departure functions.

(Refer Slide Time: 47:06)

Now suppose it is not a Vander

Waal's equation of state, suppose I say that well, it obeys some other equation of state which is
not a Vander Waal's equation of state but it is more or less it is it shows that instead of a by V
square, there is also a temperature term there. Quite expectedly because intermolecular force of
attraction depends on the temperature as well right? So for this case, just proceeding in the

482
similar manner, can we not find out the Delta h star and the Delta s star, the enthalpy and entropy
departure functions just the way we have done it for the Vander Waal's equation of state.

(Refer Slide Time: 47:52)

We start in the same particular manner, this is our basic equations. This and this are our basic
equations. We make this and this substitutions just the way we had done for the Vander Waal's
equation and then what do we get?

(Refer Slide Time: 48:10)

483
We get Delta h star equals to RT into z minus 1. Just same equation I am writing down. Infinity
to V T Del P Del T at V minus P into dV just the equation we had obtained. This can be written
down as PV minus RT as you know plus integral V equals to infinity to V, this is 2A if if you
make the necessary substitutions, then in that case, if you write the Berthelot equation, I will just
go to the equation once more for you to see it.

If you go to this particular equation, what does it give you? It gives you P equals to RT by V
minus b minus a by TV square, right? So therefore and then from here you can also find out what
is Del P Del T at constant V? This is nothing but equals to R by V minus b right? So therefore
we can very well substitute that whatever we have obtained, and so therefore here in this
particular case what do we get?

We get this as 2a by TV square dV fine? And then this becomes equals to PV minus RT minus
2a by TV which gives you RTV by V minus b minus a by TV minus RT minus 2a by TV. On
further simplification, we get bRT by V minus b minus 3a by TV. You can compare this with the
Vander Waal's expression that we had obtained. In the same way, you can proceed for in order to
find out the entropy departure function as well.

(Refer Slide Time: 50:19)

484
In the entropy departure function also, again I start with the same equations which I had derived
right? Equation was Rlnz plus V equals to infinity to V Del P Del T at constant V minus R by V
constant TdV which can be written down as in this particular case if you substitute all these
things, you get V equals to infinity to V we get R by V minus b minus a by T square V square
minus R by V to dV, right?

So here what do we get? We get RlnZ plus R lnV minus b by V infinity minus b minus Rln V by
V infinity minus A by T square V which can further be substituted as plus Rln. With all the
simplifications you get V minus b by V minus a T square V, right? So therefore if you know or
rather if you can express your enthalpy and entropy departure function then in that case whatever
be the equation of state, you can substitute it and accordingly we can you can find out the
enthalpy and entropy departure functions for any particular cubic equation of state.

The same is true for the Vander Waal's equation of state as well. If you do not know whether the
gas obeys the any particular equation of state, then definitely we have to resort to the generalised
charts to find out the enthalpy and entropy departure functions. So this completes our assignment
or rather our tutorial for the 4th week. I would welcome much more questions from you so that
for the tutorial of the next week, we can include all those doubts and we can address whatever
concerns you have. Thank you very much.

485
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 26
Concept of Fugacity

Hello everybody, so I would like to remind you that in the last class we had discussed the phase
transitions of pure substances from one physical state to another. Now just one thing I would like
to mention regarding that whatever transitions we have discussed in the last class all of these
they are referred to as the first order transitions or the transitions of the first kind because in
those particular transitions we found that the first derivative of thermodynamics potential, for
instance in this case it is G.

(Refer Slide Time: 0:59)

The first derivative is s minus Del G Del T at constant P and the volume which is Del G Del P at
constant T, so these 2 parameters they had undergone an abrupt discontinuity as a transition as a
result these transitions they are known as the first order transitions or they are known as the
transitions of the first kind.

Here I would just like to mention that in the second order transitions what happens is that, the
first derivatives of the thermodynamic potentials namely in this case it is entropy and volume the
first derivatives of the thermodynamic potential Gibbs free energy they remain constant while
the

486
second order the derivative it exhibits an abrupt discontinuity at the phase transition, for example
what is the second-order rather second derivative.

It is suppose say Del 2 G Del T 2 at constant P, what is this? This is nothing but equal to Del Del
T of Del G Del T everything at constant P. So therefore this is nothing but equal to minus yeah or
in other words this is nothing but equals to Del S Del T at constant P which is nothing but equal
to CP by T. So and in the same way we can also write Delta 2 G Del T Del P this is also this can
also be found this is equal to beta v.

So therefore for the second-order transitions we find that these 2 parameters they would be
undergoing an abrupt discontinuity while the first derivatives would remain constant and so
therefore whatever transitions we had discussed in the last class all of these they refer to the first
order transitions. Now along with that I would also like to remind you that in thermodynamics
we have been dealing with equilibrium states. Now all these equilibrium states that we deal with
all of these refer to stable equilibrium states.

We will be recalling from mechanics a very difficult diagram where this refers to the stable
equilibrium, this refers to the unstable equilibrium; this refers to the metastable equilibrium
states. Now for most of the cases we deal with stable equilibrium states while phase transitions
they refer to metastable equilibrium states and some particular equilibrium states for example
suppose water is existing below 0 degrees centigrade or in other words sub cooled water or
maybe water existing above hundred degrees centigrade for example it is the super heated water
or maybe steam existing below hundred degrees centigrade which is nothing but the sub cooled
vapor. So these particular states which are extremely unstable, so these particular states they are
known as the metastable states in classical thermodynamics.

We mostly deal with stable equilibrium states and during phase transitions we deal with
metastable equilibrium states. I would also like to mention that just like in mechanics we have
found out that the potential energy extremum in the state of equilibrium and that implies that
when that extremum is maximum the state is unstable when the extremum is minimum then this
equilibrium is a stable equilibrium.

In thermodynamics also as I’ve already mentioned in one of my previous classes stability of a

thermodynamic system in terms of the extremum of some thermodynamic property it can be

487
either maximization of entropy or minimization of energy. Well, so with this I come to the topic
which we are going to discuss today. We are going to discuss little more about the phase
transitions or rather phase equilibrium thermodynamics and here I would like to mention that the
task of phase equilibrium thermodynamics is to describe quantitatively the distribution at
equilibrium of every component among all phases which are present.

For example suppose we are performing distillation of a mixture of alcohol and water and we
would like to know that under a certain temperature and pressure conditions, how much amount
of alcohol or water would be distributed between the liquid phase and the vapor phase. Again
suppose we would like to extract acetic acid from a solution of toluene, acetic acid in toluene
using water as the solvent.

We would like to know how much amount of acetic acid would be distributing itself between the
2 liquid phases or in other words what is the maximum amount of acetic acid that can be
extracted from toluene using water under the same conditions of temperature and pressure. Now
the thermodynamic solution to such phase equilibrium problems was obtained several years ago
by a person Gibbs who we will be referring very frequently during the course of this phase
equilibrium thermodynamics.

He gave a solution to this problem by introducing the abstract concept of chemical potential
which we have defined and we have discussed some aspects of it in the last 2 classes. Now we
just know that all the chemical potential it has got a profound influence on phase equilibrium
thermodynamics it is basically an abstract concept. It does not have an imagent equivalent in the
physical world and therefore it is desirable to express chemical potential in terms of some
auxiliary function which might be more easily identified with physical reality.

In this class we would attempt to relate the abstract concept of chemical potential of a substance
through a physically measurable quantities like temperature, pressure, composition etc. Now I
am attempting to simplify the abstract equation or rather the abstract concept of chemical
equilibrium, GN Lewis he first considered the chemical potential of a pure ideal gas and this I
guess I had already derived in one of the previous classes.

488
(Refer Slide Time: 8:14)

We found out that chemical potential for ideal gas at constant temperature can be obtained from
this particular equation. He observed that under isothermal conditions the change in the abstract
thermodynamics quantity that is chemical potential is a simple logarithmic function of a
physically real quantity that is pressure and so in this particular way he could relate a
mathematical abstraction to a common intensive property of the real world but we need to
remember that this was applicable just for an ideal gas.

(Refer Slide Time: 9:10)

489
Let us see what happens when we are dealing with a real gas situation. Say for example a real
gas which is obeying the virial equation of state, what is the virial equation of state? We have
already discussed in, while we were discussing the PVT Behavior of the gases. This we write it
down in the pressure explicit form where the constants or functions of temperature only. So this
can be written in this particular form.

From here we can very well-define V as RT by P which is nothing but equal to dividing
throughout by RTP RTP by P, sorry multiplying throughout we get the equation in this
particular form plus RT B3P as a function of (T) P plus so on and so forth. Now suppose we we
try to compute dmu for this particular condition. We know that for constant temperature this is
equals to vdp, so therefore if we compute this from say state one to state 2, we try to compute
this from state one to state 2, what do we get?

The expression is something like 1 to 2 d ln p plus RT B2p (T) integral 1 to 2 dP plus RT B3p as
a function of (T) 1 to 2 Pdp plus so on and so forth. On integration we can write it down as mu2
minus mu1 equals to RT ln say P2 by P1 plus RT B2p as a function of (T) into P2 minus P1 plus
RT B3p (T) P 2 square minus P1 square plus so on and so forth. Now this is applicable for any
other equation of state as well.

490
(Refer Slide Time:
11:38)

Now we find that this equation is exact but this will be different for each and every gas and it
turns out to be more convenient particularly for calculations involved in chemical equilibrium to
maintain the form of equation which is shown in to the form of the equation which is represented
here. So what Lewis he attempted, he wanted to maintain this form of an equation by defining a
property or an auxiliary function which is fugacity where fuga In Latin means flight or escape.

491
(Refer Slide Time: 12:16)

(Refer Slide Time: 12:25)

So therefore what he did? The equation which was developed for or rather which was obtained
for ideal gas he tried to maintain this particular form of the equation for all substances whether it
is substance or a component in a mixture for all substances he tried to retain this particular
expression simply by replacing P with F the fugacity of the mixture, from where we can get just
like we had got for the case of an ideal gas we can get, on integration of this equation we can
get this as RT ln f by f0, right? It’s better we write fi referring to a component i.

492
(Refer Slide Time: 13:15)

So therefore if we do this then in the process, what have we done? We have varied the entire non-
ideality of the gas in the expression for f. In this particular case what is the expression of f, if we
assume the gas to obey the virial equation of state? This will be equal to, then d ln p plus say B2p
(T) dP plus say B3p (T) Pdp plus so on and so forth. So therefore what do we find? What have
we done in this particular case?

493
(Refer Slide Time: 13:42)

(Refer Slide Time: 13:51)

We have just retained the type of the expression which we have obtained for an ideal gas and in
and this particular equation this is varied for any component in any system whether it is a solid, a
liquid, a gas whether it is a pure whether it is mixed whether it is ideal whether it is non ideal and
everywhere we can we can use this particular function or this functional form of fugacity and we
would like to remember that this particular equation will reduce to the equation RT d ln P when
we would be dealing with ideal gases.

494
The only thing which we, there is 2 things which we need to remember in this particular respect.
First thing is we need to remember and in fact I had mentioned this earlier regarding other
properties as well here also we need to remember that the defining equation of dmu be it for a
real gas or for an ideal gas we find that we cannot compute the absolute value of mu because the
defining equation of mu is in terms of a differential equation which on integration gives a
difference quantity.

So therefore the important part here to remember is that we can define the entire equation but the
standard state that we select that is arbitrary. We can select anything as the standard state but we
need to remember that the standards states have to be same for both mu i0 and fi0. If suppose for
mu i0 I select the lower limit of integration to be at P equals to 1 bar then fi0 also has to be
computed at 1 bar, this is the first thing which we need to remember the entire thing arises
because the equations which we are dealing with in phase equilibrium thermodynamics most of
these are differential equations or the difference equations.

In fact let me remind you that phase equilibrium thermodynamics becomes much more complex
not due to the concepts of the equations but more due to the need or the importance of the
reference states which are known as the standard states. The other thing which I need to mention
is that if I have defined the quantity fugacity then it is very important and I would like to find out
this particular quantity fugacity at any particular conditions of temperature and pressure and that
has to be provided I perform this integration with respect to some particular condition where I
know the value of fugacity, is it clear to you?

495
(Refer Slide Time: 16:47)

What I mean to say is that if I want to calculate or rather if I want to find out the utility of this
particular equation then I need to integrate this particular equation and if I am integrating this
particular equation it has to start from some particular reference condition where the value of
fugacity is known. Now where do I know the value of fugacity? I know the value of fugacity at
very low pressure under which condition all substances can be assumed to be in the gaseous state
and the gas at such low pressure behaves as an ideal gas because I know when the gas behaves
like an ideal gas then under that condition the equation is d ln p.

So therefore the point is that this particular definition it is not complete unless I specify the lower
limit or I specify the limiting condition of fugacity and this particular limiting condition is
suggested in this particle form, when we are dealing with a pure gas or single component gas
then the condition is limit P tends to 0 f by P equals to 1 and suppose I’m dealing with a mixture
of gases which behave ideally at low pressure then under that condition it becomes limit P tends
to 0 f by the partial pressure of the gas at 0 at the very lower pressures.

So therefore it’s very important for us to remember that once we have defined fugacity it is
important to remember that in order to define fugacity we need to equations, the first equation as
I have already mentioned it is d mu equals to RT d ln f and along with that we need to mention
the a subsidiary equation which states this where for pure gas yi becomes equals to 1 and
condition becomes limit P tends to 0 f by P equals to 1, unless this is specified the definition

496
remains incomplete because just a differential equation without an idea of the value at any of the
end points of the integration is definitely meaningless.

Now here I would like to mention that it might appear to you that we are just going around and
around in circles, what are we doing? First we defined chemical potential; it was an abstract
quantity, we did not know what to do with it? Or rather we wanted to find out a straightforward
way of defining it. You wanted to keep matter simple, so therefore we define fugacity this is
another unknown quantity adding to our problems.

(Refer Slide Time: 20:22)

So therefore immediately it appears that we are we are just going on in circles and we are not
doing exactly we are not achieving anything useful we are not going in a straight road, road
towards understanding more of phase equilibrium thermodynamics but the reality is, in defining
fugacity, what we have actually done? We are now in a position to use all thermodynamic
functions which have been defined for ideal gases just by replacing pressure with fugacity.

497
(Refer Slide Time: 20:35)

(Refer Slide Time: 21:09)

So the only thing that we need now at this moment is to find out or rather to devise a
straightforward way to define fugacity and the other important part is, it’s very important to
remember that is particular equation this is applicable, as I’ve already mentioned this is
applicable for solids, liquids, gas, the gas is pure, mixture everything it is applicable.

So in that case the power of the definition of fugacity arises from the fact that we are in a
position to define fugacity for any particular component, for any particular system, for whatever

498
we define the limiting condition that I have proposed here for the limiting condition that is
written down here, it holds well and by defining this particular the quantity we are in a position
to use whatever thermodynamic functions we have developed a device for ideal gas.

(Refer Slide Time: 22:01)

So therefore there are 2 other things which I would like to clarify before winding up this
particular lecture. The first thing which I would like to mention is that very frequently we would
be coming across a term like f by P this term we across very frequently. In fact we get this term
more frequently than f because for most of the cases we would like to define fugacity with
respect to pressure at the low-pressure.

499
(Refer Slide Time: 22:51)

So this particular term is usually defined as a fugacity coefficient phi. The other thing which is
very important is this particular term f by f0. We find what does this term signify you tell me this
term signifies the activity or this term signifies how active the substance is in the state of interest
compared to its standard state. So therefore this particular term f by f0 this is also defined as
activity.

It gives an indication of how active a substance is relative to its standard state because we would
like to remember that f by f0 or in other words ln f by f0 or RT lb f by f0 it gives the difference
between the chemical potential of the substance at the state of interest and the standard state and
we would also like to remember another very important fact that this particular derivation was
done isothermally and therefore it is important to remember that when we are dealing with these
equations the standard state temperature must be the same as the temperature of the state of
interest while the pressure and composition need not and indeed they are intended not to be the
same. We continue with the discussions regarding fugacity, fugacity coefficient activity in the
next class as well.

500
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 27
Fugacity(Contd.)

Well, I continue my class from the place where I had left my previous class just because this
concept it is slightly difficult to understand, so I would like to revise from where I had left the
last class. In the last class what did we do?

(Refer Slide Time: 0:40)

I defined d mu as RT d ln f and this is constant temperature any substance, any system, any state
and then I also defined d mu equals to RT d ln P ideal gas pure component and usually this is for
an ideal gas. So therefore this and usually we find d of mu minus mu 0 this is equal to RT d ln f
by P, right?

Now if we have to define and from there we found that this particular quantity is known as the
fugacity coefficient. We also found out that if we have to move from one state to the other then
in that case suppose I move from one particular state mu to some other reference state, so
therefore in this particular situation or rather if I take the state as mu 0 itself this is also equal to f
by f0 where I need to remember that the standards state of mu0 and the standard state of f0 they
have to be the same but there is no other restriction on the selection of the standard states, we can
select anything as the standard state as long as we can define mu0 and f0 at the same state.

501
(Refer Slide Time: 2:34)

I would also like to remind you that in fact we can also select any hypothetical state as a standard
state as well and mostly we find that the different type of standard states that has been defined or
there, which are mostly defined in literature are shown in this particular slide where we see that
usually the standard states are selected for the different phases at say pure component gas in its
ideal state at 1 atmospheric pressure everything has to be at the same system temperature or it or
can also be pure component gas at the system pressure and the system temperature.

I just need to mention that the first standard state is usually used in chemical reactions while the
second standard state we will be using very frequently in vapor liquid equilibrium because in this
class we are going to deal mostly with vapor liquid equilibrium we are not going to deal
chemical equilibrium same thing applies for a liquid as well it can either be a pure component in
the liquid state at the pressure of 1 atmosphere or pure component in the liquid state at the
system pressure.

And again this is the, this is of interest to us since we will be dealing with vapor liquid
equilibrium, for solid it is pure component in the solid state at a pressure of 1 atmosphere. So we
find that we can use different standard states, only thing which we need to remember in this
particular case is that the standard state should be the same for chemical potential as well as for
fugacity.

502
(Refer Slide Time: 4:09)

503
(Refer Slide Time: 4:19)

And the other thing which I would like to mention here is that there was another term which we
had defined the term was defined as activity, why? Because again we find that this is another
term which we come across this term was defined as activity as I had mentioned in the last class
since it gives the indication of how active a substance is. It gives an indication of how active a
substance is relative to its standard states and this one thing which we need to remember.

504
The first thing is mu0 and f0 have to be at the same standard states the other thing we need to
remember that the standard state is with reference to the pressure and composition it is not with
reference to the temperature because we have been dealing with, the derivation was done under
isothermal conditions for each and every case this particular thing this particular condition was
maintained.

So therefore the standard states temperature must be the same as the temperature of the state of
interest it is only the pressure and composition which can vary between the 2 states. So therefore
to summarize the whole thing it's the thing which I would like to mention and this particular slide
has been prepared primarily keeping in mind the confusion which students have when they are
asked what is fugacity?

This is a problem which I have also faced several times whenever we try we ask the students
what is fugacity? They try to say that it is a corrected pressure and then they are asked why I
need to 2 to correct pressure and then the whole thing they just get confused. So therefore to
make matters simpler I have just summarized the concept of fugacity and it is just like this which
you need to remember for understanding chemical potential and fugacity it is that since it is
difficult to wish, sorry.

505
(Refer Slide Time: 6:18)

Since it is difficult to visualize chemical potential therefore the concept of fugacity it has been
introduced it can be more easily conceived, what is it? It is a corrected pressure which for a
component in a mixture of ideal gases is equal to the partial pressure of the component just as I
had shown in the last class this tends to 1 for limit P tends to 0, right? And why do you need to
correct pressure? It is we need to correct pressure just because they are typical non-idealities
which can be interpreted by their molecular consideration.

So therefore if it is if you are concerned with gas, fugacity is nothing but a measure of the non-
ideality of the gas. If it is for any other substance, it just notes the deviation of that substance
from the ideal gas behavior and by defining fugacity what have we done? We have actually it is a
conceptual aid by which we have translated thermodynamics to the physical variables.

We have actually related abstract concept of chemical potential to the concept of fugacity which
can be understood or related to the physically realizable world the only condition which we need
is to find out a straightforward way of defining fugacity but before that there are certain other
things that we need to do. First thing is we need to remember that the whole concept of fugacity
or chemical potential it came up just because we were trying to define chemical equilibrium.

We were trying to find under what conditions chemical equilibrium is achieved? What is the best
composition that we can get? What is the best separation that we can get from our process? And

506
all those things were dictated by chemical potential and accordingly what did we do in the last to
last class?

(Refer Slide Time: 8:21)

We had defined the conditions of thermodynamic equilibrium if you remember the conditions it
was T alpha equals to T beta equals to so on, it was P alpha equals to P beta equals to so on and
it was for each and every component it was mu i alpha equals to mu i beta equals to so on. So
now since we want to replace mu with fugacity we should be in a position to reframe the
conditions of thermodynamic equilibrium in terms of fugacity particularly we need to reframe
the chemical potential condition in terms of fugacity.

507
(Refer Slide Time: 9:09)

Let us see how we can we can do, we can come about do doing this? Let us start from the
beginning, simplest case we have say one component in 2 phases, the 2 phases are say alpha
phase and the beta phase. Now in equilibrium, what do we know? We know mu alpha must be
equal to mu beta, what is, how is mu alpha related to fugacity?

It should be equal to some standard state mu alpha plus RT ln f alpha by f0 alpha. Similarly what
is mu beta? It is nothing but equals to mu 0 beta plus RT ln f beta by f0 beta. Now we are stuck
up here, why? Because we cannot proceed further unless we have some idea regarding the
standard states in the 2 phase’s alpha and beta, what are the standard states? They may be same
they may not be same.

Suppose for the time being we assume that they are the same. If they are the same then what do
we have? We have this equal to this under equilibrium this equals to this and if these 2 are equal
automatically these 2 are equal because as I have said we cannot very these and these separately.
So when these 2 are in under the same standard states these 2 are also under the same standard
states.

So with these 2 equal, this and this equal and these 2 equal, what does it imply? It implies
automatically that f alpha equals to f beta but this was under the condition where we assumed
equal standard state same standard states for both. Suppose we assume that the standard states
are not

508
the same in what can they vary? They cannot vary in temperature remember because everything
we have done, we have done under isothermal conditions.

So they can just vary in terms of pressure or in terms of composition if that is so they do not vary
in terms of temperature then we can also have a equation relating the chemical potentials of the 2
standards states with the standard fugacity’s of the 2 standard states, isn’t it? We find that these 2
standard states are not the same the temperature has to be same. So either the pressures are
different or the compositions are different, the same thing should be applicable to both of them.

So therefore we are simply relating the change in chemical potential to the change of fugacity
under constant temperature conditions. So therefore the same equations which we had used
should hold for this condition as well.

(Refer Slide Time: 12:10)

So now we can replace this particular condition in these 2 equations if we replace what do we get
let us see? The first condition was mu Alpha equals to mu0 alpha plus RT ln f alpha by f0 alpha.
Second thing was mu beta was equal to mu0 beta plus RT ln f beta by f0 beta, if we subtract one
from the other what do we get? Which is equal to 0 and this minus this, this is nothing but RT ln
f0 alpha by f0 beta.

Now if you subtract from this what do we get? Can we write it down as f alpha by f beta minus
RT ln f0 alpha by f0 beta just rearranging this part? So therefore in this particular condition also

509
what do we get? We find that RT ln f alpha by f beta equals to, so therefore from this condition
also we find that since this part is equal to 0 this should be equal to this and we get this
condition.

(Refer Slide Time: 13:17)

So therefore what do we deduce from here? We find that whatever be the standard sate, we find
that irrespective of the standard states the condition of equilibrium reframed is this, for all
components i equals to 1 to C whatever components we have there is just one note of caution
which I would like to mention here that while equality of fugacity signifies chemical equilibrium
the condition that activities must, this equality of fugacity does not imply that activities in all the
phases should be equal.

The activities will be equal only for the special case where the standard state in all the phases are
the same because what is activity again if I recollect ai equals to fi by fi0. So even if fi is same
for all the states if fi zeros are different for all the states the ai is different.

510
(Refer Slide Time: 14:20)

In fact here I would like to pose a question for you, what is the difference between fugacity
coefficient and activity? Or rather under what conditions will activity and fugacity coefficient be
equal activities fi by fi 0 phi equals to fi by P, so it is very evident when fi0 equals to dv when we
define the standard state such that the standard state is the total pressure of the system under that
condition fugacity and activity are the same.

Well, so once having defined the or rather reframed the condition of equilibrium the next thing
which I would like to do before I go into the estimation of fugacities for solid, liquid, gases etc .
since whenever we define any new quantity we would like to find out the effect of temperature
and pressure on it.

So I would like to find effect of temperature and pressure on fugacity. Let us define it and we are
going to, our starting point will be the variation of chemical potential with temperature and
pressure we know that, why? Because in a in one of the classes I had spend 1 whole class rather
discussing the affect of temperature and pressure on molar Gibbs free energy. The reason for
doing that was that molar Gibbs free energy is equal to the chemical potential for a pure
component.

511
(Refer Slide Time: 16:00)

So therefore whatever equations we have derived for the molar Gibbs free energy is applicable
for the chemical potential and using the relationship between chemical potential and fugacity we
should be in a position to find out, how fugacity it varies with pressure under constant
temperature conditions? And how fugacity varies with temperature under constant pressure
conditions?

Let us start doing the equations, we know this equation and what is this equals to? Minus sdT
plus vdp we already know that, so therefore what is del ln f del P at constant T and the total
number of moles? This is nothing but equal to del mu del p at constant T and total mole and we
know what is this? This is equals to v. So therefore what is, just a minute there has to be an RT
here sorry, so therefore we deduce that del ln f del P at constant T, n is nothing but equals to v by
RT or in other words ln f by f0 equals to again if f0 is defined at P equals to 0 v by RT dp.

512
(Refer Slide Time: 17:19)

And what is the variation of fugacity with temperature? Again we start from the same thing we
know that RT, for everything we are going to start with the same thing, we integrate this
equation, again this is from mu 0 to mu, this is from f0 to f from where we get mu minus mu0
equals to RT ln f by f0, we can divide this by T, we can divide this by T. Now we know what,
now suppose we want to find out R del ln f by f0 as a function of T for constant pressure, what is
it? Del mu by T del T at constant pressure minus del mu 0 by T del T at constant pressure, you
must recall that this is nothing but equal to G.

So we had already defined it or rather derive the Gibbs Helmholtz equation. We can straightaway
substitute it here and therefore this is equals to minus h by T square plus of minus or rather
minus of minus h0 by T square. So therefore we can very well write it down as del ln f0 by del t
at constant P this is nothing but equal to h0 minus h by RT square.

513
(Refer Slide Time: 18:53)

The same equation which I have just jotted down and the derivations are here, where what is h0
equals to? This is the molar enthalpy at low pressure which is nothing but equal to Cp0 dT, what
is h equals to? It is the molar enthalpy at the pressure of interest at given pressure and therefore
what is this term? Term h0 everything is right. What is this term h0 minus h? This term is
nothing but the increase in enthalpy due to expansion from pressure P to P0 at constant
temperature.

So well, once we have understood this one more interesting thing since we know that all these
are properties of the system, so therefore we can arrive at the same particular expression by using
an alternative route.

514
(Refer Slide Time: 20:38)

I very frequently do this particular exercise just to show you that since we are dealing with
properties we can start and derive by any particular route, if you have done it correctly we will
be arriving at the same results. For example this equation we can also start in a different form
starting from the same equation, we are interested to find how this quantity varies with
temperature? We can just write it down in this particular form, right? This is equals to R ln f by
f0 plus RT del ln f by f0 del T at constant P and n, we can do it this is equal to R ln f by f0 plus
RT del ln f del T at constant P, n minus RT del ln f0 del T at constant P, n, fine.

Now we know that this term it has to be equal to 0 because f0 is equal to P0 and that is definitely
a constant quantity. So therefore what is this term then in that particular case? We can write this
term as minus s, this term is plus s0 this is equal to R ln f by f0 plus mu by T minus mu0 by T
and so therefore we can write down sorry this term was wrong, we can just write it down as this
plus RT del ln f del T at constant P, n. So therefore from there what do we get?

It is nothing but minus s plus s0 equals to, this mu by T is nothing but g by t it can be written
down as hT minus s minus h0 by T plus h0 plus RT del ln f del T at constant P, n. It is slightly
more involved thing but from here we find del ln f del T at constant P, n all these terms they
cancel out and we get this is nothing but equal to h0 minus h by RT square.

515
(Refer Slide Time: 22:58)

The same expression that I had obtained in this particular case as well, so therefore it hardly
matters which route you use what type of approach you do, since if your approach your concepts
are correct then by whatever way you do, you should be in a position to come to the correct
answer.

Well, after this the next thing which I would like to do is to estimate rather to devise or rather to
discuss different ways by which fugacity of a pure idea of a pure gas of a pure liquid can be
found out once we have understood that then we go for mixtures and there we try to find out how
the fugacity of component in the mixture can be estimated?

516
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 28
Estimation of Fugacity Coefficients

Well, so in this class we are going to discuss the estimation of fugacity coefficient for real gases.
Now here again I would like to mention we are first going to discuss the estimation of fugacity or
fugacity coefficients whatever it is, for real gases and then we are going to do it for the
condensed phase it's same for the liquid as well as the solid. After that we go for the, for
discussing the fugacity coefficient for one particular component in a mixture of gases.

(Refer Slide Time: 0:57)

Now whenever we do, we start from the same 2 equation for a real gas this is this, for an ideal
gas the equation is this because every time my standard for all particular system that I use my
standard is going to be an ideal gas. Now I subtract one from the other, so on subtraction what do
I get?

517
(Refer Slide Time: 1:13)

(Refer Slide Time: 1:20)

I get d of mu minus mu0 equals to each and every term I subtract these 2 terms I subtract these 2
terms I subtract these 2 terms, right? So therefore this gives us RT d ln f by P which is nothing
but equal to phi and this is equals to v minus RT by P dp then I can integrate each one of them,
what should be my integration? I should be integrating it from a condition where I know the
fugacity or in other words I should be integrating it from the ideal gas state which is applicable at
P equals to 0 again I recall that ideal all gases for all substances exhibit ideal gas behavior at

518
high-temperature low-pressures since the temperature is constant here, the only way to approach
the ideal gas state is by starting from a very low pressure.

So therefore at that particular low pressure say it is P equals to 0 to P equals to P, at P equals to 0

fugacity equals to pressure so therefore this is equals to 1 to f by P and her mu minus mu0 this
should be equal to one here and then it should have a value of mu minus mu0 here. So on
integration what do I get?

(Refer Slide Time: 3:02)

I get mu minus mu0, sorry it is not one its 0 here because mu equals to mu0 here, mu minus mu
0 this is equals to RT ln f by P which is equal to integral 0 to P, v minus RT by P dp this is the
standard equation which I have also mentioned here this is the standard equation which I used in
order to find the fugacity of the gases.

519
(Refer Slide Time: 3:15)

Now the first thing which appears is, this can also be written down as vdp minus 0 to P RT d ln P
integrating this is not difficult, for integrating this we have to use the PvT behavior of gas. We
have PvT Behavior or the equations of states relating the PV and T of real gases. We have
already discussed those equations and as we have found out there are several types of equations,
the cubic equations of state, the virial equations of state etc.

We can use any of those equations the only thing that we need to remember in this particular case
is the same thing which had happened when we were trying to derive the departure functions for
enthalpy and entropy that all these equations they are in terms of pressure explicit form. For
example P equals to something whereas in this particular case we would like to express v in
terms of P. So therefore in order to perform that again I would like to remind you what do we
need to do for that particular case?

We have already done it, for those particular case we can write it down vdp is nothing but equal
to dpv minus pdv, so therefore when we write down RT ln f by P then instead of vdp, we can
write it down as dpv minus pdv minus RT d ln p integral 0 to P this should be equal to v 0 to v
and here what p equals to 0 it is RT this is going to be pv, right? So therefore for such particular
conditions we find that ln f by P will be equal to Pv by RT minus1, minus1 by RT integral v0 to
v pdv where else we know Pv by RT is equals to Z.

520
So therefore this is z minus1, minus 1 by RT v0 to v pdv. So therefore in this particular form if
we use then in that case we can use any particular equation the cubic equations of state

(Refer Slide Time: 5:52)

And in this particular slide what I have done is, I have mentioned the different cubic equations of
state which we have already discussed and using this cubic equations of states I have also written
down the final forms of the fugacity or the final forms of the fugacity coefficient in terms of z
the 2 constants a and b where usually the constant b it refers to the correction due to the non-
negligible volume of the molecules as compared to the volume of the gas and a arises due to the
interaction between the different molecules.

Well, here there is one but particular thing which I had forgotten to mention in the last class,
which I would just like to mention here when we are dealing with the standard states we say that
it is the at very low pressure then in that case you must remember that under the standard state
conditions we neglect the interactions between the molecules and therefore for all the gases the
standard states are different due to the structure and the molecular nature of the different gases
and not the way that the gases interact.

521
(Refer Slide Time: 6:56)

Well, so therefore this is the first way of determining the fugacity coefficient of the gases that is
by the PvT Behavior. Now apart from the PvT Behavior this particular approach is similar to the
way we had defined the ways of estimating the departure functions from entropy and enthalpy,
we should also be in a position to determine fugacity or fugacity coefficient from compressibility
factor as well, isn’t it?

522
(Refer Slide Time: 7:35)

(Refer Slide Time: 7:46)

From the compressibility factor also we should be able to able to define it for that in what way
should be expressing this particular equation? We can express it as d ln f by P this is equal to v
minus RT by P dp this is the same thing which I have written which is nothing but equals to Z
minus 1 by P dp. Now here if I integrate it from 0 to P, in this particular case we are going to
integrate it from f by P equals to 1 to f by P and this equation also we can define it from 0 to P.

Now in this case we need to remember that Z becomes equal to one for the limiting condition. So
therefore in this particular case we can write ln f by P is nothing but equal to integral 0 to P, Z

523
minus 1 by P dp, right? And we can also write this down as z minus 1 by Pr d Pr integral 0 to Pr,
can we not write it down in this particular form? And then I have also displayed to you that the
different compressibility factor charts which gives z as a function of Pr with Tr as parameter. We
have already discussed those particular charts.

So therefore using those particular compressibility factor charts you can find out Z we can
substitute Z here and we can find out the value of fugacity coefficient if you observe this
equation very closely this and the interesting thing that you find, we find that when you are
writing this equation in the form of integral 0 to Pr Z minus 1 by Pr d Pr, if you observe we find
that from this particular equation phi it is a function of Z Pr or in other words what do I mean?

And we know Z is a function of Tr Pr, so from here does it suggest that just like entropy and
enthalpy departure functions phi also obeys the 2 parameter law of corresponding states or in
other words we should be able to generate curves like phi versus Pr with Tr as parameter and
from these particular curves in absence of any other data we should be able to find out phi from
the this particular charts.

524
(Refer Slide Time: 10:17)

(Refer Slide Time: 10:22)

525
(Refer Slide Time: 10:26)

(Refer Slide Time: 11:18)

Let us see whether this is, such charts they are available for different types of fluids they are
available in several thermodynamic books and we find that these, this particular chart it is
applicable both of these charts they are they are there for different values of PR for lower values
and higher values of PR and it is also important to note that from this particular equation we can
also use the 3 parameter law of corresponding states for greater accuracy where we can define
the 3 parameter law in terms of the acentric factor proposed by Pitzer.

526
We can write it in this particle form or in the log base 10 form which gives the equation as which
gives this particular equation and once we know this particular equation then or rather once we
know how phi0 and phi1varies with Tr and Pr? We should be in a position to find out phi.

So therefore these graphs they show us the variation of phi 0 with TrPr and also the next graphs
they show us the variation of Phi1 with TrPr. So using these curves also we should be in a
position to find out the fugacity coefficient from this generalized compressibility factor charts.
What is the next technique by which we can find it out?

(Refer Slide Time: 12:00)

527
(Refer Slide Time: 12:09)

We should also be in a position to find it out from residual volume. We should in a position to
find it out from residual volumes and this is very evident if you look at this particular, what is
this particular expression? This is nothing but the volume of a gas minus the volume of the same
gas one mole of the same gas provided it was behaving ideally under the same conditions of
temperature and pressure. So this particular term it is nothing but equal to Delta v star, right?

528
(Refer Slide Time: 12:44)

So if we have data on Delta v star then we can directly write ln f by P it is nothing but equals to 1
by RT 0 to P delta v star by RT dp but for this what do we need to know? We need to know how
Delta v star by RT varies with P. So therefore we need to, suppose it varies in some particular
way and we would like to find out the fugacity at some particular pressure P then the area of the
curve starting from P equals to 0 to P equals to P this whole area under the curve that gives us an
idea about the fugacity.

Now here I would also like to mention one particular thing, see we can use whatever you have
said we can use PV behavior we can use the compressibility factor chart but it is not sufficient to
know the Pv behavior or the compressibility factor value at the particular pressure of interest
where we want to calculate the fugacity, why?

529
(Refer Slide Time: 13:49)

Because if you observe all the equations you will find for each of the equations the integration
starts from 0 to P. So therefore if I have to integrate over this entire range either this equation or
this equation we need to know how volume varies with pressure. How compressibility factor
varies with pressure, right from 0 pressure to the pressure of interest? So it is not just sufficient
to know the behavior of the gases at the pressure of interest in order to calculate fugacity.

We need to know the behavior of the gases throughout the entire range from the very low
pressure almost 0 pressure to the pressure of interest this is one thing which makes the
calculation slightly difficult and therefore depending on the availability of the data we have been
suggesting the variety of methods which we have proposed.

Now all these methods that we have discussed they are based primarily on PVT Behavior of
gases. Suppose we do not have an idea regarding the PVT Behavior then how should we find out
the fugacity coefficient?

530
(Refer Slide Time: 15:00)

Again we can find out the fugacity coefficient remembering that dg or rather g equals to h minus
Ts mu equals to h minus Ts. So therefore at constant temperature d mu equals to dh minus Tds,
what is this equal to? This is RT d ln f, so therefore we can also find it out from thermal data as
well, right? And this particular case also again the only problem is the identification of the
standard states we start from here to here, accordingly h has to be evaluated from very low
pressure under which the gas behaves ideally.

So therefore this is equal to Cp0 dT dh minus T s0 to s ds which gives us h minus h0 minus T s

minus s0. So from which we can write it down as RT ln f by f0 if f0 equals to P then this can
reduce to RT ln phi provided f0 is equals to P and this is nothing but equal to delta h star minus T
delta s star.

531
(Refer Slide Time: 16:46)

And there is one more thing which I would also like to mention that usually what we do is, the
other thing what we do is, we usually use the Redlich Kwong 2 parameter equation of state. So in
the next class we will be first discussing how to estimate the fugacity? We have discussed how to
estimate the fugacity of gases definitely pure gases but real gases single component real gases
but if you are dealing with phase equilibrium problems then definitely the most common
problems are liquid vapor equilibrium or solid vapor equilibrium.

So therefore if we have to find out the equilibrium we need to have an idea about the fugacity of
the gas as well as the fugacity of the condensed phase under any particular condition of
temperature and pressure. Regarding fugacity of gases we had a good amount of discussion.

Next class we are going to discuss the fugacity of the condensed phase and then we would like to
take up the 2 parameter Redlich Kwong equation of state which is very frequently used for gases
as well as for the condensed phase and we are just going to, this is particularly important or it is
particularly frequently used for in the hydrocarbon industries. So we are going to end with the
discussions on how to use the equation Redlich Kwong equation of state, thank you very much.

532
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 29
Fugacity of Condensed Phase

(Refer Slide Time: 0:25)

533
(Refer Slide Time: 0:37)

Well, just as I had mentioned in the last class, in this class I am going to discuss fugacity of pure
liquid or pure solids, remember one thing the starting equation is the same in this particular case
as well, what is the starting equation? The starting equation just I had mentioned several times
earlier it is ln f by P equals to integral 0 to P v minus RT by P dp, we start from this particular
equation but we must remember that we are dealing with a liquid.

So how to perform the entire integration starting from 0 pressure where usually all substances
they exhibit or they are in the gas phase and they exhibit ideal gas behavior from there I would
like to reach a pressure where the substance has undergone a phase transition from the vapor
phase to the liquid phase and then it remains in the liquid phase from the saturated vapor
pressure to the pressure where we would like to find out the fugacity of the pure liquid.

So therefore to calculate and whatever I discuss this same thing is applicable for a pure solid as
well. So therefore to calculate the fugacity of liquid or a solid at a given temperature and
pressure, what do we do? We breakup this whole thing into 2 parts, the first part it gives us the
fugacity of the vapor’s phase from 0 pressure to the saturated pressure such that this gives us the
fugacity of the vapor phase under saturated conditions, right?

And after that we have another part which extends from P saturated to P which gives us the
fugacity of the condensed phase and we need to remember that when we have in this particular
integration we get the fugacity of the vapor at the saturated condition. Under this condition we

534
know that the vapor and liquid they are at equilibrium. So therefore this is nothing but equal to
fL saturated fine, so therefore what do I do here?

To find out the fugacity of the condensed phase I simply break it up into 2 parts the first part
gives us the fugacity of the saturated vapor which is nothing but equal to the fugacity of the
condensed phase under the saturated pressure conditions and then if my particular or the
condition of interest is far removed from the saturated pressure then we need to calculate the
additional fugacity which is generated by compression of the condensed phase from the saturated
pressure to the pressure of interest.

So therefore this can be written down as I can write this down as RT ln f saturated by P saturated
this part can be written down by f saturated by P saturated and then the other part can be written
down as P saturated to P, v condensed phase dp minus RT ln P by P saturated I can write it down
very much in this particular form. Now removing all the logarithm of and writing it down we can
write it as fC this is nothing but equal to P saturated Phi saturated exponential integral P
saturated to P vC by RT into dp, right? Where this is the saturated vapor pressure, this Phi
saturated it is nothing but equal to the fugacity of the vapor under the saturated state divided by
the pressure under the saturated state.

535
(Refer Slide Time: 4:57)

(Refer Slide Time: 5:14)

Well, this is the equation that I have mentioned here. So this is the fugacity of the condensed
phase and I would like to recall that this particular expression has been derived starting from the
same equation the defining equation of fugacity with the supposition that the fugacity is equal to
the pressure for very low pressure.

So we have started from very low pressures and we have assumed that under this condition the
gas the substance is in the gaseous state and the gas is an ideal gas. So therefore what we did?
What is the path of integration? The path of integration is compressing the ideal gas from 0

536
pressure to the saturated pressure where it becomes a saturated vapor under this condition the
saturated vapor and the saturated liquid states have got the same fugacity and so therefore
whatever fugacity value I get in compressing the gas from 0 pressure to the saturated pressure is
the fugacity of the liquid under same conditions then there is an additional fugacity generated
because the liquid has to be compressed from the saturated state to the state of interest.

(Refer Slide Time: 6:41)

537
(Refer Slide Time: 6:49)

Now here I would like to say that suppose we are dealing with very low pressures, now let me
write it down in the next page and then it is going to be easier for me fC equals to P saturated Phi
vapor saturated it's better written as phi v saturated exponential P saturated to P v condensed
phase by RT into dp, here also I would like to make the correction this is phi v saturated this is
also the vapor phase saturated, right? Now try to observe this particular equation what do we
get?

538
(Refer Slide Time: 7:15)

(Refer Slide Time: 9:19)

The first thing we get is fugacity of a pure component at temperature and pressure, suppose the
temperature and the pressure is not very high then in that case the gas can be assumed to behave
ideally when the gas is assume to behave ideally then in that case its fugacity and pressure are
equal under saturated conditions, as a result of which this term it becomes equal to one, right? So
therefore this is basically it is a deviation of saturated vapor from the ideal gas behavior.

Next we look at this exponential term, this exponential term arises because the liquid or the solid
for that matter with whatever we are dealing that is at a pressure P far removed from the

539
saturated pressure. Now at conditions which are much remote from the critical very frequently a
condensed phase can be regarded as incompressible as a result of which vC it is not a function of
P, under that condition the exponential term can be written down as vC P minus P saturated by
RT, we can write it down in this particular form.

The other thing which is important is that usually but not always these 2 correction terms they
are negligible because unless P is not far removed from P saturated this particular correction
term which is usually known as the poynting correction factor, this correction factor it is not very
high if provided the temperature is such that P saturated is low under that condition if P saturated
is low then phi v saturated is almost equal to one.

And we find that if the pressure is not very high then poynting correction factor can also be taken
as one.

540
(Refer Slide Time: 9:24)

Just to give you an idea regarding these 2 correction factors, I have a table and a graph with if
you if you observe this particular table what do we see? Here I have taken it from the book of
Prausnitz and there we have found out that the poynting correction factor for the condensed and
incompressible substances in this particular case. We find that for 1 bar pressure even for 100 bar
pressure also its not high but if you are dealing with thousand bar pressure then it becomes really
high. So therefore this implies that the poynting correction factor, it becomes important when we
are dealing with condensed phases at very high pressure.

So under that condtion this particular term this term is important otherwise it is not very
important even if we observe the variation of the fugacity coefficients with reduced temperature,
what do we find? We find that at conditions closed to the critical for those particular conditions
the fugacity coefficient differs remarkably from unity and we find that as go for lower and lower
reduced temperatures the fugacity coefficient almost approaches one.

So therefore this implies there are 2 things which are important one is phi v saturated it differs
considerately from unity as the critical temperature is approached. So under this condition we
cannot neglect Phi v saturated the other thing is if you observe phi v saturated is always less than
one this particular term phi v saturated it is always less than one which implies that this
correction term it tends to decrease the condensed phase fugacity and usually we find that the

541
poynting correction factor is the exponential function of pressure this is low at small values of
pressure but it may be very large for large values of pressure or at low-temperature.

We also find that the phi v saturated that is also for conditions far removed from the critical this
is also one. So therefore for low to moderate pressures we can safely assume that the fugacity of
the condensed phase is equal to the saturated vapor pressure and when we are dealing with
conditions like high pressure or very low temperature or for conditions close to the critical then
we need to incorporate the corrections to the saturated vapor pressure in order to obtain the
fugacity of the condensed phase and these 2 corrections they arise primarily because of the non-
ideality of the vapor phase and the fugacity generated due to the compression of the liquid of the
liquid phase from the saturated pressure to the pressure far removed from the saturated.

(Refer Slide Time: 12:30)

Now before ending fugacity there is one more thing that I would just like to mention, this is very
commonly used for the for predicting the fugacity’s of hydrocarbons and therefore I have used
this particular equation or rather I have discussed this particular equation usually we use the
Redlich Kwong equation of state which is given here, you already know it with the constant
value is given in this particular equation.

542
(Refer Slide Time: 13:07)

Now when we use this particular equation we find that the expression of fugacity is given by this
equation where we find that fugacity coefficient it’s a function of Z it is a function of BP and it is
a function of A square by B. So therefore if we know these 3 quantities we would be in a
position to find out f by P. Now usually we can use this analytical form to find it out but there is
one note of caution which we should remember for using this equation that is the value of Z that
we are using here that should be calculated from the Redlich Kwong equation of state which I
have already mentioned here.

It should be used with the Redlich Kwong we should be calculating Z from here and we should
not be using any particular experimentally observed value of Z or we should not take Z from any
other equation or chart in this equation if we want to get a correct value of the fugacity
coefficient using Redlich Kwong equation of state.

543
(Refer Slide Time: 14:12)

The other interesting thing is this also I had discussed earlier since we know that Redlich Kwong
equation of state is a cubic form, so therefore it can be used to predict the fugacities of a
saturated liquid as well as a saturated vapor.

If you are solving this particular equation we know that we get a cubic value of or rather we get 3
routes of Z out of these 3 roots the largest value that corresponds the saturated vapor the smallest
value that corresponds to the saturated liquid and intermediate value that we get that has no real
physical meaning. The only problem of finding out the fugacities of saturated liquid and vapor
using this cubic form of Redlich equation of state is that in this particular case usually the BP
that is an unknown.

So therefore how to find out BP from this particular equation, for this we use the concept that the
vapor and liquid are in equilibrium under the saturated conditions. So how do we go about doing
it? What we do in this particular case is, first what we do is we use this particular equation, in
this particular equation we try to find out BP by rather we assume different values of pressure
and we try to find out the BP value such that we use it to find out the fugacity of the liquid phase,
we use it to find the fugacity of the vapor phase and then we choose that particular pressure as
the actual pressure in which we use that particular pressure under which the fugacity calculated
for the liquid phase and the vapor phases are the same.

544
Once we have identified the pressure under which the liquid phase and the vapor phase pressures
are the same then we know that the Z corresponding to that pressure should be used here or in
other words the BP value which we have obtained from this particular equation is then used in
this equation and then we obtained the roots of Z and then the higher value is taken as a saturated
vapor and the lower value is taken as the saturated liquid.

(Refer Slide Time: 17:48)

Now other than that just like I had shown you the Lee Kesler plots we find that solving this
particular equation and equation for Z we can generate a set of curves which can give us the
fugacity the coefficient in terms of the Redlich Kwong parameters which are A square by B and
BP and we know that that these 2 curves have been plotted under 2 conditions one is the low
range of BP for which in this particular graph we have shown the fugacities of both the liquid
and the vapor and this is for the high range of BP under which condition we have shown for only
the vapor phase.

For this particular low range of BP this is applicable for BP less than 0.20, for this condition this
graph is drawn and then this is primarily for coexisting vapor and liquid regions and this is ffor
higher values. Now in this particular case using the chart are must simpler as compared to using
the equations.

545
For using the charts it is very easy we first have to evaluate the 2 parameters A square by B and
BP. Once we have evaluated this from the temperature, pressure and the criticals of the system
once these are evaluated then we read the charts and we use and we using the aforementioned
values and we try to find out the fugacity coefficient for the condition of interest we can use this
for finding out the fugacity coefficients for both the condensed phase as well as for the gas
phase.

(Refer Slide Time: 19:24)

So therefore here I would just like to mention in conclusion. Since Redlich Kwong equation is
used very frequently to find the fugacity coefficient in the case of hydrocarbons we can use the
analytical equations but they are quite tedious but they are accurate, on the contrary if we use
charts, the charts are simpler to use we just need to find out as I have already mentioned we need
to find out the parameters A square by B and BP from temperature, pressure, conditions and the
critical of the system and then read the charts using this aforementioned values.

Here I would also like to mention that the Redlich Kwong equation is less accurate compared to
the Pitzer correlation particular regions of coexisting vapor and liquid but since these equations
have an advantage of being completely analytical and they satisfy conditions of critical isotherm
at the critical point. So therefore under these conditions these few equations are quite useful with
this I end my discussions on the concept of fugacity.

546
This be the importance of standards states for a correct estimation of fugacity reframing the
equilibrium conditions in terms of fugacity estimating fugacities for gases estimating fugacities
for liquid, a practical method of estimating fugacity for hydrocarbons and after this we go to
discuss mixtures their properties, fugacity of a component in a mixture and after that we go for
discussing the ideal solutions and the non-ideal solutions.

(Refer Slide Time: 21:05)

Well, just one particular example or one particular problem has been provided here to find out
the fugacity coefficient for a Van der waal’s gas. You can use or rather you can try this problem
to find out the value and I think it is pretty straightforward you will not get stuck up in this but in
case you are stuck up we are there with you, thank you very much.

547
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 30
Mixtures

(Refer Slide Time: 0:42)

Well, hello everybody so till the last class what were we discussing, we were trying to discuss
different sort of characteristics the properties of pure components or single component or single
component solution and till date whatever we have done, we have come across a post of
equations which I have just I have just listed the summary of it in this particular slide. Now you
will be recalling that several times we have seen that the properties they demand data of say
volume as a function of pressure.

While on the other hand we find that most of the equations which we have dealt till now say the
cubic equations of state primarily they express pressure in terms of volume and temperature. So
what I have done in this particular slide? I have noted down all the equations both in pressure
explicit form as well as in volume explicit form. From this particular chart what do we see? We
find that for any particular substance regarding of whether it is pure or it is a mixture all the
thermodynamic properties of interest in phase equilibrium can be calculated from thermal data
and volumetric data.

548
We take up any particular equation we find that in both the equations they comprise of thermal
data which suggests how the properties concerned properties varies with temperature under
conditions of constant pressure or constant density and we also need volumetric data which gives
us how the particular thermodynamic function it varies with pressure or density at constant
temperature. Now we would also like to remember that all these equations which have been
given they are for single component substances when there is no phase change.

When we have additional phase change then we require additional volumetric as well as thermal
data to account for the phase change situation we also see that very frequently we would like to
express the property change in terms of the property which the substance would have in the ideal
gas state under the conditions of same temperature same pressure, a typical example of this is the
fugacity of the component.

If you observe the equations of fugacity in this particular equation and in this particular equation,
we find that in both the equations they have expressed fugacity in terms of the corresponding
value under the ideal gas state keeping in mind that the fugacity for an ideal gas is its pressure
under that condition of temperature or its partial pressure under the condition of constant
temperature and composition.

Now if you go through all these equations what do we find? We find that for all these equations
the only data that we require, see solving these equations are not at all a problem they are just
simple integration techniques. And I would also like to remind you 2 things usually we are
tempted to think that all these equations are applicable for gases, it is not so these equations
again I repeat they are applicable for any substance in pure state, mixed state whether it is a
component in a mixture whether it is ideal whether it is non-ideal whether it is a gas whether it is
a condensed phase for all particular situations the equations are applicable number 1.

The equations have been written for pure components only number 2 and the most important
thing which I would like to suggest or to bring to your attention is that all the equations which I
have mentioned here all the equations they are exact equations, whatever equations we have
mentioned they have no simplifying assumptions they are exact equations and they are not
restricted to the vapor phase only. Now for solving these equations what is the only thing that we
require?

549
(Refer Slide Time: 5:14)

For solving the equations we just need say either volume as a function of P, V sorry P, T or
maybe we require pressure as a function of V, T. Normally we have also discussed in the PvT
behavior of gases several types of equations for example we have discussed the cubic equations,
the compressibility factor chart, the virial equations of state when we should use what everything
is fine but I would also like to remind you that they need for such a large number of equations
was because no particular equation is applicable over a large range of temperature, pressure and
composition conditions.

The situation becomes much graver when we start dealing with mixtures and we all know that
mixtures are the things that we actually deal in practical situations there are very few pure
components situations that we deal. So therefore if we have to deal with mixtures if I really want
to find out the property changes in case of mixtures from these particular equations what do I
need? I need to replace all these, these are total volume of the mixtures, so then I have to express
total volume of the mixtures in terms of the molar volume of the mixture and what do I need for
that?

I need some particular way to estimate the total volume from the molar volume and here I would
just like to mention that it is not always the case that I know the molar volume of pure
component mixtures and I can just write that the total volume of the mixture is often not equal to
the equal to the number of moles times the molar volume of the pure components we will be

550
coming to it shortly but before that I would like to mention that if we are supposed to use these
particular equations for finding out mixture properties the only obstruction or the only hitch
which we face are not these equations.

These equations they are exact they give very accurate results, we can use them for any purpose,
the only problem is in order to solve these equations we need something of this sort and when we
are doing it for a mixture then we have additional variables which are the compositions or the
mole numbers of the different constitutions in the mixture.

So therefore finding such a particular equation of state which will be valid and it is not that we
have to find one particular equation of state at some particular conditions of temperature and
pressure, just go through all the equations for every equation there is an integral sign because as I
have told you we cannot predict the absolute value of properties.

(Refer Slide Time: 8:13)

So therefore we have to integrate it from one particular lower limit to the other upper limit and
during this entire limit I need to know this particular expression I need to know how V varies
with T not only at P the pressure of interest but from P equals to 0 to P equals to P. Now this is a
very difficult situation we have large number of equations of states but finding a unique equation
of state which is going to be applicable from almost very low pressure to the pressure of interest
for the different components which are present in a particular mixture is very difficult.

551
(Refer Slide Time: 9:11)

So therefore today we are going to discuss the different mixing rules and combining rules which
we use in order to evaluate mixture properties in terms of pure component properties. Now in
this particular respect I would like to mention that in reality if we have to mention, if we really
have to find out the mixture properties from the pure component properties we need to have a
proper idea of the molecule considerations the exact equations are going to come from molecular
considerations.

But unfortunately at the present moment we do not have that much idea about the behavior at the
molecular level the interactions of the different components at the molecule level. So therefore
most of the cases the rules which we suggest for finding out the mixture constants in terms of the
individual property constants they are empirical in nature since they have been found to work
well for a large number of conditions we just adopt them.

552
(Refer Slide Time: 10:09)

(Refer Slide Time: 10:20)

Now here I have just noted down a few mixture properties which or rather a few mixing rules for
which can be used or which can be used to predict the constants of the equations in terms of pure
component constants. Now here I would like to mention that suppose we take up any particular
say suppose for example we take up the Van der waals equations of state.

553
(Refer Slide Time: 10:49)

Now for Van der waals equation of state we find that the first thing which we need if we have to
use the Van der waals equation of state say v minus b or else I will write it in the proper form P
plus a by v square into v minus b equals to RT, for gaseous mixture this becomes M, this
becomes M, this becomes M, this becomes M. So I need something to find out aM, I need to find
out vM and I need to find out bM and once these can be accurately predicted then this particular
Van der waals equation can be used for mixtures under the conditions where it is applicable.

You already know that we cannot use this equation for P greater than 200 atmospheres this is not
very accurate for the close to the liquid vapor coexistence conditions but other than that for
normal conditions we can use it provided we can find out this particular constants. Now usually
there are different ways or different rules have been proposed, keep in mind that so many rules
have been proposed just because none of them work very well, all of them are empirical in
nature.

So for example if you take up b from where does b come, if you recollect b is nothing but it is
the correction due to the volume occupied by the molecules in ideal gas state we assume
negligible volume of the molecules as compared to the volume of the entire gas. So therefore bM
it should be somehow related to the size of the molecules.

554
(Refer Slide Time: 12:15)

Now there are 2 ways of expressing bM as I have written down in this particular slide the first
one, what does it assume? It assumes that we have spherical molecules and it assumes additivity
of diameter, right? From assuming additivity of diameter the bM is given in the form that I have
have written down here but usually the more popular thing which the people usually use is they
assume additivity of volumes other than diameters.

Under that case we find that the mixture b is simply an additive property of the pure component
b and this is quite a popular equation and this is usually normally used but remember one thing
choosing a correct mixing rule again I should mention it is not possible because all mixing rules
they are based on different assumptions to which assumptions are going to hold under different
conditions is not very clear.

555
(Refer Slide Time: 13:25)

Any way b it’s not very difficult you can use either but for most of the cases we use this
particular equation and for this class we will be using this particular equation.

Now let us come to the a if you recall a was introduced just to take into account the interaction
between the different molecules so therefore at moderate densities for mixtures whose molecules
are not very dis-similar in size we find that the mixing rules for b does not significantly affect the
prediction of the results but the influence of a on the results are much more important.

556
(Refer Slide Time: 14:12)

(Refer Slide Time: 14:23)

557
(Refer Slide Time: 16:17)

So it is very important to or rather to propose a proper relationship for finding out aM in terms of
the component a’s. Now in this particular aspect usually what is the most popular rule which is
used is written down here, this says that aM it is a summation of in fact it is a double integral i
and j suppose there are 2 molecules it is for a binary mixture it is a summation of aij where i and
j can be same i and j can be different.

So therefore for a just for a binary mixture for what do we get from here? For a binary mixture it
should comprise it should be a function of aii the interactions between the molecules of
component i it should be a function of ajj which is again arises due to the interaction between the
molecules of j, now these 2 they can be obtained from the Van der waals constants for
component i and component j but apart from that we would also have something like aij.

What is this aij? This takes into account the interaction between molecule i and molecule j. Now
aij can be equated to aii or ajj provided the molecules they are all similar in nature which is
usually not the case in under those circumstances when usually this case is when i and j they are
the similar molecules then there should be some way of predicting again aij in terms of aii and
ajj.

558
(Refer Slide Time: 16:27)

Now the rules which suggest how to predict the interaction parameters in terms of the mixing
parameters those rules are known as the combining rules and the rules which predict the mixture
constants in terms of pure component constants they are known as the mixing rules.

So therefore I repeat once more the mixing rules are the equations which express mixture
constants in terms of pure component constants and combining rules are the equations which
provide the interaction terms when the different components are there in terms of the interaction
parameters of the pure component constants.

559
(Refer Slide Time: 17:12)

Now usually the combining rules which we use for a is either to assume that it is an arithmetic
mean of the component is or to assume that it is a geometric mean of the component a’s. When
we assume the arithmetic mean then we come across this particular equation and when we use
the geometric mean we come across this particular equation. Usually the trend is to assume the
geometric mean this has been used by Van der waals himself, it was first proposed by Berthelot
and then we find that although it was empirical London could proposed some theoretical
justification for it so that nowadays we are quite convinced with it.

And I would also like to mention that it is not only (a) which is expressed in terms of such a
geometric mean in fact any parameter of thermodynamics which gives mixture interaction in
terms of component interaction is often expressed in terms of such an expression as aij. So
therefore we find out that for any cubic equation of state this is not only for Van der waals
equation of state for any cubic equation of state usually we will find that b is expressed by this
equation and a is expressed in terms of this equation.

560
(Refer Slide Time: 18:31)

(Refer Slide Time: 18:58)

Now here we find that that we have introduced a parameter Kij. Kij it is nothing but it is the
binary interaction parameter between components i and j. It arises due to the binary interaction it
is simply adopted for a better prediction between experiments and theory and for most of the
purposes we will find that we take kij equals to 0. It is just adopted for whenever there is good
amount of mismatch between experiment and theory.

561
(Refer Slide Time: 19:12)

(Refer Slide Time: 19:22)

So therefore when we are dealing with cubic equation of state we find that we can adopt the
mixing rules which has been written down here. Now let me tell you that there are several other
mixing rules also which have been proposed it is quite natural because we really we do not know
which mixing rule is going to help under what condition. So therefore there are large, whatever
number of equations of state are there almost for each of the cases there are equivalent number of
mixing rules, so these are some additional mixing rules which we find.

562
(Refer Slide Time: 19:50)

And then we find that in these particular mixing rules we find that the critical constants should
also be expressed, so therefore again for these mixing rules some particular mixing rules for the
critical constants TCm, VCm and PCm all those things have been suggested. Equivalent mixing
rules are also available for virial equation of state where also we find that there are mixing rules
and combining rules taken into account, the mixture property in terms of pure component
property and also the interaction parameter in terms of pure component interaction parameters.

(Refer Slide Time: 20:24)

563
(Refer Slide Time: 23:10)

So accordingly we find that in this particular case also we have proposed your this bM has been
written and then this bM has to be expressed in terms of B ig, so Bij is taking the 3 parameter
law of corresponding states which is expressed in this particular term where Bij0 Bij1 these are
expressed as shown here.

But again you must remember that all these equations they are empirical in nature, right? Now
apart from these also we find that in most of the equations we need to know vM and we need to
know zM that we have already seen, so apart from the cubic equation and virial equation of state
there is another way also by which we can find this out these are by assuming Amagat’s law of
additive volume and Dalton’s law of additive pressure.

What do these equations say? These equations say for example this equation it simply says that
V mixture is nothing but equal to Nivi since in this particular case we assume that vi refers to the
molar volume of each and every component this is primarily applicable to, it is not an ideal
mixture of gases, right? So therefore in this particular case how can we express zM as? This is
nothing but equal to PvM by NRT, correct?

Can we not write it down as P by NRT Sigma Nivi VM has been substituted by this or in other
words this can also be written down as P by NRT Sigma NI ziRT by P. I have replaced each
particular vi with ziRT by P. So therefore from here what do I get? I get zM is nothing but equal
to Nizi. We continue our discussions with Amagat’s law of additive volume and Dalton’s law of

564
partial pressure in the next class and once we wind up with the mixing rules and the combining
rules for the different equations of state we will go to some more interesting and important
properties of mixtures, thank you very much.

565
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 31
Mixtures (Contd.)

(Refer Slide Time: 0:30)

566
(Refer Slide Time: 0:32)

(Refer Slide Time: 1:22)

Well, so to continue with our discussions we had left the discussion in the last class when you’re
trying to find out zM in terms of Amagat’s law of additive volume and as I had said in terms of
Daltons law of additive pressure. Now we have already derived from the Amagat’s law of
additive volume, what did we give? We expressed zM in terms of zi where all the zi’s has been
determined at the T and P of the mixture, fine. So therefore this as I have said, this is
approximate for real gas mixtures and valid for mixture of ideal gases, right?

567
(Refer Slide Time: 1:46)

Now let us take up the other law for finding out the partial pressure all of us know this is
nothing but Daltons law of additive pressure, what does it say? It tells us that total pressure
exerted by an ideal gas mixture it is nothing but the sum of the partial pressure or in other
words the pressure of the gas mixture it is nothing but the pure component pressure.

What is this? This is the pure component pressure which is pressure exerted by Ni moles of
pure component gas i if it occupies the entire volume V at temperature of the mixture. So
therefore T mixture and T pure components are the same and the partial pressure it’s nothing it
is simply the pressure exerted by Ni moles of the gas if it occupied the entire volume.

Now just it’s important to remember that for real gas mixtures it can happen that the pure
component pressure the Pi this is usually known as the pure component pressure, okay.
This pure component pressure may not be equal to the partial pressure, it can happen if
interactions are quite dissimilar but for ideal gases we always have Pi equals to pi, fine.

Now in this case letters found find out zM prime I have just written the prime just to
differentiate the mixture compressibility factor obtained from Amagat’s law and the
mixture compressibility factor obtained from Dalton’s law.

Now what is zM prime equals to, this is nothing but equals to P the total pressure V by NRT

which can be written down as V by NRT Sigma of Pi which is nothing but equal to Sigma of Ni

568
by N PiV by NiRT, fine. Now what is this? This is also a form of z, so therefore can I not
write it down as Ni by or else I write it down as yi considering that this is a gaseous
mixture into Zi prime. So therefore from both the equations what have you got?

(Refer Slide Time: 4:37)

569
(Refer Slide Time: 4:42)

If I just write down the other equation here just for comparison we get zM is equals to,
sorry this is yi zi, fine. Now just see the difference between the 2, I can predict the
mixture compressibility factor either by using the Amagat’s law of additive volumes
which I have derived here or the Dalton’s law of partial pressure.

570
(Refer Slide Time: 4:50)

We find that from Dalton’s law of partial pressure the mixture compressibility factor is
defined in terms of the pure component compressibility factor where this is taken at system
temperature and pure component pressure this is something very important. So in this case
we express it in system temperature and pure component pressure or the pressure which
this particular gas would have exerted had it occupied the entire volume.

While from Amagat’s law we express it in terms of the pure component compressibility
factors where these are expressed at system temperature and pressure.

571
(Refer Slide Time: 6:12)

So well, this was one the other way of finding out Z is we had already discussed apart from the cubic

equations and the virial equations of state, we had also discussed the compressibility factor charts

where we have shown z is a function of Pr and Tr, right? So therefore from these equations suppose

there is some particular way of finding out the pseudo critical constants as it is said.

572
(Refer Slide Time: 6:23)

(Refer Slide Time: 6:25)

Suppose we can find out the Trm and Prm for the mixture case then using these 2 values, since we

know that the compressibility factor chart it is applicable for all particular substances so therefore

what we can do? We can find out if we can propose some pseudo-critical constants then in that case

using the compressibility factor chart also we should be in a position to find out Z.

573
(Refer Slide Time: 7:08)

Now usually there are several again mixing rules for finding out the pseudo critical
constants, we find that there is a Kays rule which gives the mixture the Pcm the critical
pressure of the mixture and the Tcm, both as additive function of pure component
critical constants from where the reduced components can be found out and this
particular graph can be used but usually we find out that although in from the Tcm value
gives quite accurate cases. Again in this particular case since the mixture critical
pressure again is a function of the interaction of the different molecules in the mixture.

574
(Refer Slide Time: 7:50)

So naturally the additive equation does not usually does not hold good and accordingly another
equation has been proposed by Prausnitz and Gunn, this particular equation along with the
additive law for the mixture critical temperature these 2 combined they give a better result and
so therefore usually what is recommended is you use this particular Tcm you use this particular
Pcm and then you try to use this particular compressibility factor chart for better results.

575
(Refer Slide Time: 8:08)

And if you’re going for a 3 parameter law of corresponding states then usually this particular
equation is used to find out the mixture acentric factor in terms of the component acentric
factors. Now so with this, what I do? I have finished up or rather I have covered the
estimation of properties of real gas mixtures or for that matter estimation of total properties
of any particular mixture in terms of the properties of the pure components.

And I have proposed different mixing rules and I have proposed different combining
rules, so now suppose I would like to predict the fugacity of the component, right?

576
(Refer Slide Time: 8:55)

So for predicting the fugacity of the mixture this is the equation, right? So therefore for
other things it is pretty straightforward if you want to predict h, s, u, a, g you can do it.
Let us take up the fugacity of the mixture. Simply for finding out the fugacity of the
mixture what we need? We need to know the molar volume of the mixture, right?

577
(Refer Slide Time: 9:20)

What is the molar volume of the mixture? This vM this is nothing but equal to the total
volume divided by capital N the total number of moles of the mixture, it is not very
difficult to find out this equation and from here we can find out the value of fM, right? But
we need to remember what is the purpose of the fugacity?

Why did we define fugacity? Fugacity was defined in order to provide a quantitative prediction of
phase equilibrium in terms of some physically realizable quantity, right? So therefore we had if
you recollect in the while we’re discussing fugacity what we would do? We reframe the entire
condition of equilibrium in terms of temperature, pressure and fugacity. What was the condition
for fugacity which refers to chemical equilibrium if you remember?

It was fi alpha equals to fi beta which says that the fugacity of each component i in the different
phases should be equal. So therefore if we really want to find out whether 2 phases are in
equilibrium what do we need to calculate? We do not need to calculate fM, on the contrary we
need to calculate the fugacity of the individual components in the different phases, right?

578
(Refer Slide Time: 10:55)

So rather than this particular equation what we would need under that condition is the
equation which I have written down here. So if I have to write down this equation what do
we find? Just for the time being you just note that this refers to fugacity coefficient of
component i in the mixture, this refers to the fugacity of component i in the mixture, right?

We will have more discussions because this is something very special and very different
from other molar properties of components in mixtures. So therefore for this, what do we
need? We need to find how the volume of this particular component the total volume Vi
of the component varies with ni? This particular data we need.

Now what is this data? For pure components we know suppose we are dealing with
pure components, so therefore under that condition this particular del V del n at
constant TP it is nothing but the molar volume very easy to find out but in this particular
case, can we simply substitute the molar volume in this particular equation?

Or will the volume contributed, the contribution of the volume of component i when it is in a
mixture be different as compared to the contribution of component i when it is added to its
own particular system. What I mean to say, that when i is added in a large amount of i,
suppose one mole of i is added to a large volume of i, it contributes its molar volume to the
mixture, the mixture volume increases by the molar volume of component i.

579
Now if the same component i, I mix it or I add to a large volume of say component J
then in this particular case is it going to contribute its molar volume or is it going to going
to contribute something different from the molar volume? So therefore we discuss all
this things regarding the partial molar properties as they’re called in the next class and
then it will be easy for us to understand how we have we can predict the fugacities of
components in mixtures and then from them that how we can find out the phase or
rather how we can deal with different phase equilibrium problems?

580
 Phase Equilibrium Thermodynamics
Tutorial
Module 5
Lecture 32

Good morning everybody my name is Alex Koshy, I am the TA for the course phase
equilibrium thermodynamics. So today I will be discussing a few problems that deals with the
contents of week 5 of this course there are a couple of problems that deals with the fugacity
coefficient estimation and there is one problem after that, that deals with the estimation of
molar volume of a mixture from Amagat’s law as well as from Dalton’s law.

(Refer Slide Time: 0:44)

So let me go to the problem first of all. So the first problem I am going to read it out, estimate
the fugacity of propane at 60 degree celsius and 17.03 bar. Use the generalized
compressibility factor correlation and Lee-Kesler data. So the critical temperature and critical
pressure are given Tc is equal to 369.9 kelvin and Pc is equal to 42.57 bar, the acentric factor
is given and it is 0.153.

581
(Refer Slide Time: 1:20)

So the first thing that we need to do is we need to list out the values that are given to us. So I
am going to do that, T is given, that is 60 degree celsius that is 333.15 kelvin, Tc critical
temperature is given and that is 369.9 kelvin, pressure is given and that is 17.03 bar, critical
pressure is given 42.57 bar and acentric factor is also given 0.153. So the first thing that we
need to find out is, we need to find out the reduced temperature as well as the reduced
pressure using these values.

We know the reduced temperature is T by Tc that is 333.15 divided by 369.9 and I have
already calculated these values I got the value as 0.9 and then I calculated Pr that is reduced
pressure 17.03 divided by 42.57 that is 0.4. So we have the value of Tr, we have the value of
Pr.

582
(Refer Slide Time: 2:40)

Now from Lee-Kesler chart we need to find out the value of log 10 phi 0, where phi is the
fugacity coefficient. So we take this graph here in the x axis reduced pressure is plotted and
all these lines corresponds to a different values of reduced temperature. So here we know that
our reduced temperature value is 0.9 and reduced pressure value is 0.4, so we start from 0.4.
We start going up and when we reaches the reduced temperature line of 0.9 we stop and then
we go horizontally to the log phi 0 value.

(Refer Slide Time: 3:36)

And we got log phi 0 value as minus 0.086. So we keep that value there. Now as we are using
a three parameter model we need to find out the log phi 1 value as well.

583
(Refer Slide Time: 4:06)

584
So we go to the next chart. Here we are having reduced pressure 0.4, we go up we reach the
reduced temperature line 0.9 and we go horizontally and we get the value of log phi 1 around
say (point) negative 0.039, so we got log phi 0 value, we got log phi 1 value. So we need to
find log phi value that is log phi 0 plus omega that is acentric factor which is equal to and
omega value we already know we have written it down here. So we write that value here, so
we get the value of log phi as minus 9.1967 into 10 raised to minus 2.

So we all know how to find out an anti-log of a value using a calculator, so we find the anti-
log of this value and we get the value of phi as 0.809.

(Refer Slide Time: 5:36)

585
Now we know phi that is the fugacity coefficient is nothing but the ratio of f that is fugacity
by pressure. Therefore fugacity that we need to be we need to estimate fugacity is equal to
phi into pressure. We know the value of phi 0.809 and we know the value of pressure that is
given earlier 42.57 bar. So solving this we get the value of fugacity f as 13.78 bar, alright. I
am sorry about this value this is critical pressure. So the pressure that is given the pressure of
the system is 17.03 bar so we substitute it here 17.03 phi into P that we get as 13.78 bar,
okay.

So what have we done here in this problem we first of all listed out the values that we have
we listed out temperature, we listed out pressure, we know the values of critical temperature
and critical pressure from these values we found out the reduced temperature as well as the
reduced pressure and from the values of reduced temperature and reduced pressure we found

586
out log phi 1 and log phi 0 and from these values using the acentric factor value we found out
log phi and from the value of log phi we take the anti-log and we found out the value of phi
and from the value of phi we just found out the value fugacity by multiplying phi with the
pressure.

(Refer Slide Time: 7:38)

So I am going to do a similar problem that is our second problem, I am going to read out the
problem determine the fugacity of methane at 32 degree celsius and 9.28 bar using the
generalized compressibility factor correlation. The Tc, Pc and acentric factor values are
given. So we proceed in the same way we did in the last problem.

(Refer Slide Time: 8:00)

587
First thing we write down the values T is equal to 32 degree Celsius that is (190) I am sorry
that is 305.15 Kelvin, P 9.28 bar, Tc 190.7 Kelvin, Pc 46.41 bar and the acentric factor 0.011,
okay. Now we calculate the reduced temperature and reduced pressure values, reduced
temperature value we got it as 1.6, reduced pressure value we got it as 0.2. Now we have the
reduced temperature value, we have the reduced pressure value so we go to the chart.

(Refer Slide Time: 9:14)

588
We proceed from reduced pressure value that is (1 point) I am sorry 0.2 we go up. Now we
go up. Now we know that the value of reduced temperature is 1.6, so we need to find out the
line for 1.6 here it is this line, so it meets somewhere here then we go horizontally to left.

(Refer Slide Time: 10:04)

So we get the value of log phi 0 we get the value of log phi 0 as minus 0.006.

589
(Refer Slide Time: 10:28)

And in the same way we need to find out the value of log phi 1, we go to the next chart we
start from 0.2 that is this point, so 1.6 curve is here it meet somewhere here and we get the
value of log phi 1 as 0.007. So we know these two values, we know the value of acentric
factor. So log phi and we get the final value of log phi as minus 5.923 into 10 raised to minus
3.

590
(Refer Slide Time: 11:42)

We take the anti-log so we get the value of phi as 0.9865 and we find out the fugacity as phi
into P as we did in the last problem and that is 0.9865 into 9.28 is equal to 9.155 bar that is
the value that we got. So we have almost familiarized with what we need to do when we get
charts and when we get reduced temperature and critical temperature values, etc to find out
the fugacity and fugacity coefficient.

(Refer Slide Time: 12:24)

So we move onto the next problem, I read out the problem. Estimate the molar volume of a
mixture of 40 mole percent propane and 60 mole percent n-pentane at 500 Kelvin and 50 bar
using one Amagat’s Law, two Dalton’s Law. So the values of critical temperature, critical

591
pressure and acentric factor all are given we do not need acentric factor here because we are
dealing with two parameter model so we simply avoid these values.

(Refer Slide Time: 12:56)

So before going into the problem we need it has been already taught but I just want to revise
it little bit so that you will be understanding this problem better. We want to look back to
what is Amagat’s law and what is Dalton’s law. So Amagat’s law of additive volume states
that the total volume of a mixture of ideal gases equals the sum of volumes of pure
components at the same temperature and pressure as that of the mixture.

(Refer Slide Time: 13:26)

592
So basically it means the total volume of the mixture Vm it is summation of molar volumes
of each components means the volumes of each component that constitutes the mixture. So
we can write Zm that is the compressibility factor of the mixture as Pvm by NRT, where N is
the total number of moles. We can substitute the value of m here and now here instead of Vi
we can write, so we can substitute this value Vi here.

Therefore mixture compressibility factor becomes, this RT and this RT cancels out, this P and
this P cancels out. This N comes into the summation part. We know Ni by N is nothing but
the mole fraction of the component in the gas.

(Refer Slide Time: 15:02)

593
So I am writing it fresh here, here this is the compressibility factor of each constituent of that
constitutes the mixture and here Zi is calculated at the same temperature and same pressure as
that of the mixture where Zm is calculated. So let me go to the problem here we are having
we are given temperature, temperature is 500 Kelvin, we are given pressure it is 50 bar, we
are given critical temperature, critical temperature of component 1. I have mentioned propane
as component 1 and n-pentane as component 2. So critical pressure temperature of
component 1 is 369.9, component 2 is 469.8, critical pressure of component 1 is 42.57 and
that of component 2 is 33.75, okay.

So we need to find out the reduced temperatures and reduced pressures of both these
components, where 369.9 is the critical temperature of the first component we get the value
as 1.35, Pr1 is equal to 50 by 42.57 we get the value as 1.175, Tr2 is equal to 500 divided by

594
469.8 that is 1.06, Pr2 is equal to 50 by 33.75 that is 1.482. So we got the value of Tr1, Tr2,
Pr1and Pr2.

(Refer Slide Time: 17:30)

595
So now we have a chart and from there we need to find the value of Z1 and Z . This is the
chart here reduced pressure is plotted at the x axis and all these lines corresponds different
reduced temperatures and the y axis is Z, it is actually Z0 but since we are dealing with two
parameter model I simply mention it as Z. So we need to find out Z1 and Z2 for both
components.

So here we know Tr1 is 1.35 and Pr1 is 1.175, same in the same way as we have done in the
last problem we just find out the value of Z1 from this graph 1.175 will come somewhere
here we go up, 1.35 there is a value of reduced temperature is coming somewhere here. So we
get the value of Z1 as I have made a small mistake here it will come somewhere over here. So
the value of Z1 is 0.84, in the similar way we calculate the value of Z2 from the values of
critical temperature and critical pressure for component 2 and the value of Z2 we got it as
0.35.

596
(Refer Slide Time: 19:06)

597
So here we know we have (calculated) we have proved it earlier for Amagat’s law Zm is
equal to summation of yi and zi. Here y1 here we know mole percent of propane is 40 and
mole percent of n-pentane is 60, so y1 is 0.4 and y2 is 0.6. So we simply do the summation
here 0.4 into 0.84 plus 0.6 into 0.35 and we get the value of Zm as 0.546.

(Refer Slide Time: 20:03)

Now we need to find out the molar of volume of the mixture we know Vm is equal to Zm
into R into T by P, so we know all the values P, T, R and everything we know we just
substitute the values, the value of P is 50 bar we need to calculate in SI units, we need to put
it in SI units. So we write it in Pascals 50 into 10 raise to 5 Pascal.

(Refer Slide Time: 20:44)

598
So we get the value of Vm value of Vm that is measured is 4.539 into 10 raised to minus 4
meter cube per mole, okay. Now we need to go to the second part of the problem that is the
estimation of the molar volume of the mixture using Dalton’s Law.

(Refer Slide Time: 21:14)

So Dalton’s law of additive pressures states that the total pressure exerted by the gas mixture
is the sum of the partial pressures of its constituents.

599
(Refer Slide Time: 21:32)

So what are we doing here total pressure of the mixture exerted by the gas mixture is the
summation of the partial pressures Pi, so Zm we know it is Pm V by NRT which is equal to V
by NRT, we substitute the value of Pm here and we bring these values inside the summation
part, here we multiply both the numerator and denominator with an Ni and then we rearrange
this this term. So we know this term corresponds to I am going over here over the next page.

This term we know it corresponds to Zi and Ni by N it corresponds to yi that is the mole

fraction. So similarly we get almost a same expression that we got in Amagat’s law in
Amagat’s law we got the means we got something like this but in Amagat’s law the Zi that is
calculated was at the same temperature and same pressure as that of the mixture but the Zi

600
calculated in the case of Dalton’s law it is as the same temperature as that of the mixture but
it is at a different pressure.

The mixture pressure is different and this Zi is calculated at a pure component pressures, so
we move with the problems. So before starting we need to estimate the values of Z1 and Z2,
so for that we need to know the pure component pressures but that we do not know. So as an
initial guess we take the pure component pressures as the partial pressures of each
component.

(Refer Slide Time: 24:21)

So P1 I am just writing it in the tabular form. So P1 we take it as the partial pressure of the
component that is our initial guess that is 20 bar and P2 is 0.6 into 50 that is 30 bar. So we
take those as our initial guess for pure component pressures.

601
(Refer Slide Time: 25:18)

602
Now we know the Pc value we know the Tc value everything, so Tr reduced temperature
value is same as that of the one we measured for the Amagat’s law case that is 500 by 369.9
this is the critical temperature so that we got as 1.35. Now using these values these P1 and P2
values we know the Pc1 and Pc2 values. So we find out the Pr1 and Pr2 values, Pr1 value we
got it as 20 by 42.57.

Now from these values we can calculate Z1 and Z2 using the chart that we have done earlier
so I am not going into it I am simply writing down the value Z1 here we got it as 0.94, Z2 we
got it as 0.68. So I am writing down the values here. Now from these values we can calculate
Zm that is 0.4 into 0.94 plus 0.6 into 0.68 w get the Zm value as 0.784.

(Refer Slide Time: 27:50)

603
Now we need to calculate Vm, we do in the same way as we have done for Amagat’s law we
know the Zm value that is 0.784 into R 8.314 into T that is 500 Kelvin divided by the total
pressure we get the value of Vm as 6.518 into 10 raised to minus 4 meter cube per mole I’m
writing it here.

Now I told you earlier we have assumed that we assumed the pure component pressures as
the partial pressures of each component. So that was our assumption we need to cross check
whether our assumption was right or wrong.

(Refer Slide Time: 28:48)

604
So we need to recalculate P1 value and P2 value using Z1 and Z2 values that we have got. So
I am doing that, P1 you know it is Z1 N1 RT by V we assume one mole. So Z1 is 0.94 I am
taking it from here, N1 is 0.4, R is 8.314 and T is 500 Kelvin and V it is the Vm value that we
have got from here it is 6.518 into 10 raised to minus 4 and we get the value of P1 as we get
the value as 23.98 bar I am writing here the value first iteration and we get the value of P2 in
the similar way, we substitute these all these values Z2 we know it is 0.68, we get the value
as 26.02 bar, I am writing down the value here.

So if we just notice we can find out that here there is a deviation from our initial guess around
4 bar deviation is there 20 to 23.98 and here also around 4 bar deviation is there 30 to 26.02.
So our initial guess was not exact means it was not very good guess.

(Refer Slide Time: 31:05)

605
606
So we need to recalculate the whole thing with our (second) first iteration values, so we know
P1, we know P2 we find out Pr1 again and we find out Pr2 again and from these Pr1 and Pr2
value we know the Tr1 and Tr2 values already that is the initial value itself there is no change
for Tr1 and Tr2 values and from these values we use the graph and we find out our Z1 and Z2
again.

I am not going deep into the calculation because there is a time limit for me, so I am just
writing down the values of Z1 and Z2 as calculated earlier, Z1 value it is point 0.93, Z2 value
it is 0.73 and we can calculate Zm as I told earlier just a minute. So I am writing down the
value Z1 the new Z1 value that I have got and Vm we know it is 0.81 that is the new value
that we have got, so we get a new value for Vm that is molar volume of the mixture, write
down the value here.

So we continue these iterations till we get least deviation between these P means these P
values that is here we got deviation of 4 bars so these deviation reduces with iteration and
finally we have to be satisfied with that point that where were these gap has come down to a
very small value.

(Refer Slide Time: 33:20)

Now as we have got the value of Vm after our first iteration, so we need to do our second
iteration. So we need to calculate P1 and P2 again from the new values of Z1 and Z2. So I am
going to that here. The new value of Z1 is 0.93 and the new value of Vm is 6.734 into 10
raised to minus 4 and we get the new value of P1 as 22.96 bar, similarly we get the new value
of P2 as 27.04 bar.

607
So we know the new value of P1, we know the new value of P2, we calculate Pr1, Pr2 again.
The values of Tr1 and Tr2 are the same that we calculated earlier. So from the chart we can
find out the new values for Z1 and Z2 we do that again.

(Refer Slide Time: 34:50)

So we got the values of Z1 and Z2 as I am writing the values of P1 and P2 here, we did the
calculations and we got the value of Z1 as 0.93 and Z2 as 0.73 we did the calculations and we
found out the value of Zm as 0.81 and if we just simply compare we can see that here the
value of Z1 is 0.93, here also it is 0.93, 0.73, 0.73, 0.81 and 0.81. So there is no difference
after the (second iteration) means after the first iteration when we come to the first iteration to
second iteration there is no difference in these values. These values are exactly the same.

608
And the value of Vm when we calculate it again we get it again as 6.734 into 10 raised to
minus 4. So since there is no difference after this iteration we fix the value Vm as 6.734
meter cube per mole into 10 raised to minus 4 meter cube per mole. So this is how we
calculate the molar volume of the mixture using Dalton’s law.

So in this problem we have found out how we calculate molar volume of the mixture using
Amagat’s law as well as Dalton’s law. So that is for that is all for this tutorial and if you are
having any doubts regarding this problems just let us know we will be helping you, so that is
all thank you, thank you so much.

609
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 33
Partial Molar Properties

(Refer Slide Time: 0:45)

So I start from the point where I had left in the last class in the last class I had left in this
particular slide I had left in this particular slide where I was trying to define fugacity for
component i in a mixture and I had also mentioned what is the importance of finding out the
fugacity of a particular component in a mixture rather than finding out fugacity of the total
mixture and when we were trying to find out what did we find?

We found out that we got stuck up in this particular property, what is this particular property?
Is this property the can this be defined as the molar volume of component i which is just the
volume of component i divided by total number of moles of component i or in other words
suppose I add this component i with when i is in a mixture of j, will the contribution of i be
the same as if I add i to a large portion of component i itself.

So therefore in order to understand these things and I would also like to mention you all these
things I am mentioning in terms of volume the same things are applicable for other extensive
properties as well, right? So therefore in order to understand this, what people did? They tried
with different solutions or rather different proportions of water and ethanol to form a water
ethanol mixture and what did they observe?

610
They found that when they added one mole of water to a large volume of water, what was the
increasing volume?

(Refer Slide Time: 2:35)

It was 18 cc and all of us know that 18 cc it is the molar volume of water then what they did?
They added the same one mole of water to a large volume of Ethanol and what did they see?
They found that in that particular case the increment in volume was not 18 cc it was 14 cc. So
then what did it deduce it meant that the volume occupied by water by one mole of water in
an environment of ethanol is not the same as the molar volume of water in pure water in fact
they found out that whenever this one mole of water was added to different compositions of
the ethanol water mixture they found that the increase in volume was by different amounts.

So therefore that automatically implied that the contribution of any particular component i to
the total property of the component or rather total property of the component in the mixture is
cannot be evaluated in a straightforward manner just by its molar property. So therefore we
need a different set of properties to define or to identify the contribution of component
property to the total property, right?

611
(Refer Slide Time: 4:24)

And these set of properties they were defined as the partial molar properties. What are these
partial molar properties, let us see? They are a measure of how much the total property
changes when the moles of component i is added keeping temperature pressure and other
mole numbers fixed.

(Refer Slide Time: 5:11)

So therefore the partial molar property was defined as, suppose my total property is Y this Y
it is a function of say T, P and mole numbers ni’s, right, right? And for the time being I just
assume binary mixture’s simply because it will be easy for you to understand that. Remember
one thing whatever we are doing far binary mixtures will be equally applicable to any number
of components n of multi-component mixtures number 1. Number 2 most of the discussions

612
will be in terms of volume simply because it is easy to measure and easy to imagine and easy
to understand, whatever I discuss will be equally applicable to partial molar properties of all
extensive functions, right?

So therefore suppose Y is an extensive property then we find that del Y del ni at constant T, P
nj not equal to i this is nothing but the molar property of y which is signified as the
component is specified and to differentiate it from molar properties a bar is put on the top and
therefore these are termed as partial molar properties which identify the contribution that a
component of the mixture makes to the total property of the mixture, right?

So therefore this is basically fused again if we start its understanding it in terms of volume it
means that suppose we add one mole of component a to a very large volume of the solution
such that the composition of the solution remains unchanged under that condition the
contribution to volume which this particular one mole of component a makes is the partial
molar volume of component a in mixture a and b, remember one thing partial molar volume it
has to be a function of composition it has to be a function of the identity of the other
components because a will behave differently when it is in a solution with b. It will be able
differently when it is in solution of c and so on and so forth depending on how friendly b, c
etc are with a.

So therefore it is important to remember that yi bar it is a function of temperature pressure

and suppose there is just 2 components it is a function of n1 and it is a function of n2. So
therefore what do we find here? We find here that in the process when we shifted from pure
components to mixtures what did we do? For pure component we had total properties we had
molar properties I had very categorically mentioned to you total properties means capital
letters, molar properties means small case letters why did I enforce?

613
(Refer Slide Time: 8:29)

It is going to be clear you are going to come across more sets of properties and unless the
nomenclatures are defined we get confused at the end. So therefore now we have come across
partial molar properties which are exhibited only in solutions how do we denote them? We
denote them by denoting the component as the subscripts every time whenever I am
mentioning the components they are subscripts you must be remembering phases are
superscripts and components are subscripts and there is a bar over there over the top which
differentiates this from the molar property of component i, right?

(Refer Slide Time: 9:09)

So therefore first thing once we define a property let us see what are the different
characteristics of the properties? How does it differ from the normal molar properties? We

614
find, here is something interesting just to show you that in what way the it is just a typical
example to show you that how the importance of partial molar properties? What have we
done? We have taken a mixture of 1 propanol and water.

(Refer Slide Time: 9:32)

We know we have taken hundred millilitres of 1 propanol and we have also taken hundred
millilitres of water, right? We know and the entire thing was taken at 20 degrees centigrade
we know rho propanol the density of propanol at 20 degrees centigrade this is nothing but
equal to 0.803 gram per cc. rho water that is also known to you this is nothing but equal to
0.998 grams per cc, right? Now with these rho’s if we calculate and using the molecular
weights, we find this hundred mole of 1 propanol in hundred mole of water this corresponds
to x propanol of 0.194, right?

615
(Refer Slide Time: 10:58)

So therefore if that is the case let us come to 0.194 from here if we go straight up we can find
out the volume of the partial molar volume of water we find that this is more or less 18
millilitre per mole, right? Then we keep on continuing straight up till we touch the partial
molar volume of propanol versus x propanol curve I could not draw it straight but if we
continue then in this particular case we find that more or less the intersection this is at 72
millilitres per mole, fine. So therefore from here what did we know?

616
(Refer Slide Time: 12:08)

We knew we have come to know that the partial molar volume of propanol at x equals to
0.194 this is equal to 72 millilitres per mole or 72 cc per mole and v bar water at x propanol
equals to 0.194 this is equals to 18 millilitres per mole, right? Now then can you calculate the
total volume of the solution it is nothing but equals to n1v1 bar or rather suppose say 1 is I
write it down suppose say 1 is ethanol 2 is water it will just make my writing faster that is the
only thing plus n2v2 bar, what is n1 equals to?

If you just substitute n1 you find it is nothing but equals to 80.3 by 60.09 this is grams this is
grams per mole into 72 millilitres per mole plus what is n2 equals to 99.8 grams by 18.02
grams per mole into 18 litres per mole just do the calculation and see what is the result you
get?

617
(Refer Slide Time: 13:22)

We find that from here we get the total volume you can take up the calculation and you can
do it yourself we find this gives you 196 millilitres. So you have started with hundred plus
hundred 200 millilitres and this was the volume of the liquid before mixing; now the solution
is 196 millilitres, okay. So this was just to show you that well, the importance of partial molar
properties we cannot just use molar properties and add them up for normal circumstances
there are situations where we can do it, we will be doing those solutions are known as ideal
solutions we will be dealing them at the end after we have understood solutions, after we
have understood how difficult or how complex the behaviour of solution can be and the need
for simplifying the situation for practical calculations.

(Refer Slide Time: 14:19)

618
For the time being we continue our discussions more on partial molar properties, right? Now
just like I have been discussing the properties the volume characteristics let us see the general
characteristics of partial molar properties.

(Refer Slide Time: 14:25)

The first thing is the thing I have written quite at the end but this differentiates the partial
molar and molar properties this is something very important we know molar properties are
always positive we can never have a negative molar property under some circumstances
partial molar properties may be negative as well.

(Refer Slide Time: 15:01)

619
A typical example when does it happen? When you start with or rather when you add some
component in a solution and the volume actually contracts for example suppose you add
some magnesium sulphate to water we find that the molar volume of vMgSO 4 in almost pure
water this is minus 1.4 grams per cc, possibly I do not remember the exact value it is
probably minus 1.4 cc per mole.

So when we add magnesium sulphate in a magnesium sulphate solution at say xMgSO 4 is

almost equal to 0 that means you are added to pure water, we find that when we are adding
this one mole of magnesium sulphate to a large volume of water it actually causes a decrease
in the volume of water. Now this happens just because the open structure of water breaks up
and therefore the ions they become hydrated and therefore they collapse this is entirely a
different story it deals with the molecular levels we need not know it. It is just sufficient for
us to know that while molar volumes are always positive partial molar volumes maybe
positive may be negative this is number 1 thing.

(Refer Slide Time: 16:10)

And for a closed substance the partial molar volume it reduces to the molar property this is
number 1.

620
(Refer Slide Time: 16:20)

Number 2 just like we have defined all extensive properties have associated partial molar
properties for example we have already dealt with partial molar volume, fine. Same way we
can deal with partial molar entropy we can have all sorts of partial molar properties just in the
same way as we have a large number of your for each and every extensive property and it is
important to know that the relationships which we have proposed for total properties and for
partial properties the same equations also holds for partial molar properties as well.

(Refer Slide Time: 17:14)

Let us see, all of us know G equals to H minus TS, fine. Now suppose I differentiate this with
respect to say Ni keeping T, P, nj constant, right? What is this? Del H del ni T. P nj not equals
to I minus T, since we are doing the whole thing at constant temperature you must

621
remember the conditions constant temperature pressure and composition this is del S del ni T,
P nj not equals to i. So what is this? This is nothing but gi bar this will be nothing but hi bar
minus T si bar. So we have got this equation.

(Refer Slide Time: 18:20)

Again suppose we remember that the cross second partial derivatives are equal, what is this sj
bar? This is nothing but del S just now as it is written this is T, P nj, fine. What is this? This is
nothing but equal to del del ni of minus del G del T at constant P, nj not equal to i , what is
this term? This is nothing but minus sj so therefore this can we not write it down as minus del
del T of del G del ni P, nj not equals to i, what is this equal to? This is nothing but equal to
minus del mui del T at constant P, nj, fine.

So therefore in the same way we can show that vi bar this is nothing but equal to your del v
del ni T,P nj not equal to i, what is this equal to? Del del ni del G del P at constant T this
whole thing at constant T, P nj again we can change the order of differentiation we get del del
P of del G del ni constant T, P, nj this whole thing at constant T, nj and then what is this equal
to? This is nothing but equal to del mui del P at constant T, nj.

So from here what do I get? I get dgi bar this is nothing but again this is nothing but the
chemical potential this is minus si bar dT plus vi bar dp. So in this particular way what do we
find? We find that we can develop an identical set of equations just like we have we had done
for total properties as well as for molar properties we can go on in this particular way and
finally in this particular process what do we find?

622
(Refer Slide Time: 20:39)

We find that in this particular process we can arrive at a set of Maxwell’s equations in terms
of partial molar properties, right? After this what is the next thing? This is the next firstly I
found out the difference between molar and partial molar properties the second thing is we
found out that all extensive properties have the corresponding partial molar properties and all
these partial molar properties they have the same relationships with one another just like the
way the molar properties and the total properties were having. What is the next thing that that
can be, what are the other properties that we can define? The other properties are that a very,
all of these are intensive properties.

623
(Refer Slide Time: 22:42)

Now since they are intensive properties we can come across or we can deduce a very
important equation which is known as the Gibbs Duhem equation in terms of partial
properties. Now let us define this equation before we stop for a break and go to the next class.
Let us start from the basic equation. What is the equation let us see?

We know that what is dV suppose we know for the volume of a mixture this is a function of
TP, I have just 2 components A and B, so it is a function of TP and nV. So at constant T and P
what is dV equals to? It is nothing but del v del nA constant TP, nB into dnA plus del v del
nB constant TP nA into dnB, agreed? Or in other words what is this equal to? This is equals
to vA bar dnA plus vB bar dnB, fine. Now suppose I keep on increasing the size of this
particular sample such that the volume increases but the composition does not change.

How can I do it? I just increase the size uniformly by a scale parameter lambda say such that
we find that dn1 this will be equal to n1 into d lambda similarly I keep on adding dn2 moles
such that this becomes n2d lambda so what am I doing in the process? I am increasing the
entire volume of the of that particular system keeping the composition constant. So what am I
doing? I am adding n1d lambda moles of 1, I am adding n2d lambda moles of 2 such that
when I add these 2 my volume becomes v into d lambda, right?

Because I am just adding I am just increasing the volume without changing the composition.
So therefore in this particular case I can, so dV I can write it down as vd lambda this is equals
to vA bar nAd lambda plus vB bar nbd lambda, agreed? Now we need to remember that as I

624
vary lambda from 0 to 1 I am going from 0 volumes, I am just adding both the components in
the similar same proportion such that I attain volume V.

So therefore as lambda is varying from 0 to 1, I am getting the volume of my solutions

changing from 0 volume to the total volume and during this process how am I doing this? I
am adding nA d lambda of component A and nBd lambda of component 2, right? And so
therefore I can write the equations in this particular form. Now as I changed lambda from 0 to
1, I can differentiate this from 0 to 1, I can differentiate this from 0 to 1.

Remember one thing I am not changing the composition of the mixture, so therefore my vA
bar ad vB bar remains constant as the volume of the mixture is increasing this is important.
So therefore from here what do I get? I get vA bar nA plus vB bar nB, fine. So what have I
done? I have increased the solution volume simply by adding or rather simply by keeping the
composition of the solution constant now suppose I differentiate this equation what do I get?

On differentiating this equation I would be getting vA bar dnA plus nA dvA bar plus vB bar
dnB plus nB dv bar and remember with what had I started? I had started from this equation,
what was this equation? It was vA bar dnA plus vB bar dnB what have I done? In both the
cases I am trying to compute the change in volume by dV amount and I know that volume is
a property of the system so therefore these 2 equations they should be equivalent.

(Refer Slide Time: 26:43)

If these 2 equations have to be equivalent then what does it reduce to? It automatically tells
you nA dvA bar plus nB dvB bar should be equal to 0, so if I am going for not 2 components
but more than 2 components what do I get? I get ni dvbi bar equals to 0 and this particular

625
equation is known as Gibbs Duhem equation this has been derived in terms of partial molar
volume but the same derivation that I have done can be applicable for any other partial molar
properties as well.

So therefore generalizing this equation, can I not write it down as? Suppose again my yi bar
is a partial molar property of total property Y then can I not write it down in this particular
form? Where yi bar is the partial molar property of any extensive property i and this also can
be rewritten this is in terms of mole numbers we can also rewrite this equation in terms of
mole fractions which give us this particular equation. So this is the generalized Gibbs Duhem
equation in terms of partial molar properties and this I would like to say is a very important
finding or a very important derivation of thermodynamics this is used for a large number of
situations we are going to discuss the importance of Gibbs Duhem equation in the next class
after I discuss little more about partial molar properties.

626
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 34
Partial Molar Properties(Contd.)
(Refer Slide Time: 0:59)

Well, to continue with our discussion on partial molar properties I ended the last class by
deriving in this particular equation when I had derived the Gibbs Duhem equations in terms
of the partial molar properties. as I mentioned this particular equation is something very
important, before anything else I would like to discuss the importance of the Gibbs Duhem
equation of course in this particle case I had denoted my property with the value of m just
keep in mind that what I was telling in terms of n the same thing I have shown in terms of m
here mi bar is my partial molar property.

627
(Refer Slide Time: 1:17)

Now from this particular equation a large number of things that we can see for example let us
start with a binary mixture, what is it? x1 for a binary mixture plus x2 dm2 bar equals to 0 I
can also write it down as dm1 bar equals to minus n2 by n1 dm2 bar which is nothing but
equal to minus x2 by x1 dm2 bar. So from this what do we find? We find that if x2 by x1 is
large then a small change in this will correspond to a large change in this number 1, right?

So therefore we find that the changes in the partial molar properties they are not independent
they are dependent upon the mole fractions. So therefore we find that if x2 by x, so therefore
if I have a large proportion of x2 then a small change in dm2 is going to bring a large change
in dm1, number 1. Number 2 what do we find? X2 by x1 is always positive so therefore an
increase in the partial molar property of component 1 reached to a correspond decrease in the
partial molar properties of component 2.

628
(Refer Slide Time: 2:32)

Or in other words as I have mentioned here dm1 is positive dm2 can be negative, fine. And
the most important thing is that it is very difficult to find out the partial molar properties.

(Refer Slide Time: 2:48)

So therefore this equation gives us a way by which if I know the variation of dm1, I would be
in a position to find out dm2 on integration it will give me m2 bar for any particular
composition x2, right? So therefore how do we go about doing this, suppose my m1 bar is
known say, right? I would like to find out my m2 bar I do not have a very straightforward
experimental way to find this out.

629
So therefore can I not write it down as dm2 bar this is equals to minus integral x1 by x2 dm1
bar from 0 to m1 bar or in other words from this integration what do I get? This is equals to
minus integral 0 to m1 bar x1 by x2 dm1 bar. So therefore from here what do I get? From
here I get that if I know dm1 bar how it varies with composition? I shall be in a position to
find out m2 bar and the important thing is for each m2 bar it can be estimated from the
method of intercepts.

Firstly just apparently if you just look at the expression, what is mbar mi bar this is delM, Del
ni at T, P, nj, fine. So therefore it automatically implies that if you have any particular graph
where your total property M that varies with the mole numbers ni say suppose it has a graph
of this sort, suppose M equals to volume so suppose what I do? I keep on adding for
component i or for component a say I keep on adding moles of a, say 1 mole 2 moles 3 moles
4 moles I keep on adding and for each particular addition I start finding out the increase in
volume, right?

Accordingly I can find out different points on this curve I can construct this particular curve.
Now if I want to find out the partial molar volume at some particular compositions say at this
particular composition what do I do? I I can simply draw a tangent and find the slope of this
curve but finding out the slope of this curve in this particular way it is not at all a very easy
technique, okay.

(Refer Slide Time: 5:50)

It leads to a large number of errors, so therefore the proper way or much more accurate way
of finding out the partial molar properties the experimental determination that is usually done

630
determination of mi bar the usual method is known as the method of intercepts. Let us see
what is this method first? And then we will see how it works?

(Refer Slide Time: 5:59)

Now definitely you must remember that direct numerical differentiation just I was telling this
can be done that but this is not very accurate so therefore this method of intercepts has been
adopted.

(Refer Slide Time: 6:06)

Now what is this method of intercepts? We know that the molar volume of a mixture, what is
this? This is nothing but the, again we use it for a 2 component system say component A and
component B, what is V? This is equals to nA plus nB I can write it down or in other words I

631
can write it down as v into nA plus nB, agreed? What is vA bar equals to by definition? It is
Del V delnA at constant T, P nB in this particular case.

So this should be equal to differentiation of this, so therefore this should be del del nA of v
into nA plus nB at constant T, P nB. On differentiating what does this give you? This gives
you v plus nA del v del nA T, P nB plus nB del V del nA T, P nB, right? Because del nA del
nB gives you 0. So there what is this? This is equals to v plus nA plus nB that is the total
number of moles into del v del nA at T, P nB, agreed? Now from chain rule let us if we can
simplify this a little more further.

(Refer Slide Time: 7:52)

What is this let us see? Del v Del nA at constant what have we done?

632
(Refer Slide Time: 7:57)

This is at constant nB and T, P of course it remains

constant. (Refer Slide Time: 8:14)

So therefore can we not write it down from chain rule as Del v delxB at constant nB delxB
delnA again at constant nB this is possible, what is this tell me? Is this not equal to Del v Del
xB at constant nB into del del nA of xB is nothing but nB by n plus nB and what does this
give you? Is this is not equal to minus of Del v Del xB into nB by nA plus nB whole square,
simply differentiation that is what I am doing nothing else.

633
(Refer Slide Time: 9:05)

So from here what is this equal to? From where did we start? Now this particular equation,
suppose I substitute it here so therefore what do I get then? Then I get this is vA bar the
equation vA bar was equal to v.

(Refer Slide Time: 9:26)

If I just write down the equation this is v plus nA plus nB into del v del nA T, P nB, right? So
if I substitute this particular path from here then what do I get? In this particular case I get vA
bar this is equal to v minus xB because nA plus nB whole square and this cancels out del v
del xB I get this. So therefore vA bar is this.

634
(Refer Slide Time: 10:17)

So suppose so suppose I have plotted in this particular case what have I done? In this
particular case I have plotted the volume of the component the molar volume of the
component as a function of its composition, right? So from there what do I have?

635
(Refer Slide Time: 10:44)

I can very well write down this particular equation as v equals to vA bar plus xB del v del xB
I can write it down in the same way if I had started with component B and not component A
then I would have landed in a equation as vB equals to v plus 1 minus xB del v del xB from
where I know this v would have been equals to vB bar minus 1 minus xB del v del xB, it is
alright v equals to, right? Now from this equation and this equation what do you find? We
find that suppose we are plotting v versus xB.

(Refer Slide Time: 11:33)

The thing which I have done in this particular graph we have plotted molar volume as a
function of xB or x2 whatever it is, right? And we would like to find out the partial molar
volume at say some particular say point say at x2 equals to 0.3 suppose I would like to find

636
out. So this is the point at this particular point I would like to find out the partial molar
volume, what do I do? I draw a tangent to the curve and from this equation what do I find?

(Refer Slide Time: 12:13)

From this particular equation we will find that at xB equals to 0 what is v equals to? V equals
to so therefore this reduces to the partial molar volume v equals to vA bar under what
condition? When x yeah, v equals to vA bar for xB equals to 0 and v equals to vB bar at xB
equals to 1, isn’t it? Just check it up this is the thing that you are supposed to get.

(Refer Slide Time: 13:01)

637
So therefore the thing which we do is we simply draw the tangent and then we find the
intercepts on the two extreme axis and once you find the intercept we know that merely from
the intercept at x2 equals to 0 that gives us the partial molar volume of component a or
component 1 and the intercept at x2 equals to 1 that gives us the partial molar volume of
component 2.

So therefore this is one particular way of finding out the, this experimental way of finding out
partial molar volume of any of the components and then again I would like to remind suppose
you are finding just the partial molar volume of any particular component the other
component can be found out from Gibbs Duhem equation. So therefore what are things that
we discussed? We first understood what are partial molar properties, we discussed the
characteristics of the partial molar properties and we also found out how by the method of
intercepts really from experimental data partial molar properties can be evaluated. Now
before I end there is one particular partial molar property which is particularly important.

(Refer Slide Time: 14:33)

We continued our discussion with the partial molar volume just because it is easy for us to
understand and to imagine that particular quantity but the partial molar property which is
going to be of importance to us is the partial molar Gibbs free energy gi bar, what is this
equals to? This is equal to del G delni TP nj not equal to i and we have already defined this is
nothing but equals to mu i and again just the way that we had discussed earlier in the same
particular way we can very well be defined that dG equals to we can define it in this

638
particular way this is nothing but at constant temperature and pressure, so this will be Sigma
ni TP nj not equal to i, right into your dni.

So therefore here we can write it down as Sigma mu dni this can be written down where
again mu this is a function of TP for binary mixtures it is a function of n1 n2 otherwise it is a
function of ni if we have to deal with in this particular way again we find out that this
particular term this does not only show how your Gibbs free energy varies with mole
numbers. If we can write this equation in this particular form and then we differentiate dU
equals to if we elaborate this particular differentiation and say I will just do it for you minus
Pdv minus vdp plus sdT just differentiating it plus again that dG comes.

So therefore for dG we have vdp minus sdT plus if it is binary mixture we have just 2 terms
muB, agreed? Now if we cancel out the terms this term this term this term this term if all of
these they cancel out then in that case what do we get? We get dU equals to minus pdv plus
Tds plus mu A dna plus mu B just the way we have done for pure component systems. So
therefore at constant V and s what do we know? dU is nothing but equals to Sigma mu i dni,
fine.

(Refer Slide Time: 17:32)

So there again in the similar way if you continue what do we find? We find mu i therefore it
is delG del ni at constant T, P nj not equals to i this is equals to del u del ni at constant s, v nj
not equals to i. This is equal to delH del ni just the way but we have just added this particular
term and in terms of Helmholtz free energy it is T, V nj not equals to i. So from there we find
that this particular term it is not only the partial molar Gibbs free energy but it also shows,

639
how other thermal parts other thermodynamic properties they vary with change in
composition under different interactions between the system and the surroundings.

So therefore mu i occupies a unique position just because of that we do not apply any sort of
bar or anything on it we just keep it as mu i for component i in solution for a pure mixture we
simply denote it by mu this is sufficient for you to remember. The other thing which is of
importance in this particular case is that we very well-know that dG equals to minus dw non-
PV this also we had defined under some particular conditions. So therefore what does this
imply? This implies this is equals to nA dmuA or in other words if I write it in general it is ni
dmu i.

So from this what do we deduce? From this we can deduce that suppose we can also bring
about some particular change in the partial molar properties which gives us a change in the or
rather suppose in an electrochemical cell we can perform reactions at 2 different sides due to
which the mole fraction or the composition of the mixture can change and from there we can
derive some amount of non-PV work which will be a function of the chemical potential.

So therefore chemical potential there in that case we find therefore that it has got some very
special significance as compared to the chemical potential of, so we can just write it down
that d mu A dnA plus mu B dnB just repeating what I have said this is equal to dw non-PV.
So therefore from here it implies that additional non-PV work can be done just by changing
the composition of the system and this is what is practiced in an electrochemical cell where
chemical reaction is arranged to take place at 2 distinct sites at the 2 electrodes and electrical
work that the cell performs can be related to its change in composition or to the changing of
the chemical potential of the reactors, right?

So therefore this was all about partial molar properties we have defined a large number of
partial molar properties and from there what have I told you? We have come to know that
from these partial molar properties they have some particular relationships they obey all they
have all the characteristics of total properties and molar properties we have found out that
experimental determination the only thing only question which I keep before you which we
will be discussing in the next class is, whatever properties we have defined we have defined
that they have an additive characteristics all other properties.

640
(Refer Slide Time: 21:49)

Does partial molar fugacity have identical properties or does this have some unique
characteristics of its own? Think about it we are going to deal with partial molar fugacity in
the next class. How it is unique as compared to other partial molar properties and this is of
special concern to us because we started discussing partial molar properties just because we
wanted to find out the fugacity of component i in a solution and from there the entire problem
started. So therefore it is very important to understand the partial molar fugacity of
component i in solution we will be dealing with this in the next class, thank you very much.

641
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 35
Partial Molar Fugacity
Well, in the last class we had started discussing the characteristics of mixtures and then we
found out that in order to properly define mixture properties in terms of single phase or rather
the pure component properties we had to propose a number of mixing rules and combining
rules whether we were using the cubic equation of state, the Virial equation of state or the
compressibility factor charts.

So we have to propose a number of mixing rules which gave us the constants characteristics
of the mixture in terms of the pure component constants and then we found out that in those
particular this empirical laws there were some interaction parameters which took into account
the interaction between unlike molecules and in order to express them in terms of interactions
between light molecules say for example there is a mixture constituting of component a and
component b.

We would like to express the interaction of component a and b in terms of the interactions
between molecules of component a and the molecules of component b.

(Refer Slide Time: 1:53)

So in other words we would like to express aij in terms of aii and ajj and this was possible by
means of combining rules so we have devised a number of combining and mixing rules and

642
based on those we had tried to define the mixture properties in terms of the pure common
properties.

And then we found out that suppose we would like to define particularly fugacity of a
component i in a mixture of say c number of components where i extends from 1 to c then in
that case and I had also explained why it is so very important to define the fugacity of
component i in a mixture because that is what decides the phase equilibrium conditions. Now
under that conditions we have found out from the basic equation RT ln 0 to Pv mixture minus
RT by P dp this was for the mixture case.

Now if I try to find out the fugacity of component i in the mixture, so therefore what does it
become? It becomes fi which is actually the partial mole or the contribution of component i to
the mixture product. So normally for any and this should be I will be explaining all the terms
this should be 0 to P and here it should be vi bar where vi bar it is nothing but I had already
described it was del V del ni at constant TP nj not equals to i minus we are often tempted to
write this down as minus RT by Pi.

We are tempted to write it in this particular form. Now let us see what should be here and
what should be here before we proceed? Now what from where did this term come? This
term is actually vi bar 0 what does it mean? It means the partial molar volume of component i
in the ideal gas mixture, a mixture of ideal gases there what is this vi bar0? Now what is this
equals to? This should be equal to del v del ni where this is for the ideal gas TP nj not equal
to i, right?

Now what is the total volume of the mixture? Total volume is RT by P into n1 plus n2 say
nith component it was still nN component, right? Now suppose I differentiate this with
respect to component ni, what do I get? I get del V0 del ni at constant T is constant P is
constant all other mole numbers are constant. So therefore what does it reduce to? I can just
get this one differential term when I am differentiating it with respect to ni RT by P where all
other terms they cancel out.

So therefore under that condition this reduces to RT by P. So therefore it is very important to

remember that basically the term which I use here it is the partial molar volume of any
particular component i in this case this RT by P came because it was the volume of the total
gas mixture in the ideal gas state because it is important to remember that whenever I define
fugacity it is with respect to P equals to 0 where the gas behaves as an ideal gas and therefore

643
it is important to remember that v for the molar volume for a pure gas or a single component
gas that is equal to RT by P under ideal gas state, right?

(Refer Slide Time: 6:21)

In the same way, so I will just repeat it once more in the same way v for an v0 ideal gas
naturally 0 means ideal gas that is equal to RT by P. Interestingly suppose this same gas it is
in a mixture with other gases where I have to define the partial molar volume just note the
partial molar volume since it is molar it is small case it refers to component i subscript i, it
refers to the ideal gas state superscript 0 this is a partial molar volume there is a bar.

So just remember if unless you give all the notations all the necessary details are provided in
this particular nomenclature if you miss any of them then the notation becomes incomplete.
So therefore I keep on emphasizing the importance of correct depiction of different
properties. Again I repeat the primary reason for this is that we are using or rather we will be
dealing with a host of different sets of properties just for your convenience if I can write it
down total properties that means properties for the entire mass or for the whole system they
are given in terms of capital M.

Molar property the corresponding molar property it is nothing but the total property divided
by the number of moles contained in that particular component whether it is a mixture
whether it is a pure substance whatever it is this will be denoted by m. Next we go for partial
molar properties these have meaning only when any component i, suppose the total properties
for component i, so I can write it down as Mi this is Mi this is mi, okay.

644
Now when this component i is dissolved in a solution with other components then only the
partial molar property comes into being otherwise no. So therefore when it comes what is
this? This is a molar property small case, molar property of what; i suppose it would have
been in a phase Alpha. Suppose I have 2 phases each comprising of say n components of n
components of pure substance say in phase Alpha n components of pure components in say
phase beta.

Then in that case the partial molar property of component i in the Alpha phase would have
been shown in this particular form. Suppose if I had 2 phases but at the moment since we are
dealing with just mixtures either in say the vapor phase or in the liquid phase so for the time
being I am not mentioning this but moment we are going to mention phases and components
please remember component is subscript, phase is superscript and small case means molar
and bar means partial molar property, right?

So therefore the partial molar property of component i in the ideal gas state that is when
suppose I assume that the gas mixture which I have is a mixture of ideal gases, for that
particular case I find for a mixture of ideal gases this is also equals to RT by P very
frequently written to confuse this therefore it is very important to remember. So therefore
accordingly what is the equation for finding out the fugacity? It is RT ln the fugacity how I
am going to denote this fugacity? We are going to discuss.

Normally if we feel that they are there should be a bar but whether I would give a bar here on
something else that I will be discussing a little later. And this fugacity it has to be expressed
in terms of the fugacity of this particular component i in the ideal gas state. What is the
fugacity of this particular component i in the ideal gas state? If this i would have remained as
a pure component than the fugacity would have been equal to the pressure in the ideal gas
state but when it is mixed with a large number of components then it manifests itself as the
partial pressure of component i in the gas phase.

So therefore this gives us the fugacity of component i in any particular state other than the
ideal gas state and therefore in order to find this out what do I need? I need to integrate it
from very low pressure under which this particular real gas behaves as an ideal gas and the
fugacity it is the partial pressure from this state to the pressure P where we normally wanted
and this is equals to vi bar minus vi bar 0 which is nothing but equal to RT by P dp. So till
this much we have defined in the last class.

645
(Refer Slide Time: 11:30)

I have also noted this down here with a proper definition of partial molar properties. Now
when we go to find this then what did we find?

(Refer Slide Time: 11:50)

That in order to calculate fugacity we need to know the partial molar volume of component i
in the mixture not only at the pressure of interest but since it is inside the integral we need to
know this particular value or how this varies with pressure right from P equals to 0 to the
pressure of interest, right?

646
(Refer Slide Time: 12:06)

So we need to know the partial molar volume and also what is the other thing we know which
I have already mentioned here?

(Refer Slide Time: 12:23)

The partial molar volume it is not only a function of this will be a function of temperature
that we can keep constant the total pressure and it will be a function of the mole numbers of
all the components or in other words it’s a function of composition we had already seen in the
last class that this is different for different compositions of the same set of components
making up the mixture and it is also important and the composition it depends upon the
nature of the or molecules of component i.

647
It is dependent upon the nature of the other molecules which are present, so therefore finding
vi bar in terms of P over the entire integration from 0 to P that is something very difficult in
any case we had already discussed that even finding out a generalized equation of state even
for a pure component over a large range of conditions comprising the 2 phases was anyhow
difficult and we have already got a favour of the large variety and number of equations that
we deal in order to describe the PVT behaviour of gases.

The tremendous number of equations which had come up as I had already mentioned was just
because no one equation was found to be suitable for the entire range of conditions that we
would be dealing, right? So therefore in this particular case also it was quite difficult. So
naturally what people then tried to do? People then tried to find out whether any particular
simplified model can be used in order to calculate the partial molar fugacity of component i
in the mixture before we go that it is important for us to realize the unique position of the
fugacity of component i in the mixture as distinguishing from the partial molar property of
any other compound for any other component.

(Refer Slide Time: 14:24)

For any other component what did we find? We found out that partial molar properties they
are additive in nature or in other words when I write the total property it is nothing but equal
to ni mi bar where m is an extensive property and in terms of molar property we can write xi
mi bar that we have already seen, okay. We had tried to express it in terms of volume,
enthalpy, entropy everything what about partial molar fugacity?

648
Let us see how the fugacity of the mixture in what way is it related to the fugacity of
component i in the mixture. Does it obey do we have a relation like fM equals to xi say fi bar
or something of component i in the mixture. Do we really have such a equation? Or such an
equation does not hold for fM, let us for everything that we do we are going to start from the
basics.

(Refer Slide Time: 15:18)

This is very important because we have to start everywhere everything that we do this is the
basic equation from which we are going to start why? Because moment you try to jump steps
you will find that with standards states etc you are going to get completely confused, right?
So therefore this is something very important and so therefore let us see what we can do
about it. So therefore let us start from the very basic equation which is given by this particular
form.

649
(Refer Slide Time: 16:09)

Now what is that equation? What is the basic defining equation RT d ln fM which is the
fugacity of the total mixture that is equal to vM dp, right? This one we know. What is vM
equals to? Just if we can write it in this particular terms v mixture is nothing but Sigma xi vi
bar, fine. So instead of vM can I not write it down at as xivi bar dp, fine.

Or suppose for a gas mixture suppose because usually we will deal with gases and then we
will come to liquid in that particular case RT d ln fM, normally again let me define that mole
fractions this is another definition which you need to know mole fractions in gas phase they
will be denoted by y mole fractions in the liquid phase will be denoted by x, right? So
therefore what do we find? That the relationship between d ln fM with a partial molar volume
is given in this particular form, now suppose I try to expand it then what do I get?

650
(Refer Slide Time: 17:35)

Say for the time being just for simplicity for 2 component system which is gas phase mixture
at constant temperature and pressure. If I elaborate that what do I get? RT d ln fM this is
equals to y1v1 bar dp plus y2v2 bar dp, agreed?

Now what is this y v1 bar dp can you tell me? This v1 bar dp it was nothing but y1 RT d ln
f1, agreed? Which is the partial molar fugacity of component 1 in the mixture plus y2 RT d ln
f2, the RT if they cancel out what do we get? We get d ln fM is nothing but y1d ln f1 plus y2
d ln f2, now we start integrating this. If we integrate this from say P equals to 0 where fM
equals to P to the point of P of interest where they for P equals to 0 to the point of interest at
P.

Definitely we will also be rather we will also be integrating these particular terms also this
will be in terms of y1P to some particular situation f1 this will be y2P to say f2. So in that
particular case on integration what do we get? We get ln fM by P equals to ln f1 by y1P y1
plus y2 ln f2 by y2P.

651
(Refer Slide Time: 19:48)

Or in other words for removing the ln terms what do we get? We get fM by P this is equal to
f1 by y1 P whole to the power y1 f2 by y2 P whole to the power y2, is it clear to you?

So therefore what is the relationship between partial molar between the fugacity of the
mixture and the partial molar fugacity of the individual components it is in no way related to
the way the any other partial molar property m is related like yi mi bar so this relation does
not hold for the molar fugacity and since it does not hold it is very important to distinguish
the partial molar fugacity from any other molar properties.

So what do we do for this? Just to keep on reminding us that the partial molar fugacity does
not causes the properties of other partial molar properties it is something different from them,
how do we do? We do not denote this with a bar that we have been doing it we denote this
with the cap it just a hat to mark that this is different from this.

652
(Refer Slide Time: 21:06)

So therefore what we do for each partial molar fugacity it is important that we denote them
with a cap while we are denoting the partial molar volumes with a bar.

(Refer Slide Time: 21:12)

So therefore from this particular equation from this particular equation what are the things
that we see? We see from this equation that more or less the fugacity of component i in a
mixture it is something different from the fugacities of or rather it is different from the other
partial molar volumes number 1.

And also it is important to remember that the molar volume of component i in pure state is
equal to the partial molar volume of component i in the mixture. So with this finally what we

653
have done we had tried to define the partial molar fugacity in terms of PvT behaviour of
gases when the component i is in a mixture and from there we have realized that since it is
very difficult to find out the partial molar volume of component i in mixture not only at the
pressure of interest but over the entire range of pressure from 0 to P. So therefore it was felt
necessary to suggest or to devise a simpler way of trying to estimate the partial molar
fugacity of component i when it is in a mixture.

(Refer Slide Time: 23:10)

Now how to go about this? What people found out was that they started with gaseous
mixtures, in gaseous mixtures what did we find? In gaseous mixtures we observed that again
suppose I start writing for gaseous mixtures what do I get? First let me generalize this, so
therefore this fM by P it can be better written as not Sigma but product of over i equals to 1 to
c fi by yiP whole to the power yi this is something which is quite interesting and quite
different.

So the next thing is in order to find out rather since it is very difficult to or rather although it
is an exact equation but it is very difficult to find out fi bar, so we wanted to find out some
other way by which we can find out fi for the different phases and then we can start solving
the other phase equilibrium problems. Now for that what did we take up? We tried to find out
something simplifying something much simpler.

So what is the simplest thing that comes to our mind? The simplest thing which comes to our
mind is the as I had mentioned mixture of ideal gases. Now there what we have already seen
suppose we are dealing with a mixture of ideal gases then in that case again suppose what do

654
I find? This is nothing but equal to RT d ln now very confidently I will be putting a cap and
please do not forget the cap because this just shows the partial molar fugacity here I would
like to mention once more that when I put a cap this is partial molar fugacity of component i
in mixture and when I just mention fi it is the fugacity of component i in pure state.

So you will find in several textbooks they try to put a 0, a star and lot of things just to make
matters simple I am not going to add up anything when it is a partial molar property it has a
cap, when it is in a pure state it has no cap and therefore the only thing which and the
subscript i it denotes component i, right? So for a mixture of ideal gases what do we find? We
find for a mixture of ideal gases this is equal to vi bar minus RT by P the standard equation
from 0 to P.

Now when we are dealing with ideal gases then we know this is vi0 which is nothing but
equal to RT by P. So for a mixture of ideal gases this term cancels out and we know fi equals
to yiP, right? But this is a very restrictive assumption or rather it cannot be generalized very
easily, why? Because this assumes that there is no interaction between the molecules.

Now while this can be imagined or this can be accepted for gaseous mixtures we would like
to propose a generalized expression to calculate the partial molar fugacity in terms of your
compositions or something for all types of solutions be it liquid solution, solid solutions,
gaseous solutions everything, right? So therefore using the idealization for the mixture of
ideal gases will become too restrictive.

(Refer Slide Time: 26:34)

655
Than what people thought or tried to do they tried to find out the situation for an ideal
mixture of real gases whether this could be a slightly more generalized portion while being
quite simple.

Now generally we find that the ideal mixture of real gases are those which confirm to the
Amagat’s law of additive volumes, right? So in the next class we will be dealing with the
Amagat’s law of additive volumes and whether it gives us a much more simplified version of
calculating the partial molar fugacity in terms of composition, we continue in the next class.

656
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 36
Non Ideal solutions
(Refer Slide Time: 0:47)

Well so to continue with your discussions, what did we observe in the last class? We
observed that well to find out fugacity by using this particular equation, the equation I had
already written down the equation, so we finding out fugacity by using this particular
equation this was not at all very easy and so therefore this is the corrected this is the equation
written properly. So therefore it was it was important to find out a simple model from which
we can find out fi from known measurable parameters in terms of the composition.

657
(Refer Slide Time: 1:09)

So we found out that if we consider a mixture of ideal rather a yeah mixture of ideal gases
then we know that fi bar is nothing but equals to yiP and this is too restrictive because it
neglects the interaction between the molecules it tells us that that the interaction between the
molecules is negligible which is quite restrictive particularly for dealing with liquid solutions.
So then after that what we wanted to do we thought of doing let us go to an ideal mixture of
real gases. Now what does this Amagat’s law it suggests it basically suggests that the
volumes can be added over the entire range of the volumes can be added over the entire range
of pressure that we consider.

So therefore from here we know that the total volume of the mixture is nothing but equal to
your nivi, okay. Or in other words we assume vi bar 0 equals to, it’s not 0 it can be for a real
gas that is equal to vi. So this on this particular bases we write down the total volume in
terms of the molar volumes this Amagat’s law of additive volume states or it states small v it
it can be equal to xivi this is this is what it states.

Now suppose we use this particular assumptions or rather this particular consideration, let us
see what is the expression of the partial molar fugacity that we get? Again we start from the
very basics, suppose we have for a pure substance for a pure substance component i we can
write it down as RT d ln fi the pure state so it is just a subscript nothing else this is equals to
vidp and then in the ideal gas state it is P it is RT by P dp, right?

658
On subtracting what do we get? R T d ln fi by P equals to vi minus RT by P dp integral, sorry
this integral and this is also integral, right? So therefore from there we get this integration is
from P equals to 0 where fi equals to P to P equals to P where the value is fi, here the value of
fi is equal to the total pressure, right? So therefore we get RT ln fi by P equals to integral 0 to
P vi minus RT by P dp this is what we get for a pure substance, a pure component i.

(Refer Slide Time: 4:01)

Now let us see what do we get for this same component i in mixture? For component i in
mixture, what do we get you tell me? In this particular case this becomes R T d ln fi this is
equals to vi bar dp and R T d ln Pi this is equals to RT by P dp, we know it. On subtraction
what do we get? We get RT d ln fi by yiP is nothing but equals to vi bar minus RT by P dp if
we integrate it again from P equals to 0 where fi bar equals to yi by P2P equals to P where fi
bar equals to fi this side also 0 to P, what do we get? We get R T ln fi by yiP equals to integral
0 to P vi bar minus RT by P dp, this was for component i in mixture.

659
(Refer Slide Time: 5:12)

And for the pure component i what do we get? I have just now derived for pure component i
in the derivations i is here.

(Refer Slide Time: 6:13)

So for the pure component i what do I get RT ln fi by P where fi this is vi minus RT by P dp

and what is this fi equals to? This fi is the fugacity of pure component i, one thing we need to
remember very well this fi bar fi everything is evaluated at the conditions of temperature and
pressure remaining same. It for both the cases the pressure is P and the entire integration this
and this have been done at constant temperature only when the pressure and temperature
conditions of this equation and this equation are the same then we can combine the two.

660
Suppose this particular gas mixture it obeys Amagat’s law of additive volume than what
was the Amagat’s law of additive volume I have already written down it's vi bar equals to vi.
So therefore if it obeys the Amagat’s law of additive volume then this term and this term they
cancel out.

(Refer Slide Time: 6:23)

And then ultimately what do we get? We get on cancellation we get fi by yiP equals to fi by P
or in other words the partial molar fugacity of component i in a mixture is nothing but equal
to the mole fractions times the molar or rather the fugacity of component i in the pure state at
the same temperature and pressure as that of the mixture. So therefore this is applicable we
find this has been derived from Amagat’s law of additive volume this is applicable provided
fi and fi bar at same T and P, right?

661
(Refer Slide Time: 7:09)

So therefore what is fi then in this particular case? I have written down the ideal solution
model here, this is ideal solution model where what is fi equals to? Just remember it is
fugacity of pure liquid i evaluated at temperature and pressure of solution. The only thing we
need to know that this fi cap and fi are evaluated at the same conditions of temperature and
pressure. Accordingly from this suppose I express fi bar by pi equals to fi bar and so therefore
we know for the ideal solution model phi I phi i bar equals to phi i both being at the same
temperature and pressure which automatically implies that the partial molar fugacity
coefficient it is a function just of temperature and pressure it is not a function of composition
which is quite restrictive assumption.

662
(Refer Slide Time: 8:14)

Yet we find that from this particular restrictive assumption it is although it is restrictive it
gives us a very simple equation a very simple model which tells us that the partial molar
fugacity of component i in a mixture or in a solution is the mole fraction of that particular
component in solution times the fugacity of that component in the pure state at the conditions
of same temperature and pressure as that of the solution and finding out pure component and
finding out your component fugacities in the in for a real gas for a liquid these I have already
been discussed and those particular equations can be used to find out fi. Once we know fi we
know the composition of the mixture of the solution we can find out fi bar. So therefore this
particular rule for this particular law this was first suggested by Lewis M Randall and
therefore it is known as the Lewis Randall model for ideal solutions.

Now here it is very important to note that it is less restrictive as compared to each of ideal
gases, right? What does it assume? It is applicable when is, just try to understand it is
applicable for a situation where the intermolecular attractions between unlike molecules are
similar to the intermolecular attraction between like molecules. For example suppose I have a
binary mixture most of my discussions will be in terms of binary mixture just because it is
easier to conceive that and so that for a binary mixture there are say 2 components,
component a and component b the Lewis Randall model will be applicable when the
interactions between aa molecules and the interaction between bb molecules and the
interaction between ab molecules are similar in nature, right?

663
(Refer Slide Time: 10:09)

So when does this happen? This can happen only for conditions this can happen under what
conditions the first condition is when the physical properties of the components they are
similar this is number 1, number 2 is when the chemical components they are also similar this
is something very important and we know that, see whenever any component is in excess
what happens? Suppose again a mixture of a and b they were large excess of a there we add
some amount of a.

So therefore a feels at home in the environment of a there is very less b. So therefore

whenever we find that at any particular pressure if the component is in large excess it obeys
the Lewis Randall rule. We know that at low pressure gases behave ideally so they normally
they obeyed the Amagat’s law of additive volume. So for gases at low-pressure it is
applicable usually it is valid for all gas mixtures and we find that at any pressure if the
component is in large excess it obeys the Lewis Randall rule. This is evident from this
particular figure where I have plotted the vapor mole fraction as function of the liquid mole
fraction.

What do I find? We find that at, when the mole fraction of this particular component it’s close
to 1 we find that more or less, the Lewis Randall rule it is valid under this particular
condition. For this particular case I can write down the fi bar this is equal to yi fi but let me
remind you that this is done for a for a very low pressure. Now for low pressures what
happens? For low pressures we know that fi it is equal to the total pressure of the component
in the vapor phase, right?

664
So therefore if it is a vapor it is a pure component vapor then we know that this particular fiv
is nothing but equal to the total pressure of the system. What about the liquid composition? In
the liquid phase we have this fiL this will be equal to xi fi L. Now if you recall at low to
moderate pressure what happens to this particular fiL this becomes when the poynting
correction factor and the correction factor due to non-ideality of the of the gas, so the vapor
phase is there it can be neglected this reduces to pi saturated, right? So therefore from here
what do we get? Under equilibrium conditions we know fiL equals to fiv so for low to
moderate pressures what do I get?

The equation becomes something like xiPi saturated equals to yi into P and I guess this
equation, all of you already know you are aware of it you have been doing this from your
school days this is nothing but the Raoult's law, right? So therefore under what conditions are
these applicable? I repeat at low pressure for gases we behave ideally this is usually valid for
gaseous mixtures it is valid for any pressure if the component is in large excess when we find
that the component is in large excess then the linearity is maintained in this particular
condition.

(Refer Slide Time: 13:50)

And we also notice for this particular condition we find that more or less the can write it
down as your fi bar it is proportional to the composition of the mixture and we know that for
when the when the component is in large excess for this particular portion more or less this
particular proportionality constant this is equals to the fugacity of the pure component
situation.

665
(Refer Slide Time: 14:17)

Now before I proceed I would just like to bring your attention to another particular aspect of
this graph, what do we observe? We observe that the graph it exhibits some sort of a
deviation from Lewis Randall rule which we will be discussing later this is an, this shows that
it exhibits non-ideal behaviour over a large range of composition we are not going to deal
with it at the moment. At the moment we will just be dealing with the ideality portion. We
find that it is ideal under this particular condition.

(Refer Slide Time: 14:54)

Here the relationship is linear and we know that this particular relation holds under that
condition, right?

666
So therefore for this particular condition we can write fi bar is proportional to xi, fine. Now
let us notice the other extreme when the when the component is in very less amount. Just
observe this portion we find that in this case the slope or the nature of the curve is completely
different as compared to this case but if you slightly extrapolate this thing you will find that
and if you keep on extrapolating this we find that in, at very low concentration that that when
the concentration is almost equal to 0 under that condition also there is the linearity part
holds.

(Refer Slide Time: 15:38)

Or in other words this equation this equation holds under that case but for very low
concentrations the linear the proportionality constant is not equal to the pure component

667
fugacity at the temperature and pressure of the mixture. So therefore we know that for high
concentration of component for high concentration of component f the partial molar fugacity
is proportional to the mole fraction or fi bar equals to ki xi this ki is nothing but equal to the
pure component fugacity.

We also know that for very low concentration of the component means a xi is almost equal to
0 under that condition also we find fi bar is proportional to xi but in this particular case or in
other words fi bar equals to ki xi but in this particular case this ki is not equal to fi. On the
other hand this is some other proportionality constant which is known as the Henry’s constant
or it gives the Henry’s law for in infinitely dilute solution.

(Refer Slide Time: 17:01)

So therefore we have the ideal solution which gives you and the ideal dilute solution for both
the cases what do we have? fi bar is proportional to xi but for ideal solution or in other words
fi bar equals to ki xi but for ideal solution ki equals to the pure component fugacity. For ideal
dilute solution it is the Henry’s constant of component i in the in the solvent. So this is
something very important which I wanted to show you.

668
(Refer Slide Time: 17:42)

Well, what about the other thing? We found out then it is usually valid for gaseous mixtures
as I have mentioned. It is rarely valid for liquid solutions because remember one thing what is
the basis of the Lewis Randall rule the Amagat’s law of additive volume. What is the basis of
Amagat’s law of additive volume? It assumes that the interactions between the molecules are
same when it is in the pure state as well as in the solution this is quite applicable for gaseous
mixtures where in any case the interactions are very weak. So therefore for gaseous mixtures
very safely you can assume the Lewis Randall rule even if the gases are not ideal remember
the basis is ideal mixture of real gases it is not an ideal mixture of ideal gases.

But for liquid solutions where the interactions are very important, for very few situations you
will get that interactions between like molecules is similar to the interaction between the
unlike molecules or in other words interactions between aa molecules and bb molecules will
be different from the interaction between ab molecules for most of the cases of a binary
solution of a and b, right?

669
(Refer Slide Time: 19:08)

So therefore in this particular case we find that in liquid solutions Lewis Randall rule it is
applicable when you find that the component molecules are similar in size and in chemical
nature. For example isomers, adjacent members of a Homologues series for example benzene
toluene mixture or heptane hexane for such mixtures it is applicable and also for isotopically
substituted compounds.

(Refer Slide Time: 19:21)

For example suppose you have C6H6 H6D6 deuterium, so for such types of things only the
ideal solution model is applicable for most of the cases it is not applicable but we need to
remember just like ideal gases ideal gas equation it is not applicable for most of the cases yet

670
we have it and when the gas deviates from ideality we have tried to incorporate some sort of a
correction factor in order to incorporate the deviation from ideality.

We had incorporated the compressibility factor the normal ideal gas equation for one mole of
the gas was this and for a real gas we suggested equation in this form where z is a correction
to the to the ideal gas equation to take into account the real gas behaviour. In this particular
case also knowing fully well that it may not be applicable for liquid solutions we consider the
ideal solution model we are going to discuss the characteristics of ideal solution model and
once we have understood ideal solution model we would try to see what are the corrections
and in what way the corrections can be incorporated into the ideal solution model to account
for the behaviour or to account for the non-ideal behaviour, right?

(Refer Slide Time: 20:45)

So therefore let us discuss the properties of the ideal solutions before we proceed. So
therefore what are the different ideal properties of ideal?

671
(Refer Slide Time: 21:05)

There is one interesting thing which I would like to say well, this is something very
interesting that usually we feel that when the gas is at higher pressure where it exhibits liquid
like density the fluids they tend to mix with little or no change in volume there is quite
evident. So we are all always tempted to think that therefore when the gas is at high pressure
then it should be opening Lewis Randall rule and therefore we can apply Lewis Randall rule
for gases at high pressure.

Although at the first thought it seems to be probable or seems to be correct, at second thought
we are reminded that this is not always correct, why? If we observe this graph what do we see
this is a compressibility factor chart for Nitrogen Butane mixture at 171 degree centigrade,
what do we find? That at 690bar more or less we find that it is a linear function. So therefore
we are tempted to think that well at 690 bar we can use the Lewis rule because under is
particular condition vi is linear function of mole fraction.

But we have we need to remember that the v is inside the integral. If you remember the
equation is R T d ln fi by yi P which is equal to 0 to P vi bar minus RT by P dp this particular
vi bar it is inside the integral sign, right? So therefore it is important that vi bar or the
Amagat’s law should not only be obeyed at 690 bar it should be obeyed from 0 pressure to
690 bar pressure. At 690 bar the Amagat’s law is valid, right?

So therefore you can you might think at 690 bar we can use it but if you slightly come to
lower and lower pressures you find that below 345 bar the linear variation or the additive
property of volume is no longer applicable in this particular case. So therefore in this

672
particular zone if you find in this particular zone we will find that the additive the additive
property it fails completely but at the same time we need to remember that we want the
additive property to be valid right from 0 pressure to 690 bar pressure.

So therefore just being applicable or just obeying the Amagat’s law of additive volume at any
pressure does not guarantee that that we can use the Lewis fugacity rule at that pressure
because for using the Lewis fugacity rule we need to consider this particular integral where
the volume is inside the integral sign and therefore the law of additive volume is required to
be followed not only at the pressure of interest but over the entire range of pressure from 0 to
the P. So therefore this has to be considered with caution.

(Refer Slide Time: 24:05)

Well, so now let us come to the property of, what are the properties of ideal solutions? Let us
see we have already assumed that it the mixture it satisfied the Amagat’s law of additive
volumes.

673
(Refer Slide Time: 24:15)

What is that particular law? It is vi bar equals to vi, fine. What is the Lewis fugacity rule? It is
fi bar equals to Xi say fi, fine.

(Refer Slide Time: 24:42)

Now if you recall in the last class we had or maybe in few classes back we had tried to we
had we had found out the variation of fugacity with pressure and the variation of fugacity
with temperature.

674
(Refer Slide Time: 24:47)

There what did we found?

We found out that del ln fi del
P at constant T and total
mixture this would be equal
to vi bar by RT is same way
from using this equation
we can write down del ln fi
del P at constant T, n
this is nothing but
equals to vi by RT, so we know that if vi bar equals to vi from where we got the equation. So
therefore the first property is that ideal solutions are formed with no change in volume. So
therefore there is no change in volume when ideal solutions are formed, right? What are the
other properties? We will be discussing the other properties in the next class?

675
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 37
Ideal solutions(Contd.)
(Refer Slide Time: 0:27)

Well, to come back to the properties of ideal solutions so therefore the first property we had
already discussed in the last class where I told you that the ideal solutions are formed with no
change in volume, right?

(Refer Slide Time: 0:36)

Let us see what are the other properties? They will all come from the properties of fugacities
that we have dealt with. I had already mentioned to you how fugacity changes with changes

676
with temperature, right? So therefore in that particular case suppose we write it down what do
we get del ln fi by del T at constant P this is nothing but equals to hi bar minus hi0 by RT
square and what about this term also? If you refer to this is hi0 minus hi by RT square, right?

(Refer Slide Time: 1:20)

So therefore from this particular equation what do we get?

(Refer Slide Time: 1:32)

We get this, now from here what do we get? We find out that the partial molar enthalpy of
component i is equals to the molar enthalpy of component i in other words which implies that
the components mixed with no absorption or evolution of heat, they mix with no absorption
or evolution of heat, right?

677
And that is quite expected, why? Because I have already mentioned that this is applicable
when the interactions between unlike and like molecules are the same it automatically implies
that when they are mixing there is no extra energy added or no extra energy evolved during
this particular process. So therefore if this is correct then what about the molar enthalpy of
mixing, what is the molar enthalpy of mixing? The molar enthalpy of the components after
mixing minus the molar enthalpy of the component before mixing.

After mixing what is the molar enthalpy of the components this is Xihi bar Sigma and what
were the molar enthalpy of the component? What was the enthalpy of the total enthalpy of the
component when they were kept separate to each other that was Xihi the total proportion or
rather for one mole of the mixture when the components were kept separately for each
particular component this was the molar enthalpy taken together and after they mix the molar
enthalpy changed to partial molar enthalpy so therefore the enthalpy of molar enthalpy
change of mixing is given by this particular equation which can otherwise be written down as
Xihi bar minus hi we know that these are equal so therefore the molar enthalpy of mixing is 0
in the same way just like we can also write down the molar sorry the molar volume change of
mixing is also equal to 0.

(Refer Slide Time: 3:33)

Now what do we know? We know h equals to u plus pv or in other words we know u equals
to h minus pv or Delta u equals to delta h minus P delta v the whole thing is being carried out
at constant pressure. So therefore in this particular case suppose I take that before mixing the
or rather the if I take the molar internal energy change of mixing this should be equal to

678
Delta h mixing minus P delta v mixing already we have proved this equals to 0 this equals to
0 which automatically implies that delta um equals to 0.

(Refer Slide Time: 4:11)

You can just notice that I have mentioned all these equations. So therefore we find, from here
what do we find? We find that when ideal solutions are formed they are formed with no
change in volume they are formed with no evolution and absorption of heat and no change in
the internal energy of the solution quite naturally. What about the Gibbs free energy change
of mixing? Does that also disappear or there is some Gibbs free energy change of mixing
when the 2 components are mixing even when they form an ideal solution.

679
(Refer Slide Time: 4:57)

Let us see what is Delta gm equals to this Delta gm this should be equal to suppose this is
again suppose I take up the capital Gm, right? This will be equal to your G of pure liquids
before mixing minus G of liquids in solution, right?

Now what is the G of pure liquids let us see? For the time being let us consider a binary
mixture two components solution, okay. So therefore before mixing what was this? This was
nA muA of pure liquid, so therefore I do not put anything just in case you get but for after
this I am not going to write pure liquid plus nB muB, fine. What is G mixture after this A and
B have been mixed? This will be nA muA again I would like to remind you that mu also has
the same property of fugacity. So they are also shown by a cap here. So therefore in this
particular case since I have written down G pure liquids I have written down G pure mixture.
So therefore from here I should be in a position to find out Delta G mixture.

680
(Refer Slide Time: 6:24)

Just subtracting what is Delta G mixture in this particular case you tell me? This is nothing
but equal to nA muA bar minus muA plus nB muB bar minus muB, fine. Now you tell me
how are this muA muA bar related to one another.

Is MuA bar not equal to muA plus RT ln XA, isn’t it? Does this relationship hold or not?
Again I would like to remind you from where did this relationship come, it came from the
Gibbs Duhem equation, what was the Gibbs Duhem equation? What did I find out? If you
remember xi dmu i equals to 0 this was Gibbs Duhem equation, right? So therefore in this
particular case suppose I would like to integrate it from your Xi or in other words this is X1 d
mu1 plus X2 dmu2 equals to 0. So therefore this comprises of component 1 this comprises of
component 2.

Now suppose I integrate it from X1 equals to say one to that is the pure component to X1
equals to X1, suppose I perform that particular integration then we will find that this gives
us your muA minus muA bar which is nothing but equals to RT ln XA. Same way we can
write down muB is nothing but equal to muB plus RT ln XB, right? So therefore from here
what do we know? We know that G mix minus G pure liquids which is nothing but equal to
Delta GM this is nothing but equal to nA your RT ln XA plus nB RT ln XB.

So therefore what is the molar Gibbs free energy change of mixing this is nothing but XA RT
ln XA plus XB RT ln XB. So therefore for N number of components, we can simply write
Delta gm it is Sigma XI ln Xi RT i equals to 1 to N. If you recollect this is similar to the

681
molar Gibbs free energy change of mixing for ideal gases as well, right? So we find that
Delta gm is not equal to 0 if the components form an ideal solution.

So therefore if Delta gm is this, what about your Delta sm? The entropy change of mixing,
we know Delta sm should be equal to Delta gm minus it is by T minus delta h by T, isn’t it?
We know g equals to h minus Ts so therefore Ts is nothing but equals to h minus g or in other
words s equals to this. So this is going to be plus this is going to be minus, fine. We know this
term your Delta hm 0 is 0. So therefore what is Delta sm? It is nothing but minus of Delta gm
by T which gives you as in this particular case this gives you minus R sigma Xi ln Xi, right?
And in this case I would like to tell you that since Xi is less than one Delta gm will always be
negative if the solution has to be formed and again I would like to remind that we are dealing
with solutions which are miscible in all proportions and again we know that Delta sm is
greater than 0 for a solution to be formed these are with in line with what we have discussed
till now.

(Refer Slide Time: 10:30)

So therefore I have jotted down all the properties that I have mentioned and these are the
properties of ideal solutions we find that the ideal solutions they are accompanied by no
volume change of mixing, no absorption or evolution of heat, no change in internal energy
but with an increase in the randomness of the solution and just because there is an increase in
randomness of the solution that is why this the 2 components they mix spontaneously to
form an ideal solution and it is this particular entropic effect which gives you a negative
Gibbs free energy change of mixing due to which the solution mixes it is not due to the
enthalpic effect it is due to the entropic effect.

682
Now therefore so suppose I give you a solution and I tell you to find out whether it is ideal or
not we can definitely take we can measure the 2 volumes of the component that we are
taking we mix them up and then we measure the volume of the solution from we find that
there is no change in the volume of this solution we can assume that the solution is ideal but
in this case there is something that you need to remember there are certain instances where
we find that under specific composition the solution might behave ideally that in no way
implies that the solution exhibits ideal behaviour over the entire range of composition.

(Refer Slide Time: 12:24)

For example suppose say we find that, we take up a solution of say your benzene and ethyl
alcohol, for this particular solution suppose we take this particular solution and we try to plot
the molar volume as a function of say benzene, right? Where A is benzene, we find that for
from 0 to one if we take the we more or less find that the graph it varies in this particular
form we find that initially there is a increase in the the volume of the solution on mixing and
then there is a decrease of the volume of solution on mixing and at one particular XA we find
that the volume of mixing gives us 0.

Or in other words at one particular XA we find that there is no volume change on mixing the
volume initially and after mixing are the same. So suppose we take up these 2 components
and we try to see the volume change on mixing at this particular composition then
automatically we think that the solution is ideal but it is not ideal it deviates from the Lewis
fugacity rule but if you are testing it under one composition you might get P or you might get
the wrong result.

683
Same thing can be applicable for enthalpy change as well and so therefore it is very important
that when you are testing for ideality of a solution you perform the experiment over a large
range of compositions, so that if it is valid over a large range of compositions then it is ideal
or in other words the safest test for finding out is, if it is at low to moderate pressure then I
have already mentioned that for low to moderate pressure, what is fi? fi is the fugacity of pure
component i in the at the temperature and pressure of the mixture.

Suppose I am dealing with a liquid solution than for low to moderate pressure the poynting
correction factor is 0 and the non-idealities due to the non-idealities of the gas phase that
particular correction factor is also 0. So therefore fi it reduces to Pi saturated and under that
condition what is this equal to? This is Xi into P or in other words, this is the partial pressure,
isn’t it? So for low to moderate pressure this reduces to Pi equals to Xi Pi saturated which as
you all know this is nothing but the Raoults law, right?

So the safest thing is if you plot the partial pressure as a function of molar volume over the
entire range of pressure and you see that you get a linear curve in this particular fashion then
you know that the solution obeys ideal solution model. So therefore an ideal or a very safe
test to test the ideality of solutions is not to test the volume change of mixing or the enthalpy
change of mixing at very few mole fractions.

We can do it over a large range of mole fractions or else the safest thing is just find out the
partial pressure and plot the partial pressure and you find the linear plot. In other words we
will find that in fact suppose there are 2 components maybe component 1 and component 2
and then suppose we plot the partial pressure of both the components here. So for X1 what do
we have? We find that when X1 equals to 1 then the partial pressure is equal to the pressure
of the component; this is equals to P1 saturated.

And when X1 equals to 0 this is going to be P2 saturated, right? Now usually one is our more
volatile component so therefore at constant temperature if I try to find out the X1 at if I keep
on varying X1 and I try to find plot rather the partial pressure then I get a curve of something
of this sort. For component 2 the curve will be something of this sort and therefore the total
pressure in this particular case it varies from P2 saturated to P1 saturated in a linear fashion.

So the simplest way to do is what? You find out you mix 2 components in different
proportions find out the total pressure over the mixture plot all the points and if you find that
all the points more or less fall in a straight line which varies from P2 saturated at one end to

684
P1 saturated at other end then we know that the solution is ideal this is a sure test of ideal
solution.

Well, so once we have understood an ideal solution and we have understood under what
conditions ideal solutions are applicable? What is the next thing that we should do? We
should find out ultimately we are dealing with phase equilibrium thermodynamics, so what
we should do? We should find out the phase equilibrium relationships for ideal solutions in
the vapor and the liquid phase. What do I mean? I mean suppose I have got the, a simplest
solution a 2 component solution component a and component b say and they are there, there
is a liquid solution in equilibrium with its vapor phase. So therefore component 1 and 2 they
are present both in the liquid phase as well as in the vapor phase.

(Refer Slide Time: 18:05)

Now for the liquid phase suppose I say that in a liquid phase the solution is ideal then in that
case I know fi liquid this is Xifi in the, I say that the vapor phase also exhibits the properties
of an ideal solution or ideal mixture. So in the vapor phase what can I write? yifi this is L
remember this is v. Again I remind the phases are given as subscripts the component numbers
as superscripts, component number as subscripts and these are the partial molar properties
these are the pure component properties, right?

So therefore in the liquid phase this is the equation, in the vapor phase this is the equation. If
I assume that the vapor and the liquid phases are in equilibrium the vapor and liquid phases
are in equilibrium, when is it possible? We know we have already framed the conditions of

685
equilibrium for the 2phases in terms of fugacity we have already done. The fugacity of each
component must be same in all the phases or in other words this should apply.

If we write it down in terms of pure components this particular function should apply at low
to moderate pressures I have already mentioned at low to moderate pressure what happens? fi
v equals to total pressure over the mixture, fi L is equal to the saturated vapor pressure of the
pure component at the temperature of the mixture, so it is Pi saturated at the temperature of
the mixture. So on substituting these, what do I get? I get at low to moderate pressure the
equation for phase equilibrium reduces to this. This is the most important equation that we
use for finding out the phase equilibrium problems and this along with the additional
constraints that Sigma Xi should be equal to 1 and Sigma yi equals to 1 this whole set gives
us the different phase equilibrium problems that we come across for 2 component systems of
a 2 phase mixture, right?

(Refer Slide Time: 20:25)

So in this particular case therefore if it's2 component then what are the, what is F? What is
the minimum number of, what is the degree of freedom in this particular case?

686
(Refer Slide Time: 20:44)

We know F equals to C minus P plus 2 this is 2 this is 2 so therefore F is also equal to 2,

right? So therefore what are the total number of variables the temperature of the solution, the
total pressure, the mole fractions in the liquid phase, mole fractions in the vapor phase.

(Refer Slide Time: 21:16)

And we know that x1 and x2 they are related by X2 equals to 1 minus X1 these 2 are related
by y2 equals to 1 minus y2 these corresponds to this particular condition, right?

687
(Refer Slide Time: 21:25)

So therefore these are related so therefore basically we have four variables out of these we
know for a 2component system just 2 variables can be varied that will the other 2
automatically gets fixed up.

(Refer Slide Time: 21:48)

So what are the things that we can vary the different things which we can vary are, we can
vary T and x. We can we have T and x we can vary P and y, we have T and y we can vary P
and x and so on and so forth. So there therefore these are the 4 sets of things that we can vary
and these are the 4 ways that we can vary the 2 variables and accordingly if we vary the 2 the
other 2 has to be calculated accordingly. So before we do a number of problems the next
thing which I would like to discuss are some phase diagrams for binary mixtures.

688
(Refer Slide Time: 22:26)

Well, before that it is important for you to note that for a binary mixture F was equal to 2. For
n components n components what it becomes? F equals to C minus P plus 2 this is N this is 2
therefore F equals to N. So therefore for N number of components we find that the degrees of
freedom is N and out of here we have T, P x1 to xN again bounded by the constraint of Sigma
Xi equals to 1. So therefore X1 to XN minus 1 if I know then automatically I know XN and
y1 to yN minus 1 which will automatically give me yN. So this is my range of variation out
of which I can very just N number of components it can be the total N minus 1 composition’s
temperature, N minus 1 composition pressure or N minus 1 composition is in the vapor phase
temperature, N minus 1 composition vapor phase at pressure.

So any of these permutation combinations are done usually they are done either at constant
temperature or at constant pressure and accordingly we generate either isothermal xy data we
will be doing a few problems or maybe isobaric xy data, right? And so therefore in the next
class what we are going to do is, in the next class we are going to discuss the different phase
diagrams primarily at constant temperature and constant pressure.

689
(Refer Slide Time: 24:19)

At constant pressure temperature we generate the Pxy plot and at constant pressure we
generate the Txy plots, we will be discussing these in the next class.

690
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 38
Ideal solutions(Contd.)
(Refer Slide Time: 0:40)

Well, so therefore as I have mentioned in the last class today we are going to discuss the
phase diagrams for 2 component systems in equilibrium with one another in the vapor and the
liquid phases, right? So first let us see about the situation when we do it under constant
temperature, when we do it under constant temperature say we know T and we know X1,
right? So from there we are required to find out P and we are required to find out y1.

So I had already given you the set of equations. So what are the set of equations that were
given? Suppose you know T then you are in a position to find out P1 saturated P2 saturated,
how you are supposed to find out P1 saturated and P2 saturated? Remember it was either the
Clausius Clapeyron equation or the Antoine equation. Usually we use the Antoine’s equation.
What is the Antoine’s equation?

Just to recollect this is equals to A minus B by t plus C. A, B, C they are constants

characteristics of that particular compound. We will be taking up the specific example, say for
example we take an example of say Pentane and Heptane we know that these are adjacent
members of the homologous series. So therefore they form an ideal solution, right? And for
this particular mixture I would we would like to generate a PXY diagram.

691
So the Antoine constants for Pentane they are given I think the problem is there at the end we
or I will just give it to you. For pentane suppose it is pentane, for pentane A, B, C in case of
pentane it is 6. 8 7 632 and B is 1075.78 and C is 233.205. For Heptane we find A is 6.89386
this (B) is 1264.37 this (C) is 216.64, right? So I know for pentane what are the Antoine
constants? So from these Antoine constants I can find out P.

Suppose pentane is my component 1 say, right? This is component 1 this is component 2

because otherwise I have to write P pentane every time, so this is P1 saturated. Just calculate
it out and see what you are going to get this is more or less just check it, it is possibly 2
21929.6 torr, In the same way for Heptane we can find out P2 saturated, what is it? 304torr,
right the moment you know P1 saturated P2 saturated.

From Raoult's law what do you know? You know the partial pressure P1 is nothing but X1 P1
saturated assume a X1 you know P1 saturated you can find out P1, right? You know P2 it and
find out 1 minus X1 P2 saturated, right? What else do you know? You know the total pressure
P is equals to P1 plus P2. So therefore you can calculate it out say suppose for X1 equals to
0.1 I will just give you the value so you can check it up.

Your X1 equals to .1 you get more or less P equals to 486.56 torr, moment you know what is
P, P equals to P1 plus P2. What is P1? P1 is nothing but equals to y1 P therefore from here
you can find out the mole fraction in the vapor phase it is nothing but equals to P1 by P. So
therefore we find out that for X1 equals to 0.1, y1 usually comes to as 0.4377, fine.

(Refer Slide Time: 4:43)

692
So therefore in this particular way we can keep on assuming different X1 values and we can
we can find out different P values, right? We can do it for example, so and this is 0 this is 0
and this is 304 torr, right?

(Refer Slide Time: 4:46)

Which is the vapor pressure of the, if you not this is the vapor pressure P2 saturated I believe
this is P2 saturated.

(Refer Slide Time: 4:52)

Naturally when X1 equals to 0 then suppose I will just give you a few representative values
for 0.4 it is 0.8236, this is 1034.24 then for 0.8 this is 0.9655 and this is1764.48 again for one
what do you get? This y1 will also be equal to one and (2129.6) this is P1 saturated, right? So

693
therefore in this way we can we can we can generate x, y P-x-y data at constant temperature
and let us plot it. What sort of plot do you expect?

(Refer Slide Time: 5:34)

We find that suppose we plot P versus x, as I have already mentioned what is P equals to? It
is P1 plus P2 we are just dealing with the same 2 or 3 equations and we are trying to do this.
What is this? X1 P1 saturated plus X2 P2 saturated or in other word this is X1 P1 saturated
plus 1 minus X1 P2 saturated or in other words this is equals to P1 saturated minus P2
saturated into X1 plus P2 saturated, what does this equation suggest?

This equation suggest that when you are plotting P versus X1 it should be a linear curve the
linear curve should have P2 it starts from P2 saturated. Here we have already discussed P1
saturated here always remember that when we are talking about X1 we are always referring
to the more volatile component this is something very important. So when we refer to the
more volatile component naturally P1 saturated at constant temperature is higher than P2
saturated this happens at constant temperature.

So therefore we find P it varies from here to here and the variation of P versus X1 it has to be
linear, right? And also if you find out the variation of P1, P1 is nothing but X1 P1 saturated.
So therefore naturally if you are varying then in that case also we know that P1 versus X1
should be linear and at X1 equals to 0, P1 equals to 0 at X1 equals to 1, P1 equals to P1
saturated. So therefore this is the variation this is P versus X1 this is the variation of P1
versus X1.

694
What about P2? P2 we know this is 1 minus X1, P2 saturated. So for X1 equals to 0, P2
equals to P2 saturated. For X1 equals to 1, P2 equals to 0, right? So therefore this curve
should be something of this sort, so this is P2 versus X1, fine. So therefore if you are plotting
we find out that the partial pressure of component 1 the partial pressure of component 2 and
the total pressure all these 3 they are linear functions of the mole fraction of the more volatile
component there also linear functions of the mole fraction of the less volatile component.

If we would have taken the less volatile component we could have plotted X2 here than just
the total nature of the curve would have been just inverted because this would have then been
something of this sort that is only difference nothing else. So therefore we find that in the
liquid phase the pressure versus mole fraction your curve that is linear.

(Refer Slide Time: 8:54)

What about the vapor phase? In the vapor phase suppose we take then in that case what do we
get? We get y1 this is equals to X1P1 saturated by P. So therefore we find that the relationship
between y1 and P that is not linear, right?

695
(Refer Slide Time: 9:06)

So therefore we find that while the P versus X1 plot is linear the P versus y1 plot this is not
linear this is something of this sort. So in this graphs now let us see we have plotted P versus
X1 this is P versus X1 plot, this is P versus y1 plot, right? Now at very high pressure beyond
this is a saturated vapor pressure curve beyond the saturation vapor pressure, what do we
have? We have liquid, fine. We have liquid at high pressure this is sub-cooled liquid at high
pressure, right?

So therefore suppose we have a liquid in the liquid region any particular point it can be
denoted by, how many variables? In this particular case again in if you if you start to write C
minus P plus 2, P is one, C is 2 so therefore F will be equals to 3.

696
(Refer Slide Time: 10:07)

So therefore in order to identify any point in the single phase liquid region we need at least 3
variables we need P, we need X1 and in this particular case the temperature is already
constant. Now here if I start expanding the liquid mixture or rather at constant temperature.
So when we start doing it at constant temperature when we are reducing the pressure we are
basically following this vertical line, we keep on expanding we find the liquid keeps on
getting expanded but it remains in the single phase region there is no change in composition.

Just a pressure over the liquid it keeps on getting lesser and lesser this continues till the liquid
strikes the saturated liquid region when it strikes the saturated liquid region, what happens?
The first vapor bubble appears and what is the composition of the vapor bubble? The vapor
bubble is in equilibrium with the liquid phase. If it is in equilibrium with the liquid phase then
it lies at constant temperature and pressure both the vapor and liquid are at constant
temperature and pressure.

697
(Refer Slide Time: 11:33)

So naturally the vapor phase should have a composition on this particular horizontal line
where it strikes the P versus y1 curve. So therefore the first vapor bubble which appears that
will have a composition which lies on the horizontal line drawn from X1 where it cuts P
versus y1 curve, agreed? So therefore and then again we keep on continuing our expansion
process, when we start we keep on continuing the expansion process what do we find?

We find that from the single phase liquid region we have entered the 2 phase liquid vapor
region this particular region which is bounded by the saturated liquid curve and the saturated
vapor curve this is the zone of heterogeneous region where vapor and liquid both the mixture
they exist, right? And remember one thing we are just going down a vertical line therefore the
overall composition of the mixture is constant but this mixture after it crosses the saturated
liquid line it separates out into a vapor phase and the liquid phase.

698
(Refer Slide Time: 12:39)

The composition of the liquid phase is given by the horizontal line from this point on the
saturated liquid line and the corresponding vapor phase composition is given by the
horizontal line extended to the saturated vapor curve, right?

So therefore we find that already overall composition it keeps on remaining constant as we go

down and down but we find that the vapor composition and the liquid composition they keep
on changing, right? This continues till we strike the saturated vapor curve, moment we strike
the saturated vapor curve we find that the more or less the last drop of liquid has vaporized
and then we enter the gas the single phase vapor region and again my composition in the
vapor phase is the same as that of X1 this case I denote it with y1, right?

Though I had started with a composition of X1 in the liquid region I perform isothermal
expansion and strike the saturated liquid region saturated liquid curve and after that I enter
into the liquid vapor region it is important to remember that in the liquid vapor region the
overall composition of the mixture remains constant quite naturally from conservation of
mass but the proportion of liquid and vapor they keep on changing.

We find that the proportion of liquid and vapor it keeps on changing and while the proportion
of liquid decreases the proportion of vapor it keeps on increasing, right? This continues while
we are in the heterogeneous liquid vapor region and then once we cross the liquid vapor
region we have entered the vapor region with overall composition as y. Now here I would
like to mention that suppose I have started with one other point say X2 I would have
continued in this particular point.

699
Here also when it strikes the saturated liquid region the vapor composition the composition of
the first bubble which appears would have been given by this particular straight length, okay
where it cuts the saturated vapor region. So therefore what do we find? We find that this
entire saturated vapor curve it basically gives us the composition of the first bubble which
appears when the saturated liquid is or it first converts to the vapor phase. So therefore this
particular curve therefore is known as the bubble point curve.

In the same way suppose we would have started from here and we would have started
compressing the vapor phase we will find that when the compressed vapor phase reaches the
saturated vapor region the first drop of liquid the first dew which appears that would have a
composition which would be given by the horizontal line from here when it crosses touches
the saturated liquid region, right? So therefore we find that on saturated vapor curve we find
that they all of them they give the composition of the first dew which points and whereas this
gives us the composition of the first bubble which forms therefore the top one is known as the
bubble Point curve the lower the saturated vapor curve is known as the dew point curve,
right?

(Refer Slide Time: 16:33)

And this particular horizontal line which we find which connects the liquid and vapor
compositions in equilibrium these lines are known as tie lines, fine. So this was the Pxy
diagram which we were discussing. Now very frequently we know that it is easier to maintain
a constant pressure as compared to a constant temperature so therefore very frequently we are
interested to generate Txy diagram as compared to Pxy diagrams.

700
(Refer Slide Time: 17:30)

So let us see at constant pressure, suppose we would like to generate a Txy diagram, how
would it look like in this particular case?

(Refer Slide Time: 17:43)

Now remember one thing for this particular case again let us start from the let us state
previous problem, right? The previous problem was we were we were trying to find out we
were trying to generate the Pxy curve at 70 degrees centigrade at that condition only I had
given you P1 saturated etc.

The next problem suppose we try to generate the Txy curve at say P equals 101.325kpa or
760torq, fine. So in this particular case what do we know? We know the total pressure. Now

701
corresponding to this pressure can we not find out the boiling point of component 1 which is
nothing but pentane and the boiling point of component 2 which is nothing but Heptane,
right? Now moment we can find out t1 saturated and t2 saturated, how do we find out?

Same Antoine’s equation in the Antoine’s equation we know that log P saturated equals to A
minus A minus B by t plus c I have already written so you know this because this is the total
pressure at this pressure you would like to find out the boiling point so you can very easily
find this out. In this particular case you will find that the boiling point of pentane it is 36.04
degrees centigrade and that for Heptane it is 98.43 degrees centigrade, right?

Now it is very evident that the 2 phase region has to lie between these 2 temperatures, isn’t it?
So it has to lie between 36 to 98.43 degrees centigrade. So what we do? We choose temp any
particular temperature between these 2. For that particular temperature say it is 36.4 it is
greater and this is 98.43 choose the temperature here and then for that temperature we find
out P1 saturated we find out P2 saturated once we can find out we choose one particular X1
and then we know one minus X1, from here what do we do?

We can find out the total pressure and then from there we can from total pressure we can find
out y1, isn’t it? Same way we can find out 1 minus y1 so this is the sequence in which the
calculations proceed once we can do it then for each particular temperature that we have
selected we can find out one particular x1 we can find out one particular y1 and where x1 it
varies from 0 to 1 and y1 also corresponding y1 also varies from 0 to 1.

(Refer Slide Time: 20:32)

702
So therefore we can plot a txy plot also in this particular way where on the x axis we are we
plot either X1 or y1 and we plot temperature here. This particular in this particular situation
where X1 equals to 0 this should correspond to the boiling point of phase 2. Since phase 2 is
the less volatile component so its boiling point will be higher so this is t say boiling point of
this is component 1 component 2, right? And this should extend to the t boiling point of
component 1 which is naturally it is more volatile it should be less, right?

(Refer Slide Time: 21:19)

And then and in this particular case we find that the relationship it is not linear the
relationship is going to be non-linear because t it is related to X1 from X1 we P1 saturated
from there we get X1. So therefore in this particular case the relation between t versus x1 or t
versus y1 it is not linear.

703
(Refer Slide Time: 21:38)

So more or less the type of curve which we get it is something of this sort, this is the t versus
x1 curve this is the t versus y1 curve.

(Refer Slide Time: 21:47)

I have shown this in this particular figure. Here also we proceed in the same way, at very low
temperature what do we have? We have only the liquid region, suppose we have liquid at this
particular composition X1 this is at a at some particular temperature and pressure in this sub-
cooled liquid region we know that we have 3 variables we can vary the composition
temperature pressure everything independent and for this particular case this has been drawn
for a constant pressure conditions. So therefore we can vary t versus x1 and locate our any
particular point in the sub-cooled liquid region.

704
From here what do I do? I start heating the liquid mixture at constant pressure, when I start
heating at constant pressure what do I do? I am basically following this particular vertical line
I follow this particular vertical line and then I touch this particular point in the saturated
liquid curve, moment I touch this point what happens? The first bubble appears what is the
composition of the bubble? Again it will be given by the horizontal tie line connecting this
point to the saturated vapor curve. So this is the composition of the first bubble.

Suppose I would have started from this particular point again I would have isobarically
heated it when the first bubble appears the composition would have been given by this
particular point. So therefore what do I find we find that the this particular saturated liquid
curve it is the locus of all the points where the first bubble appears or in other words this is
known as the bubble Point curve as has been mentioned here, agreed?

Now once we cross the bubble Point, remember one more thing which I forgot to mention for
pure components we have a boiling point, at the boiling point the liquid changes to the vapor
phase the vapor condenses back to the liquid phase. For a true component mixture we do not
have a definite boiling point of the mixture exceed point under a very specific condition
which we will be discussing later on while for usually we find that the point from where the
liquid starts vaporizing and the point from where the vapor starts condensing they are not the
same point.

The vapor starts condensing when it when it first strikes the dew point curve the liquid starts
vaporizing when it first touches the bubble Point curve and usually there is a good amount of
difference as it is as you can see here between the dew point curve and the bubble Point curve
the greater will be the difference when the boiling points between the 2 your component’s
will be more. Now this is very important while a pure liquid is characterized by a definite
boiling point which is also the condensed the condensing point also but in this case the
bubble Point and the dew point are not the same they are usually different.

For ideal solutions very interestingly when we go to an extreme state of ideality we find that
these 2 points they coincide. So this is this is an ironical fact that more we move from ideality
we find that that dew point curve or rather the dew point and the bubble Point of the 2
mixture they come closer and they coincide at one particular point which I can mention now
but we will be discussing it later which is the point of isotropy where the liquid solution
exhibit extreme deviation from ideality, right?

705
So therefore in this particular case then what do we get? Just in the same way we find that the
saturated liquid line is the bubble Point curve, fine. Then now once we cross the bubble Point
curve we have entered the liquid vapor region this is a heterogeneous liquid plus vapor
region. Now when we are crossing this particular region we find that as we go to higher and
higher temperature we find the amount of liquid that keeps on decreasing and the amount of
vapor it keeps on increasing this keeps on happening, okay.

And the thing is, as when we are inside this liquid vapor region what do we find? The
horizontal line which crosses the required composition and connects the bubble Point and the
dew point curve these 2 extremes they give us the composition of the liquid phase and the
vapor phase in equilibrium. Here we get the composition of the of the vapor phase which is in
equilibrium with the liquid phase, right? So as we enter more and more into this region we
find that the overall composition of the mixture remains same but X1 changes y1 changes and
but at every point X1 is in equilibrium with y1.

And more interestingly we will find suppose we take up the the conservation of mass overall
mass and component balance we find that the proportion of the liquid and vapor changes in
the inverse ratio of the distances of the point from the dew point and the bubble Point curve,
what do I mean? Suppose I take up the overall mass balance I know F the initial feed was
equals to L plus v, F - I write it as ZF, why? Because it can be a liquid or it can be a vapor this
is equals to LXF plus VyF, right?

So from here what do we get? From here I get L by V the proportion L is the moles of liquid
form, V is the moles of vapor form keeping in mind that both of them add up to form the total
volume of the feed which I have introduced. So from this substituting this here we get L by V
is nothing but equals to y minus ZF by ZF minus X. So therefore what do I get? I get 2 things
from the Tie lines I get the equilibrium values of the composition in the liquid phase and the
vapor phase.

I also get the proportion of liquid and vapor which is formed from this particular feed they
are in the inverse proportional to the distance of the Tie lines, what I mean is this particular
distance it gives me the moles of liquid form and this particular distance gives me the moles
of vapor form. So therefore as we move in we find that this distance decreases and this
distance it increases. So therefore if I continue in this particular way what do we find?

706
That finally when we come here we have all vapor and this particular here we have all vapor
and then we have no liquid, okay. And in the same way suppose we would have started from
this particular point and if I would have condensed the vapor then in that case what would
have I got? I would have found out that if I keep on decreasing the temperature of the vapor
under isobaric conditions then we find moment the vapor it touches the saturated vapor curve
the first dew appears and the composition of the dew is given by the Tie line drawn from this
point to this point. So therefore this particular curve accordingly is the saturated vapor curve
is known as the dew point curve, right? And again from here if I keep on reducing the
temperature I enter into the liquid vapor region and in this particular way it continues.

So therefore here what do I find? We find that this that Txy diagram is characterized by a
single phase region, a sub cool liquid region it is below the bubble Point curve it is
characterized by a single phase superheated vapor region beyond the dew point curve and the
liquid vapor region between the bubble Point curve and the dew point curve and we will find
that while we are moving in the 2 phase region we find as we move from the bubble Point
curve to the dew point curve we find the proportion of liquid decreases the proportion of
vapor increases and finally we enter from a sub cooled liquid region to a saturated vapor
region with the same overall composition of the mixture but in that heterogeneous region the
compositions were different.

And using this we can explain the principle of fractional distillation which we will be doing
in the next class, thank you very much.

707
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 39
Ideal solutions(Contd.)
Well, we were we were discussing ideal solutions and in ideal solutions we had I had already
discussed the P-x-y diagrams which is shown here and then I had also discussed the Txy
diagram and these diagrams they are particularly useful for describing the process of
rectification which is distillation a number of times which we use very frequently in chemical
engineering in order to separate 2 particular components with small differences in boiling
point when the difference is very large then it is simply vaporized.

We start vaporizing the mixture and then condense it then in that case we can get a good
enough separation but when the boiling point differences are not very large we need to repeat
this process of boiling and condensation a large number of times.

(Refer Slide Time: 1:13)

Let us see how this particular process works. Say suppose we have started with a liquid of
composition, we start heating we keep on heating it till we reach the bubble point curve, once
reached the bubble point curve when we start vaporizing we will find that the vapor
composition is given here we already find that the vapor is richer in more volatile component
as the liquid phase. Now we condense the whole vapor, what we do?

We simply cool it we do not do anything and then again we start vaporizing, moment we start
vaporizing the first vapor again we find it is richer than the liquid with which we started and

708
it is much more richer than the liquid mixture which we had originally taken. Now suppose
we keep on repeating this we simply condense the whole thing and again we vaporize it,
right? This is the composition of the vapor we simply condense the vapor and again we start
to vaporize it.

So in the process what have we got? In the process we find that we have we and we have
come to a liquid solution which is much richer almost which comprises of the pure
component of the more volatile component where we had started with a mixture which had
much less proportion of the more volatile component. Now in this particular process is the
basic principle underlying the process of rectification which we will be learning in great
details when you go further into your training of chemical engineering?

(Refer Slide Time: 3:08)

So by this time what are the things that we have done? We have dealt with Pxy we have dealt
with Pxy diagrams we have dealt with Txy diagrams and now from this Txy diagrams we
find that locating the Tie lines with we can generate the vapor and the liquid, the liquid and
the vapor in equilibrium with one another and this particular compositions they are then
plotted in the x1 versus y1 curve. Now here I would like to mention one thing again I would
like to remind that what is x1?

x1 refers to the more volatile component, this is something very important that you need to
remember throughout you can perform the entire phase diagrams with the less will tell
component the only difference is this the curves will be just they will just be reversed that is
only difference but by convention we usually perform phase diagrams using the more volatile

709
component for the composition of the more volatile component in the x axis and accordingly
we plot either P in the Y axis or T in the Y axis for Y1 in the Y axis. Now under this
condition we always find that Y1 is always greater than X1 as expected because we are
dealing with a more volatile component, right?

(Refer Slide Time: 4:12)

Now in this particular case we can I can just mention in this case Pi this is equals to Xi Pi
saturated or in other words for 2 common mixtures I can write down the ratio of their partial
pressures as P2 saturated X2 by P1 saturated X1 plot this particular ratio this P2 by P1 this is
known as alpha which we define as the relative volatility. What is this relative volatility? It is
the relative escaping tendency of this second phase with respect to the first phase, right?

So therefore this is the relative escaping tendency of phase 2 compared to phase1, right? So
therefore this is termed as alpha, so therefore we can write this down as alpha into X2 by 1
minus or alpha into X2 by 1 minus X2 we can write it down in this particular term. From
where we can find out that Y this is Y, let it be Y2 it does not matter.

From here we find out that Y1 by 1 minus Y1 this is nothing but x1 by 1 minus x1 P1
saturated by P2 saturated, from here what do we get? We get Y1 equals to alpha X1 by 1 plus
alpha minus 1 x1 from where we know that y1 will always be greater than x1, right?

710
(Refer Slide Time: 6:23)

And therefore we find that we always get a plot which is which is sort of a hyperbolic curve
and for all particular cases except for x1 equals to 0 and x1 equals to 1 y1 will always be
greater than x1. Well so till now what are the things that we have done? We have discussed
ideal solutions, we have discussed the Lewis Randall rule, we have developed the Raoult's
law from the Lewis Randall rule, we have discussed under what conditions ideal solution
model can be applicable, we have also discussed under what are the properties of ideal
solutions no heat absorption or evolution etc . Increase in randomness of mixing quite
naturally otherwise they will not mix spontaneously. Then we have discussed different phase
diagrams for binary ideal solutions the vapor and the liquid phases both are both exhibit ideal
solution model.

(Refer Slide Time: 7:36)

711
(Refer Slide Time: 7:37)

Well, I had also discussed how to how to construct the Pxy, Txy diagrams and in the next few
slides I have just noted down the procedures for finding out the dew point temperature and
also the bubble point temperature etc. So these procedures can be followed and different
problems can be dealt with.

(Refer Slide Time: 7:46)

712
(Refer Slide Time: 7:48)

(Refer Slide Time: 7:51)

The different problems are given at the end of this particular lecture series. So you can try out
these problems one or 2 problems I have already done before you. You can try out the other
problems and in each problem you will find that the answers are also provided for your
benefit. Now there is one particular type of your problem which we have not dealt with till
now which I would like to deal these problems is known as Flash calculation problems, what
are these?

713
(Refer Slide Time: 8:34)

Very frequently we encounter a situation where we find that more or less we introduce any
particular feed it can be vapor feed it can be a liquid feed since we have we just to make
matters general I would like to mention the feed composition as ZF, right? And then it is
introduced into a partial vaporizer or a partial condenser whatever the case may be. From
here what do I get? It flashes I get a vapor stream with the amount of vapor formed it's V
moles the your composition is given by y and pressure and temperature I also get a liquid
stream which again with the total fraction or rather total moles of liquid being L the
composition x and temperature and pressure conditions.

So from here what I have done? I have just flashed any particular liquid or a vapor mixture
for partial separation into a vapor stream and a liquid stream. For this case from the overall
material balance I know this and I also know from the component material balance ZFi equals
to Lxi plus Vyi, right? And with this we also know the equations yiP equals to xiPi saturated
or in other words this is can be written down as Kixi where Ki is nothing but Pi saturated by
P, right?

So therefore I I have these 2 equations and I have got a set of Ki equations as we say it, from
there we can find out xi as ZFi we can just calculate these and find out this is equals to L by F
plus 1 minus L by F into Ki. So in this particular case usually what are known? We know
pressure we know temperature we know they composition of the feed and what are we
required to find out? We are required to find out composition in the liquid phase, composition
in the vapor phase, amount of liquid formed, amount of vapor formed.

714
So how do we proceed in this particular case? We know the temperature once we know the
temperature we can find out each Pi saturated, agreed? Once we know the Pi saturated we
know the pressure then from there should be able to find out Ki, Ki is Pi saturated for P. So
for each case you can find out the Ki once can find out the Ki then what we do? We guess
either a L by F or a V by F whatever it is. If we guess it we can find out xi.

We can find out xi for all the components i, fine. And then once you have calculated xi for all
the components we then check-up Sigma xi equals to 1. If we find it is not equal to one that
again we start with a different assumed temperature continue the process till we get xi equals
to 1 if we get Sigma xi equals to 1 we know that my guess temperature is correct and I have
come or I have reached the correct mole fractions in the liquid phase. Once I have done this
then from there what I do? I find out yi for the individual components.

(Refer Slide Time: 12:20)

So in this particular procedure it has also been written down here. So accordingly we can
perform these particular equations or rather these calculations and we can perform a few
Flash calculations in this way there are few problems which have been given at the end of
the assignment I guess there is one problem of pentane, hexane and Heptane this is the
problem, right? So in this particular problem also suppose we find that these 3 they formed
the ideal condition, okay.

715
(Refer Slide Time: 13:00)

And this is given problem. This particular case the way I have told you the x the
compositions are given ZF’s are given.

(Refer Slide Time: 13:27)

So therefore ZF1 this is equals to 0.2 ZF2 equals to 0.35, ZF3 equals to 0.45 we know that
they Z's are actually axis here because introduce a ideal rather we introduce a liquid solution
here the temperature is given as 95 degree centigrade, the pressure is given as 150kpa.

So what are we required to find out? We are required to find out xi’s and yi’s, right? So what
we do? Again the way I have said T equals to 95 degrees centigrade I calculate the P1
saturated, P2 saturated, P3 saturated in this particular case from the Antoine’s constant. The

716
Antoine constant you can note down the ABC, so for this case n pentane is my component 1
so for this case A is 6.876, B is 1075.78 and C is 233.205.

For n Hexane this is 6.91, 1189.64 and this is 226.28. For n Heptane what do we find? n
Heptane this is 6.894, 1264.37, 216.64 so from here we can find out the P1 saturated, we can
find the Pi saturated we find that this is 3967.8torq this is 1613.5torq this is 686.6torq. We
know the pressure the total pressure it is 150kpa. So therefore we can find out the Ki’s here.
What are the Ki’s? In this case it is 3.5266 and then this is 1.4311, this is 0.6103.

So once we have done it we can find out the Xi’s, right? For finding out the Xi’s we need to
guess L by F, say suppose I start with my first guess of L by F as 0.121 and find out the
corresponding Xi, say in this particular case suppose I start with 0.121 my X1 will be equal to
0.0621 and accordingly my X2 will be 0.2533, X3 is going to be 0.6845 more or less my
Sigma Xi is 0.999 which I can take as one this particular case. Accordingly I find my Yi the
Yi’s in this particular case if I write it down here the Yi’s will be 0.219, 0.3633 and 0.4178.
So therefore the fraction which was vaporized one minus L by F is nothing but equal to
0.879.

A few other problems have also been provided here we can try out the problems. So in this
particular class what have I done? I have discussed different problems of concerning ideal
solutions for example I have discussed if Px is given how to find out Ty. Ty is given how to
find out Tx. If Tx is given how to find out, If Px is given how to find out Ty or if Py is given
how to find out Tx and I have also discussed Flash distillation problems there are a few
problems which are there in this particular slide the answers are provided you can work them
out to have more feel of ideal solutions.

Now suppose we go to next part of our lecture which deals with non-ideal solutions. What
you expect what is going to happen for non-ideal solutions? What are the things that we
expect in this particle case? Let us see, so in non-ideal solutions let us see what happens.

717
(Refer Slide Time: 17:41)

In ideal solutions I have already drawn this particular case. We have, what did we find? We
have found that for ideal solutions the partial pressure and the total pressure they vary
linearly with mole fraction. Definitely the solution does not behave ideally and along with
that what are the other things that we have found?

(Refer Slide Time: 18:15)

We found 3 features of ideal solution, first feature is that partial pressure and total pressures
they vary linearly with mole fraction number 1. Number 2 we have observed that Delta vM
equals to 0, Delta gm equals to xi, RT sigma xi ln xi and Delta sm was equals to minus R
sigma xi ln xi these are the things that we observed. If it is non-ideal then definitely there will

718
be some evolution or absorption of heat and when the 2 components mix then definitely we
expect that either there will be contraction of there will be expansion in volume.

So therefore what do we expect? We expect that this might be not equal to 0, this might not
be equal to 0 and so if these, one of these are not 0 then this again becomes not equal to 0 etc
it continues in this particular way. Now remember one thing non-ideal solution can have only
one particular property different for example it can have Delta vM not equal to 0 while the
other characteristics they follow the ideal solution model, okay it can happen.

(Refer Slide Time: 20:09)

Or it can happen that maybe more than one particular characteristics are not been exhibited
by the non-ideal solution. So therefore for non-ideal solution firstly if these things happen
then quite naturally your pressure cannot vary linearly with xi it is going to deviate from this
and the deviation can be either way. We can find that the pressure is actually higher than the.
There can be 2 things one is the pressure can be vapor pressure can be higher than that
calculated from the ideal solution model. It can be same for all the cases.

And accordingly since this happens, so naturally the total pressure will also be exhibiting will
be also be higher as compared to the total pressure calculated from your ideal solution model.
So therefore in this case what do we say? We say that this exhibits positive deviation from
Raoult's law. We can also have other things we can say that moment we mix the 2
components what happens? There is a contraction in volume it can happen.

And under that case what do we find? Suppose, see it all depends on the interaction of the
molecules repeatedly I have said that when we assume that the interaction between unlike and

719
unlike molecules are the same then we have ideal solutions. Suppose we see that that the
interaction between unlike molecules it is greater than the interaction between like molecules
then what happens? Then in that case when we mix than the interaction it increases.

When the interaction in the liquid phase increases naturally the tendency to go into the vapor
phase that decreases and under that condition what do we have? We have a negative deviation
of the Raoult's law, every case you are going to have a negative deviation from Raoult's law.
And when we find that the intermolecular interaction between unlike molecules are less as
compared to the intermolecular interactions between like molecules then when we mix then
what happens?

They are not very comfortable so there is a greater tendency of vapor formation, so as a result
the saturated partial the saturated pressure becomes higher as compared to that calculated
from the Raoult's law. So therefore we have a positive deviation, so when the unlike
interactions are less as compared to the like interactions we have positive deviation from
Raoult's law when the unlike interactions are higher as compared to like interactions we have
negative deviations from Raoult's law.

(Refer Slide Time: 22:53)

The entire thing can be summarized by a formula which has been provided by Van Laar. Now
this particular person as we know that he has worked throughout with solutions and he has
proposed a formula which accounts or rather which combine Raoult's law and deviations
from it. What is the formula that he has given? It’s quite an interesting formula.

720
He suggest P2 it is equal to x2P2 saturated which is Raoult's law where the correction factor
as exponential one minus x2 whole square Delta u0 by KT, right? Where what is Delta u0
equals to? Delta u0 is the interaction energy which takes into account the differences in
nature of the interaction of like and unlike molecules, right?

(Refer Slide Time: 24:29)

So therefore in this particular equation if I write the equation once more here just to define all
the properties it is the correction term to Raoult's law is one minus x2 square Delta u0 by KT
where K is nothing but the Boltzmann constant, T is the temperature as we all know this is
the temperature and this Delta u0 this is the interaction energy this takes into account the
differences in nature of interaction between like and unlike molecules.

Now when Delta u0 equals to 0 then naturally we find that interaction between like and
unlike molecules similar. So therefore the exponential term it becomes almost equal to one
and P2 can be given as x2 P2 saturated which is the case for ideal solution. When we find
Delta u0 is greater than one under that condition we get positive deviation from Raoult's law
and when does this happen? This positive deviation happens when you are unlike this
happens when unlike molecules attract more weakly as compared to like molecules, right?

Under this condition as I have explained. So therefore evaporation is facilitated and vapor
pressure is higher than that calculated by Raoult's law and if Delta u0 is less than one then in
that case as I have said at unlike molecules attract more strongly and so therefore the total
molecular interaction increases and the evaporation of molecules becomes more difficult and

721
we have negative deviations from Raoult's law. So therefore we can have 2 types of
deviations from Raoult's law.

(Refer Slide Time: 26:45)

Now what to do under this case? If you have some deviations what are what is the thing?
How we are going to account for the deviation? For everything you will notice that we take
help of or rather we go to my to the ideal the gas state and try to find something. In the gas
state what did we find? We found Pv was equals to RT for one mole of gas definitely most of
the gases they were not ideal and as I had mentioned we had introduced a compressibility
factor to account for non-ideal behaviour. So in this case also we found that for ideal
solutions this is the situation. So therefore when it is not ideal people thought why not
introduce a correction factor here such that this correction factor would take into account the
deviations from ideality.

And the correction factor which was introduced here that is known as the activity coefficient
and this particular correction factor they found that this gamma i can be greater than 0.

722
(Refer Slide Time: 27:33)

For this particular case gamma i should be greater than 0, here gamma i should be less than 0,
right? So therefore what are the things which happened when a solution does not behave
ideally? There are 2 things first thing is we find that there is either positive deviation or
negative deviation from Raoult's law. So therefore we find that in this particular case the
linearity is not followed and this is modified by introducing a gamma i here where we can
write down Pi equals to gamma i XiPi saturated.

So therefore the equation for non-ideal case it becomes this. Along with that what happens?
Just the way I have written down the there are some volume changes of mixing.

(Refer Slide Time: 28:36)

723
Definitely under that cases we will not have Delta m equal to 0 we will have some Delta m
some something some Delta Delta Vm which may be either greater than 0 or it may be less
than 0 but it will not definitely be equal to 0 and this excess property which the solution has
just because it is not ideal is known as the excess property of the solution.

(Refer Slide Time: 29:11)

So therefore with advent of non-ideality what are the things that that we have come across we
have come across an activity coefficient and we have come across another set of properties
which are known as the excess properties something similar to concept as the departure
functions or the residual properties which we had described for real gases you will remember
under that condition the residual properties was real gas value minus ideal gas value.

In this case the excess property, suppose it is we denote it as M excess it is the M real value
minus the M ideal value. So the property in the real solution minus the property in the ideal
solution again when we went we have defined the set of property excess properties we have
total excess properties which will be given by capital letters with a superscript E we will be
having, naturally when we have excess properties all of these are extensive we will be having
molar excess properties will be given by a small case letter.

And very sadly we will also be having a partial molar excess property which will be given by
this particular set and we also find out that just the way of total properties molar properties in
this case also they obey all the equations which are applicable for molar properties and total
properties.

724
(Refer Slide Time: 30:51)

For example in this particular case also we will have u excess equals to h excess minus pv
excess. We will have g excess equals to h excess minus Ts E excess and so on and so forth.
So in the next class what we do? We deal with non-ideal solutions and for dealing with non-
ideal solutions we will be dealing with excess properties and we will be dealing with activity
coefficients and we need to remember that the activity coefficients has come up because the
solution is not ideal. The excess properties have come up because the solution is not ideal. So
therefore there should be some relationships between the excess properties and the activity
coefficient.

So in the next class we will be trying to establish relationships between activity coefficient
and excess property and since we have defined a new property gamma i we would like to find
out how gamma i it varies with temperature, pressure and composition. We would be deriving
an equivalent Gibbs Duhem equation for gamma i and then we would like to find out, how
we can determine gamma i from known experimental parameters because finally the
prediction of gamma i would help us to calculate the fugacity in the liquid phase from where
we can proceed for further phase equilibrium problems in non ideal solutions. So we enter
into the domain of non-ideal solutions in the next class and it will make your journey of
thermodynamics slightly more difficult but I hope you will be enjoying it with that, thank you
very much.

725
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 40
Non Ideal solutions
Well, to continue with your discussions on solutions you will recall that in the last class we
had discussed ideal solutions and there are several properties of ideal solutions that we said
and for instance we had said that suppose we have for ideal solutions we had the things that
we had shown for ideal solutions if you recall we had plotted the partial pressure versus
partial pressure and the total pressure versus the mole fraction, right?

(Refer slide time: 0:43)

And what did we find? Please remember whenever we are plotting for this class this is not a
standard convention for this class just to avoid any confusion we would like to plot the mole
fraction of the more volatile component in the x-axis with that we plot the partial pressure of
component 1, partial pressure of component 2 and the total pressure over the solutions. Now
since this is a more volatile component, what do you expect?

Naturally the saturated vapor pressure of component 1 will be higher as compared to the
saturated vapor pressure of component 2. So therefore we find that at X1 equals to 1, suppose
this is P1 saturated then for X1 equals to 0 this is going to be say P2 saturated and we know
that for ideal solutions what is the what is Raoult’s law for ideal solutions? Just to recollect
that Raoults law is just the Lewis Randall rule for low to moderate pressure where the
fugacity term can be replaced by pressure, right?

726
So the Raoult's law, if we write it down the Raoult's law for ideal solutions this was X1 P1
saturated this was equals to P1. So since we know that this particular P1 saturated this is
nothing but the function of temperature and this particular plot, this is done under constant
temperature conditions. So naturally P1 saturated becomes fixed as a result of which what do
we find? We find that P1 is proportional to X1, right?

In the same way what is P2? This is nothing but equals to X2 P2 saturated, what is X2?
Remember we are dealing with binary mixtures. So for binary mixtures P2 equals to 1 minus
X1 into P2 saturated, so in this case also we know this is P2 saturated minus X1 P2 saturated.
So in this case also we find that P2 it is inversely proportional to X1, right? So therefore if we
plot say P1 as a function of X1, what do we expect?

We expect to get a linear plot like this, right? This plot is P1 versus X1 same way if you are
plotting P2 versus X1, what do you expect? We expect to get a plot like this; this plot is P2
versus X1, all these things we have discussed in the last class, fine. And what is the total
pressure P equals to? It is obviously P1 plus P2, what is it? It is X1 P1 saturated plus 1 minus
X1 P2 saturated which gives you as X1 P1 minus P2 saturated plus P2 saturated, so what
does it imply?

The curve of P versus X1 that should (ha) that should have an intercept of P2 saturated at X1
equals to 0 or in other words P equals to P2 saturated at X1 equals to 0. P equals to P1
saturated at X1 equals to 1. So therefore what does it imply? The P versus X curve should be
a linear plot extending from P2 saturated to P1 saturated. So this was what we have obtained
for an ideal solution if you recollect.

For again I would like to repeat that this whole thing was for low to moderate pressure where
f can be replaced by the pressure term fine so therefore for this particular case we found out
that this is the situation and this is the law which governs ideal solutions. Again I would like
to repeat that we had already discussed that ideal solution model is applicable for solutions
over a very rare narrow range of conditions if you recollect.

I had already mentioned the range of conditions for which ideal solution model is applicable
and very frequently we come across as chemical engineers we come across solutions which
are non-ideal and that can occur when the molecules are not similar in size, the
intermolecular attraction between say aa and bb components it is not similar to the

727
intermolecular attraction between ab components and so on and so forth this happens very
frequently.

Now when this happens what do you expect? What are the things that you expect should
happen? First thing is in that particular case can we directly write down just in this particular
form.

(Refer slide time: 5:42)

Can we write down this particular equation P1 is proportional to X1? This should not happen
when it is in non-ideal solutions. So therefore what will happen? We find that there will be
either the partial pressure will be greater than that calculated from the ideal solution model or
it will be less than that calculated from the ideal solution model.

And remember one thing for every solution ideal or non-ideal at X1 equals to 0, we this
particular equation is, at X1 equals to 1 when a component 1 is in excess we will always have
P1 equals to X1 P1 saturated or in other case under that condition P1 always approaches P1
saturated. So far even for any non-ideal solution also for the excess component we find that
the Raoult's law is applicable when the composition of the excess component approaches
unity that is a universal truth that always happens.

So under ideal conditions also even if we have a negative or positive deviation from Raoult's
law we will find that at the extreme cases under this particular condition Raoult's law will be
applicable for component 1 it will also be applicable for component 2. I had also mentioned if
you recollect that for very low concentrations under that condition also we will find P1 is

728
proportional to X1 but under that condition as I told you the proportionality constant is not
equal to the saturated vapor pressure of that particular component.

If if you if you extend this particular linear portion you will find that the intercept is at a
different point as compared to P1 saturated and we had founded and we had told or rather we
had termed this particular condition as an infinitely dilute ideal dilutes solution and for such
ideal dilute solution we had found out that the proportionality constant it is equal to the
Henry’s constant of component 2 in component 1 or vice versa.

So we have dealt with 2 types of ideal behaviours one was ideal solution where both the
components obey Raoult's law when it is in excess quantity and ideal dilute solution this is
particularly applicable for those particular components in solution which do not remain in the
phase of the solution when it is in large excess, what do I mean? I mean say suppose carbon
dioxide in water, what is the phase of the solution, liquid. Carbon dioxide when it is in much
excess the solution does not remain in the liquid state. So therefore carbon dioxide does not
remain in the phase of the solution and therefore for such situation we find that that ideal
dilute solution is applicable for the solute part which is carbon dioxide and ideal solution
model is applicable for the solvent part which is water in this case, right?

So therefore just like we have deviations from ideal solutions we will also have deviations
from ideal dilute solutions the method of dealing with the deviation from ideality is the same
for ideal solutions as well as for ideal dilute solutions, what do we observe?

729
(Refer slide time: 9:14)

We observe that the curve it deviates from ideality just as I have shown in a better
representation here it deviates from ideality and it can either be positive deviation or it can be
negative deviation. For positive deviation the total pressure calculated is greater than that
which should have been obtained for Raoult's law, for negative deviation the total pressure is
less than the sum of the partial pressures, right? Now how to account for this? You will again
I will say that for Raoult's law this was the law which was applicable.

Now what to do for a non-ideal solution? For everything remember we again and again resort
to ideal gas behaviour, why? Because it has got a sound theoretical basis in the Kinetic theory
of gases, so for gases what do you do?

730
(Refer slide time: 10:13)

We find that for gases we have a relationship like Pv equals to RT which implies, what does it
implies? This implies that suppose we plot Pv by RT as a function of temperature say we
should be getting a straight line at 1.0. For most of the situations what do we find? We find
that usually the graph is something of this sort and it keeps on deviating from ideality in this
particular way. So we find that Pv by RT it deviates from ideality, this is pressure it deviates
from ideality at high pressure and at low-temperatures, all these isobars are for different
temperature.

So what did we do? We found that the Pv by RT terms it deviates from ideality and therefore
we had incorporated correction factor which is termed as the compressibility factor. What
was the compressibility factor? It was the actual volume occupied by the gas divided by the
ideal volume which we it would have occupied had it been ideal under the same conditions of
temperature and pressure.

So what is this? This is nothing but v actual by RT by P for one mole of gas or in other words
what is z equals to? This is nothing but Pv by RT for the real gas and therefore what did we
do? This was the ideal gas equation, right? And for real gas what did we do? For real gases
we just modified the equation by incorporating a correction factor z in this particular case.

731
(Refer slide time: 12:09)

(Refer slide time: 13:04)

In the similar way just take taking note from the ideal gas equations here also we try to do a
similar thing, what is the similar thing? This is the case for ideal solution, so therefore when
the solution is non-ideal what should happen? Very simply just to keep matter simple we can
simply incorporate a correction factor here. The correction factor is conventionally defined by
gamma and it is termed as the activity coefficient, right?

So therefore in order to account for non-ideal behaviour, what we do? We simply incorporate
or modify the Raoult's law by incorporating a correction factor gamma and you will recollect
that if you are dealing with higher pressures then in that case you will be modifying the
fugacity expression by incorporating this particular correction factor this particular correction

732
factor gamma. So therefore this is the way we try to do or rather we try to modify the ideal
solution model which is pretty simple and can be obeyed by several solutions we try to
modify it just by incorporating gamma.

(Refer slide time: 13:39)

So therefore now in this case what does my vapor liquid equilibrium become? The vapor
liquid equilibrium if you recollect we had already told you that started with fiL for
component 1 is equal to fiV. For low to moderate pressures usually remember for such
conditions usually the gaseous mixtures behave as ideal solutions and that is quite a rational
thing, why? Because the ideal solution model if you recollect it had its genesis from the
Amagat’s law of additive volumes which are obeyed by ideal mixture of real gases, right?

So therefore for most of the cases we can write fiV equals to Pi which is nothing but yiP,
right? What is fiL equals to? Had the solution been ideal it would have been xiPi saturated
since it is not ideal we will have a gamma i here. So therefore what happens for low to
moderate pressure non-ideal liquid solution? And we usually have ideal gaseous mixture
remember ideal gaseous mixture does not imply a mixture of ideal gases it implies a ideal
mixture maybe of ideal gases or of real gases.

So for this case what will we have? We will be equating these 2 or rather we will be equating
these 2 expressions, what do we get? We get yiP equals to gamma i Xi Pi saturated, agreed?
So therefore we find out that if we really want to find out the vapor liquid equilibrium some
this particular condition then we have to encounter an additional term gamma and if we can

733
find out this particular additional term, in terms of some known measurable properties some
known thermodynamic relationships then definitely we can use that particular equation here.

Once it is incorporated here then the calculations of vapor liquid equilibrium they come
similar to the ideal solutions and I will just I will just recall that whatever different types of
problems that we have dealt for the case of ideal solutions are applicable for non-ideal
solutions there will be bubble pressure calculation, bubble temperature calculation, due
pressure calculation and dew temperature calculations such as we had discussed Tx is Py has
to be found out Ty is known Px has to be found out and so on and so forth just the way we
had done.

(Refer slide time: 16:28)

And for everything we have to start with referring this particular equation and the only hitch
which we have here apart from whatever we had done for ideal solutions is calculation of this
particular gamma this is the only new thing that we have incorporated in order to account for
non-ideality in the liquid solutions equation. So now let us see how to find out Gamma in
terms of known measurable parameters. What do you expect? Where with what parameters
should gamma be related?

Let us analyze and find out, see what happened because the solution has shifted from ideal to
non-ideal behaviour first thing which is happened is it does not obey Raoult's law. Let us jot
down one by one what has happened, the first thing is again I will try to recover I try to
emphasize that we are dealing with low to moderate pressure, why? Because repeatedly I am
referring to Raoult's law not Lewis Randall solution or in other words I am referring to partial

734
pressure, total pressure etc instead of fugacity that is the only reason. If you are going for
higher pressure I will be talking about fugacity instead of pressure.

(Refer slide time: 19:05)

So therefore for low to moderate pressure what are the things which happens if the solution
was ideal, what would you have expected? Obeys Raoult's law, you agree with me. What was
the second thing? Naturally if it obeys Raoult's law Pi equals to Xi Pi saturated. Third thing is
if you remember we had found out some property changes of mixing. What did we tell under
that condition? We had we had told you that delta v mixture was equals to 0.

Or in other words x sigma xi vi bar minus vi that was equal to 0 where if you recollect vi bar
is the partial molar property of component i in solution. Since everything is in liquid so I am
not putting this particular L superscript it is implied and this is the molar volume of the same
component under the conditions of same temperature and same pressure as that of the
solution. Similarly we had found out delta hM equals to 0 we had found out delta uM equals
to 0, what else did we find out?

Delta gM was that equal to 0? If you recollect that was not equal to 0, we found out that Delta
gM was sigma xi, Delta gM it was nothing but RT sigma xi ln xi and Delta sM this was
equals to minus R sigma xi ln xi. We had derived all those things and I would also like to
recollect if you remember that when we were dealing with mixing of ideal gases we found
out that the entropy change was mixing was given by this particular equation when we had
devised a reversible path for calculating the entropy change and we had calculated the

735
entropy change for mixing of some ideal gases. So this was exactly the expression that we
that we had obtained.

(Refer slide time: 20:07)

Now tell me one thing this whole thing was for ideal solutions. For non-ideal solutions what
are the deviations that you would expect? Naturally first is does not obey Raoult's law, quite
natural. As a result what do we have? Pi equals to gamma i xi Pi saturated and naturally what
are the other things that you expect? Naturally in this particular case see the volume there was
no volume change in mixing, why? Because definitely the molecular sizes and the shape were
similar in the 2 components or in the N number of components which were there, so in this
particular case when it is behaving in a non-ideal fashion naturally the molecular sizes and
shapes should not be similar in that case, okay.

So therefore quite naturally Delta vM will not be equal to 0, expectedly Delta hM will not be
equal to 0. Delta uM will not be equal to 0 similarly Delta gM will not be equal to this it is
just the same thing if I write it down so none of these things should happen, right? What
should happen then under that particular condition? That particular condition we find that the
molecular sizes and shapes are not equal as I was telling so Delta vM is not equal 0.

Definitely the intermolecular interactions between molecules of component a and of

component a and of solution ab are not the same. So naturally when they are mixing there
will be either some absorption of heat or some rejection of heat. Quite naturally, so there will
be some it will be accompanied by a heat of mixing that will not be equal to 0 and so on the
other things follows. So what does it imply?

736
It implies, definitely it means that for real solutions there will be a nonzero component of the
properties of mixing and this particular nonzero properties arise because the solution does not
behave ideally, you get my point. If the solution would have behaved ideally then all these
properties would have reduced to this they are not reducing to this just because the solution is
real and not ideal. So therefore there has to be some amount of excess properties of all this
particular extensive properties that we have written down and these excess properties are
coming just because the solutions are real.

(Refer slide time: 22:51)

Again recollect the real gas and the ideal gas behaviour, for that particular case what did we
say? We said just because the gas was not real, what came up? We had defined this equation
and there had come up a series of residual properties or departure functions if you recollect
and if you will also be recollecting that and they were they were shown by delta P star or
delta M star whatever it is better whatever you wish you can you do can write it down.

These were the departure functions which were the actual properties minus the ideal gas
properties at same temperature and pressure. And we found out that each of the residual
properties are related to Z. Quite naturally because Z has been introduced to consider real gas
behaviour and these residual properties have come up because the gas is real and not ideal. So
therefore there should be some relationship between the residual properties and the
compressibility factor and as a result of which we were we had tried to express each and
every residual properties or departure functions whatever term you wish you can use in terms
of the compressibility factor Z, right?

737
(Refer slide time: 24:12)

So therefore in a similar way is it, does it not occur to you that the behaviour of non-ideal
solutions can be predicted or can be defined much more easily if we say that since the
solution is not ideal since the solution is real so therefore it should have a set of excess
properties defined as ME or PE in whatever way you wish to define the property of a
solution. So therefore we find that just because it is real it should have a a set of excess
properties.

It should have a excess volume v excess it should have h excess it should have u excess and
similarly definitely it should have g excess it should have s excess, why does it have this
excess property? Because the solution is not real so therefore how do we define v excess?
Quite naturally should be the actual volume of the solution minus the volume of the solution
had it been ideal, right?

And therefore or in other words it should be the volume change occurring in the solution due
to mixing minus the volume change which would have occurred had the solution been ideal.
We know that the volume change which occurs in ideal solution is 0, so naturally your excess
properties nothing but the volume change on mixing, right? So this should also be i, so in the
same way we should be able to define each and every of the excess properties that we have
that rather each and every excess property of the extensive properties that we are dealing with
and somehow we should there should be some relationship between Gamma and the set of
excess properties and definitely that should give us some hint about estimation of gamma in
terms of easily measurable properties, right?

738
(Refer slide time: 26:16)

So therefore what do we do now when we start dealing with non-ideal solutions? We have
defined just to account for the non-ideality of the solution.

(Refer slide time: 26:42)

We have defined a set of excess properties, right? And we find out that these excess
properties they arise just because the solution is not ideal. So therefore let us have a proper
definition of these particular excess properties. How do we define these excess properties
then? Naturally, if we define where any particular property with P, remembering that P should
be referring to any extensive property.

739
So therefore and when we are denoting the property with the capital letter it refers to the total
property and under that condition the excess property PE it it should have be it should be
equal to the difference between the actual property of the solution minus the property which
the solution would have, had it been ideal very important under the same conditions of
temperature, pressure and composition and just like we have defined the total excess property
definitely we should be in a position of defining a molar excess property as well, agreed?

We are dealing with solutions, so therefore for each component we should be in a position of
defining partial molar this is the molar property minus the partial molar property of rather it is
better if I write ideal solutions at i at the bottom. Let me write it down as I that will be better.
So therefore this should be equal to P excess P bar excess under the ideal condition, right?
This is P by, agreed? So therefore just like we have a total excess property we have a molar
excess property and for each component we have a partial molar excess property which is the
actual molar excess property, partial molar excess property of component i in solution minus
the partial molar excess property of component i had the solution been ideal.

So what do we find? What have we done in the process can you tell me? In the process again
unfortunately for you we have defined a set of another set of property list, this property list
they are known as the excess properties and excess properties I repeat it is the difference
between the actual property of the solution and the property which the solution has, had it
been ideal under the same conditions of temperature, pressure, composition and remember
one thing departure functions, excess properties etc they all refer to extensive properties,
okay.

And therefore just like we have total excess properties or total departure functions for real
gases we have a molar excess property we have a partial molar excess property. So in the next
class what we do we try to express the excess properties or rather the activity coefficient in
terms of excess properties and let us see for what comes out or whether we can find out
activity coefficient in terms of easily measurable parameters.

740
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 41
Non Ideal solutions(Contd.)
(Refer Slide Time: 0:37)

So therefore let us see what are the excess properties that we were dealing with? Just as I
have told you, so therefore if we go to the previous slide it will be clear to you that what do
we have then in this particular case? We have v excess what is v excess it is v actual minus v
ideal. Actual ideal everything refers to solution. What is v actual then? It is Sigma xi vi bar,
what is v ideal you know? It is xi vi, so therefore this should be xi vi bar minus vi had the
solution been ideal vi bar would have been equal to vi and vE would have been equal to 0.

This case this is not equal to 0, therefore what is this? It is actually the volume change on
mixing of the solution. Same way we can write down hE just the same way we can repeat it h
actual minus h ideal, right? Which is nothing but equal to Sigma xi hi bar minus hi which is
nothing but equal to Delta h mixing for a real solution same thing uE will be equal to U delta
U mixing. Let us see what is gE equals to?

For gE, what is this? Again in the same way it should be g actual minus g ideal, okay. So
therefore what is this remember one thing this is the Gibbs free energy of the solution actual
or the real solution, this is the Gibbs free energy of the ideal solution both at same T, P,
composition. Now remember one thing as I had already mentioned we cannot define absolute
values of any properties we always define the property changes, right?

741
So therefore this actual Gibbs free energy of the solution is better expressed as the actual
change in Gibbs free energy on forming the solution, isn’t it? So we had 2 components ab we
mixed them due to this mixture there was a change in Gibbs free energy, what is this Gibbs
free energy change for a real solution? What is this Gibbs free energy change for an ideal
solution? The difference between the 2 should be a better measure of the excess property as
compared to the Gibbs free energy of actual solution minus the Gibbs free energy of the ideal
solution, why?

Because we cannot define absolute value of Gibbs free energy for any particular substance in
any particular state we always need a datum as you remember. So therefore this can be
expressed as Delta gM actual minus Delta gM of mixing ideal, agreed? Which can also be
referred to as Xi gi bar minus gi bar ideal this is actual. Now if you recollect what was Delta
gM bar ideal? If you recollect, so therefore gi for an ideal solution what was this? Your gi bar
ideal or rather gi ideal. Let us take it; it was RT Sigma xi ln xi that we have already derived.

So therefore what should this become? This should become xi gi bar actual minus Sigma Xi
RT ln, so we can we can take this out or else we can define this as Sigma xi gi bar minus gi
minus RT Sigma xi ln xi. This should have been gi bar actual minus gi, agreed?

Okay, so therefore what did I do this Delta gM actual that that is actually the Delta gM
change for the actual solution which is given by this the , what happens during formation of
the solution? The molar Gibbs free energy changes to the partial molar Gibbs free energy for
each component i. So this is for the actual solution and this is for the real solution, right?

742
(Refer Slide Time: 5:10)

Now tell me one thing, how is this gi bar minus gi related? This so I will just repeat it once
more, so that it is easier for you to understand g excess this is equal to Sigma Xi gi bar minus
gi minus RT sigma Xi ln Xi, fine. What is this gi bar minus gi? This is why because the pure
component i was taken into a solution that is why this particular change occured. Can we not
write it down as RT ln fi bar by fi? What does it show?

The fugacity change of pure component i when it is taken into solution. So what is my
standard state in this particular case pure component i. So from there and the standard states
as same for both the left inside and right-hand side of the equation, so from the pure
component i, I am taking it to the solution state as a result of which it’s Gibbs free energy is
changing to gi bar and its fugacity is changing to fi cap, agreed?

So therefore this particular equation this particular exact equation can be written relating the
molar Gibbs free energy or the partial molar Gibbs free energy with the fugacity and the
partial molar fugacity, agreed? I can do this, so what is this again what is this fi bar equals to?
For a real solution this is equals to gamma i Xi fi. So therefore can I not write it down as RT
gamma i Xi fi by fi? Both the cases remember the fi are the same it is the pure component
fugacity at temperature, pressure of the solution.

This was the fugacity of the pure component and this is also the fugacity of the pure
component, can I not write it down in this particular way? So therefore what is my gi bar
minus gi then in this case this is RT ln, Sigma gamma i plus RT sigma ln xi. I can write it
down in this particular way, fine. And so therefore if I substitute this in this particular

743
equation, what do I get? Excess Gibbs is nothing but Sigma Xi gi bar minus gi minus RT
sigma Xi ln Xi.

What is, I just substitute gi bar minus gi so what do I get? This is equal to Sigma Xi RT can
be taken outside ln gamma I plus RT sigma Xi ln Xi minus RT sigma Xi ln Xi. What do I get
as a result of this? I find that the gamma i can be related in a very simple way with this is the
relationship of excess Gibbs free energy with gamma, right? And remember one thing just
because the solution is not ideal. So therefore there will be some excess Gibbs free energy
and that decides whether the solution forms spontaneously or not and as a result of this non-
ideality gamma has come into picture.

So therefore we find that that the gamma of the of each and every component can be related
to the molar Gibbs free energy change of the solution in this particular summation form. Now
what is this gE tell me? gE is nothing but equals to your X into gi bar E, yes or no? Agreed?

(Refer Slide Time: 8:59)

So therefore how can I get this gi bar E? This gi bar E it is nothing but equal to del gE del xi
at constant T, P xj not equal to i, agreed. Or in other words can I not write it down in this
particular way del del Ni of? Say suppose it is a your binary mixtures. So therefore del del
this is g1 E is del del N1, N1 plus N2 into gE at constant T, P, N2. And and what is this gE
equals to? This is nothing but equal to del del N1, N1 plus N2 X1 ln gamma1 plus X2 ln
Gamma2 constant T, P, N2.

744
Simply differentiate this and what do you get? You just differentiate it you are going to get g
by gi bar E that is the excess, I will write it down excess Gibbs free energy change, excess E
excess partial molar, why? The bar is there excess partial molar Gibbs free energy change of
component i in solution, what does what is this? This is gi bar E equals to RT ln gamma i. So
if you can calculate the excess Gibbs free energy the partial molar excess its free energy of
component I then you should be in a position to calculate the activity coefficient of that
particular component, agreed?

(Refer Slide Time: 10:54)

So therefore we find that we have come across these 2 equations which I have already written
down here if you observe. We find the molar Gibbs excess Gibbs free energy of the solution
is given in terms of composition or activity coefficient by this equation and the partial molar
excess Gibss free energy change of component i in solution is nothing but equal to ln of
gamma i, right? So therefore this should be one particular way of finding out gE of finding
out gamma i and it it’s important for you to recollect that just like we had done for ideal
solutions we can always write down or rather we can always find out the excess entropy
change of the solution.

What it is going to be? This is nothing but equals to hE minus, if I write it down what is s
equals to? Recollect g equals to h minus Ts. So therefore s will be equal to g by T minus h by
T, right? And therefore once this equation was written for a for a molar quantity we can
always add a superscript s for the in extensive properties and this is the relationship between
the excess properties. So therefore we can easily find out Gibbs free energy excess entropy in

745
terms of excess Gibbs free energy and excess enthalpy and once we do it we find that the
excess entropy is given by this particular equation, right?

(Refer Slide Time: 12:50)

So therefore the interesting part which we found out was the first thing which we found out gi
bar E by RT equals to ln gamma i gamma should also be related to other excess properties,
how would is related let us see?

(Refer Slide Time: 12:54)

How gamma varies with temperature pressure and composition etc. Just try to do it yourself
and then see what do you get?

746
(Refer Slide Time: 12:59)

(Refer Slide Time: 13:03)

Suppose you try to find out del ln gamma I del T, right? At constant pressure and
composition, what is this equals to? You already know what is ln gamma i? It is nothing but
equals to gi bar E, please do not forget there are 3 things which have to be mentioned at
constant P and x, right? And what is this gi bar E this is nothing but equals to del del T of mu
i cap by RT at constant P and x, right? We can write it down in this particular way.

So therefore from here what do we get then? From here we find out or in other words this can
also be written down as, we can start it in a slightly different way and which will be easier for
you to understand. Let us start from the basics then it will be easier for you to understand.
This we have already found out, right? So therefore what is ln fi L by fi L? Pure component

747
nothing but equal to ln gamma i Xi. Now this is equal to RT, so if we have a RT here you
have a RT here you have ln gamma i plus RT ln X Xi, agreed? So what is del ln fi L by fi L.

Say del P if we take at constant temperature, recollect what is this equal to? This will be
equal to vi bar L minus vi L by RT, if you remember dg equals to minus sdT plus vdP. So
therefore this is what precisely we have done.

(Refer Slide Time: 15:09)

So in the same way we, can you not write it down as l, del ln gamma I Xi del P? This should
be equal to vi bar L minus Vi L by RT. What is vi bar L? It is the partial molar excess volume
of component i in the liquid solution. What is this? It is the molar volume of component i in
the liquid state for an ideal solution we know that if the for an ideal solution what do we
know? For an ideal solution vi bar L ideal is nothing but equal to Vi L.

So therefore what is this then in that case? It is the partial molar volume change for the case
of a real solution minus the partial molar volume change for an ideal solution. So therefore
what is it then? This is nothing but vi excess by RT vi bar excess by RT, agreed? Which is
what? This vi bar E excess, what is this? Excess partial excess E bar means partial molar
volume, agreed? For component i which is nothing but actual molar volume of component i
in solution minus the, in real solution real is within bracket minus molar volume of
component i in ideal solution, very important at same T, P, x. This you have to keep on adding
because the standards states have to be the same, right?

748
(Refer Slide Time: 17:31)

(Refer Slide Time: 17:37)

And therefore and what is this Del ln gamma I by del P? So we can we not write it down as
del ln gamma I del P at constant T plus del ln xi del P at constant T. We know that Xi the
mole fraction does not change with pressure. So therefore del ln gamma I del P is nothing but
it can be related to the excess partial molar volume for component i which I have also written
down. Same way suppose we would like to find out the temperature, the dependence of
gamma i on temperature. It should again go on in the same way.

749
(Refer Slide Time: 17:56)

It should be del ln fi del T at constant P this is nothing, if you recollect hi 0 minus hi by RT

square . Same way, can we not write down for the partial molar fugacity hi 0 minus hi bar by
RT square? So therefore can I not write and what do I know? I know fi bar by fi equals to
gamma i Xi, right? So therefore ln fi bar by fi is nothing but equal to ln gamma i plus ln Xi
agreed? So therefore can I not write this down as del ln gamma i?

Again I know that Xi is not a function of temperature it been composition does not depend on
temperature. So can I not write this down as del ln fi bar by fi del T at constant P? Is this not
equal to minus of hi bar plus hi by RT square hi0 is the same for both the equations. What is
this equal to? Isn’t it equal to minus hi bar excess by RT square? What is this? This is nothing
but the partial, I will write it down hi bar E it is the partial molar enthalpy of component i.
For partial there is a bar partial molar excess, See I am also missing a lot of things partial
molar excess enthalpy of component i which is nothing but the partial molar enthalpy of i
minus the molar enthalpy of component i at the same conditions of temperature and pressure,
right?

750
(Refer Slide Time: 19:49)

So variation with temperature can be related with the partial molar excess enthalpy of the of
the solution and the gamma variation with pressure can be related with partial molar excess
volume of component i in the solution and we know that the you’re the gamma i it is related
with the with the excess Gibbs free energy of each and every partial molar Gibbs free energy
of each and every component in the mixture by this equals to ln gamma i and the excess
molar Gibbs free energy of the mixture by this particular equation.

So therefore we find as we have anticipated that excess properties arising to the real the
deviation of the solution from the ideal behaviour and we have used gamma to account for
this particular deviation. So therefore gamma can be related with the excess properties and
we have shown how it is related to excess molar Gibbs free energy of the solutin excess
molar the partial molar enthalpy of each component, partial excess partial molar volume of
each particular component.

751
(Refer Slide Time: 21:09)

Now since we have defined another set of excess properties it is very interesting to find out
that these particular excess properties they also obey the same equations as that of the the
molar properties as that of the total properties whatever equations we have defined in terms of
total properties in terms of molar properties you just add a superscript E and you find that
each and every particular relationship is applicable and all these equations you can actually
derive and you can find it out that all these equations which you have derived for the your
molar and total properties are applicable for total excess properties, for partial molar excess
properties for molar excess properties, right?

(Refer Slide Time: 22:03)

752
Total excess properties given by this, your molar excess property is given by this, partial
molar excess property for each component is given by this. So these are the 3 symbols that
we will be using we just should be little careful about the symbols since we are dealing with a
large number of properties, right?

(Refer Slide Time: 22:09)

There is one more thing which is very interesting we have defined rather we have tried to
incorporate the non ideal behaviour by incorporating gamma we have related to the excess
properties of the solution. Now whenever we have been dealing with solutions we found out
that for mixtures there is a very important relation which tells us that all properties of the
mixtures are not ideal, what was that particular equation?

It was the Gibbs Duhem equation; we have derived the Gibbs Duhem equation in terms of
chemical potential, right? The same can be derived for in terms of fugacity and we should be
also in a position to derive Gibbs Duhem equation for activity coefficient, why do we need to
do it? We need to do it because we have seen that if Gibbs Duhem equation gives us one
particular clue by which we can find out any property of component a if the same property for
component b is known provided it is a binary mixture. If it is a N component mixture if we
know the property for N minus 1 components we will be in a position to find the same
property for other nth component provided the entire thing is being performed under the
conditions of same temperature, pressure and composition.

So therefore since finding out gamma is so very important gamma is so very important as far
as vapor liquid equilibrium for nonideal solution are concerned and we come across nonideal

753
solutions very frequently so therefore it is important to find out gamma or rather to express
Gibbs Duhem equation in terms of gamma. So that for a binary solution which we usually
come across is we know gamma 1 we will be in a position to find out gamma 2.

(Refer Slide Time: 23:55)

Let us start from the beginning and see how we can derive the Gibbs Duhem equation in
terms of gamma? Remember the basic equation it was SdT minus Vdp plus Ndmui equals to
0 this we had found out for single component, agreed? For multi-component what should
what should happen? The same equation but this should be Sigma should come up Nidmu i
equals to 0, this is for multi-component case.

In terms of molar property everything becomes small, plus Sigma xi d mu i equals to 0 in

terms of molar properties, agreed? So therefore in terms of excess molar properties again let
us write down, can I not write it down as mu i is nothing but gb gi bar E equals to 0, right?
And again I know what is this gi bar E. We have derived it it is nothing but RT ln gamma i,
agreed? So therefore instead of this di bar g I can just write down RT ln gamma i then what
happens to the Gibbs Duhem equation?

754
(Refer Slide Time: 25:24)

The Gibbs Duhem equation then becomes sEdT minus vE dp plus Sigma i equals to 1 to N of
whatever xi d RT ln gamma i equals to 0, agreed? We can write it down in this particular way,
so therefore and now if we expand this part what do we get? We get sEdT minus vEdp I am
just deriving it so that it will be easier for you xi R ln gamma i dT plus Sigma xi RT d ln
gamma i equals to 0, agreed? So therefore what do I get finally sEdT minus vEdp plus R
written in a proper shape xi ln gamma i dT plus RT Sigma xid ln gamma i this is equals to 0,
fine.

What is sE equals to? It is nothing but equals to hE minus gE by T we have already done it.
What is this gE? Again I would like to (re) recollect gE by RT it is equal to Sigma xi ln
gamma i, right? So therefore this particular term gE by T, can this not be written down as R
Sigma xi ln gamma i? Let us substitute it here what do we get? Do we not get sEdT minus hE
by T dT minus R? Then this term and this term they cancel these 2 terms this and this they
cancel out and we get minus R Sigma i equals to 1 to N Xi ln gamma i dT, fine.This
particular term, this term then can be written down as this term can be written in this
particular form, okay. And so therefore on substituting this particular this and this in this
equation, what do we get?

755
(Refer Slide Time: 27:43)

We get equation like this hE by T dT minus R I have just substituted Xi ln gamma i dT minus
vEdp plus R sigma Xi ln gamma i dT plus RT Sigma Xi d Ln gamma i equals to 0. We
immediately find that this and this and this they cancels out and when all these cancels out,
what is the final equation that we get? Final equation it becomes hE by T dT minus vE dp
plus Rt Sigma Xid ln gamma i equals to 0. This is the final equation this is the Gibbs Duhem
equation in terms of activity coefficient gamma, right?

So therefore this is the equation that that we get under this particular condition.

756
(Refer Slide Time: 28:51)

And then remember one thing that when we are dealing with properties this is what I have
written down this is the final equation that I have written down so this is the equation that we
have obtained, right? Now here I would like to mention one more thing before I go further.
Since this is a property therefore we can have this particular derivation in an alternative form
as well.

(Refer Slide Time: 29:33)

Instead of starting from this particular way we could have also started the Gibbs Duhem
equation in terms of mui and we could have proceeded. At constant temperature and pressure
Gibbs Duhem equation in terms of mui, in this particular case we have started with gi bar in

757
terms of fugacity or in terms of mu, suppose we write we can also start with this particular
equation, okay. In terms of fugacity this becomes Xi d ln fi bar equals to 0, right?

We know fi bar is nothing but equal to gamma I Xifi everything is for is for the liquid phase
if you take, so therefore can I not write it down in this particular way gamma i d, Xi d ln
gamma i Xi fi Sigma this is equals to 0, right? So therefore we if you break it down what
happen? This is Xi d ln gamma i plus sigma Xi d ln Xi plus Sigma Xi d ln fi equals to 0 this
is a easier derivation. So that is why I am just doing it for you.

Now try to understand we have started with constant temperature and pressure. At constant
temperature and pressure, what is fi equals to? It is a fugacity of the pure component liquid at
the same conditions of temperature and pressure and we know that from low to moderate
pressure it is a saturated vapor pressure. So when T and P are constant fi automatically
becomes constant. So therefore even if it is at low to moderate pressure or at any pressure we
know that this component becomes equal to 0.

What about this part? This is nothing but equal to Sigma Xi dXi by Xi, isn’t it? Or in other
words this is Sigma dXi. What is this? This Sigma of dXi means the variation of the total
number of moles we know the total number of moles of the component is constant. So
therefore automatically this becomes equal to 0. So therefore what does it imply? It implies
that at constant T and P automatically Xi d ln gamma i equals to 0.

(Refer Slide Time: 31:43)

758
Just the same equation which I have derived here but this was the generalized expression at
constant T and P this goes out. So therefore this is the equation which I have derived in this
particular case as well, right?

(Refer Slide Time: 31:53)

So therefore this is the generalized equation which we have we find that at constant
temperature naturally this term goes off and this is the equation at constant pressure this term
goes off and therefore this is the equation at constant temperature and pressure I have derived
this equation in two particular ways and we find that this is the equation. Now this equation
just like I had told you previously this equation is very important particular importance or can
be better appreciated if we can write it down in terms of a binary mixture.

759
(Refer Slide Time: 32:31)

In terms of a binary mixture what do how can we write down the equation it is X1 d ln
gamma 1 plus X2 d ln gamma 2 equals to 0, agree? So therefore from this particular equation
there are a number of important information that we can get which you are going to discuss in
the next class.

760
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 42
Non Ideal Solutions(Contd.)
Well to continue with our discussions on nonideal solutions before I proceed to for today’s
class I would just have a short recap of what we have already done. Just taking into
consideration the fact that nonideal solutions are quite complex means when students first
deal with them they often get confused with the different sorts of terminologies one extra set
of properties that have come up. So I will have a quick recap of what I have already taught in
the last class and then we proceed further.

What did we do? Initially when I started solutions I dealt with ideal solutions. I would like to
mention I have dealt with 2 types of solutions one was ideal solution one was ideal dilute
solutions. Most of my discussions are limited and will be limited to ideal solution and maybe
at sometimes I will touch upon ideal dilute solutions.

(Refer Slide Time: 2:00)

And there what did we find again as a recapitulation we found out that more or less Lewis
Randall rule is applicable for ideal solutions where we get that the partial fugacity of
component i in a solution it can be a vapor mixture, it can be a liquid solution doesn’t matter
the law is the same this is equal to xifi if this refers to a liquid solution it can be denoted by a
superscript L and if it is a vapor solution than it can be represented by a superscript V, right?

761
And we found out that under equilibrium conditions fiL equals to fiV, so long our situation
was fine and then we found out that the situation becomes simpler at low to moderate
pressure, why? Because under those conditions we found out that the Lewis Randall rule it
reduces to Raoult's law as a result what do we find? We find that the f’s can be replaced by
P’s the pressure terms. As a result of which we find that for low to moderate pressure it
becomes Pi equals to xi Pi saturated and in this case Pi equals to yiP, okay.

Recalling that for the vapor phase at low to moderate pressure or in other words limiting P
tends to 0 f of pure component of the vapor phase this is the total pressure if you recollect and
for the liquid phase what did we derive? We derived for the liquid phase this was equal to P
saturated and then there was some other correction terms taking into account the corrections
would account for the non-ideality of the vapor phase as well as the compression of the liquid
from the saturated pressure to the pressure of interest which was incorporated in an
exponential term of an termed as the poynting correction factor, right?

So therefore for low to moderate pressure this was the situation and this is the rule according
to which we can write down we can combine the 2 and we can write it down in this
particular form for an ideal solution. Then we went for a nonideal solution and according to
this we also found out that if we plot pressure versus x1 then in that case what do we expect?
We get the total pressure curve as something of this the P2 versus x1 curve like this and the
P1 versus x1 curve in this particular fashion. We found that all the 3 they are linear because
partial pressure of component 1 partial pressure of component 2 as well as the total pressure
all of them vary linearly with the mole fraction of either of the component.

It’s important to remember that when we are plotting these phase diagrams we are always
plotting the mole fraction of the more volatile component in the x-axis it can be the mole
fraction of the more volatile component in the liquid phase or in the vapor phase but just by
convention we try to maintain the fact or rather maintain the practice that we will be plotting
the more volatile component composition in the x-axis.

You will often come across problems maybe I can also give you can get such problems in
the textbooks which you are referring and you’ll find that in those problems often we find
that the less volatile component composition is used to plot in the X axis. It’s interesting to
remember that the curve remains same only when we are plotting the less volatile component
quite natural its vapor pressure will be less. As a result of which under that condition what are
we going to get?

762
For that condition if it is x2 or y2 then in that case we will be getting something like this. The
curve will just change because it is important to remember that the less volatile component
will have a lower vapor pressure as compared to the more volatile component this is the only
thing which is going to happen, right?

(Refer Slide Time: 6:27)

After this what we did? We went to nonideal solutions and in nonideal solutions what did we
find? We found out that in nonideal solutions there can be either positive deviations from
ideality or there can be negative deviation from ideality, okay. When we have a positive
deviation from ideality the total pressure calculated is greater than that calculated from
Raoult's law when there is a negative deviation from ideality the total pressure calculated at
any particular composition is less than that predicted from Raoult's law.

763
(Refer Slide Time: 7:10)

As a result it was felt necessary to modify the equation which we had derived this equation or
this equation for ideal solutions in order to incorporate nonideal behaviour.

(Refer Slide Time: 7:22)

And this was done by introducing a term yi in the by introducing a term gamma i in the
equation of Lewis Randall equation.

764
(Refer Slide Time: 7:42)

And for low to moderate pressure what happen? The equation reduces to Pi equals to gamma
I Xifi liquid if it is a liquid phase and it is also important to remember that mostly non-
ideality they exist in liquid solutions, gaseous mixtures are usually ideal, why? Because the
Lewis Randall rule is based on Amagat’s law of additive volumes which I have mentioned
several times in my previous classes.

(Refer Slide Time: 8:09)

So therefore for non-ideal solutions we find that - thee non-ideality is taken care of by
introducing this particular activity coefficient in the Lewis Randall rule and we found out that
once can find out gamma i then the vle behaviour of nonideal solutions and ideal solutions
are naturally the same.

765
(Refer Slide Time: 8:30)

The only difference is that in this particular equation we will be having a gamma i here there
is no other change. So therefore we have introduced one other property of component i when
it is in solution. So the next interest was to express Gamma i in terms of known parameters or
in terms of parameters which we have already discussed.

(Refer Slide Time: 9:00)

In the last class we have made those particular derivations and we found out that the ln of
gamma i it is nothing but equal to the partial molar excess Gibbs free energy of component i
in solution, right? You also found out that the variation of gamma i with temperature can be
related to the partial molar excess enthalpy of component i in solution and its variation with
pressure can be related to the partial molar excess volume of the component i in solution.

766
So this was what we had derived in the last class and after this, what we did was we derived
the Gibbs Duhem equation in terms of activity coefficient and I stopped there mentioning that
particularly for binary solution the Gibbs Duhem equation gives us large number of important
information.

(Refer Slide Time: 9:59)

So let us start from that particular point and see what are the important information’s that we
have, okay. So therefore I believe I had derived rather we have derived this particular
generalized equation Gibbs Duhem equation in terms of activity coefficient and this was the
equation which we had derived and in this particular equation at constant temperature and
pressure naturally this becomes 0 this becomes 0 and this is the Gibbs Duhem equation which
for a binary mixture reduces to this.

767
(Refer Slide Time: 11:03)

Now let us see what important information we can get from this equation? Say for instance if
I write it down as, I will just write down and derive it once more X1 d ln gamma1 plus X2 d
ln gamma 2 equals to 0, can we not write it down as X1 del ln gamma 1 del X1 at constant
temperature, pressure plus X2 del ln gamma 2 del X1 at constant temperature, pressure this is
equal to 0.

Or in other words del ln gamma1 del X1, is this not equal to minus X2 by X1 del ln gamma2
del X1 at constant T, P. We know that since it is a binary mixture d X2 equals to minus dX1.
So this can be written down as minus X2 by X1 del ln gamma2 del X2 at constant
temperature and pressure automatically what do we see? We see that the first thing is slope of
ln gamma1 versus X1 is just opposite to the slope of ln gamma2 versus X1 this is the first
thing that we see here right number 1.

Number 2, can we write down this particular equation? So let me write down the first thing is
slope of ln gamma 1 versus X1 opposite to slope of ln gamma2 versus X1. What is the next
thing that we know from here? We know that as X1 tends to 0, what happens you tell me? X1
tends to 0 means X2 tends to 1. Naturally under that condition what happens? Your gamma 2
tends to 1 and ln gamma 2 tends to 0.

So therefore what do we see? We see that when we are plotting gamma 2 versus X1 then at
X1 equals to 0 the gamma 2 becomes equal to one and similarly we also see at X1 tends to 1
or X2 tends to 0, what happens? Gamma 1 tends to 1 and ln gamma 1 tends to 0, quite

768
natural whichever component is in excess it’s natural that Gama 1 tends to 1 at X1 tends to 1
and gamma 2 tends to 1 at X2 tends to 1, fine.

(Refer Slide Time: 13:40)

What is the other thing that that we can observe from here? Just observe and see from the
equation which I have written down. The equation implies that, let me write down the
equation d ln gamma 1 d X1 constant T, P this is equals to minus X2 by X1 del ln gamma 1
del X1, right? Now or in other words if I write it down in this way also X1, can I not write it
down as X1 by gamma 1 d gamma1 d X1, is this not equal to minus X2 by gamma 2, d
gamma 2 d X1, can I not write it in this particular form?

I can write it down or else I can also write it down as, in this form also I can write it down
that 1 by gamma1 d gamma1 d X1, is this not? Because what happens, is this not equal to 0 at
X2 tends to 0 or X1 tends to 1, what does this imply? It implies gamma1 versus X1 has a
horizontal slope at X1 tends to 1, fine. So therefore there are 3 observations which we find
from here.

769
(Refer Slide Time: 15:29)

What are the 3 observations? The first thing if we observe the equation this particular
equation, what do you find number 1 is gamma 1 and gamma 2 they are not independent of
one another. So therefore they are more or less related to composition in this particular
fashion.

(Refer Slide Time: 15:44)

Next is that if we know one we will be in a position to find the other in what particular way
we will be doing a problem and finding this out that suppose we know one say X1 d ln
gamma 1 plus X2 d ln gamma 2 equals to 0. So suppose I know gamma 1 I would like to find
out gamma 2. So what is this d ln gamma2 minus X1 by X2 d ln gamma1, right? Now I
would like to find out gamma 2 at some particular composition X2.

770
Starting from X2 equals to 0 and here this also X2 equals to 0 to X2. So therefore ln gamma 2
at X2 minus ln gamma 2 at X2 equals to 0 is nothing but 0 to X2 minus X1 by X2 d ln
gamma 1. So therefore suppose I know gamma1 or gamma1 variation with X1 then we can
find gamma 2 gamma 1 this is number 1. Suppose we take up a problem in this respect, say
for instance I take up a problem I know that in a binary mixture say suppose the problem is
that I have got a binary mixture, right?

In the binary mixture I know gamma1 is represented by an equation as ln gamma1 equals to

AX2 square, where A is a constant, fine. I would like to find out or I would like to express
obtain expression for gamma 2 in the same solution. So therefore I have a binary mixture of
component 1 and 2 gamma1 adequately represented adequately represented by this particular
equation where A is a constant obtain expression for gamma 2 of component2 in the same
solution, how do we think we should be proceeding for this?

Just if you if you look at this particular equation we can we can very well find it out from
here. We know that d ln gamma1 we already know that d ln gamma 1 dX1 this is equals to
minus X2 by X1 d ln gamma 2 dX1, right? This is nothing but as I have mentioned this is
Gibbs Duhem equation. So from here what do we know? We know ln gamma1 it’s given it’s
already AX2 square which can be written down as A into 1 minus X1 whole square, right?

So therefore in that case what is d ln gamma1 d X1 equals to? It is minus 2A into 1 minus
X1, agreed? So therefore I know this particular term, right? So what is d ln gamma2 by dX1
we can simply substitute this here and we can find it out.

(Refer Slide Time: 19:09)

771
So therefore this reduces to minus 2A into 1 minus X1 equals to minus X2 by X1 d ln
gamma2 d X1, this is nothing but X2 this is also X2 here it is minus here also it’s a minus, so
the minuses they cancel out, what do I get? I get 2A x1 is nothing but d ln gamma2 d X1,
agreed? So I would like to find out an expression for gamma 2. So we know d ln gamma 2 is
nothing but 2A X1d X1, fine. We will be performing the integration from X1 equals to 0 to
X1 equals to X1, right? And we know that at X1 equals to 0, what is the value of gamma 2?
X1 equals to 0 means X2 equals to 1 and therefore at X2 equals to 1 Gamma 2 has to be
equal to 1. So therefore this is going to extend from ln gamma 2 has to be 0. So therefore this
has to extend from gamma 2 equals to 1 to gamma 2.

As a result of which if you perform this integration what do we get? We get ln gamma2 this is
nothing but 2 into A X1 square by 2 which is nothing but equal to AX1 square. We find that
we will be dealing more with these with such type of equations.

(Refer Slide Time: 20:23)

We find that more or less then in that case your variation of ln gamma1 with composition

772
(Refer Slide Time: 20:28)

and ln gamma 2 with composition they are more or less symmetric to one another, agreed? So
therefore just imagine I had started from or rather I had the expression of gamma variation
with composition for any 1 component in a binary mixture and we found out since we knew
Gibbs Duhem equation using the equation we could find out the variation of the activity
coefficient of the other component as a function of composition.

So in this particular way this is just one very typical example can take up a lot of examples
and you will find that for most of the situations if you know the gamma variation for 1
component in whatever be the functional form you can find out the gamma variation for the
other component this is for a binary mixture if it is just like what I had repeated in the last
class. If I if I’m dealing with multi-component mixture then if I know the gammas for the n
minus 1 component I can find the gamma for the nth component using the Gibbs Duhem
equation. So this is the first thing which is very important.

773
(Refer Slide Time: 21:41)

The next thing as I have already mentioned the next thing is when we know that gamma 1
tends to tends to 1 at X1 tends to 1. So therefore we know that at X1 tends to 1, suppose I
would like to plot gamma or gamma1 as a function of X1, what do I know? That at X1 equals
to 1 gamma1 also has to be 1, right? Suppose we start and suppose say there is a positive
deviation from Raoult’s law, so therefore the equation should be something of this
sort.Similarly for gamma 2 the equation should be something of this sort.

(Refer Slide Time: 22:19)

What is the other thing that I know? The other thing which I know is that at X1 equals to 1
year d gamma1d X1 the slope 0 or in other words the curve should be horizontal.

774
(Refer Slide Time: 22:27)

Or in other words we know that at this particular point when we are plotting gamma 1 the
gamma 1 has to be 1 at X1 equals to 1 and the curve has to be horizontal at this particular
point the same applies for gamma2 at X1 equals to 0, right?

(Refer Slide Time: 22:54)

So therefore these are some of the important information which the Gibbs Duhem equation
provides to us, right? So I had already mentioned I have also shown the plots gamma
variations with composition if you observe this particular thing as I have told you gamma1
tends to 1 as X1 tends to 1, this is true whether we have a negative variation from deviation
from Raoult's law or a positive deviation from Raoult's law.

775
That we have a positive deviation the value of gamma is greater than one, so therefore we
find that at X1 equals to 1 gamma1 is equal to one. In this case also as X1 equals to 1
gamma1 equals to 1 number 1. Number 2 the slope is horizontal here for gamma1 horizontal
here for gamma 2. Here also the slope is horizontal here for gamma 2 the slope is horizontal
here for gamma 1.

So therefore these are the 3 important this was the first observation which we had slopes
equal and opposite the second observation was this and the third observation is this. And the
fourth thing I have already mentioned I have done a problem and shown you that how gamma
1 or rather how, if you know gamma variation of 1 component with composition how the
gamma variation of the other component can be found out.

(Refer Slide Time: 24:26)

So therefore with this I think we have discussed sufficiently regarding the importance of
gamma, now we have to go for finding out and we know that that this also I have repeated we
know that once we can find out gamma then we will be able to tackle the vapor liquid
equilibrium data and gamma as we know it is a function of temperature, pressure, the liquid
phase composition very importantly which is not mentioned here it is also a function of the
vapor phase composition.

Because the gamma has to be obtained from this particular equation which deals with both
the liquid phase as well as the vapor phase. It is important to remember that although gamma
is a property of the liquid solution in order to find it out we need to have data on both the
liquid phase as well as the vapor phase composition.

776
(Refer Slide Time: 25:11)

Now how to find out gamma? What are the different normal ways of finding out gamma,
right? So usually what we do is, the problem types yesterday also I have told you that these
are the different problem types and whatever problems are given these are the equations
which have to be used to solve them.

(Refer Slide Time: 25:31)

We have already done this for ideal solutions the extra thing which needs to be done here is to
incorporate the or rather is to know how to calculate gamma from known input parameters.

777
(Refer Slide Time: 25:34)

So well the thing which is usually done is that see it’s mostly gamma data it’s very difficult to
find out how a real solution actually behaves or rather how gamma in the real solution can be
related to the properties of the real solution, why? Because see for a real solution when it is
deviating from ideality why it happens. Firstly the size and shape of the molecules can be
different, so as a result of which there is a non-vanishing volume of mixing.

Second thing is the intermolecular interactions as I have said their different for the like pairs
and the unlike pairs. So therefore actually if we have to find out a particular model in order to
predict gamma from the liquid phase compositions then in that case it makes a very good
knowledge of the intermolecular potential etc and this is not very easy. So therefore usually
the things which are done are either a semi-empirical or empirical correlations are used.

Well, and these empirical correlations they are developed from experimental data. Now
whenever you’re developing any empirical correlations from experimental data we need to
remember certain things, what are those certain things? The first thing is that it must
subscribe to certain limiting conditions number 1. Number 2 is that more or less whenever we
have proposed something from experiments it is very important to test whether it is
thermodynamically consistent and that we will be discussing later it will be done with the
help of Gibbs Duhem equation.

Now let us see what is the easiest model which can be proposed which subscribe to the
limiting conditions of a real solution and can be thought of describing the real solution
behaviour or it can be related to the composition in the real solution. Usually these particular

778
models or the equations which are proposed they are proposed from or rather they usually
express the excess Gibbs free energy of the solution in terms of composition.

(Refer Slide Time: 28:08)

What I mean to say is, that in this particular case what we do is we usually instead of directly
expressing ln gamma1or ln gamma 2 as a function of x1 x2 usually what we prefer to do is,
we would prefer to express Gibbs free energy as a function of x1 x2 once this can be
expressed we know that ln gamma 1 is nothing but g1 bar E by RT and this can very well be
found out by expressing your del del N1 of N1 gE by RT at constant temperature, pressure
and N2, right?

So usually we try to express or rather we try to find out activity coefficient from some
particular models which are known as excess Gibbs free energy models which express excess
Gibbs free energy of the solution as a function of its composition, why we do it? Because we
really do not know that or rather it is not very convenient to find out the limiting situation or
limiting condition for ln gamma1 and ln gamma 2.

We just know that ln gamma1 must be equal to 0 at X1 equals to 1 this is the only thing that
we know ln gamma 2 must be equal to 0 at X1 equals to or rather at X2 equals to 1 this is
only thing that is just one limiting condition. On the other hand when we’re dealing with
excess Gibbs free energy models what do we know? We know that the excess Gibbs free
energy should disappear both at both when either of the component is in excess.

Or in other words the limiting condition in this particular case is this tends to 0 at X1 tends to
1 at X1 tends to 0 or in other words X2 tends to 1 under both these conditions. So therefore

779
the simplest expression which expresses the excess Gibbs free energy in terms of composition
should become equal to 0 either when X1 becomes equal to 0 or when X2 becomes equals to
0. So therefore what can be the simplest expression?

Simplest expression can just be this a product of the compositions where A is parameter
which has to be determined experimentally and this particular equation this is the simplest
Gibbs free energy model this particular expression it is known as 2 suffix Margules equation.

(Refer Slide Time: 31:14)

And here I have shown you that for this particular condition what should be the characteristic
of the Gibbs free energy the excess Gibbs free energy as a function of composition and the
characteristics of ln gamma as a function of composition. If you observe you find that in this
particle figure we find that g by RT it is symmetric about 0.5, why? Because we have taken A
equals to 1 in this particular case, right?

780
(Refer Slide Time: 31:50)

And this shows the corresponding variation of gamma1 and gamma 2 as a function of
composition, just note the difference with the previous curves which I have shown you the
previous curves it was just gamma versus X1 as a result of which we have obtained curves of
this sort.

(Refer Slide Time: 31:58)

In this particular case it is ln gamma versus X1, so therefore we find that ln gamma reduces
to 0 at X2 equals to 1, ln gamma reduces to 0 at X1 equals to 1, right? So therefore we find
that the simplest equations which can be used in order to find out the excess Gibbs free
energy is nothing but the simplest expression is a product of the 2 compositions.

781
(Refer Slide Time: 32:35)

And from here if we find out the expressions for ln gamma 1 and ln gamma 2 which as I
have told you we can find it out from this particular expression we will find that in this case
ln gamma 1 equals to A X2 square ln gamma 2 equals to A X1 square.

(Refer Slide Time: 32:57)

e
So this is the simplest equation and for this particular equation we will find that while for A
equals to 1 while g by RT is symmetric about X1 equals to 0.5 you will find that the activity
coefficient curves they are mirror images when ln gamma is plotted against X1 and this
particular parameter A this can be 0 when it reduces to an ideal solution this can be greater
than 0 or this can be positive which signifies g by RT is positive or a positive deviation from
Raoult's law.

782
This can be negative which signifies a negative deviation from Raoult's law and this
particular
equation which is known as the 2 suffix Margules equation as has been mentioned here this
particular equation we find this is quite accurate or it is adequate to represent the behaviour of
liquid solutions when the components are of similar shape, the similar size and similar nature.

And for such cases it is good but for most of the situations we will find that the case is
slightly more complex and we need to modify the simplest expression which we have shown
in order to predict the behaviour of slightly more complex liquid solutions. We are going to
deal with it in the next class.

783
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 43
Non Ideal solutions (Contd.)

(Refer Slide Time: 0:29)

Well to continue with our discussions on non-ideal solutions where I had ended the last class
was I had told you that usually since it is very difficult to have a proper equation of state to
predict the behaviour of real solutions, so therefore usually the gammas they are predicted
from empirical or semi-empirical equations and the simplest empirical equation which is used
is known as the Margules 2 suffix equation which gives you g by RT in this particular form.

(Refer Slide Time: 1:02)

And from there we find out that more or less my ln gamma 1 and ln gamma 2 they can be
represented in this particular form gE by RT this is equals to A X1 X2 and from there if we

784
are finding out ln gamma 1 which is nothing but gi bar E by RT which is del del N1 N1 plus
N2 gE by RT and if you differentiate it you find that ln gamma 1 equals to AX 2 square
similar way you can either differentiate it in terms of N2 and find out ln gamma2 or you can
use the Gibbs Duhem equation once you know gamma 1 to find out gamma 2 the choice is
yours you can do either of them and we find that ln gamma2 equal to AX1 square this in
particular case.

(Refer Slide Time: 1:55)

And we have also depicted the curves of excess Gibbs free energy of a solution as a function
of composition and also the ln gamma1 and ln gamma2 as a function of composition and I
had already also mentioned in the last class that although the 2 suffix Margules equation it is
good for very simple spherical molecules similar type of attractions etc but for most of the
cases then we need to modify this particular equation in order to incorporate the non-ideal
behaviour of solutions.

785
(Refer Slide Time: 2:25)

Now what are the different ways by which it is done? One particular way was proposed by
Redlich and Kister, what they have done is they have they have simply extended the number
of terms here if you observe. For this case if you observe that the first equation or the first-
term it represents the 2 suffix Margules equation and the other terms they are they are just the
series expansion in X1 minus X2 these are introduced if you observe closely you will find
that it has got some similarity with the virial equation of state modifying the ideal gas
equation and depending upon the complexity of the situation you can keep on increasing the
number of terms in this particular case.

And in that way the number of parameters to be included in the model depends upon the
complexity of the solution and if you observe these terms you find that the A it is symmetric
in x just the A of the 2 suffix Margules equation, B if you observe B it is actually it is
asymmetric in x and it tries to skew the parabola either to the left or to the right and if you
observe the third term you will find it is again symmetric in x, what it tries to do? It tries to
either sharpen the parabola or it tries to flatten the parabola.

So in this particular way the contribution of the first 3 terms ABC of the Redlich (ki) Kister
equation is shown in this particular in this particular graph where we see that the first-term it
is the contribution to excess Gibbs free energy due to 2 body interactions of molecules of
unlike components and the second and third terms, these are the second and the third terms
they refer to the 3 body interactions of unlike molecules, right?

786
And so therefore if we have expressed g by RT by this equation the activity coefficients can
automatically be expressed in the in the way we have shown and from there we can express
ln gamma 1 by gamma 2.

(Refer Slide Time: 4:47)

We can express this by dividing one with the other we can this ln gamma 1 and ln gamma 2
it can be expressed in this particular equation and we can we can we can subtract the 2 and
we can find out how we can express ln gamma 1 and ln gamma 1 by ln gamma by 2 this can
be expressed as A into x2 minus x1 plus B 6x1 x2 minus 1plus C x2 minus x1 into 1 minus 8
x1x2 plus so on and so forth and all these terms ABC they are temperature dependent
adjustable parameters because we need to know that the composition dependence of gamma
that should be a function of temperature and pressure. Since it is a liquid phase property it
should not be a very significant function of pressure but it should be a significant function of
temperature.

So therefore this ABC they are temperature dependent adjustable parameters to be

determined from experimental data, right? So therefore we have some experimental data on
the real liquid solution from which we can determine them and the number of terms which we
need to include as I have already said that depends upon the complexity of the solution. So in
that particular way we find that the Redlich Kister equation it is quite flexible.

And it can adequately represent the excess Gibbs free energy of the solution. We just need to
include more and more terms if we have to make it more complex or rather when we have to
deal with more complex situation.

787
(Refer Slide Time: 6:45)

Now there is one other way also apart from the Redlich Kister equation it is the Wohl’s
equation. These Wohl’s equation also it takes into account the 2 body and the 3 body
interactions and it gives us some type of rather it helps us to predict the excess Gibbs free
energy of a binary liquid solution by taking into account the three body and the two body
interactions between the like and unlike molecules as we have shown here this first-term
takes into interaction the 2 body interactions and the subsequent terms they take into account
the 3 body interactions in this particular case.

(Refer Slide Time: 7:40)

Now based on all this we find that several algebraic equations have been proposed to
determine activity coefficients. Since from thermodynamics it's rather there is a sufficient

788
guidance to formulate between gamma and x. So therefore large number of algebraic
equations are proposed which satisfy Gibbs Duhem equation they are thermodynamically
consistent and they are used or rather they are just a correction to the 2 suffix Margules
equation in order to predict gamma under more non-ideal conditions.

For very less non-ideality 2 suffix Margules equation is fine when we have to go for much
further or rather when there is greater amount of non-ideality we need little more equation the
most generalized formulation to predict non-ideal behaviour is has been proposed by Redlich
Kister and Wohl has also proposed and both of them they have considered 2 body 3 body
interactions that they have tried to propose. That we find that apart from that if we use some
algebraic equations which are basically empirical equations they do not have much
thermodynamic foundation but they are (thermody) there found to be thermodynamically
consistent number 1.

And they subscribe to the limiting condition which I have mentioned then in that case those
equations they are usually used for finding out the activity coefficients the most common
equations which will be important for this particular thermodynamic course are the 3 suffix
Margules equation and the Van laar equation these 2 equations they are the most common
equations where the Gibbs free energy excess Gibbs free energy model is given by these
equations and the ln gamma1, gamma2 expressions are given from here.

(Refer Slide Time: 9:53)

Now apart from this we find that these equations they are fine but they do not incorporate the
temperature dependence of gamma. So therefore in order to incorporate the temperature

789
dependence of gamma we have some more equations which have been suggested for example
Wilson equation is 1 where the adjustable parameters in this particular case the gamma sorry
the lambda 1 to and the lambda 2 1 they are functions of temperature and we find that this
particular equation the Wilson plot this is very good for mixtures where the components
differ from each other in molecular size and interactions between the like and unlike
molecules are different.

(Refer Slide Time: 10:21)

So we find that the Wilson’s plot its I will just like to tell you that for this class the 2 suffix
Margules equation that 3 suffix Margules equation as well as the Van laar equation are
important, for the other equations you need not remember the equations but under what
conditions those equations have been proposed that you need to remember because in future
if you are supposed deal with non-ideal solutions it is important that you (ac) actually know
that for which conditions which equation is applicable but along with that also remember that
till today also several equations are being proposed to predict the excess Gibbs free energy
and the activity coefficients of (Ide) sorry non-ideal solutions because it is very evident that
suppose we are working with a mixture of hydrocarbons we find one equation is applicable
definitely the same equation will not be applicable when you are working with aqueous
solution of organic acids.

So naturally with each different compound or each different component of the solution newer
and more and more different equations are necessary because all of them finally boil down to
the intermolecular interactions and the molecular shapes and sizes which are different for the
different cases, right?

790
(Refer Slide Time: 11:40)

But again I repeat I will just be discussing under what condition the more complex equations
are applicable but for this particular class we will be dealing with just 3 equations the 2 suffix
Margules equation, the 3 suffix Margules equation as well as the Van laar equation, fine. So
for most of the cases which are required for you these 3 equations are fine but it is important
to remember that they do not or rather they do not express the constants as temperature
dependence. For them the same particular constants of the equations are applicable no matter
or at what particular temperature and pressure conditions that we are working, fine. So
therefore in order to incorporate the temperature (dep) dependence of activity coefficients we
need to go for more and more complex equation.

791
(Refer Slide Time: 13:15)

One such equation is the Wilson equation which we where we find that the Wilson equation it
is applicable for 2 conditions one is, the components differ from each other in molecular size
and secondly interactions between like and unlike molecules are different, right? For these 2
conditions more or less they are applicable because the lambda 12 and lambda 21 they are
related to the pure component molar volumes and characteristic energy differences etc.

So therefore the advantage is the third advantage is, it predicts temperature dependence of
activity coefficients gamma1 and gamma2 and therefore these are developed through
concentrations of molecular behaviour the only disadvantage which it has is, it is not
applicable to systems where ln gamma exhibits maxima or minima this particular aspect we
will be dealing separately after we have we have finished dealing with the normal situations.
And it is also unable to predict limited miscibility these are the drawbacks of the Wilson plot.

792
(Refer Slide Time: 14:39)

Well, other than that what we have? We have a 4 suffix Margules equation which tries to
modify further the 3 suffix equation by incorporating it's it’s basically a Redlich Kister
equation which has 3 instead of 2 terms it is evident that we have a non-random two liquid
equation the NRTL equation this is applicable to partial miscible and completely miscible
systems and in this particular case also we find that here we do not have much advantage over
the Van laar or the Margules equation for moderately non-ideal solution but it is it represents
the excess Gibbs free energy of strongly non-ideal and partially miscible solutions quite
satisfactory.

(Refer Slide Time: 15:22)

793
After that we find that apart from these there are some other equations also which are which
have been proposed. For example we have the UNIQUAC which is nothing but the universal
Quaci chemical equation and these equations we find that in fact they have been developed
by Prusnitz and he was one of the contributors to this equation we find that they it primarily
comprises of a combinatorial part and the residual part.

And we find that combinatorial part this accounts for the compositions the size and shape of
the constituent molecules. So the combinatorial part it basically it comprises of pure
component properties only and the residual part if you see that accounts for intermolecular
forces and it contains 2 adjustable parameters.

And after that we find apart from the UNIQUAC equation it we there are also some other
types of equations which have been proposed they are known as the group contribution
method, what are these group contribution method? See if you assume that each particular
component of the solution are made of individual molecules then there are huge number of
molecules and then for each particular solution you have to find out the how the non-ideality
is coming into picture? Other than that if you assume that each particular solution is made up
of a number of radicals or number of groups of molecules and it is the contribution of the
groups which lead to a non-ideality of the solutions then in that case we can have large
number of solutions which have the same type of groups.

For example Methyl group can be present in a large number of liquids or maybe the acetate
group can be present in a (la) acetic acid in a large number of acetate and we assume that the
behaviour of these particular groups they are independent of the behaviour of the other
molecules. Therefore in this particular case the advantage is if you know the group
contribution for the group existing in one solution you can safely assume that this particular
group contributes in a similar way when it is in a environment of some other type of
molecules.

794
(Refer Slide Time: 17:50)

So therefore the other way of finding out or evaluating the activity coefficient this is known
as a group contribution method, right? Where we assume that each component it is made up
of not only individual molecules but different functional groups like say the functional groups
can be (func) each particular component of the solution are made up of functional groups like
CH3,OH etc and then a solution of the component is treated actually as a solution of the
groups.

And then the gamma i’s they are determined by properties of the functional groups rather than
the properties of the molecules. So this is the basic idea that we have several thousands of
chemical compounds but there are very few functional groups constituting these compounds.
So that the gamma in large number of solutions the basic idea is that the gamma in large
number of solutions can be calculated from parameters characteristic of functional groups and
the only assumption here is that it just assumes the only assumption here of the group
contribution method is that contribution of 1 functional group in a molecule is independent of
that made by other functional groups, right?

So therefore on this basis it expects that this makes our life little more simpler because if we
know the contribution of 1 functional group in any molecule we can safely assume that the
same contribution is made by the same functional group in a different molecule and there are
2 models which use this particular method one is the analytical solution of groups which is
known as the ASOG and it is the Uniquac functional group activity coefficient or it is known
as the UNIFAC.

795
So therefore I would not like to go into much details of these particular things. It is as
important for you to remember that the large number of different algebraic equations, large
number of models which have been proposed, the primary reason for the huge number of
models are that it is very difficult to use one particular model for the wide range of solutions
that we have considering the different shapes, sizes, orientation and the intermolecular
interactions between the different molecules constituting the solutions and it is important to
remember that may be one particular molecule it behaves in a in one particular way in an
environment and behaves in a completely different way in a different environment.

So therefore depending upon the conditions different equations have been solved on the sorry
have been proposed and when we find that the molecules are not (co) completely randomly
mixed possibly due to hydrogen bonding etc then we go for the NRTL. When we find that
gamma should be is not does not remain constant with temperature so in that case we have to
we should go for Wilson plot etc but again I repeat that despite the large number of models
which are present for your particular case you will be primarily concerned with 2 suffix
Margules equation 3 suffix Margules equation and Van laar equation. So before I end this
class I would like to discuss the estimation of gamma from using the 2 suffix is very simple,
using the 3 suffix and the Van laar equation for that what we do we try to arrange the
equation in such a form that it gives is very simple or rather the method of solution becomes
very simple.

So I would like to discuss this particular method of solution before I end this class. So that if
you are given some amount of data thermodynamic data say at a constant temperature usually
the data given as X or Y or X and Y at either isothermal conditions or under isobaric
conditions, right? And you are required to test the thermodynamic consistency of data or
suppose you have data at one particular temperature or one particular pressure you are
required to generate that the whole set of data or the whole set of gamma data for the solution
with the compositions varying from almost 0 to or X1 varying from 0 to 1.These are
generally the type of problems that you are supposed to deal with, right?

796
(Refer Slide Time: 23:20)

So for that case what do we do? We are going to take up the Van laar equation and Margules
equation and discuss the method of solution. So let us take up either of the equations and see
in what way we can express it. For example suppose we are taking up the Van laar equation,
what is the equation? If we write down the equation we will find ln gamma 1 is say suppose I
have you can since they are constant you can express them in any particular way it can be AB
it can be A12, A21 etc.

For my case what I do here is, I have expressed Van laar constants as B12, B21 and the 3
suffix Margules constants as A12, A21 just to differentiate that they belong to constants for
different equations. So the Van laar equation is ln gamma 1 is 1 1 plus B12 X1 by B21 X2
whole square, right? ln gamma 2 this is nothing but B21 by 1 plus B21 X2 by B12 X1 whole
square. Let me see the 3 suffix Margules equation ln gamma 1 X2 square into say A12 plus 2
into A21 minusA12 into X1.

ln gamma2 equals to X1 square A21 plus 2 into A12 minus A21 into X2, right? Now if you if
you observe these 2 equations, what do you find? You find that for this equation if we write it
down as ln gamma 1 by X2 square this becomes A12 plus 2 intoA21 minus A12 into X1.
Similar way ln gamma 2 by X1 square this becomes equal to A21 plus 2 into A12 minus A21
into X2, sorry into X1. So therefore does it not imply that if I have plot of ln gamma 1 by X2
square versus X1 then in that case it should be a straight line and that particular straight line it
should (ha) it should have a intercept of A12 at X1 equals to 0, fine.

797
So therefore from this particular straight-line I should be in a position to find out this. Similar
way I can also plot ln gamma2 by X1 square as a function of X2 and I can also generate a
straight line from this straight line from the intercepts at X2 equals to 0 and X2 equals to 1 I
can find out A21 and A12, right? So in this particular case what do I find? That at X1 equals
to 0 my intercept is A12 and at X1 equals to 1 my intercept is A21.

In this case we find that at X2 equals to 0 my intercept is A21, here my intercept is A12,
agreed? So therefore we find out that from here is we have data at one particular temperature
or pressure X X data and (gam) gamma versus X data then you can very well plot a straight
line from the intercepts we can find out A12 A21. In a similar way we can also arrange this
particular equation just observe this equation and try to find out in what way we can arrange
this equation such that from such similar simple geometrical constructions we can find out
the constants B21 and B12 we are going to discuss it in the next class before the next class try
to think out, how we are going to express the Van laar equation? Such that we can have such
similar (geom) such a simple geometric form from where just by construction or just by
plotting one parameter against another we should be in a position of finding out the constants
of the equation.

This is the first thing I want you to think, the second thing is suppose we are using gamma 1
and gamma 2 and plotting in this particular fashion then what do we find? This equation is
going to give me one set of A12 A21 this graph is going to give me another set of A21 A12,
how to combine the 2 and how to find out the average value that should be used? So with
these 2 assignments for these 2 thoughts or rather these 2 topics for you to think I end this
class we are going to discuss about this the next class, thank you very much.

798
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 44
Non Ideal Solutions(Contd.)
(Refer Slide Time: 0:44)

Well to continue our discussions on nonideal solutions or more specifically on the van laar
and the 3 suffix Margules equation we were trying to find out, how to find out activity
coefficients using these 2 before I start I would like to mention once more that both the van
laar and the 3 suffix Margules equation they are firstly simple they contain only 2 adjustable
parameters and they are used extensively to correlate gammas.

Now amongst these 2 it is important for you to remember that Margules equation it is usually
used for symmetrical systems when A12 is almost equal to A21 and Van laar equation it is
used usually for unsymmetrical systems these are just as guides as when you should apply
which one and it is also important for you to remember that van laar equation although it is
used for un-symmetrical systems it is not applicable for ln gamma exhibiting Maxima or
minima under this condition van laar equation is not applicable, right?

And we find that since both of these equations this van laar and 3 suffix Margules they
contain only 2 adjustable parameters. So therefore we find that the constants which are
nothing but A12 and A21 in one case or the B12 and B21 in the other case they can be
determined from a single measurement from a single measurement of VLE composition plus
P1 saturated, P2 saturated data, so we will we are we will be doing a few problems shortly to

799
just to show you how from a single measurement of VLE composition along with either the
Antoine constant or data on saturated vapor pressure of component1 and 2.

Just for a single measurement we can find out these constants, we can find out these constants
as a result of which we can generate gamma values using either the van laar or the 3 suffix
Margules equation, right?

(Refer Slide Time: 3:23)

Now coming back to our original discussion, what did we find out in that particular case? We
had written down the van laar and the 3 suffix Margules equation in a term which will be
friendly or rather in a form where we can find out the constants very easily. So we found out
that if we write down the 3 suffix Margules equation in this particular form then we find that
that more or less ln gamma1 by X2 square should be linear with X1 and from the intercepts at
the 2 extreme positions of X1 equals to 0 and X1 equals to 1 we should be in a position to
find out 2 constants, right?

Same thing applies for gamma 2 ln gamma2 by X1 also. Now I had given you a problem if
you remember that it’s always possible that A12 A21 determined from this particular graph
may not be same as the A12 A21 determined from this particular graph. So therefore under
that condition what do we do? We try to combine ln gamma1 and ln gamma2 and try to
generate a single equation from where we can find out A12 and A21.

800
(Refer Slide Time: 4:42)

What is the way of combining ln gamma1 ln gamma2? If you recall the original expression of
g by RT, what was it? g by RT was equal to if you remember it is nothing but X1 ln gamma1
plus X2 ln gamma2, isn’t it? So what can we do for this particular case? We can write it down
as if you substitute the expressions of ln gamma1, ln gamma2 then in that case we find that
this will be equal to X1X2 A21 X1 plus A12 X2, right?

And from there we find that if we have combined these than in that case we can plot your
suppose we plot gE by RT, X1 X2then what do we have suppose we are plotting it in terms
of X1 g by g by RT X1 X2 this whole thing taken on this particular site. So in that case what
do we have? We have a single equation from which we can find out A12 and A21 just from
the 2 extremes.

So therefore this gives us A12, this gives us A21 an average A12 and A21 constituting the
variation of both gamma1 and gamma 2 with composition, right? So what do we do from
here? We can if we have one data on X1Y1 etc and then from one set of data we can find out
these 2, we will be doing a problem shortly.

801
(Refer Slide Time: 6:32)

And once or rather I can show you the problem in this particular case it is a ethanol it is
ethanol benzene problem at 40 degrees centigrade you are required to generate theoretical
VLE data, say for example you are you are required to use the 3 suffix Margules equation.
You know P10 or rather you know P1 saturated you know P2 saturated you have data on X1
you have data on P, right? And you’re supposed to find out the y1 values using Margules
equation as well as van laar equation which we will discuss shortly.

And then you need to compare with the experimental data, fine. So therefore you already
know P1 saturated P2 saturated, now from here what do we know? You can find out gamma1
equals which is y1P by x1 P1 saturated. So therefore you can find out the individual gammas
in this particular case for example say for instance the gammas in this particular case it will
be something of this sort you can check it up and can find out.

The corresponding gamma 2’s also can be found out and once you have, so how do you find
out gamma 2 and gamma 1? You are first supposed to find out gE by RT by X1X2, so what
do you do? Once you know these things you can find out gamma 1 gamma 2 and then from
there you can find out gE by RT as sorry X1 ln gamma1 plus X2 ln gamma 2 then you know
the gE by RT for what did different X1 Y1 values?

For example in this case it’s about say 237.3, 219.4, 196.3 and so on and so forth. You can
just calculate once you have calculated this then what do you do? You simply plot gE by RT
X1 X2 as a function of the different X1s that you have. You find out the straight line and
from this particular straight-line you find out A21 and A12 once A21 and A12 have been

802
found out then from there you can generate the gamma values for the different compositions
and once you have done that then you can find out y1 for each case and you can compare
with the experimental y1 values, right?

So therefore this is the way that that you are supposed to work out in this particular case. We
know that for just check up you will be finding A12 into is particular case as 2.181, A21 will
be 1.37, so once you know them you find yis or y1s whatever it is and then you can proceed
and you can find it out, fine. So this was about the Margules equation.

(Refer Slide Time: 9:33)

Now what about the van laar equation, just look at the van laar equation the way I have
written down. In this case what do you expect? How do you expect to proceed in the for the
case of van laar equation.

(Refer Slide Time: 10:15)

So in this particular case just look at the equation, can we write it down in this form? Say
suppose if we can write it down as ln gamma1 which is equal to I will just write down the
equation once more it was B12 by 1 plus B12 X1 by B21 X2 whole square and this was ln
gamma 2 it was B21 by 1 plus B21 X2 by B12 X1 whole square. Just these 3 equations it will
be good if you can remember them, agreed? So therefore can we just write it down in this
particular form?

Just check up, if you can do it in this particular form root B12 by B21 X1 by X2 and same
way one by root ln gamma2 this will be 1 by root B21 plus root b21 by B12 X2 by X1, right?

803
So in this particular case what can we do? We can plot 1 by root ln gamma 1 as a function of
X1 by X2 and then from there we can find out from the slope and the intercept. Similar way
we can plot out one by root ln gamma2 as a function of X2 by X1.

And then from both of these we know that the slope in this particular case it is going to be or
in other words we know that at X1 equals to 0 the intercept here will be root B12 and at X1
and so therefore from here we can find it out and again from here we can find out B21 again
there is a problem that from the slope and intercept the values that we have got they might be
different when they are using this equation or this graph and this graph, so what we need?

We need a proper averaging or a proper summation using the g by RT value, right? So for this
case again suppose we use gE by RT equals to X1 ln gamma1 plus X2 ln gamma 2 then what
you find in this particular case? You find that suppose we plot RT X1 X2 by gE then in that
case this will be equal to 1 by B21 X1 plus 1 by B12, sorry B12 this X1 is here this is X2,
right? So therefore in this particular case if we plot RT X1 X2 by gE as a function of X1 then
we find that we get a straight line is intercept here will be B12 and intercept here will be B21.

(Refer Slide Time: 13:14)

So therefore in a very similar manner if we take up this particular data. Here in this particular
case also you can very well from here again find out gamma 1 gamma 2 gE by RT is the same
then instead of this particular thing in, here you need to plot X1 X2 by gE by RT as a function
of X1 and then from the intercepts you are in a position to find out B12 and B21, once you

804
found them out then from there you can find out the gammas from the gammas you can find
out y1 and you can compare with the results in this particular case.

So you can try out this problem and you can find out the, what are the more or less how
comparable you find the experimental data with the calculated data that you find and then in
and you can compare the thing which you need to do is you need to find out y calculated,
okay. And then once you find out y calculated for each case then in that way the variation can
be found out from Delta y bar equals to yi experiment minus yi calculated the Sigma of the
whole thing divided by n.

So you are required to find out this thing this Delta ybar for 3 suffix Margules and for van
laar compare the 2 and then you can comment on the suitability of the 2 equations.

(Refer Slide Time: 15:02)

In the similar way there is also one more problem you can take it up the previous problem it
had given you data it was an isotherm thermal data, here we have got a Acetone\CCl 4 mix
system and we are providing you isobaric VLE data and in this case the P1 and P2 saturated
are not provided the Antoine’s constants are provided from which you can find out the P
saturated for Acetone and P saturated for CCl4.

And then if you’re using van laar equation and 3 suffix equation in the same way you can
find out the constants A12 A21, constants B12 B21 once you know those particular constants
we assume that they do not vary with temperature as a result of which at X1 at these 2 values

805
you can find out when you know the X1 from there you can find out gamma 1. You can you
already you have found out gamma 1 gamma 2 using this particular data.

And then from these gamma 1 gamma 2 you can find out A12 A21 in the other case you can
find out B12 B21, once is particular thing is available then you can find out gamma 1 for any
particular condition gamma 2 for any particular condition and from there you can find out y1
equals to and then once you know y1 you know everything you can find out temperature and
you are supposed to compare with experimental data and find out how accurate it is.

So this is a general way by which we find out the activity coefficients we can use the 2 suffix
Margules equation when the situation when it is not very far removed from ideality when it is
slightly more removed we slightly modify the 2 suffix Margules equation and try to use the 3
suffix Margules or the van laar equations. For the purpose of the class it is sufficient for you
to be involved with these 3 types of equations for the other equations it is just important to
know when we use the other equations.

For example Wilson’s plot or rather Wilson’s equation it is used when there is a temperature
dependence of the constants are quite significant the NRTL is used when the 2 particular
components they are not randomly mixed and so on and so forth. But for this particular class
we will be concerned with 2 suffix Margules equation, 3 suffix Margules equation and the
van laar equation.

(Refer Slide Time: 17:48)

806
Now apart from this I would also like to mention one more interesting thing. When there is a
remarkable or significant deviation of the solution from ideal behaviour, what do you expect
under that condition? Let us see what happens. For that particular condition we find that
whether it is a negative deviation or positive deviation we find that see the curve has to start
from sorry the curve it has to start from P1 P2 saturated it has to end to P1 saturated. Now
when there is significant deviation from ideality then in that case we find that very frequently
the curve it either passes through maxima when there is positive deviation from Raoult's law
or it can pass through minima when there is a negative deviation from Raoult's law, under this
condition what happens?

When there is such a significant deviation from Raoult's law we find that there is either a
maxima or a minima and if we plot the corresponding temperature curves we find that in
those particular TX plots also we find that the temperature or the boiling point of the mixture
will be either much greater than the boiling point of the individual components or it will be
less as compared to the boiling point of the individual components. So accordingly we find
that at this particular point where we encounter a maxima in the PX curve for example here
or we encounter minima in the Py curve, what do we find?

(Refer Slide Time: 19:22)

That for this particular situation can we not write down dP by dX1 equals to 0. What is this
P? It is just the summation of the partial pressures, so therefore we can write it down in this
particular way. From where we can write down dP1 d X1 is nothing but equal to dP2 d X2,

807
agreed? From Gibbs Duhem equation what do we know? Just remember one thing repeatedly
during our entire discussion on ideal as well as nonideal.

Particularly for nonideal solution for everything we are coming back to Gibbs Duhem
equation. So that shows the significance of this equation as far as phase equilibrium
thermodynamics is concerned. From Gibbs Duhem equation what do we know in terms of
fugacity? One more thing I would like to mention we are using Gibbs Duhem equation either
in terms of chemical potential or in terms of fugacity or in terms of activity coefficient just to
find out some data when other data are known or to test thermodynamic consistency of VLE
data which will be discussing in the next class or maybe to prove certain things which are not
very self-evident.

As we will be doing in this particular case, from Gibbs Duhem equation in terms of fugacity
we know that the equation is this. This equation from low to moderate pressure what
happens? Fugacity becomes partial pressure Pi equals to 0 for binary mixtures or binary
solution whatever you say this reduces to X1 d ln P1 plus X2 d ln P2 equals to 0, right? So
from here what do we get? We get X1 d ln P1 d X1 plus X2 d ln P2 d X1 equals to 0.

Or in other words can we write it down as X1 by P1 this equals to minus X2 by P2 d P2 dX1,

agreed? Or in other words can we write it down as dP1 d X1 by d P2 d X2 equals to X2 by P2
by X1 by P1 this is equal to one. Now what is this P1 by P2 equals to? This P1 by P2 it is
nothing but equals to y1 by y2, isn’t it? Because it’s important to remember that we can
substitute this P2s with y2P. So therefore what do we get? We get X2 by Y2P by X1 by Y1P
this is equal to one.

From here we get X2 by Y2 equals to X1 by Y1 or X2 into 1 minus Y2 will be X1 into 1

minus Y1 and from this equation what do we get? We get that at the point of Azeotropy
where this particular equation is applicable x equals to y or in other words X1 equals to y1 or
X2 equals to Y2.

808
(Refer Slide Time: 23:01)

So therefore what do we find? We find that at the point of Azeotropy definitely there is a
maximum or a minimum in the pressure versus X1 plot. When there is a maximum quite
naturally the boiling point of the solution under this particular condition it will naturally be
less as compared to the boiling point of the pure components, so therefore for positive
deviation of Raoult's law, what do we get?

We get a minimum boiling mixture in the same way for negative deviation of Raoult's law we
find that temperature of boiling point of the solution at the point of minimum pressure will
be higher as compared to the boiling point of the pure components. So therefore we are going
to get a maximum boiling Azeotrope under this particular condition, fine.

809
(Refer Slide Time: 23:49)

And accordingly we find that under these 2 conditions under the conditions of your minimum
boiling Azeotrope for positive deviation we have a minimum boiling Azeotrope and at this
particular sorry at this particular condition we find that there is a maximum in the pressure
curve there at this particular condition the boiling point is much less as compared to the
boiling point of the pure component. And for this particular condition we find that the
equilibrium curve it coincides with the diagonal line as a result of which X equals to Y under
this condition.

(Refer Slide Time: 24:32)

810
In the similar way we find that when there is a negative deviation from Raoult's law under
that condition what do we find for negative deviation from Raoult's law we have a maximum
boiling mixture for maximum boiling Azeotrope and here also it is X1 equals to Y1 under
this condition.

(Refer Slide Time: 25:13)

Now this gives you a very fortunate situation what do we know? We know that X1 equals to
Y1 at point of Azeotropy, right? What is gamma1 equals to can you tell me? It is Y1P by X1
P1 saturated, when X1 and y1 are same that in that case gamma1 it is simply equal to P by P
saturated. So therefore you see that when there is a large deviation from Raoult’s law and we
actually encounter Azeotropes be the maximum or a minimum Azeotrope we find that
gamma 1 at the point of Azeotropy can be obtained as a ratio of the total pressure and the
saturated pressure of that component.

This is gamma 1 this is gamma 2, so once at the point or once we can locate the point of
Azeotropy we can find out gamma 1 and gamma 2 at that particular point then for these 2
points once at the point of Azeotropy gamma 1 and gamma 2 can be found out and we have
selected a proper excess Gibbs free energy model it can be either Van laar or Margules. So
therefore if we know gamma we know X gamma 1 gamma 2, we know X1 we know we
know y1 at the point of Azeotropy then we should be in a position to find out either A12 A21
or B12 B21 depending upon the model that we have selected and once we can find this out
then from these we can find out gamma 1, gamma 2 at other conditions from which we can

811
find out Y1Y2 if P is not given then P if T is not given then T under that particular condition,
right?

(Refer Slide Time: 26:40)

So therefore the calculations become pretty simple under these conditions and I have just
jotted down, they will be doing a few more problems on 2 suffix Margules equation and the
under conditions of Azeotropy these problems are pretty simple and they will be doing a few
of them in your assignments and if you have further doubts regarding these problems you are
always welcome to ask us during our either you can upload the questions and we will be glad
to answer those questions.

Now here I would like to show you the rather I have just jotted down the methodology which
is used in order to solve the vapor liquid equilibrium problems under isothermal conditions.

812
(Refer Slide Time: 27:16)

Under isobaric conditions.

(Refer Slide Time: 27:19)

And also at high pressure, now at high pressure what happens? We cannot replace fugacity
with pressure under this condition. So therefore for these particular conditions more or less
usually when we are dealing with hydrocarbons what we do is we try to express Ki in terms
of yi and xi. And we find that this particular Ki this accounts for the non-ideality in the liquid
phase it accounts for the non-ideality in the vapor phase and very frequently several Ki plots
have been suggested for hydrocarbons.

813
We are in a position to rather you will be in a position to find out the plots and so therefore
yeah, just I will be undoing it I will be doing it once more.

(Refer Slide Time: 28:25)

Yeah, so several K plots are available for different pressure and temperature conditions at
low-temperature range and high-temperature range from which K can be found out and the
VLE at high pressure can be worked out accordingly.

(Refer Slide Time: 28:37)

And well the bubble point and dew point calculations we have been discussing for binary
mixtures, if you’re going for multi-component VLE data the same procedure is going to
apply just we need to remember that instead of working with 2 components we will be
working with more
814
than 2 component and for this particular part the determining the for this particular cases we
find for this particular cases we find that determining this Sigma Xi equals to 1 and these
particular checks are particularly very important for this particular cases, right?

(Refer Slide Time: 29:31)

And very frequently what we do in this particle case is that I have written down the algorithm
for bubble Pt estimation and also we have we have written down the algorithm when we are
using the Peng-Robinson equation of state.

(Refer Slide Time: 29:40)

815
(Refer Slide Time: 29:47)

I have also written down the steps for finding out the dew point estimation the algorithm for
finding out the dew point due pressure calculation.

(Refer Slide Time: 29:48)

And also for the Flash calculations, right? So therefore you can use this particular algorithms
and you can you can find out the VLE data at high pressure and the VLE data for multi-
component system remembering very well that the VLE data at low pressure that has been
discussed for binary systems is applicable in this particular case.

816
When we go for multi-component systems then in that case we have more number of
equations, more number of equilibrium relationships and as a result of which with these more
number of equations we can find out the different gammas when we are dealing with high
pressure instead of pressure we have to have to use fugacity but for this particular class
usually we will be concentrating on binary solutions at low to moderate pressure, fine.

And when we know that there is a point of Azeotropy the calculations becomes simpler. For
the purpose of this class it will be important for you to remember that we will be primarily
considering once more I repeat 2 suffix Margules equation 3 suffix Margules equation and
van laar equation of state, thank you very much.

817
 Course on Phase Equilibrium Thermodynamics
Prof. Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture 45
Deviations from Ideal Dilute Solutions

Well, so long we were discussing ideal solutions and deviations from ideality and we found out
that that we have introduced a term Gamma and we have tried to express Gamma in terms of the
excess properties of the solutions the excess properties arising due to the deviations from ideality,
right? The excess properties are the properties of the solution due since it is deviating from
ideality and assuming that these excess properties are referred to as the properties of the real
solutions minus the properties of the had it, this had the solution been ideal at the conditions of
(temp) same temperatures, same pressure and same composition, right?

(Refer Slide Time: 01:20)

Now what about the ideal dilutes solutions? If you remember we had discussed that if we take up
say suppose pressure versus X1 plot for an ideal solution I had told you that the plot will be
something of this sort, this is the partial pressure P1 versus X1, right? For most of the cases as
we had mentioned the plot is something of this, fine. So therefore we find that when the
component is in excess then the slope is almost linear and intercept is P1 saturated.

818
We found that at X1 equals to 0 also the slope is more or less linear but if we extend it then it the
intercept here is not at P1 saturated it is at some other point and therefore we said that for both
the cases it is linear or in other case we found we proposed that or we found out that for ideal
solution your fi bar is proportional to xi, okay. And in the same way we can say that for ideal
dilutes solutions also fi bar is proportional to xi, what is the difference?

Suppose in this case we take the proportionality constant as ki so in this particular case your ki or
in maybe say Ri or something that is equal to the fugacity of the pure component at this
conditions of temperature and pressure of the solution, right? And for low to moderate pressure
this becomes equals to Pi saturated for ideal dilute solution in this particular case Ri it is not
equal to fi, what is it? It is equal to rather this is equal to H21 which is the Henry’s constant of
component 2 in component 1; we had already discussed these things.

So that Ri this is equal to limit Xi tends to 0 fi bar by Xi is equals to H21 and limit Xi tends to 1
fi bar by Xi which is equals to fi. Now suppose there is (jus) just as if when there is deviation
from ideal solution that in that case what do we write? We write fi bar equals to gamma i Xi fi,
similarly for deviations from ideal dilute solutions what should be put up? We should have a
some sort of an identical activity coefficient and this identical activity coefficient should
incorporate the non the deviations from ideal dilute solution. So in this particular case we should
have fi bar equals to say gamma i star Xi H21, agreed?

Now you should remember that when we have an ideal solution then in that case for Xi of every
component tending to 1 fi reduces to or fi bar reduces to the saturated vapor pressure of that
component for low to moderate pressure or the pure component fugacity of that component at the
same temperature and pressure of the solution.

819
(Refer Slide Time: 04:40)

What I mean to say is when we are talking about an ideal solution then in that case for each and
every case for each and every component we can write gamma I tends to 1 as Xi tends to 1 this is
true for all components, isn’t it?

But when we are dealing with an ideal dilute solution then in that case definitely when it is when
it is dilute with respect to one component other component is in much excess. So suppose it's a
binary component then for the solvent plot we have gamma i tends to 1 as Xi tends to 1 and for
the solute part, what do we have? Gamma i star maybe it's a binary mixture, so this is gamma 1
X1 and gamma 2star tends to 1 as X2 tends to 0.

So what do we find? We find that while the normalization of activity coefficient is symmetric it
is same for all the components for the case of ideal solution in this particular case the
normalization of gamma is un- symmetric while we find that the solvent if it obeys the Raoult’s
law the solute it obeys the Henry’s law and therefore while for an ideal solution this is known as
symmetric normalization of activity coefficients in this particular case we call it as un-symmetric
normalization of activity coefficients, right?

And so therefore again I would repeat that symmetric normalization it tells you that Gamma i
tends to 1 as Xi tends to 1 and un-symmetric normalization because in this particular case the
normalization for the different components are different so therefore this is gamma 1 tends rather
gamma 2 tends to 1 as X2 tends to 0 gamma 1 tends to 1 as X1 tends to 1, right? And in the

820
similar way this should be gamma 2 star and therefore we have 2 types of gamma normalization
1 for ideal solution 1 for ideal dilutes solution. And what is the relation between gamma2 and
gamma2 star of any particular component.

(Refer Slide Time: 7:21)

We know gamma2 is f2 by X2 f2 we know gamma2 star this is equals to f2 by X2H21. Now

what is gamma2 by gamma2 star? This is equals to H21 by f2. So therefore this limit X2 tends to
0 gamma 2star equals to 1, right? And therefore limit again limit X2 tends to 0 gamma2 this is
equal to H21 by f2 this is equals to Gamma2 by gamma2 star or in other words gamma2star by
gamma 2 equals to f2 by H21 equals to limit X2 tends to 1 gamma2star, isn’t it?

So therefore what do we find? We find that limit X2 tends to 0 gamma2 this is equal to this and
limit X2 tends to 1 gamma2star this is equal to f2 by H21 equals to gamma2star by gamma2 and
this we need to remember that limit X2 tends to 1 gamma 2star which is given by this. This is a
hypothetical situation because X2 tends to 1 for that particular case the component 2 does not
exist in the phase of the solution as a result of which this is a hypothetical state.

So therefore when there is deviation from ideal dilutes solution and in that particular case.

821
(Refer Slide Time: 9:11)

We can also make the correction considering some form of an activity coefficient just to
differentiate it from the activity coefficient we have been using for deviation from ideal solution
we have named it as omega, gamma2star you can use any other name if you wish for such cases
we have un-symmetric normalization of activity coefficients.

(Refer Slide Time: 9:37)

And if we want to find out gamma2star from gamma2 we can use either of these equations where
this particular equation is hypothetical since we know that gamma2star or rather 2 does not
remain in the phase of the solution at the conditions of temperature pressure of the solution.

822
And in the same way like we have tried to relate the activity coefficient gamma2 or gamma in
terms of Gibbs free energy and the partial molar volume and the partial molar enthalpy we can
do a similar endeavor and relate gamma2star with the partial molar excess Gibbs free energy
considering it to be the difference or they excess property to be the difference between the actual
property minus the property at in finite dilution for the same temperature and pressure of the
composition. So by this we have covered ideal solutions, ideal dilute solutions as well as
deviations from ideal solutions, deviations from ideal dilutes solutions.

Next the thing which is left is so far we have been discussing everything assuming that the 2
component are completely miscible over the entire range of composition but there are situations
where we find that the 2 components are not miscible over the entire range of composition they
can either be completely immiscible over the entire range which is again an extreme state or
there can be partial immiscibility which is much more real situation where we find that for very
low and very high concentration the 2 components are totally miscible and within the
intermediate composition there is the 2 components are immiscible such systems are known is
partially miscible systems.

So therefore we are going to deal with situations as partial immiscible systems, completely
immiscible systems and why do substances prefer to, prefer not to mix when do substances
prefer to mix. We already know that the, for constant temperature and pressure conditions this is
governed by Gibbs free energy. So definitely the 2 component would like to separate out when
the Gibbs free energy of the 2 particular liquid separated is less as compared to the Gibbs free
energy of the solution.

So therefore we are going to deal with those conditions when 2 particular substances or 2
particular components form an immiscible pair when do they form a miscible pair? And the
phase diagrams the Pxy, Txy diagrams for completely immiscible and partially miscible
substances and since no discussion on thermodynamics can end without having or without
discussing thermodynamic consistency of VLE data. We will be spending a short amount of time
by discussing how exactly the Gibbs Duhem equation is used to test thermodynamic consistency
of VLE data with that more or less the class or the course on thermodynamics will be coming to
an end. So we will be discussing these things in the next class, thank you very much.

823
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 46
Tutorial

Hello everybody, so today before I proceed further I would like to go for a recap of whatever
we have done for solutions both ideal and nonideal we will try to solve a few problems and
then we will go for some other new topics.

(Refer Slide Time: 0:37)

Now as a recap let us see what are the equations or rather what are things that we have
already studied?

824
(Refer Slide Time: 0:55)

The first thing is for phase equilibrium the basic equation is the fugacity’s of any component i
is same in the vapor phase and the liquid phase, okay. So this is applicable for any component
i where i can be component 1 to component C. So for the time being we are considering a
heterogeneous closed system, okay. If you recall when we were when I had first started
considering equilibrium initially we had considered homogeneous closed systems, then we
had considered your homogeneous open systems and then we had considered heterogeneous
closed system where each heterogeneous closed systems comprises of 2 homogeneous open
systems.

In this case also there are 2 homogeneous open systems one is the liquid phase the other is the
vapor phase each phase comprising of C number of components the component number is
denoted by i. For each particular component we have one such equation. So therefore we
have C number of such equations. Now what is fiL equals to? If you recall this is equal to
gamma i Xi into Fi. What is this fi? This is the fugacity fiL Fugacity of liquid in standard
state.

What is fiV equals to? This is equal to phi i v the partial rather the partial fugacity of
component i in the vapor phase into yiP. So therefore this is basically the correction for non-
ideality in the vapor phase and this is the correction for non-ideality in the liquid phase. Now
if we equate these two, what do we get? On equating these 2 we get gamma i Xi fiL equals to
phi i v yi into P, agreed? Where we know gamma i as you know its activity coefficient Xi is a
mole fraction of component i in the liquid state yi is a mole fraction of component i in the

825
vapor phase P is the total pressure and fiL as I have already mentioned its fugacity of liquid
in the standard state.

Now we can select any particular standard state but for most of this of the situations just for
convenience what do we select? We select pure component liquid at the system pressure and
at the system temperature. Now from thermodynamics the only requirement is maintenance
of the same temperature of the solution and the pure state. We could have selected any other
pressure or any other composition but since it is convenient we select the standard state P
equals to P system pressure and xi equals to 1, so this is the pressure and composition under
which we select a rather we define fiL.

So naturally what this becomes? This becomes the pure component fugacity at temperature
and pressure of solution, this is by convention we take this and out of this is mandatory and
this we could have taken something else and is important to remember that very frequently it
can happen that this particular state it is fictitious because the component which we are
selecting may not exist in the liquid state at the system temperature and pressure but it hardly
matters whether it is real or fictitious.

And we have already found out how we can estimate pure component fugacities of both the
vapor and the liquid. So therefore what is this equal to? If you remember this is the saturated
vapor pressure of the component and this has to be corrected by 2 factors, one arises due to
non-ideality in the vapor phase and the other arises due to completion of the liquid phase
from the saturated pressure to the pressure of interest this particular factor it is known as the
poynting correction factor. Its P is very close to Pi saturated then naturally this term it
disappears off, fine.

826
(Refer Slide Time: 5:22)

So therefore now if we substitute this fiL in this particular equation and equate it with this,
then finally what do we get? Finally we get an equation as has been mentioned, this is the
equation that we are finally going to get, right?

(Refer Slide Time: 5:55)

So therefore if I just write down the equation for your convenience we find the equation is
gamma i Xi Pi saturated phi i saturated exponential viL P minus Pi saturated by RT this is
equal to Phi i v yi into P, right? Or in other words we can also write it down as gamma i Xi Pi
saturated this is the fugacity coefficient of component i in the vapor mixture, fugacity
component of component i under the set pure component i under the saturated conditions into

827
yiP exponential minus viL P total pressure minus Pi saturated by RT, please remember this is
applicable for each component i.

Well, this particular equation this is the thermodynamically most rigorous equation and this
forms the basis for vapor liquid equilibrium and in this particular equation we find gamma i.

(Refer Slide Time: 6:55)

What is gamma i? Gamma i is a function of temperature, pressure and mole fractions, we

know phi i v is a function of temperature, pressure, mole fraction in the vapor phase,
naturally the saturated fugacity coefficient is a function of temperature and pressure is a
function of temperature because P saturated is a function of temperature and we know P just
the same thing as been written down here. So therefore this is the basic equation using which
we are going to solve different vapor liquid problems.

828
(Refer Slide Time: 7:34)

(Refer Slide Time: 7:48)

Now for the different vapor liquid problems, for we can encounter different sort of vapor
liquid problems and the problems which we encounter we have already discussed day they
can be bubble pressure calculation, dew pressure calculation, bubble temperature calculation,
dew temperature calculation no matter whatever the problem be this is the basic equation
which we are going to write.

829
(Refer Slide Time: 8:10)

Now for most of the cases we find that we deal with low to moderate pressures, what happens
at low to moderate pressure let us see? Does this equation reduce to something simple you
already know what happens suppose we are dealing with low pressure, say less than 1 bar.
What do, what’s the first thing that you expect is going to happen?

First thing which is going to happen is that the vapor phase starts behaving ideally, when the
vapor phase starts behaving ideally, naturally this becomes equal to one and this also
becomes equal to one, when we are dealing with low pressure naturally P will be close to Pi
saturated. So this becomes 0 as a result of which this whole term this becomes equal to one.
So therefore what happens for low pressure?

Low pressure we find gamma i Xi Pi saturated you are already aware this is equals to yiP. For
moderate pressures say less than 10bar suppose what happens under this condition?
Definitely under this condition this does not disappear this does not disappear but although
we cannot assume the vapor phase to behave as ideal gas we can always assume that the
vapor mixture behaves as an ideal mixture or in other words it obeys the Amagat’s law of
additive volumes accordingly what do we have under this condition? Under this condition we
can have phi i v is equal to phi i v which is equal to phi i saturated this we can safely assume.

So when we can safely assume this naturally what happens this whole term it becomes equal
to one and again we find that for moderate pressure also P will not be very far removed from
Pi saturated as a result of which this term also becomes 0 and exponential term becomes

830
equal to one. So we find that for both low pressures and for low to moderate pressures this
particular equation holds and when we are dealing with ideal solutions I need not repeat it
once more it’s very you know it very well that XiPi saturated becomes equals to yiP which is
nothing but the famous Raoult's law with which we have been dealing for quite some time,
right?

So therefore now depending upon the situation, it’s a very complex or rather if you are
dealing at high pressure we should be dealing with this equation. If you’re dealing with low
to moderate pressure we should be dealing with this equation. If you’re dealing with ideal
solution we should be dealing with this particular equation.

(Refer Slide Time: 10:46)

No matter which condition we are dealing with does not matter the problem types are as I
have mentioned they can be bubble pressure, dew pressure calculation, bubble temperature
dew temperature calculations.

831
(Refer Slide Time: 10:51)

Whatever be the situations we will be having i such equations or i such equations along with
those i sorry not i’s C equations C number of this or C number of this equations.

(Refer Slide Time: 11:23)

Along with that we are going to have two additional constraints which also you are well
aware Sigma xi equals to 1, Sigma yi equals to 1 or very often when we have to take up some
trial and error cases then whether we have guess or guess values are correct or not we
understand by these particular checks.

832
(Refer Slide Time: 11:30)

Now let us take up each problem maybe the first time let us take up isothermal data and let us
see how we can find out or rather how we can calculate.

(Refer Slide Time: 11:39)

833
(Refer Slide Time: 11:49)

For suppose you’re the first type which I have told you say T and xi are given. Suppose the
first type remember whatever be the type and if supposed to find P and yi, whatever be it we
will be dealing with the or rather we will be dealing with the same type of equations just we
will be rearranging them according to the condition.

For example suppose Y and Xi are given, what are we supposed to do? If T is given we can
find out Pi saturated for each component i it will nothing be but equal to Ai minus Bi by t
plus Ci, so therefore we can find out t. Sorry we can find out Pi saturated for each
component. Once we can find out Pi saturated we know Xi and suppose we have some
fragmentary data on yi and P then in that case with those at least if you have data even on one
particular point also Pxy, yy data at constant T then we are in a position to find out gamma i.

What is gamma i? It is nothing but yi P by xiPi saturated for that component i. Moment we
have found out gamma i, now we need to select the excess Gibbs free energy models. For this
particular class although I have discussed the large number of models, for this particular class
we will be confining our attention either to 2 suffix Margules equation or 3 suffix Margules
equation or van laar equation. You need not memorize the equations the basic equations will
be provided to you but derivations from the equations you are supposed to do it on your own.

So therefore once we know gamma i then for one particular value then from here we are in a
position to find out. We are going to select either the van laar or the 3 suffix Margules
equation accordingly we will be in a position to find out either A12 or A21 or B12 or B21

834
remember one thing that these are these are refer to the constants of the 3 suffix and the van
laar equations of state. Once we know this we know that we have already assumed that they
are independent of temperature and pressure so that we can use the same constants to find out
gamma i for other Xi values. Accordingly we can generate the whole set of PXY data from
the fragmentary set of data.

(Refer Slide Time: 14:11)

Now let us take up a few examples just to verify just to ratify what I have been discussing so
far. This is one of the problems which I have, in this problem what do we find? there is a
propanol and chlorobenzene, temperature is given 95 degrees centigrade we are required to
find out Pxy data from just as I have mentioned 3 suffix Margules and the van laar equation.
What are the data which are given then in this particular case?

835
(Refer Slide Time: 14:42)

T equals to 95 degrees centigrade we already know, right? And then once we know this it’s a
binary mixture, we are in a position to find out the saturated pressure of propanol the
saturated pressure of chlorobenzene if you calculate this you will be finding that the saturated
pressures are 681.77 Torr you can check it up just see that units etc are consistent otherwise
there is no place where you can make a mistake, right?

(Refer Slide Time: 15:28)

So you have got this, once you have got this then for the first set of data you know x1, you
know y1, you know P you know T.

836
837
(Refer Slide Time: 15:47)

So therefore you are in a position to find out gamma1, right? Just the way using this
particular equation here only I can, okay I will write it down here one thing I would like to
mention that for my case I have assumed propanol to be component1 chlorobenzene to be
component 2. Usually you will find lot of problems we just specify that use this as
component 1 use this is component 2 if it is not specified we already know that we usually
plot all the phase diagrams with the more sorry the more volatile phase as component 1 and
the less volatile phase as component 2.

So therefore in this particular case we can since this is got a higher saturated vapor pressure
we take this as component 1 we take this as component 2. So now we can find out gamma1
we know y1 is nothing but equals to 0.599 P we have already know it's 518 if you just
observe this you can find and we know X1 is equals to0.212 and we know the saturated vapor
pressure as this 681.77, right? So if you calculate you find this is 2.1468 and accordingly
your ln gamma1 this become 0.764.

Same way we can we can calculate it for gamma2 just substitute the proper values you find
gamma 2 is going to be 1.0609 and ln gamma2 is going to be 0.0591. Now you have you
have got ln gamma1 and gamma 2 once you have got this then you can find out the excess
Gibbs free energy of the mixture from experimental data, what is this? This is nothing but
equals to Sigma Xi ln gamma i. For a binary mixture it is X1 ln gamma 1 plus X2 ln
gamma2. So you’re in a position to find because you know gamma 1 gamma 2 X1 X2 you

838
can find this out. So with this your entire experimental data is more or less ready with which
you are supposed to work, right?

(Refer Slide Time: 18:01)

So once you have got this more or less your entire till gE by RT the entire thing is ready for
you, agreed? Now once this part is ready, now here we have already mentioned that you are
you are supposed to use the van laar equation as well as the 3 suffix Margules equation. So
therefore from the experimental data we have already found out gamma1 X1 this part was
applied. We were in a position to find out gamma 1 gamma 2 then ln gamma1 ln gamma 2
and then till gE by RT we could find out.

839
(Refer Slide Time: 18:45)

Now suppose initially we take up the 3 suffix Margules equation, now all of us know the 3
suffix Margules equation is given by this particular formula. So from here what do we find
out? We find out that for 3 suffix Margules equation if we plot this particular term gE by RT
x1x2 is a function of x1 it should give you are linear plot and from the linear plot from the
intercept at x1 equals to 0 what do we get? A12 and then x1 equals to 1 what do we get? A21.

(Refer Slide Time: 19:59)

So therefore once we have found out gE by RT we know x1 we know x2 we are in a position

to find this out. Once we are in a position to find this out, so therefore we can always plot gE
by RT X1X2 as a function of X1 this will go till 0 to 1, right? And then from the intercepts

840
you can find out that at X1 equals to 0 your A12 in this case it is going to be 1.231, A21 is
going to be 1.295 in this case.

(Refer Slide Time: 20:01)

Moment you have found out A1 A2, now you can use again this particular equation you
know A12, A21 for each particular x1x2 you are now in a position to find out ln gamma1 ln
gamma
2. Once you found out ln gamma1 ln gamma 2 from these you can find out gamma1 and
gamma 2. Moment you have found out gamma1 gamma 2 then using the data of x1 and P or
x1P1 saturated you will be in a position to find out y1 and P and this is precisely what we
have done.

(Refer Slide Time: 20:45)

841
If you find it out say let us take up the first data, right? For the first data we already know
A12 A21.

(Refer Slide Time: 20:53)

The first data if you observe here, the first data is x1 equals to 0.212 y1 is given P is given,
right?

(Refer Slide Time: 21:14)

So for this particular data we find ln gamma1 using that particular equation we get it is going
to be x2 square into 1.235 plus2 into 1.295 minus1.235, isn’t it? into x1, agreed because if

842
you remember x2 square is nothing but A12 plus 2 into A21 minus A12 into x1. So therefore
on further simplification we find this is 1.235 plus 0.12X1, agreed?

Same way you can substitute for ln gamma 2, you get this is going to be just like the since it
is already written I would like to mention this is A21 plus A12 minus A21 into X2 which is
nothing but X1 square into 1.295 minus 0.12 X2, agreed? So for so therefore from these 2
equations you if you know X1 A21 and A12 we have already found out, you can find out ln
gamma 2 for different X1s. We can also find out ln gamma 1 for different X1s.

So for X1 equals to 0.212 you can find out ln gamma1 would have been equal to, you can just
substitute them instead of X1 you can put 0.212 and then finally you find out gamma 1 as
2.1873 gamma 2 as 1.0554, what is P equals to? P will be equal to 522.79 Torr and what will
y be equal to y1 will be equal to it is gamma1 X1 P1 saturated by P. Substitute all the values
gamma 1 is equal to 2.1873, let me substitute 2.1873. What is X1? 0.212, what is P1
saturated? That also we have already calculated it is 681.77.

What is the total pressure? It’s already written down there it is 522.79, so what is y1 that you
get, 0.6047 agreed?

(Refer Slide Time: 23:41)

So therefore for the first case you are in a position to find out your X you are in a position to
find out all the values. You are in a position I have done it here and it is provided here, so
therefore you gamma 1 gamma 2 y1P everything has been there using 3 suffix Margules
equation; instead of 0.212 you can repeat the calculations with 0.43. Same way you are going

843
to proceed you are going to find out you know A12 A21, so you are in a position to find out
ln gamma1 ln gamma 2 from ln gamma 1 gamma 2 you can find out gamma 1 and gamma 2.
Once you notice can find out P you can find out y1, so if you repeat this calculations this is
the final Pxy data that you will be in a position to generate using 3 suffix Margules equation.

(Refer Slide Time: 24:46)

Now suppose I tell you to repeat the same way van laar equation, the procedure is completely
same just the equations are different. For van la equation if you remember, this is the
equation which we have, right? So in this equation if you’re going to plot this particular term
with x1 again in this case we are going to get a straight line here, okay.

And so if you’re plotting here what do we get? Suppose we are plotting X1 X2 by gE by RT

as a function of X1 ranging from 0 to 1 this case also we will be getting a straight line where
this part the at X1 equals to 0 you get 1 by B12 from here you can see and here you get 1 by
B21, right? From which from the reciprocal values you’re in a position to find out the actual
B12 B21.

844
(Refer Slide Time: 26:18)

So in this particular case if you’re repeating it, you are going to see that for your case B12
will be equal to 1.2346, B21 will be equal to 1.2970. Now once you once you know this then
you are again in a position to find out gamma 1 for the for the different cases ln gamma 1
values are given, ln gamma 2 values are given so once these 2 are given you know B12, you
know x1 x2 you can find out these. Once you can find out gamma 1 and gamma 2 you will be
in a position to find out sorry from the ln values you’re in a position to find out gamma1 and
gamma 2.

(Refer Slide Time: 26:25)

Moment you know gamma 1 and gamma 2 then you can use it to find out y1 and P. Say for
example suppose we start with X1 it’s the same X1 value the first value 0.212, right? From
845
here if you do you find out gamma1 is going to be using the van laar equation it is going to be
2.1869, gamma 2 is going to be 1.0554. From here you can find out P, what is P going to be?
It is 2.1869 into 0.212 into 681.77 plus your 1.0554 into 1 minus 0.212 which is 0.788,
248.48 which gives you the P as 522.74

Torr. (Refer Slide Time: 27:43)

What is y1 equals to then? y1 if you calculate you will find out that this is going to be 0.6047.
So therefore just the way I have I have I have done the calculations for the first case 0.212
this is the situation I have found out gamma1 gamma 2, P Torr and y1, right? So therefore I
can repeat it for the other x1 and generate the Pxy data using the van laar equation.

(Refer Slide Time: 28:03)

846
So therefore in this particular way what have we done? We have generated the Pxy data for
using the 2 suffix sorry 3 suffix Margules equation.

(Refer Slide Time: 28:11)

We have generated the data from the van laar equation and now if you compare these
particular data with the original data you can understand for yourself which one is much more
accurate. Well, this was when you are dealing with isothermal VLE data. Now in chemical
process industries we find that distillation operation are usually carried out under isobaric
conditions. So it’s often much more convenient to deal with isobaric data than with
isothermal data.

We will be dealing with this in the next class. We’ll be taking up a specific problem and see
how we are just going to rearrange the equations which we which we have rather discussed in
this class in order to obtain isobaric VLE data, thank you very much.

847
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 47

(Refer Slide Time: 0:41)

Well, to continue our discussions regarding the isobaric data. Now in this case what do we
have? We have P we know P and possibly we know X1 we will be in a, we would be required
to find out T and y1 say, right? So therefore in this particular case there is small amount of
trial and error which is involved. Now for most of the cases what we do? We assume some
particular T for that T is you find out Pi saturated and Pi saturated for the different
components.

Once you know Pi saturated you know P, so therefore you will you will be in a position to
find out the constants of the of the particular excess Gibbs free energy model that you’re
going to assume. So from there you find out either A12 or B12 B21, again once you have
found them out you are now in a position to find out gamma i for the other conditions, right?
And so therefore from that particular gamma i if you know X i and you know yi then you’re
in a position to find out the P and the mole fractions in the vapor phase.

If you know the mole fractions in the liquid phase you’re going to calculate it and going to
see whether the constraint that I had mentioned Sigma Xi or Sigma yi equals to 1 it depends
on what you’re finding out. Are you finding out the liquid phase composition or the vapor
phase composition? Find out the compositions for each of the component and see if the
composition adds up to 1. If it adds up to 1 your temperature assumption was correct if it
does
848
not go for a different temperature assumption. The direction in which your assumption should
proceed will be obvious once you find out whether your Sigma Xi is greater than one or less
than one.

(Refer Slide Time: 2:21)

The methodology of for finding it out is mentioned in this particular slide but unless we take
up a specific problem this is not going to be clear to you.

(Refer Slide Time: 2:27)

So therefore we go for a second problem here. This is an acetone carbon tetrachloride

mixture it’s almost the same as the previous one, just the previous one was an isothermal data
this was an isobaric data. You’re supposed to use both van laar as well as the 3 suffix
Margules equation

849
and you’re supposed to find out t and y1 your P is is given here it is 450Torr and the X1s for
3 different X1s you’re supposed to find it out.

In this way also the way we proceed it is alright. The small amount some fragmentary data is
given, so based on this fragmentary data you can take up this 55.29. Once you know t degree
centigrade you know the Antoine constants for acetone for carbon tetrachloride. So from here
you are in a position to find out Pi saturated. You find that for acetone it is 739.18Torr and for
CCl4 it is 372.36Torr, right?

So therefore these 2 you have found out and obviously here in this case also you find that
acetone has a higher saturated vapor pressure and we designate this as component 1 and we
designate this as component 2.

(Refer Slide Time: 4:12)

So now once we have found out these now we are in a position to find out gamma 1.

850
(Refer Slide Time: 4:15)

For the first condition let us say X1 equals to 0.049 y1 equals to 0.189.

(Refer Slide Time: 4:40)

So therefore for this particular condition we can find out gamma 1 it is 0.189 into 450Torr
divided by 0.049 into739.18, the P saturated gamma 1 it is 2.3482. Same way you can find
out gamma 2 0.811 into 450 by 0.951 into 372.36 this becomes 1.0306. Same approach as we
have done the previous problem this is X1 ln gamma 1 plus X2 ln gamma 2 this particular
case this is 0.0705.

851
Now suppose we start dealing with the van laar equation of state. For the van laar equation of
state what we get? We need to plot your x1 x2 by gE by RT its not a big deal you can
calculate it. It’s going to be 0.6610.

(Refer Slide Time: 5:45)

Now once this is calculated then you can repeat this particular calculation for other conditions
as well, right? So in this particular table I have written down all the calculations which have
been made the just the I have demonstrated the calculations for one particular data for this
particular data I have demonstrated the calculations for other data the calculations are going
to be the same. So this is the total table for the entire calculations that I have performed. Once
this has been done then from here you can find out you are required to find out X1 as 0.447.
So for this case y1 P1 saturated everything can be obtained experimentally.

Now we have to find out theoretically, how to do it? Once you know this x1 x2 by this
particular value. So therefore initially let us assume some t, let us assume the t which is
already provided here which is 44.06. So we can assume t equals to 2 44.06 X1 is 0.447.

852
(Refer Slide Time: 7:13)

Gamma 1 we have we have already found out it is 1.24 and gamma 2 equals to 1.28 these
things we have already found already found out from here the gamma1 and gamma 2 values.

(Refer Slide Time: 7:20)

So therefore from here we are in a position to find out that for this particular case gamma1 is
equal to it is equal to 1.2555 gamma 2 equals to 1.2804, right? So if we assume t equals 44.06
we arein a position to find out p1 saturated which is nothing but equal to 494.83Torr, P2
saturated you can find out 245.34Torr from which you can find out y1 it is just gamma1 X1
P1 saturated by P. So it is 1.2555 into 0.447 into 494.83 divided by450 equals to 0.6171.

853
Accordingly y2 it's equal to 0.386 add up y1 and y2 it's 1.0031 you can take this as more or
less close to one. If you want to go for slightly better values then in that case you have to
again change t. Now you know that since Sigma y1 is greater than one you should reduce the
t in this particular case.

(Refer Slide Time: 9:03)

So let us go for a second assume t. Let us take this second assume t say as t equals to 42.48
degrees centigrade. Find out P1 saturated, 466.42Torr, P2 saturated 230.70Torr, y1 you can
find it out, again y2 you can find it out add the 2 you get more or less Sigma yi equals to 1.
So therefore this is your assumed thing is correct.

(Refer Slide Time: 9:38)

854
So therefore from here we can find out we can repeat this for different values of for the
different x1 values which are given and after repeating we find that the calculated t degree
centigrade are this and the corresponding y1 values are this. And you can you can compare
them and you find that there is although small there is a mismatch between the calculated and
experimental values. Now this was when we were using the van laar equation.

(Refer Slide Time: 10:22)

(Refer Slide Time: 10:30)

Suppose we would like to do for the 3 suffix Margules equation. The situation is going to be
the same, we are going to in this particular case instead of finding out gE by RT this we will
be finding out gE by RT X1 X2 and we will be plotting that, right?

855
(Refer Slide Time: 10:52)

So therefore we can find this out and we can plot it out, if we plot out gE by RT X1X2 as a
function of X1 again from the intercepts we know that at X1 equals to 0 we will get A12 here
we are going to get A21, right?

(Refer Slide Time: 10:56)

So therefore we can proceed in the similar fashion and we can find out A12 and A21. In this
particular case also we find that A12 in this particular case is going to be 1.165 A21 is going
to be 0.735 and then once we know this we are we can find out X1 gE by RT as I have
mentioned and gE by RT X1 X2 and you can prepare this table.

856
(Refer Slide Time:
11:37)

Once you have prepared this table then you can find out ln gamma 1 ln gamma 2 for different
x1 x2 values.

(Refer Slide Time: 11:53)

857
(Refer Slide Time: 11:58)

And you can take up or you can assume a value of t, naturally the assumed the value will be
closed to the experimental value and for each particular assumed value you can find out y1 y2
and you can find out the summation, see if the summation is close to 1 or not.

(Refer Slide Time: 12:50)

Or else we go for a different sort of an assumption, say for example in this particular case let
us take up X1 equals to 0.447 I think have already mentioned gamma1 is 1.2696 gamma 2 is
equal to 1.2736 t equals to say at the first assumption, first assumption is the experimental
value naturally as I have told you. For this case we have to start from the very beginning from
P1 saturated and P2 saturated these are the values that we get from here we get y1 equals to
0.6203 we get y2 equals to 0.3816 more or less we will find that Sigma yi equals to 1.0.

858
(Refer Slide Time: 12:55)

So accordingly we can calculate for the different cases also and we get the yi calculated. Now
when we compare the experimental versus the calculated value for both the van laar and the
Margules equation in both the problems that we have dealt with we find that the van laar
equation appears to be much more accurate in these particular cases.

Well so what did we do? We have been discussing the reduction of isothermal an isobaric
data from some fragmentary experimental data if we have that. Now we know that the
calculations of gammas and the calculation of the corresponding constants for the excess
Gibbs free energy model it becomes much simpler if we know that there is an Azeotrope
formation in at any particular position, right?

(Refer Slide Time: 13:50)

859
Why does it happen? We have already discussed at the Azeotropic point what happens X1
becomes equal to y1 and naturally under this condition y1 becomes equals to P by P1
saturated, isn’t it? So therefore we get y1 equals to P by P1 saturated gamma 2 becomes P by
P2 saturated. So therefore it we are finding out of gamma 1 gamma 2 becomes very easy at
the Azeotropic point. Once we have found this out we can find out ln gamma 1 ln gamma 2
and then we can select any of the excess Gibbs free energy model it can be van laar or the
Margules.

So from here we find out the constants from both these we can find out the constants A12
A21 or B12 B21 as the case may be. Once we know this we can find out gamma 1 gamma 2
for other compositions and once we know that we are in a position to find out P and we are in
a position to find out y1.

(Refer Slide Time: 14:53)

Let us take up a specific problem in order to discuss this Methanol acetone we know that it
that it forms an Azeotrope at 760Torr with x1 equals to 0.2 and t equals to 55.7 degrees
centigrade it’s given here. We are required to predict the Pxy data under this condition.

860
(Refer Slide Time: 15:21)

(Refer Slide Time: 15:24)

Finding out gamma 1, so therefore what are we supposed to do under this particular
condition? We can find out the P1 saturated and P2 saturated and we can find out the gamma
1 and gamma 2 for this particular case isn’t it? So therefore at t it’s given t equals to 55.7
degrees centigrade, right? For this particular case, what is P saturated for methanol? This is
equal to 530.97Torr, P saturated for acetone 749.65Torr, agreed?

What is gamma methanol say? Under this particular condition this becomes 760 by 530.97
1.4313 which gives ln gamma methanol as 0.3607. Similarly for the acetone case we get this
as 760 by 749.65 which gives us 1.0138 which gives us ln gamma acetone as 0.0137.

861
(Refer Slide Time: 16:39)

Now we can either take up the van laar equation or the Margules equation. The matter
whatever we take up, if we take up van laar equation we can find out B12 and B21 we know
gamma 1 we know x2 x1 at the Azeotropic point. So for the Azeotropic point we were in a
position to find out B12 and B21 once we can find out B12 and B21 at the Azeotropic point
we know that it does not vary with composition. So therefore we can find out ln gamma1 ln
gamma 2 for other x1 x2 values as well and then from there we can generate it, right?

(Refer Slide Time: 17:09)

So in this particular case we find our B12is 0.4786 B21 was 0.7878 so therefore we have
found out everything for the Azeotropic point.

862
(Refer Slide Time: 17:36)

Now let us go to some other composition say at X1 equals to 0.1 from this we can find out
gamma1 I have already shown you the equations from where you are supposed to find out
gamma1 and gamma 2.

(Refer Slide Time: 17:55)

So therefore in this case we find gamma 1 equals to 1.5219, gamma 2 is 1.0032, so P in this
case 757.62Torr, y1 comes to as 0.1067, right?

863
(Refer Slide Time: 18:02)

Now in the in a similar way you we can find out the gamma1 gamma2 P and y1. For other
composition in the liquid phase and this in this way we can generate the Pxy data for using
the van laar equation in much more simpler fashion provided we know that there was an
Azeotrope at a specific condition of x1 equals to 0.2. Same thing can be repeated for the
Margules equation as well.

(Refer Slide Time: 18:59)

For the Margules equation also we can find out A12 and A21, how to do it? What is A12? 2
ln if you reduce the equation you find that this is going to be 2 ln gamma2 plus x2 minus x1
ln gamma1 by x2 square this becomes 0.4752. Similar way we can find out A21 just in the

864
similar way 2 ln gamma1 by x2 minus x1 minus x2 ln gamma2 by x1 square this become
0.69.

(Refer Slide Time: 19:25)

Once we know these 2 then naturally we know A12 below A21, we are in a position now to
find out ln gamma1 ln gamma2 for different values of x2 and x1.

(Refer Slide Time: 19:36)

This has been done here, right? So here we have repeated the same thing. For example
suppose we want find out ln gamma 1 its X2 square you can just compare since it is A12 plus
2 into A21 minus A12 into x1 this reduces to X2 square 0.4752 plus 0.4422 x1, right?

865
Same way can find out ln gamma 2 it will be X1 square into 0.6963 minus 0.4422x2, right?
So we can find this out.

Once we know this then at say X1 equals to 0.1 gamma1 this becomes 1.5231 gamma2
becomes 1.0030 P you can calculate 757.58torr y1 it becomes 0.1068, right? In the similar
way you can calculate it for other values as well for 0.3, 0.5, 0.7, 0.9 and in this way you can
generate the entire set of data using 3 suffix Margules equation.

(Refer Slide Time: 21:05)

Well, we can take up one more problem as well concerning the Azeotrope, say for instance
this particular problem here it is given that for a binary mixture both the components they
obey the 2 suffix Margules equation, right?

866
(Refer Slide Time: 21:22)

(Refer Slide Time: 21:32)

When they obey the 2 suffix Margules equation we know ln gamma 1 equals to AX2 square
ln gamma 2 equals to AX1 square and you are required to predict the condition of Azeotrope
and the corresponding Azeotropic the condition and the corresponding Azeotropic
composition, how to proceed?

867
(Refer Slide Time: 21:48)

What are the things that you know? You know at the Azeotropic point y1 equals to x1 at
Azeotropic point you know it already, right? Accordingly what is gamma 1 you know we just
now did P by P1 saturated you know gamma2, P by P2 saturated you know what is ln
gamma1 is ln of this; ln gamma 2 is ln of this. Now suppose you subtract one from the other
what will you get? You will get ln P2 saturated by P1 saturated is nothing but A into x2
square minus x1 square, isn’t it?

And this is nothing but equals to A into 1 minus 2x1, agreed? So therefore we find that the
condition of the Azeotrope should be related to this particular condition and where does the
and from this particular condition you know that A into 1 minus 2x1 equals to this. So from
there you can find out x1 this is nothing but half into 1 minus 1 by A ln P2 saturated by P1
saturated. So this is the corresponding composition where the Azeotrope forms, this is the
liquid phase composition of component 1 where Azeotrope forms.

Now under what condition does it form? We know that the Azeotrope has to form between 0
and between the composition of X1 lying between 0 and 1. So therefore Azeotrope forms for
0 less than x1 less than 1, right? So for x1 equals to 0 what do you get? You get x1 equals to 0
this part has to be equal to 0 for x1 equals to 1 you know 2 has to be equal to this particular
part. So therefore once you have obtained them you find that for X1 equals to 0 A should be
ln P2 saturated by P1 saturated and for X1 equals to 1 A should be minus ln P2 saturated by
P1 saturated.

868
So therefore when does this system form an Azeotrope can you tell me? System forms an
Azeotrope between X equals to 0 to 1. So it has to form an Azeotrope when A has to be
greater than ln P2 saturated by P1 saturated. Now we have a minus value we have a plus
value so it is safest if we can denote it as A has to be greater than the absolute value of ln P2
by P1 saturated for an Azeotrope to be formed.

Well, so there is one other case also where the calculation of gammas or in other words the
VLE calculation becomes much simpler. Usually it’s quite well practiced for hydrocarbons
and this is known as the K factor. We can use the K factor and find out the VLE data for
hydrocarbons in quite a simple fashion.

(Refer Slide Time: 25:29)

Now what is this K factor? Let us see we find that formation of hydrocarbons usually the
intermolecular forces are quite weak. Now since the intermolecular forces are quite weak, so
therefore under this particular condition we find that this was the original equation that we
had taken. Now when the intermolecular forces are weak, again we can assume the vapor
phase to behave as an ideal gas or ideal mixture.

869
(Refer Slide Time: 26:01)

Although it is important to remember that individual gases they do not behave as ideal gases,
so naturally for that case what happens? For that particular situation we find that phi i v
becomes equals to this particular term this phi i v it is equal to the pure component phi i right
and therefore and usually we find that the liquid phase it behaves ideally.

So therefore usually for hydrocarbons we find that this term becomes equal to one this term
becomes equal to one this more or less this becomes equal to the pure component vapor phase
and this term also it disappears off.

(Refer Slide Time: 26:41)

870
So therefore very frequently what is done is Ki for each particular component the K factor is
defined which is nothing but the ratio of the mole fractions in the vapor and the liquid phase.

(Refer Slide Time: 26:48)

(Refer Slide Time: 26:51)

Now from this equation we find what should be the K factor it is gamma i which is usually
not there Pi saturated phi i saturated by phi i v into P exponential term definitely remains
because the liquid behaves as an ideal mixture but if you’re dealing with higher pressures
than the poynting correction factor can be found out. So therefore what did we find?

By assuming that phi i v equals to phi i v or in other words the gas mixture behaves as an
ideal solution we found out that we are position to find out Ki from all the properties of the

871
pure components of or rather we can find out the Ki from the pure component properties of
component i in the vapor phase and in the liquid phase and therefore very frequently what we
find? That Kis have been calculated for a large number of hydrocarbon gases and then they
are tabulated as functions of T and P.

We find naturally since they are properties of the pure components, they are functions of the
pure component properties so therefore they are functions of T and P and lot of effort has
been put to plot the K values of individual hydrocarbons as functions of T and P either in the
tabular form as monograms.

(Refer Slide Time: 28:44)

Now the monogram which has been proposed by DePriester and then it has been modified
later is as shown in this particular here, here we find that the monograms are presented for the
low-temperature range and high-temperature range and these with the help of these
monograms where we find that we for each and every hydrocarbon gases if we know the
temperature we know the pressure, we are in a position to find out the K values the
calculation becomes pretty simple.

872
(Refer Slide Time: 29:08)

e
We will just be doing 2 problems and finish off this part. First problem is a vapor mixture,
now here it is written that this particular vapor mixture it has y methane it is 0.2 it’s written,
written that y methane is 0.2, y ethane this is 0.3, y propane it is 0.5, t is given as 30 degrees
centigrade, what are you supposed to find? You are supposed to determine the pressure at
which condensation begins. How to proceed with this particular case?

Since P is not given you’re not in a use this particular K factor chart, you have to guess some
particular P.

(Refer Slide Time: 30:08)

So let us see for the time being let me take the first guess as 2.15MPa.

873
(Refer Slide Time: 30:30)

Once I have guess this t equals to 30 degrees centigrade we can draw a straight line from 2.15
megapascal it’s roughly here and it’s about the temperature is 30 degrees centigrade. So if we
draw a line we are in a position to find out the y factors for methane, for ethane and for
propane.

(Refer Slide Time: 30:53)

Let us see what are the y factors? We find that for methane the K factor it is 8.1, for ethane it
is 1.82 if you draw this straight line you will be able to locate it and for propane it is 0.62. So
once you know this once you know the y values it is very easy to find out the corresponding
mole fractions in the liquid phase, isn’t it? This comes to 0.0247 this comes to about 0.1648
and for the propane case it's it reduces to 0.8065 if you take the Sigma X1 in this case it’s
almost

874
equal to 0.996. So therefore we can say that my assumption of pressure was more or less
correct. For greater accuracy you need to alter this slightly and then if you if you get Sigma xi
greater than 1 then by linear interpolation you can take the intermediate pressure, okay.

(Refer Slide Time: 32:12)

Accordingly you can find out the dew pressure, once you know the dew pressure then with
the dew pressure and with the temperature you can find out the K values once more and by
using the K values even find out the composition of the first liquid which appears this we had
done when the vapor phase composition was given.

(Refer Slide Time: 33:47)

875
We can also proceed similarly if the liquid phase composition is given as well there is one
more problem which we can do in this particular case we find that x of we find that in this
particular case x ethylene is provided and x propylene is provided it is there, this is C2H4 and
C3H6 this is 0.25 we know this is 0.75 we know the initial temperature was minus 40 degree
the pressure is given it is 1MPa.

Now we it’s required to find out the temperature at which the first vapor is going to appear
again the same thing there we had assumed the pressure here we are going to assume the
temperature. Say for the time being we assume some particular temperatures they are close to
minus 10 degree centigrade or something maybe minus 8.5 or something, actually minus 8.5
is the correct answer I know.

So you can start with minus 10 degrees centigrade then you can assume minus 6 or 5 and then
by linear interpolation you can find out the actual temperature once you know this
temperature you know the pressure again you are supposed to draw, it’s not a straight line you
are supposed to draw straight line and then you are supposed to find where it cuts the
ethylene where it cuts the propylene.

(Refer Slide Time: 33:58)

So in this particular case we will find that your K ethylene if you observe, your K ethylene is
going to be 2.55 provided I have assumed it as minus 8.5 degrees centigrade P equals to
1Mpa and we find that your K Propylene is going to be 0.49, right? So from there you can
find out your y ethylene that is going to be 0.6375 and your y propylene it is going to be
about 0.3675 so the sigma yi you’re going to get is 1.0050.

876
So therefore it can say that the bubble temperature was nothing but equal to minus 8.5 degree
centigrade and the corresponding vapor composition can be given by the compositions that
we found out.

(Refer Slide Time: 34:52)

The only thing which we need to do is that we have to start with a guess value; it’s not very
difficult to guess it properly because it has to lie between the boiling points of the 2 liquids.
So we can start with and more or less we know the compositions, so we can start with the
weighted mean boiling point and we can go and we can proceed.

So in this particular way what I did was? We discussed different non ideal solutions, the
methods of generating the complete Pxy or the Txy date from fragmentary data and how the
situation gets simplified when we are dealing with Azeotropic when there is an Azeotrope
formation and with the use of K factors, how the situations gets simplified when we are
dealing with hydrocarbons mixtures. Now we have been dealing with large amount of rather
we have been discussing for quite some time regarding the VLE data of different nonideal
solutions.

It is important for us to know that since we are dealing with so much of experimental data
whether the data that we have collected or generated is thermodynamically consistent or not
whether the data is reliable whether we can proceed with the data. So in the next class we
deal with thermodynamic consistency of VLE data before we proceed with any other topic,
thank you very much.

877
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 48
Thermodynamic consistency test of VLE data
Well, hello everybody so long you must have a lot of dose about nonideal solutions activity
coefficients, how to find out activity coefficients, Azeotropes, K factors, van laar, Margules
and Wilson and NRTL lot of things. Just it’s important for you to remember that for the
purpose of this particular course you just need to remember the 2 suffix Margules equation
and you need to know how to solve problems using the 3 suffix Margules equation and the
van laar equation.

In your assignments and in your tests the basic equations for ln gammas will be provided and
so that you can proceed but you need to know how to do problems just the way I have done
the problems in the last class. Now before ending the last class what did I say? I said that it’s
very important for you to know that when you are dealing with the experimental data that
you’re dealing with thermodynamically consistent data.

What does thermodynamics consistency of any data mean? It means that the data obeys
Gibbs Duhem equation for any partial molar property or for any partial molar excess
property. It’s important for you to remember that if any particular thermodynamic data it does
not obey Gibbs Duhem equation it is definitely incorrect but when it obeys Gibbs Duhem
equation there is no guarantee that it is completely correct.

878
(Refer Slide Time: 1:43)

But it can often happen that by a fortunate cancellation of errors we find that the particular
quantity it obeys or rather it obeys the Gibbs Duhem equation under that condition we can
use the data it can be considered to be reliable for or thermodynamic purposes but it is not
correct such a case is unlikely but it cannot be ruled out completely therefore we are not
dealing with thermodynamic correctness of data we’re just trying to find out how we are
going to ensure that the data are thermodynamically consistent?

879
(Refer Slide Time: 2:30)

Our main guide is the Gibbs Duhem equation, what is the Gibbs Duhem equation let us see?
All of us we know that Gibbs Duhem equation it can be written in terms of any particular
partial property where we know this mi bar refers to any particular partial molar property for
the extensive property M. Now we know that the same equation can be written in terms of
molar partial molar excess property as well.

And we know that this equation it has got 2 very important applications which we have
already discussed in our previous classes, the first important application is that suppose we do
not have the complete thermodynamic data then in that case suppose we have data for one
suppose we are dealing with C number of components we have say the activity coefficients
for C minus 1 components we can find out the activity coefficient for the cth component
using the Gibbs Duhem equation this we have already discussed.

The other important thing is, suppose we have data over the entire range of composition then
in that case we can use the data to find out if the experimental data which is available for any
particular directly measured partial property whether it is thermodynamically consistent or
not, how do we go about doing this?

880
(Refer Slide Time: 4:06)

So what is the Gibbs Duhem equation? We have already derived this particular equation and
we know that this is the generalized Gibbs Duhem equation, right? And usually I have
already mentioned that at constant temperature and pressure this particular term cancels out
this particular term these 2 cancels out as we get this particular equation.

And for our practical purpose we find that mostly we will be dealing with the equation for
binary mixture. So for a binary mixture under condition of constant temperature and pressure
the equation reduces to this particular form. And we now therefore that is if we have the
complete VLE data what you mean by the term complete VLE data?

881
(Refer Slide Time: 5:03)

It means you have data on P, T, x1 and y1 if you have the complete VLE data then for a binary
system that we can use this equation to find out whether the data is consistent or not.

(Refer Slide Time: 5:28)

Now just by looking at this equation can you tell me how we can use this equation to find out
the thermodynamic data? What is the simplest method? Just by looking at the equation it is
very evident that you’re x1 d ln gamma 1 by dx1 is equal to x2d ln gamma 2 by dx2, so what
does it suggest?

882
(Refer Slide Time: 5:53)

It suggests that if you’re having a plot of say ln gamma 1 versus x1 and if you’re having a
plot of say ln gamma 2 versus X1 then in that case suppose we are having this particular plot,
right? This is 0 this is 1and at X1 equals to 1 your ln gamma1 that becomes equal to 0, right,
say suppose? And this is a ln gamma2, so this is one particular method which is evident from
here you select a number of points here and once you select a number of points you can draw
tangents and you can find out the slopes at each and every point.

(Refer Slide Time: 6:47)

For the same compositions you can find out the slopes of ln gamma 2 versus X2 as well and
then once can find out the slopes then you can use this particular equation to find whether the
data is thermodynamically consistent or not. This particular technique it looks to be pretty

883
accurate and it also appears to be quite simple at least theoretically it appears to be quite
simple.

But there are 2 things which you need to remember firstly finding out slopes in this particular
way it’s not very easy and second thing is that the experimental data they will inevitably be
having some amount of scatter, so therefore it’s not so easy and feasible to find out slopes at
different particular points very accurately and then to check up whether the equation whether
this particular equation is going to be valid or not.

So therefore usually this equation it is used for some very serious errors in the data for
example you can just plot ln gamma1 as the function of x1 ln gamma2 as a function of x2 and
find out whether both the slopes of positive or not or else if you know that ln gamma1 versus
x1 is horizontal at any composition you can just check that whether at that composition ln
gamma 2 versus X1 is horizontal as well.

Just these gross checks can be done from this particular slope technique but in order to check
the thermodynamic consistency of data we find that a better method is the integral area test or
the integral test or the area test which was proposed by Redlich and Kister and also by
Herington independently the test the data as a whole.

(Refer Slide Time: 8:37)

Let us see what this integral of the area test is and how it works? So therefore for this, what
we do? We write down the basic equation which relates molar Gibbs free energy to activity

884
coefficient. What was that particular equation if you remember? We have been using this
equation since the time we have introduced nonideal solutions.

So for a binary mixture what happens to this equation? This also we have done a large
number of times. If you observe you will find that for that we do not use a large number of
equations but we used the same equations in different forms to serve our purposes like for
example this equation we have been using very frequently and in different forms may be for
binary mixtures, for multi-component mixtures we use this.

So for a binary mixture what do we get? This equation reduces to X1ln gamma1 plus X2 ln
gamma 2 if you differentiate this particular equation, say with respect to X1 what do we get?
We get d gE by RT dX1 this should be equal to ln gamma1 plus d ln gamma1 dX1 plus ln
gamma 2 dX2 dX1 plus X2 d ln gamma2 dX1, what is this term?

We know dX1 equals to minus dX2 therefore this term is nothing but minus 1. So from here
what do we get? We get ln gamma1 minus ln gamma 2 into dX1 plus X1 d ln gamma1 dX1
plus X2 d ln gamma2 dX2, what is this? This is one of the terms of the Gibbs Duhem
equation. What is this? At constant temperature and pressure we get that this particular term
equals to, this suppose I write this term as a then this a term becomes equal to 0 provided we
are working at constant temperature under isothermal isobaric conditions.

So therefore we know that for this particular condition we find d gE by RT dX1 this is dX1
by dX1 this is equals to ln gamma1 minus ln gamma2.

885
(Refer Slide Time: 11:33)

Or else what can we write this down as? We can write this down as d gE by RT equals to ln
gamma1 by gamma 2 of dX1. Now we can integrate this particular equation, isn’t it? We can
integrate this particular equation say from X1 equals to 0 to X1 equals to 1. Here also we
integrate it from X1 equals to 0 to X1 equals to 1. What do we know? We know that at X1
equals to 0 we have pure component X2.

So naturally when we have pure component X2 gE by RT becomes equal to 0. At X1 equals

to 1 what do we have? We have pure component 1, isn’t it? When we have pure here we have
pure component2. When we have pure component 1, naturally under that case also gE by RT
becomes equals to 0. So therefore what is the fate of the left hand side of the equation?

The left inside of the equation when we are integrating it from 0 to 1 this term it disappears
off, right? So therefore what do we find? We find that X1 equals to 0 to X1 equals to 1 ln
gamma1 by gamma 2 dX1 this should be equal to 0 for gamma1 and gamma 2 data to be
thermodynamically consistent, isn’t it?

And so therefore this is what it is the area test and remember one thing we can and perform
the area test for any particular partial molar property but the most relevant partial molar
property is the partial molar Gibbs free energy which is given as gi bar E this is the most
relevant property and therefore the Gibbs Duhem equation expressed in this particular term is
most important since we know that gi bar E is related to the activity coefficients and it’s very
important to find out whether the activity coefficients are thermodynamically consistent or
not.

886
And this can be found out if we have gamma1 and gamma 2 values that we can find out ln
gamma1 for each case we can find out the ln gamma2 for this case and then we can find out
ln gamma1 by gamma2 for each case and then we can plot ln gamma1 by gamma 2 as a
function of X1 and we can more or less we get something of this sort the find out the area
above this area below this and then we find whether the areas are equal or not.

Accordingly we can conclude whether the data that we have it is thermodynamically

consistent or not.

(Refer Slide Time: 14:30)

So therefore with this it is quite evident that the basis for testing thermodynamic consistency
of VL E data is given by this particular equation.

887
(Refer Slide Time: 14:51)

And we also know that we are dealing with experimental data. Now since we are dealing with
experimental data, so therefore it can always happen that due to experimental uncertainties
we will definitely not get the area to be exactly equal to 0. So therefore keeping this in mind
normally it is thought that the area above the x axis minus the area below the x axis divided
by the total area has to be less than equal to 2percent.

(Refer Slide Time: 15:17)

So this is generally the situation which is which we get. Now there are certain things till this
much is sufficient for this class but while you’re using some particular technique to find out

888
thermodynamic consistency it is also important for you to remember the drawbacks of this
particular technique.

(Refer Slide Time: 16:30)

Now in this particular technique what do we do? We have firstly we have assumed constant
temperature and pressure, suppose we do not assume constant temperature and pressure then
the total equation then we actually start with the Gibbs Duhem equation Then once we start
with the generalized Gibbs Duhem equation, I will just write down the generalized equation
the actual equation it should be X1 equals to 0 to X1 equals to 1 ln gamma1 by gamma 2 dX1
this should be equal to actually hE by RT square dT ranging from the temperature at X1
equals to 0 to the temperature at X1 equals to 1 minus vE by RT dP while pressure the
corresponding pressure at X1 equals to 1 to X1 equals to 0, this is the generalized equation at
constant temperature and pressures since both these terms they disappear off. Therefore we
can take this to be equal to 0 but actually this forms the basis of thermodynamic consistency
test for VLE data.

We know that for isothermal cases, what happens? The first term it disappears off, right? For
isothermal cases we know that this term should disappear off and we know that for low
pressure cases this term should disappear off. So therefore for isothermal low-pressure VLE
data the thermodynamic consistency test is given by gamma1 by gamma 2 dX1 equals to 0.
So if experimental data satisfy this then we are happy but remember one thing if we are
dealing with rigorous testing, then in that case what we should be doing?

889
We should be plotting vE by RT as a function of P and then we should we should be finding
out the area here we should also be plotting hE by RT square as a function of T where T it
ranges from T at X1 equals to 0 to T at X1 equals to 1. Here it should range for P at X1
equals to 0 to P at X1 equals to 1 and then we should be finding out these particular data and
then we should be comparing and then only we can say that really the data is
thermodynamically consistent.

But generally what happens? We do not have sufficient data on the excess property of volume
and the excess property of the hE as a result of which what we do? We generally assume
these to be 0 and for the purpose of your undergraduate class this equation is sufficient for
you.

(Refer Slide Time: 19:11)

Now when do we have this equal to 0 can you tell me? We have this equal to 0, naturally
when your excess volume is 0 or in other words the molecules of the 2 components they are
similar in size. Only under that condition we can have v equals to 0.

890
(Refer Slide Time: 19:16)

Can you tell me under what condition we have hE equals to 0? Naturally when we mix the 2
components and there is no enthalpy generation, when will it happen? Firstly when the 2
components are chemically similar and there are no Azeotropes formation and the boiling
points of the components are also close to one another. Under that condition we’re going to
have this case.

(Refer Slide Time: 20:08)

But under all other conditions it is very difficult to assume this particularly it is important for
you to remember that when we are dealing with gamma 1 by gamma 2 actually what are we
doing just try to understand, this is given by x1f1 0 by phi2 v y2 by x2f2 0. Therefore what
are we doing?

891
We are actually cancelling out the pressure term and that is the term which is very important
because that can be measured very accurately and it’s very easy to know the total pressure
conditions. So therefore when we are doing this area test we are actually cancelling out the
pressure term and that is the main limitation of this test and therefore if we have to do it
accurately the way I was telling we have 3 data we have for isothermal data we can vary Pxy
we should be varying we should be generating data on Pxy and on Py as well as on Px.

And then we from the 2 data which should be finding out the third compare that with the
experimental value only under that condition we will be convinced that the data are
thermodynamically consistent but for your purpose it is sufficient for you to remember that if
you plot ln gamma1 by gamma 2 versus X1 and find the area to be 0 your purpose is served
again say that the data is thermodynamically consistent.

If you recall that we have done a large number of problems in the last class regarding finding
or generating Pxy data Txy data etc. Now it’s very important for you to test whether those
data were thermodynamically consistent or not. So we have we are taking up one particular
problem from the set of problems that we had.

If you remember that in this particular problem you had already calculate it gamma1 gamma
2 and so therefore from this particular data it’s very easy for you to calculate ln gamma1 by
gamma 2, right?

(Refer Slide Time: 21:53)

892
So I will be writing down the corresponding ln gamma1 by gamma 2 in this particular case. If
you remember the data were 0.297, 0.447, 0.5650, 0.650 and am going to write a good
number of data’s, so that you can check up whether you have done it correctly or not. X1
0.926, 0.945 the corresponding ln gamma1 by gamma2, gamma 1 and gamma 2 we had
already calculated in the last class.

So therefore you can calculate this 0.2234 this was minus 0.415, minus 0.2245 then it was
minus 0.3496, minus 0.4668, minus 0.6396 this was minus 0.6908 and minus 0.7056.

(Refer Slide Time: 23:09)

So with these, what you can do? You can plot ln gamma1 by gamma 2 versus X1, right? And
we find that in this case the area about the x-axis this particular portion this area it
corresponds to 1760 whereas this part of the area it corresponds to 2340 at as it is written. So
therefore if you subject it to the criteria that we had that area above X minus the area X
divided by the total area then we find that in this particular case your A minus B by A plus B
this gives you as 0.1415 which is definitely greater than 0.021.

So with this, what do you conclude you find out that in this case the data is not
thermodynamically consistent we had wasted a good amount of time in finding out the data
and in using the 2 suffix Margules and the van laar equation but in reality the data was not
thermodynamically consistent at least from the fragmentary thing that you have done you it
appears that it is not thermodynamically consistent.

893
But it is very difficult for us to conclude it in such a simple fashion because for this particular
case you are required to find out in this case the hE by RT square, right? hE dT by RT square
just as I had said and this has to be found out and this has to be its required to verify whether
this is equal to 0 or not.

If you find that this is equal to 0 then only we can conclusively say that area test or the it
shows that the data is not thermodynamically consistent otherwise it is very difficult for us to
predict this and so therefore this possibly highlights the limitations of the area test to test the
thermodynamic consistency of data and it is important for us to remember that the
assumptions of excess volume energy of mixing as well as the excess enthalpy change of
mixing is also always not very accurate, right?

But anyhow it’s very important for us to remember that even though ln gamma1 by gamma2
as the area test has got some limitations there are some very important applications or rather
are there some very important implications of this particular term ln gamma1 by gamma 2.

(Refer Slide Time: 26:05)

Let us see what are the important or rather what are the interesting things which we which
this particular term ln gamma1 by gamma 2 has? We find that this is a very effective
parameter in defining the different types of mixtures into simple, more complex and most
complex solutions, how? Suppose we plot ln gamma1 by gamma 2 as a function of X1, we
find that suppose for simple mixtures like say hexane and toluene in this very lowercase we
find that it is more or less a straight line and we immediately say that this particular solution
at the pressure condition it is straight line.

894
On the other hand suppose we plot the law gamma1 by gamma 2 versus X1, for benzene
isooctane we find that in this case it is curved. It is interesting to note that for both the cases
we have been dealing with one particular aliphatic hydrocarbon one particular aromatic
hydrocarbon despite that we found that in this case we had obtained a linear curve this case
we have obtained a curved solution.

And this can be attributed to the difference in the sizes of the molecules constituting benzene
and isooctane and hexane and toluene while the molecules are similar in size while for
hexane and toluene there is much difference between the sizes for benzene and isooctane. So
although we are dealing with similar type of similar type of components we find that the
graphs of ln gamma 1 by gamma 2 versus X1 are not the same while hexane toluene confirms
to simple solutions.

We find that benzene isooctane it does not confirm to simple solutions under this particular
pressure condition and interestingly when we go for some other mixtures we find that it’s not
only that it is curved there is also a point of inflection, for example in the case of ethanol
Methylcyclohexane, now what does this complex it imply?

(Refer Slide Time: 28:26)

From the graphs when we know that this equation is simple, this is more complex, this is
most complex how do we use this information? We use this information if you recollect that
we had we had referred to the Redlich Kister equation which was an extension of the 2 suffix
Margules equation it simply added more number of terms and by that we could have a much
more realistic expression or equation to express gE by RT as a function of compositions.

895
Now from this equation we will find that we can we can find out ln gamma1 by gamma 2 and
from there we can find out ln gamma1 by gamma 2. So if you compare this expression as
well as the graphs here, what do we see?

(Refer Slide Time: 29:07)

If we compare this particular equation as well as the curves here we find that for the first
curve it is sufficient to use just one particular term in ln gamma 1 by gamma 2. So therefore
the first curve represents those solutions which can be accurately predicted by 2 suffix
Margules equation. For the second case we find that for this particular case we have to
include the first and second case for the first and the second terms.

And we find that for the last case which is the most complex or highly complex mixtures an
accurate prediction can be obtained if we if we include the first 4 terms in this case. So
therefore we find that ln gamma1 and gamma 2 is water good amount of or rather it has got a
very good amount of importance in classifying mixtures as simple, more complex and most
complex but at the same time it is important to remember that this classification is more or
less arbitrary because there can be solutions which may due to fortunate cancellations of
errors again which may be predicted by one or 2 parameters but actually in reality they are
much more complex.

A typical example of this is the mixtures of 2 alcohols say methanol and ethanol we find that
very frequently they can be well predicted by just using one or 2 parameters or the Redlich
Kister equation but in reality they are much more complex due to hydrogen bonding.

896
(Refer Slide Time: 30:47)

Another very interesting example we can find in the case of acetic acid and water. If you
observe the excess Gibbs free energy curve as a function of composition you find that it
resembles the 2 parameter Margules equation therefore you will be tempted to say that acetic
acid water it forms a simple solution but when you are when you go compare the hE curve
and the TsE curve immediately it shows in reality it is a complex mixture and the similarity of
the excess Gibbs free energy curve with the 2 suffix Margules equation is obtained just due to
a fortunate cancellation of errors.

So therefore with this I conclude this class and before ending I would like to mention that
most of the solutions that we deal in chemical engineering solutions of non-electrolytes they
can very well be predicted by including 2 to 3 parameters in the Redlich Kister equation. So
therefore most of them they are solutions of intermediate complexity. So with that I would
like to emphasize that we had discussed the area test for discussing thermodynamic
consistency of experimental data and I had also discussed the limitations of the area test it is
very important to remember that it is applicable only when your gamma1 gamma2 they do
not vary remarkably with temperature and pressure and for most of the cases for a much more
rigorous testing we need to calculate or rather we need to consider the excess volume change
and the excess enthalpy change of mixing of the solutions.

And apart from that I also discussed the other importance of ln gamma1 by gamma 2 this
particular parameter it is not only used for finding out the thermodynamic consistency of data
is also used for classifying data or rather classifying solutions are simple, complex and more

897
complex and they give us an idea regarding the number of parameters of the Redlich Kister
equation which should be included in order to predict the VLE data of that particular mixture.

And again I would like to emphasize that this is quite arbitrary although it is it is generally
used it is quite arbitrary for several particular solutions like say acetic acid water methanol
ethanol etc and it can give some misleading results due to fortunate cancellations of errors.
Now with this we say we are going to finish our topics of solutions both ideal and nonideal.
So far we have been dealing with completely miscible systems we are going to discuss some
more interesting aspects of VLE in the next class, thank you very much.

898
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 49
Retrograde condensation

Well, so we have come to the almost the end of your nonideal solutions and there is just one
part which there is one very interesting thing which I would like to discuss before we go on to
our next topic on partially miscible and completely miscible systems. Now we have already
discussed when we were discussing miscible systems we have already discussed Pxy
diagrams Txy diagrams and xy diagrams.

(Refer Slide Time: 0:48)

Let us see what happens or rather what are the characteristics of pressure temperature
diagrams under constant quality conditions.

899
(Refer Slide Time: 1:01)

For pure substances if you see we find that the pressure versus temperature curve particularly
on the liquid vapor coexistence liquid vapor coexistence curve we find that it increases the
pressure increases monotonically with temperature and the curve finally ends at the at the
critical point as we have discussed quite a number of times and we know that this critical
point it is the point of highest temperature and pressure where the 2 phases can coexist and
beyond the beyond this particular condition the 2 phases they become indistinguishable.

And I have shown the typical PT the curve for a single phase or rather for a single component
system for both water and carbon dioxide; we find that for both the cases the characteristics
of the liquid vapor coexistence curve is the same.

900
(Refer Slide Time: 2:12)

Now let us take up the case for a simple binary mixture. Let us take up the case of a simple
binary mixture this case if you observe a typical pressure temperature diagram, what do we
find? We find that at a fixed composition each curve it is marked by a bubble point curve it is
marked by a it is enclosed by a Dew point curve and the bubble point curve and the dew point
curve they meet at the critical point. We find that the bubble point curve it comprises of only
liquid phase dew point curve it comprises of fully vapor phase and in between there are a
large number of curves which are actually constant quality lines, right?

Now if you closely observe we find that while we were discussing pure components we found
out that the liquid and vapor can coexist under equilibrium conditions at the highest point of
the critical point but if you observe the 3 diagrams which I have given here you can very well
find out that the liquid and vapor they can coexist at temperatures higher than the critical
point at pressures higher than the critical point.

So therefore the critical point for the case of a mixture is not necessarily the highest point
with respect to temperature and pressure the liquid and vapor may coexist at temperatures and
pressures higher than the critical point.

901
(Refer Slide Time: 4:08)

And the point of existence of liquid and vapor at the highest pressures this particular pressure
is known as the maxcondenbar pressure or the cricondenbar pressure. In the similar way we
find that the highest temperature at which the liquid and vapor can coexist, it is known as the
max condentherm temperature of the cri condentherm temperature. So therefore we find that
unlike single phases definitely the bubble point curve and the dew point curve they intersect
at the critical point the single phases, the bubble point curve and the dew point curve they are
the same and therefore the critical point is a point of highest pressure and temperature
conditions under which the liquid and the vapor can coexist.

In this case also the critical point is the point of intersection of the bubble point curve and the
dew point curve but it does not signify the highest temperature and pressure condition under
which the liquid and vapor can coexist. They can coexist at temperatures and pressures higher
than the critical point the highest temperature at which the liquid and vapor can coexist is
known as the cricondentherm or the maxcondentherm temperature the highest pressure at
which liquid and vapor can coexist is known as the maxcondenbar pressure or the criconden
bar pressure.

And there are some very interesting things which happen in this particular region which is
bounded by the critical point the maxcondenbar point and the maxcondentherm point. We
find that there are 2 or 3 ways by which we can locate the MP and MT with respect to the
critical point, one is both of them can be either on the bubble Pt curve which is the case in
this case this is not very common case both of them can exist on the beyond the critical point

902
on the dew point curve which is quite a common case and the third one is while Mp lies on
the bubble Pt curve Mt lies on the dew point curve.

(Refer Slide Time: 6:51)

So therefore MP and Mt they can be located with respect to the critical point or they can
replace with respect to the critical point in 3 ways one is both lying on the bubble Pt 1 is both
lying on the dew point and the third is one lying on the bubble Pt while the other lies at the
dew point. Now very curious and peculiar phenomena occurs in the region which is bounded
by MP critical point and the MP, Mt and the critical point, what happens under this particular
condition let us see?

Now suppose we deal at a much lower temperature pressure conditions, what do we find? We
have already discussed it suppose we start with the vapor phase and we start compressing the
vapor at constant temperature, so therefore what happens the vapor it follows this particular
vertical line and we find as the vapor gas gets compressed at one point it reaches the dew
point curve, when it reaches the dew point curve the first dew appears.

Now if you continue this particular compression process we find that greater and greater
amount of liquid is formed and finally when we reached the bubble Pt curve the last drop of
vapor it disappears and then we have entered a single phase liquid region this is true for all
the different phase diagrams that I have shown but suppose we deal at the condition or rather
under a condition which is bounded by a critical point Mp and Mt there we find that the
counterintuitive phenomena occurs.

903
What happens? Suppose we take up the vapor phase at State one say and we start expanding
the vapor phase under isothermal conditions naturally when temperature is constant and we
start expanding it follows this particular vertical line. So as we continue this particular
expansion we find that the vapor it first reaches the point 2 which lies on the dew point curve.
So naturally under this condition the first dew is formed. If we continue this expansion
process we find that the proportion of liquid it keeps on increasing as we move from point 2
to point 3. If you observe we have shifted from a higher to a lower quality line. We continue
this expansion process and we reach point 4 again the proportion of liquid increases just the
thing which had happened under this condition.

But if we again continue the expansion process we find that from 4 to point 5 actually the
proportion of liquid has decreased. If you observe closely you will find that from point 4 to
point 5 you have actually moved from a lower quality to a higher quality line and if you
continue this at point 6 you again reach the dew point curve and beyond that when you
continue this expansion process you result again in the vapor phase.

So we find that for conditions where the constant temperature line it crosses the dew point
curve not once but twice and never touches the bubble Pt curve under this condition we find
that if we expand the vapor phase then we do result in a liquid phase but at the end we again
land up in a vapor phase, counterintuitive phenomena which does not occur at lower
temperature pressure conditions this particular phenomena is known as retrograde
condensation and here it is known as retrograde dew formation at constant temperature this is
what that has happened here.

The same thing can not only happen by expanding the gas but it can also happen by cooling
the gas suppose for example that we start from point 8 and in this case also we ensure that the
point where we have started that is again bounded by the dew point curve Mp, Mt suppose
the point 8 is at a much lower pressure conditions where it is not bounded by this zone. In this
case suppose we start heating the, in this case we have a liquid phase if we start heating the
liquid phase we find that the first bubble forms and then if we continue this heating process
finally we end up in a vapor phase.

Now for example instead of starting at this point we start at a higher pressure temperature
condition the condition which is bound or which is beyond the critical point. So therefore
under this condition since it is beyond the critical point naturally we do not have a liquid
phase like this we have a vapor phase in this particular case. We start heating the vapor phase

904
what happens? We first come and touch the dew point curve moment we touch the dew point
curve the first dew appears.

Now if you continue this particular heating process the isobaric heating process we find that
the proportion of liquid first increases and then it decreases finally the system it again touches
the dew point curve and then it goes into the single phase vapor region, so therefore this is
known as retrograde dew formation at constant pressure. Now in this particular situation or in
this particular system where both the Mp and the Mt they lie on the dew point curve under
this condition we have retrograde dew formation at constant pressure, retrograde dew
formation at constant temperature and these are known as retrograde condensation of the first
kind.

Just the reverse thing happens if instead of on the dew point both Mp and Mt they lie on the
bubble Pt curve. In this case also suppose we are operating at lower temperature pressure
conditions we find that we take a vapor we start heating the vapor at constant pressure the
first dew appears when it touches this point and then if we continue we find the proportion of
liquid increases and finally we have entered the single phase liquid region but suppose we are
operating under this particular condition.

So for this particular condition what do we have? Again it was not heating it was
compressing, anyhow we have compressed vapor and we have come to the liquid phase I was
wrong in telling, anyhow. And if we continue this compression process say instead of starting
from say point A we have started from point 9 and we continue the compression process at
constant temperature.

What happens? We find as we keep on compressing we reach the dew point the first dew
appears it continue with the compression process the proportion of liquid first increases then
decreases finally the vapor is reformed when we reached point 12. Same thing can be
observed when we continue isobaric cooling of the vapor phase. If we continue isobaric
cooling under some other condition we find that you can keep on cooling the vapor phase
finally from the vapor we can come to the liquid after traversing all the constant quality lines
and we go in the direction of decreasing constant quality lines.

If suppose we start at point 1, right? Which again forms in the zone bounded by Mp M(t) and
the critical point C, what do we do? We start cooling the vapor we continue the cooling till
we reach the dew point curve the first dew appears then we continue the cooling process the

905
proportion of liquid increases this continues till we come to state 4 and then if we still
continue the cooling process we find that again from state 5 we are shifting towards higher
quality lines and finally at point 7 the last drop of liquid disappears and we have entered into
a single phase vapor zone.

Now this particular phenomena which was just the reverse of the other this is known as the
retrograde condensation of the second kind where we get retrograde bubble formation at
constant pressure or we can also obtain retrograde bubble formation at constant temperature.
So the first kind retrograde condensation of the first kind occurs when both Mp and M(t) lie
on the dew point curve and the retrograde condensation of the second kind occurs when both
of them are lying on the bubble Pt curve and very interestingly when we find that Mp lies on
the bubble Pt curve and Mt lies on the dew point curve we find that for this particular
situation we can get the retrograde condensation of the first kind as well as the retrograde
condensation of the second kind.

For example if you in this case, if you start you’re cooling the vapor from state one, what do
you find? That if you start cooling the vapor from state one and to continue the cooling we it
first touches the dew point curve the proportion of liquid increases and then after awhile the
horizontal line it starts going from a lower quality to a high quality line and finally the vapor
is reformed here.

So therefore this particular horizontal line this suggests retrograde condensation of second
kind where the similar to what we have seen in this figure and if we follow this isothermal
compression at constant temperature then under this condition we find that initially with the
vapor phase we get the first drop of liquid the proportion of liquid increases and then it
decreases and finally we land up into the single phase vapor region.

So therefore while this path refers to retrograde condensation of the second kind this vertical
part refers to retrograde condensation of the first kind. So therefore when the Mp and Mt are
2 sides of the critical point we find that both the types of condensation can be obtained this
particular case.

906
(Refer Slide Time: 17:47)

(Refer Slide Time: 18:14)

Well, what was the need of discussing this entire thing? The need of discussing these entire
phenomena was because it is of great importance in natural gas wells, natural gas wells as we
know they are very high temperature and very high pressure and for such a case we find that
suppose we are dealing with usually the conditions are at point 1, okay. So therefore if we are
dealing with high pressure and temperature conditions under at point 1 and we keep on
drawing the natural gas from it we find that the pressure keeps on decreasing but the
temperature remains constant.

So what is basically happening in the natural gas wells? You basically the gas inside the well
it is following this vertical path 1, 2 as long as it lies above point 2 we have no problems but

907
moment in the process of drawing out the vapor suppose we reached point 2 then what
happens liquid starts forming when liquid starts forming this liquid it seeps into the
underground strata and it gets lost we cannot recover this liquid, number 1.

Number 2 when there is a presence of liquid near the bore pipe it reduces the gas flow rate
therefore naturally the output of the gas decreases and the third important thing is when we
have both the vapor as well as a liquid phase inside the reservoir, naturally they would like to
be in a state of equilibrium and therefore the vapor which is being withdrawn from the bore
pipe it is in equilibrium with the liquid phase inside the reservoir therefore the composition
of the output gas from the well it keeps on changing continuously with time and therefore to
overcome this disadvantage it is essential to maintain the pressure at the well at a value which
is greater than the dew point value corresponding to point 2 here.

So therefore we would always prefer to operate natural gas well along this particular vertical
line but above point 2. So it is very important to know the dew point pressure corresponding
to point 2 and ensure that that we are operating well above the point in order to alleviate the
disadvantages and the limitations which I have already mentioned when liquid appears in the
well.

Well, apart from this there is also another very interesting situation where we can get
retrograde condensation and that happens under constant temperature and pressure
conditions. Let us see when it happens? We have already discussed Pxy, Txy and xy diagrams
at low pressure for completely miscible systems. So suppose we would like to generate the
Txy, xy diagrams at higher pressure let us see what happens?

908
(Refer Slide Time: 20:18)

If you observe this particular phase diagrams you will see that as you go for higher and
higher pressure the zone of existence of 2 phase that decreases and this is of great concern to
the designers of distillation columns because moment zone of two-phase decreases it is very
difficult or rather it is impossible to separate any particular mixture into 2 components. We
would again land up in separating it in into 2 particular liquid mixtures.

Now under this condition suppose we concentrate on the phase diagram at P1 and the phase
diagram at P4. Let us see if you observe them more closely what do we get?

(Refer Slide Time: 21:35)

909
Suppose we observe these more closely then we find that at high pressure what do we get? At
higher pressure if you observe we find that suppose we have taken up a constant temperature
line here we have we have taken the line such that it does not cross the bubble Pt curve but it
crosses the dew point curve twice, right? So therefore under this condition if we if we keep
the temperature and pressure constant and vary the vapor phase composition we have started
with a vapor phase and then some amount of liquid form and the vapor gets reformed at B
and we get pure rather we get only vapor at state 2.

So therefore path 1, 2 this definitely denotes retrograde condensation of the first kind. Since it
is crossing the dew point twice and this is happening when we are dealing with a constant
temperature line and we are varying the composition of vapor phase. We can also get a
similar situation if we are operating again along isothermal conditions at a slightly different
temperature and pressure such that in this particular case instead of crossing the dew point
twice, we in the process of isothermal in the isothermal process we are crossing the bubble Pt
curve twice.

So in this case what we do? We are crossing the bubble Pt twice by varying the liquid phase
composition and by this process we are able to get retrograde condensation of the second kind
where we have achieved this by varying the composition of the liquid phase or rather by
increasing the composition of the liquid phase and travelling along a constant pressure line.
We need to remember that the vertical lines they are constant temperature and pressure lines
because this curve has been generated at constant pressure and the horizontal lines denote
constant temperature lines.

So therefore we find that although at lower pressures, at lower pressure conditions say at
pressure P1 it was not possible to obtain any such sort of retrograde condensation but keeping
everything constant if we just increase the pressure we find since the 2 phase region has
decreased here. So in this case we could demonstrate the retrograde condensation of both the
first kind as well as the second kind. The first kind could be demonstrated when we were
traversing a horizontal path along 1, 2.

In this case the condensation was achieved by varying the vapor phase and we were
traversing the constant temperature line and also since the line it crossed the dew point twice
and the bubble Pt never. Also the same thing happen if we just vary the temperature at the
same pressure conditions i n this particular case retrograde condensation can be
achieved by

910
increasing the liquid phase composition and traversing the constant pressure horizontal line in
this case such that the line it traverses the bubble curve twice and the dew point curve never.

So this is one particular interesting thing which we can obtain if we are dealing with high
pressure vapor liquid equilibrium for completely miscible systems, thank you very much.

911
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 50
Partial and complete immiscibility of liquid mixture
Well, so long we were dealing with liquid solutions where the both the components or the
multi-component mixtures which we were considering they were miscible with each other
over the entire range of composition. Now when they were miscible we found that they could
exhibit ideal solution behavior, they could also exhibit nonideal solution behaviour, what
happens, how to account for those things and they have been discussing and in the end
discuss very peculiar counterintuitive phenomena the retrograde condensation which happens
for binary and multi-component liquid solutions.

(Refer Slide Time: 1:03)

Well today we will be shifting to slightly different system, systems which are not miscible
over the entire range of compositions. We will be dealing first with partially miscible systems
where the solution or the components they are miscible over certain ranges of composition
and they are immiscible over other range and then we will try to find out why liquids are
miscible under certain conditions, they are not miscible under certain conditions, the phase
diagrams for partially miscible systems and also the phase diagrams for completely miscible
systems.

912
(Refer Slide Time: 1:42)

Now let us discuss a few salient features of it. First thing is why are components rather what
happens due to which the components they do not mix with one another. Now we have
already discussed while we were discussing the thermodynamic potentials we have already
discussed that under constant temperature pressure conditions the, any particular system it
will act such that Gibbs free energy minimizes in the course of the process.

(Refer Slide Time: 2:29)

Or in other words mathematically if we put it we have seen that the condition of

thermodynamic equilibrium under conditions of constant temperature and pressure is given
by this particular equation, okay. And therefore it is quite evident that when there are 2

913
components which mix definitely they will be miscible if after mixing the Gibbs free energy
of the solution is less as compared to the Gibbs free energy of the original components.

(Refer Slide Time: 2:52)

For example this particular situation is given by curve 2 in this figure. In this particular curve
we find that a single liquid phase will be obtained over the entire range of composition
starting from X1 equals to 0 to X1 equals to 1.

(Refer Slide Time: 3:40)

Now for such a binary system we find that Gibbs free energy it is a function of as we know
it’s a function of temperature pressure and composition we have discussed the different

914
relationships etc and for the system to be stable as I have already discussed (dG) T P has to be
equal to
0. Now if we consider the dependence of g by RT over composition.

(Refer Slide Time: 3:43)

Just the curve which I have shown in this particular figure and here we find that as I have
already discussed for a completely miscible system the Gibbs free energy of mixing always
reduces it is always less as compared to the Gibbs free energy of the individual components.
Now under certain conditions if it happens that on mixing instead of following curve 2 the
solution or rather the mixture it follows curve 1.

That means in curve 1 we find that over this particular portion starting from X1 equals to 0 to
say X1 equals to X1 prime we find that there is a decrease in the Gibbs free energy change of
mixing, okay. So over this particular range we know that the solution or rather the 2
components are miscible with one another but what about this particular portion which is
accompanied by a hump in the curve.

In this portion we find that system it can attain the lower value of Gibbs free energy if instead
of mixing with one another they remain as 2 separate components or 2 separate liquid
solutions whose compositions will be given by X1 prime and X1 double prime. So naturally
we find that the minimum value of G is not given by the solid curve but by the dotted line
which joins the 2 points say point C and say point D.

So therefore the system it can attain the minimum Gibbs free energy if instead of a solution it
exists as a mixture of 2 phases where the composition of the 2 phases will be given by X1

915
prime and X1 double prime respectively. So therefore any mixture which has an overall
composition in this particular range it will not remain as a solution but it will break up into 2
liquid phases with the composition as I have mentioned. And therefore this is dictated by the
condition of stability or the condition that the Gibbs free energy of a system has to decrease if
a spontaneous reaction has to take place.

Now let us see what will be the vapor liquid equilibrium diagram for such a system we have
already discussed the VLE diagram the Txy and Pxy diagrams for completely miscible
systems and also the xy diagrams. Let us see what will be the vapor liquid equilibrium
diagram under this condition? What is a system that we have here?

We have one vapor phase, one liquid phase and another liquid phase. Both the liquid phases
they have some, suppose it is a binary mixture that means it comprises of 2 components then
what do we have? In the vapor we will have both the components, in one liquid phase then
will be a predominance of 1 component and a lower proportion of the other component and
vice versa for the second liquid phase.

(Refer Slide Time: 7:07)

Or in other words we have a vapor phase, we have a liquid phase denoted by L1 which
primarily comprises of component 1 where component 2 is dissolved in it and we have
another liquid phase 2 which primarily comprises of component 2 and small amount of
component 1 dissolved in it, right? So these are the 3 phases that we have unlike the 2 phase
systems that we have been discussing so long.

916
So therefore if there are 3 phases, what is F in this particular case? When we know that for a
binary system we know C equals to 2, we know P equals to 3, so therefore for all the 3 phases
to coexist F has to be equal to 1. So therefore from here we can conclude that the 3 phases
can coexist under one particular composition, temperature and pressure conditions.

(Refer Slide Time: 8:42)

Under all other temperature, pressure, composition conditions either 2 of the 3 phases are
going to be, there all the 3 phases cannot coexist. So let us see what is going to be the phase
diagrams for such particular systems, we find for such systems as I have already mentioned
that since F equals to 1 for the 3 phases to coexist, so we find that the 3 phases can coexist
only under this particular condition which is denoted by C and this is a typical Txy diagram
that we have drawn since it is at the Txy diagram this is drawn under constant pressure
conditions and the point where the 3 phases can coexist that is given by the temperature T star
at the constant pressure P.

Definitely T star will be different when we deal with different pressure conditions. Now what
do we find here? We find here that the vapor of particular composition is in equilibrium with
the liquid L1 whose composition will be given by xD and a liquid L2 whose composition will
be given by xB above T star what do we find? Above T star we can either have only a vapor
phase here.

We can have only L2 phase here and we can have L1 phase here or else we can also have a
mixture of L1 plus the vapor phase L1 means predominantly component 1 with a small
proportion of component 2 dissolved in it or they can have L2 and the vapor phase in

917
equilibrium where L2 is predominantly component 2 with a small proportion of component 1
dissolved in it.

So this is a situation about T star below T star what can we have? Below T star either we can
have one particular liquid solution which is L2 which is predominantly component 2 with
small amount of component 1 dissolved or we can have a single liquid phase L1 which is
component 1 with small amount of component 2 dissolved in it or we can have 2 separate
liquid phases L1 and L2 where L1 comprises predominantly of component 1, L2 comprises
predominantly of component 2 and depending upon the overall composition the compositions
of the 2 liquid phases can be obtained from the horizontal Tie line or the intersection of the
horizontal Tie line with the demarcation boundaries of L2 with L1 plus L 2 and L1 with L1
plus L 2, right?

So therefore if the composition is given by say Point M for the composition xM in this
particular case we find that it separates out into L1 and L2 where L1 composition is given by
XS and L2 composition is given by XR. Suppose instead of that we find that more or less the
overall composition is given by say XP for example. So in this particular case what happens?
we find that the system it exists as a liquid phase comprising of 2 liquid solutions where one
solution L1 is given by a composition say this particular composition say X1 prime and L2
given by a which is denote or rather which has a composition of X2 prime.

Now for this particular same composition suppose we go for higher temperature, temperature
higher then T star say at this particular point, here what do we have? We have the same
overall composition but we have varied the temperature. We have only a vapor phase, right?
From this vapor phase if you start reducing the temperature the overall composition remains
the same.

And we find that the vapor it keeps on cooling till it reaches point 2. At point 2 the first drop
of liquid appears or the first dew appears this is the dew point of the vapor mixture at a
temperature T and an overall composition of XP. The dew which is formed has a composition
corresponding to X4 or rather under this condition the dew which is formed it comprises of
L1 which has a composition of X4.

Now with the same overall composition if we continue this cooling process, what are we
going to have? We find that it follows the path 1 2 and 3 say in due course during the cooling
process the mixture it reaches point 3 where it is separated into a vapor phase of composition

918
y5 and a liquid phase of composition X6 we can keep on continuing this particular cooling
process we find that on further cooling the composition of the vapor phase is given by say
y10 and the liquid and the liquid composition say given by X10.

So therefore we find that as we continue this cooling process the vapor composition travels
along GC while the liquid composition while there is just a 1 particular liquid which is in
equilibrium with this particular vapor that liquid comprises of predominantly component 1
with small proportion of component 2 and its composition keeps on varying along GD as the
cooling process continues and this continues till we reach the horizontal line given by CD at
this point we find that the vapor phase completely disappears and it separates out into 2 liquid
phases whose compositions are given by XB and XD.

A component liquid L1 with composition XD and a liquid L2 with composition XB, now if
you still continue our cooling process we find that we will be again we will left with 2 liquids
only and the compositions of the liquid will be varying along the curves ED and AB. Now
this was when we started with a composition of vapor whose overall composition can be
given by y1. Instead of that, if we start rather with a much lower vapor composition say we
start at y7 and we continue the same particular cooling process at constant pressure, pressure
is kept constant we continue the cooling process, what happens? The vapor keeps on getting
cooled and it follows the paths given by 7, 8, 9, vertical line in this particular phase diagram,
right?

Now in the cooling process when the vapor reaches points 8, what do we see? We find that
the first dew appears this dew comprises of L2, okay. And the composition of L2 is given by
X10 we continue the cooling process while we continue the cooling process with the overall
composition remaining same we reach say point 9. Now under this condition the vapor
composition is given by y12 while the liquid composition it is given by L11.

If we keep on the continuing this cooling process further, we find that the vapor composition
travels along FC while it is in equilibrium with a single liquid comprising predominantly of
component 2 and its composition varies along FB and finally when we touch this horizontal
line BC we find that there is no amount of vapor left and the system which we are left with
comprises of 2 liquid solutions the composition of one of them is given by XB and the
composition of other is given by XD.

919
So therefore we will find that the which particular phase is going to exist or rather which
phases are going to exist under constant pressure conditions it depends definitely on the
temperature but also on the overall composition of the mixture that we have at hand. For
example suppose we start at some particular point say F this is say point 1, right and we start
cooling it, what do we find?

We find that as we continue the cooling process it follows the path say 1 2 and then it enters
the 2 phase region and there we have L2 plus V. We continue the cooling region and we enter
a single liquid phase which comprises predominantly of phase 2. So in this particular case the
Txy behaviour is more or less similar to the behaviour of Txy that we have studied while
dealing with nonideal and ideal solutions.

Same thing is going to happen if we if we deal with a very high overall composition of
component 1. Say we have started with say small a, we start cooling it, we find that initially
there is a vapor which is in equilibrium with the liquid phase. Now this continues and we
have a liquid and vapor region where we find that the composition as we have already
discussed composition on the vapor phase varies along GC composition of the liquid phase
varies along GD.

Now this continues till we reach point 6, at point 6 the last vapor disappears off and the
composition of the last vapor was given by y5. If we further continue the cooling process we
find that we have not entered the 2 phase liquid region rather we have entered the single
phase liquid region or we have entered the region of a single liquid solution. So therefore
with this we find that it depends whether we are working at a range lower than T star then we
will be having only liquid where we may have a single liquid phase in the composition lies in
this range it may also have a single liquid phase of a different liquid if the composition lies in
this range.

Or we may have a mixture of 2 liquids if the composition is given by curves AB and the
composition lies between the zones of ABD and E. If you’re working at a temperature higher
than T star then we may either have a single vapor phase. We may have a single L2 liquid we
may have a single L1 liquid or we may have a mixture of L2 and vapor phase.

We may have a mixture of L1 plus vapor phase but only at this particular point we can have
the 3 phases coexisting together. So therefore suppose we start with our initial composition at
say 13 and we start cooling this particular mixture, what do we find? We find that there is a

920
vapor phase, right? This vapor phase it keeps on getting cooled but it remains in the vapor
phase and it continues the cooling process this keeps on happening till we come to the point
C.

At C suddenly the entire vapor condenses and forms 2 liquid phases here. So therefore we
find that the vapor liquid equilibrium for partially miscible systems comprising of 2 liquid
phases and one vapor phase it has different zones of coexistence of the liquid and vapor there
is one particular zone of coexistence of vapor and liquid L2, one particular zone of
coexistence of vapor and liquid L1, one particular zone of coexistence for L1 plus L2 and just
one point of coexistence of the 3 phases.

(Refer Slide Time: 22:19)

Now suppose we have drawn it at one particular pressure, suppose we would like to repeat
the experiments and see what happens if we are if we are drawing this Txy diagram at
different pressures, what do we find if we go for higher and higher pressure? There are 2
things that we observe first thing is the T star which we have seen the T star it keeps on
increasing as we increase the pressure or the 3 phase equilibrium temperature as it is called
that increases as increase the pressure.

The other thing that we find we had observed if you notice the previous slide I will just show
in the previous slide if you just notice the previous slide where we had drawn you will be
finding that in this particular case AB and ED they are they slightly inclined towards one
another which shows that the solubility of the phases increases with temperature.

921
As we increase the temperature greater amount of component 1 gets dissolved in L2 and a
greater amount of component 2 gets dissolved in L1 as a result we find that when we are
going for higher and higher pressure we find that these 2 curves AB and DE they start
inclining towards one another more and more. As a result the horizontal line which marks the
zone of existence of the 2 liquid phases that keeps on getting smaller and smaller till at one
particular unique temperature it reduces to a point which is known as the critical solution
temperature.

What happens beyond the critical solution temperature? We find that more or less the
horizontal portion has disappeared and the Txy diagram resembles the diagrams or rather the
phase diagram that we had developed for completely miscible substances with a minimum
boiling point or it resembles the Txy diagram of a minimum boiling Azeotrope.

Now in this particular case since the solubility increases with temperature we find that there
is a upper limit to the range of existence of the horizontal portion or the upper limit to the
range of existence of the 2 immiscible liquid phases. Accordingly this is known as the upper
critical solution temperature there are cases where we find that the solubility actually it
decreases with pressure under that condition we have a lower critical solution temperature but
let me tell you that is much less common, more common phenomena is solutions which
exhibit upper critical solution temperature but we have lower critical solution temperature or
rather solutions exhibiting lower critical solution temperature where there is hydrogen
bonding for such cases there are also certain systems which I have written down where we
find that they exhibit both the upper as well as a lower critical solution temperature. So this
diagram it shows the phase stability in four binary liquid mixtures which exhibit increase in
solubility with temperature decrease in solubility with temperature and also exhibit an upper
and a lower critical solution temperature.

Well, there is one more thing that is quite interesting and I would like to mention it, the first
thing is, so therefore what do we find? We find that the condition of instability of the binary
liquid mixture it depends on 2 things firstly it depends upon the non-ideality of the solution
and it also depends upon the temperature.

Now for the simplest case we find when gE is given by the one parameter equation or in other
words when gE is given by the 2 suffix Margules equation for such cases we find that the
consolute temperature this is given by the equation A by 2R the derivation is pretty simple
but I will not be going into the details of this. Since there is some constraint of time
also, so

922
therefore we will find that the consulate temperature is given by A by 2R where the excess
Gibbs free energy is given by the 2 suffix Margules equation.

And from this equation it appears that Tc should always be a maximum but in reality we
know that it can be a maximum or minimum and although upper consolute solution
temperatures are more common we do have lower consolute solution temperatures in
mixtures where components form hydrogen bonds.

(Refer Slide Time: 28:12)

And we also find that if gE is given by this equation then usually the composition
corresponding to Tc is given by X1 equals to X2 equals to half but usually we find that gE is
not given by such a symmetric distribution or gE is usually not so very symmetric with X1
and X2 and therefore the coordinates of the consulate point is often not at the composite
midpoint. Well there is one more interesting thing that I would like to point out and we will
be discussing it in the next class.

923
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 51
Partial And Complete Immiscibility (Contd.)
Well, to continue with our discussions on partially miscible systems we had already
discussed the phase diagram for miscible systems we also tried find out why certain
systems are miscible under certain conditions and then immiscible under certain
conditions but I would just like to elaborate slightly on this particularly I would like to
discuss the effect of pressure on the zone of or the tendency of miscibility and
immiscibility.

Just like I had discussed in the previous class 2 components are going to mix with one
another provided this leads to a lowering of the Gibbs free energy of the system. If on mixing
the Gibbs free energy of the system increases definitely the components will prefer to stay
separated and only the Gibbs free energy decreases we are going to form a miscible
solution, right?

(Refer Slide Time: 1:55)

Now what decides whether the Gibbs free energy is going to increase or decrease
when the 2 components are brought close or rather when they the 2 components are
contacted with one another. If you recall what did we, we have already discussed
that delg say delg mixture Del P at constant T and x, What was this? This was
related to the volume of mixing if you remember, okay.

924
And this is the main equation which tells you that what is going to happen when the when the 2

components are brought close to one another. Now 2 components will be mixing as I have

925
already mentioned provided delta g mix is less than 0 and del2 g mix Del x2 at
constant temperature and pressure is greater than 0 for all x, only if this happens
then only the 2 liquids are going to be miscible in all proportions.

(Refer Slide Time: 2:44)

Now we know that Delta g mix it’s a function of pressure and therefore when we find that
while increasing pressure it may lead to miscibility of the 2 components just like I had
shown in this particular case it may also lead to immiscibility of the 2 components. So
very naturally from this equation it is evident that that should depend upon the sign of V
mix. Suppose we bring 2 components and mix them together and we find that V mix is
actually greater than 0 then we know that when the 2 components are mixed and the
variation of the g mixture with pressure is on to be positive, so at higher pressure we are
going to get a positive value or a higher value g mix.

926
(Refer Slide Time: 4:14)

So naturally when we are going at higher pressure a situation is going to arise when
initially at a lower pressure it had formed initially at a lower pressure it was completely
mixed and when we increase the pressure when delta V is delta V mix is greater than 0
this is a mixture Delta V and this is also g mixture. So therefore when Delta V is greater
than 0 when we go for higher pressure we find that Delta g mix is actually increased and
there is a hump as a result of which partial miscibility results and in this particular
composition range given by X1 prime and X1 double prime the 2 components separate
out into 2 liquid phases while they are miscible beyond X1 prime and beyond X1 double
prime.

Similarly suppose we have one particular system which is immiscible at low at low pressures
the curve which or rather the system which is given by the curve a denoting that Delta g
mixture as a function of the composition, from this particle curve we find that there is partial
miscibility between X1 prime and X1 double prime under the condition denoted by curve a
and we know that for this case Delta V mixture is negative, is less than 0.

Now for such a case suppose we increase the pressure, what happens? Delta G
mixture reduces and as a result we find that the solution or the system it was partially
miscible at lower pressure but it becomes completely miscible at higher pressures.
So therefore from such simple variations within we can very well understand that the
qualitative affect of pressure on phase stability of binary liquid mixture it depends
upon the magnitude as well as the sign of the volume change of mixing and
therefore if you really want to perform quantitative calculations at some fixed
temperature we need to have information of the variation of Delta v mixture with XP.

927
(Refer Slide Time: 5:58)

We need to have this particular information along with Delta g mixture as a function of x
at a single pressure only if we have this data we can find out that under what conditions
the 2 components will be mixing under what conditions the 2 components will not be
mixing? They will be forming immiscible systems and if by increase of pressure we
would land up into a miscible system or an immiscible system it all depends upon the
sign and magnitude of the volume change of mixing of the 2 liquids.

928
(Refer Slide Time: 6:30)

Now let us go to the extreme of this particular situation we have been dealing with
partially miscible systems we have been dealing with completely miscible systems. Now
let us go to completely miscible systems, again under this condition I would like to
remind you that in reality it’s very difficult to have a completely immiscible system. Such
systems…

(Refer Slide Time: 6:50)

929
(Refer Slide Time: 6:53)

Well, before that I would just like to mention that this is the corresponding Pxy
diagram corresponding to the Txy diagram here. It’s interesting to note that while in
the Txy diagram we found that the solubilities they were increasing with pressure. In
this particular case the lines are more or less vertical which indicates that pressure
insensitivity of liquid solubility which we already know.

(Refer Slide Time: 7:15)

Well, now to continue with our discussions with completely immiscible systems. Again as I was

repeating at that time usually completely immiscible systems are not there in practice but they

can be approached by several systems. So what should be the typical Txy and Pxy diagram in

930
this condition? Again when there are completely immiscible systems we know that
for 3 phases to coexist F equals to 1.

So therefore if we are drawing Pxy at a constant temperature or Txy at a constant pressure

then there will be just one point where the 3 phases can coexist just like we have discussed
in partially miscible systems, what is this point? If you observe these 2 curves this point is
given by the point B, only here we have the 3 phases existing this is denoted as I have said
by a 3 phase equilibrium temperature T star and a 3 phase equilibrium pressure P star,
right?

Now other than this what happens to the other regions? Suppose we are operating at a
temperature lower than the 3 phase equilibrium temperature, we do not have any vapour
under this condition. We have to completely immiscible liquid systems if we’re operating
at temperature above T star we can have either a vapour phase or we can have the
vapour in equilibrium with one of the liquids pure liquid, mind it. It’s not a liquid solution.

In this case L1 means pure liquid L1 unlike the previous case where for partially
miscible systems I had mentioned that L1 is a solution of 2 in 1. In this case L1 refers
to pure liquid 1 or in other words depending upon the overall composition we can
have a vapour phase in equilibrium with pure liquid L2. Now in this case also we find
that suppose we start from state1 and we start cooling the vapour at constant
temperature, what happens? Cooling continues.

And since it is a constant pressure cooling, so naturally it follows the vertical line say 1, 2, 3
as cooling proceeds the vapour comes or reaches the point 2. At 2 the first drop of liquid
appears, what is this liquid? It is pure liquid L1 mind it, right? We continue the cooling
process, we get greater and greater proportion of liquid L1 and the vapour composition it is
given in this case by y5 we keep on continuing the cooling process further we find that the
vapour composition varies along path CB and we get more and more of clear liquid L1.

And finally when we have touched this particular horizontal line the entire vapour has
condensed to give us liquid L1 and the remaining portion is L2 which forms a
separate layer, it does not mix with L1. On the contrary if we have a vapour of overall
composition say y7 and we start cooling it we find that in this case also it follows a
vertical line naturally constant pressure cooling and when it reaches point 8 the first
dew of liquid which appears is pure liquid L2.

931
And if we continue the cooling process we get greater and greater proportion of liquid L2
and the vapour composition it keeps on decreasing along the path AB and finally when
we continue the cooling process and rich this particular horizontal line say this horizontal
line is JK, so when reaches this particular horizontal line we find that the entire vapour
has disappeared and we get pure liquid L2 and definitely whatever component 1 was
there in the vapour that condenses and separates out into a different liquid L1 and
suppose we start rather we have the vapour of composition say yB and we continue the
cooling process we find that we have only a single phase vapour region as long as the
cooling continues then suddenly at the 3 phase equilibrium temperature entire vapour
disappears and 2 liquids are formed, the proportion of the 2 liquids are same as the
vapour or the same as the proportion of the 2 components in the vapour phase, right?

So therefore this is the phase diagram and since this is comparatively simple we can
develop the mathematical equations which will be required in order to construct this
particular phase diagrams, what are the things that we need? We need to first locate the
point B and then we need to find out the locus of curves BC, the locus of curves AB and the
horizontal line.

(Refer Slide Time: 12:58)

We assume for a time being that we are dealing with low pressure systems, for low
pressure systems it’s the standard VLE relation which we have written n number of times
in our entire course is given by gamma i Xi the standard relation this is equals to yiP.
Now if we apply this relation to the point B say, at point B, what is there? We find at point
B vapour is in equilibrium with liquid L1 and vapour is in equilibrium with liquid L2.

932
So there will be 2 equilibrium relations, for the first case what will we have? For the first case
it should be gamma1 X1 P1 saturated will be equal to y1P and again we remember that in
this case X1 is equal to1 we have pure liquid. Since X1 equals to 1, naturally gamma1 will
also be equal to one. So at point B, what is y1 equals to? It is P1 saturated by P.

Similarly at point B the equilibrium relationship governing the equilibrium of vapour and
liquid L2, this is again going to be the same relationship with the same considerations
which gives y2 is going to be P2 saturated by P and therefore once we know that the y1
star and y2 star at the 3 phase equilibrium temperature, we should be able to find out the
P star under this condition which is nothing but equal to P1 saturated plus P2 saturated.

(Refer Slide Time: 14:41)

And we observe one more interesting thing we find that at any particular specified
pressure say for example this phase diagram is drawn at a constant pressure for
such a case if 2 immiscible liquid phase is are in equilibrium with the vapour phase
we find that the 3 phase equilibrium temperature will always be less than the
saturation temperature of either liquids at that pressure, is it clear?

We will always find that TB will be lower than the equilibrium saturation temperature
of liquid L2 and the equilibrium saturation temperature of liquid L1 this is always
going to happen.

933
(Refer Slide Time: 15:23)

So therefore under this condition we have found out what is the composition of y1 star
we have found out the composition of y2 star and we have also found out what should
be P star under this condition. And this P1 Sat and P2 saturated they both of them they
are functions of T star. So therefore we are in a position to locate point B.

Now suppose I would I would like to concentrate on this particular region. In this
particular region L1 plus V what is the , we know that it’s pure liquid L1 for this region we
just need to know the locus of this particular curve and we need to know the locus of this
curve. Along this curve we do y1 is in equilibrium with pure liquid L1. So therefore for this
curve can we not write down that again the again from the same equation these 2 cancel
out, can we not write down that along curve CB, y1p is equals to P1 saturated.

Or in other words for this particular curve y1 will be given by P1 saturated by P and
y2 will be given by 1 minus P1 saturated by P or in other words P minus P1
saturated by P, agreed? And what about curve AB? For this case again P2 saturated
should be equal to y2P from where we know y2 should be equal to P2 saturated by
P and y1 should be P minus P2 saturated by P. So therefore once we can locate the
3 phase equilibrium point, the P star, we can locate the T star because under that
particular T star the P1 saturated and the P2 saturated that we get and the y1 star
and the y2 star that we get, these 2 y1 y’s, Sigma yi star should be equal to one.

So therefore we can locate P star from the saturation pressures, we can locate or rather we can

estimate the mole fractions of the 2 phases and once we find that the mole fractions sum up to

934
1, we know that our assumed temperature also corresponds to the 3 phase
equilibrium temperature. So therefore from this we can find out or locate the point B.

Once we can locate the point B we can very easily find out the slopes of CB and AB,
the slope of CB or the composition variation along CB is given by this particular
equation and the composition variation along AB is given by this particular equation.

(Refer Slide Time: 18:21)

So with these we can, it’s quite easy to construct the Txy and the Pxy diagram for completely
miscible systems. Well, this completes the vapour liquid equilibrium discussions of different
systems we have considered completely miscible, partially miscible and completely
immiscible systems. We have among completely miscible systems we have considered ideal
solutions and non-ideal solutions and we have also tried to find out how from pure common
properties we are in a, we shall be in a position to find out the properties of the solutions and
for both the liquid phase as well as the vapour phase.

In the next class we I will be giving you a short discussion on equilibrium studies or
equilibrium considerations for other mass transfer processes. Since distillation is one of the
most important mass transfer processes, so therefore thermodynamics primarily deals with
the vapour liquid equilibrium rather of the vapour liquid equilibrium of different systems but
definitely apart from distillation we also have to deal with other processes like absorption,
adsorption etc.

As you read this processes you will be exposed into greater details regarding the equilibrium

equations or the equilibrium considerations for this process but here I would just like to

present a very brief discussion of the equilibrium considerations for other processes. So we

935
do that in the next class and that is going to end up our course on phase equilibrium
thermodynamics.

936
 Course on Phase Equilibrium Thermodynamics
By Professor Gargi Das
Department of Chemical Engineering
Indian Institute of Technology Kharagpur
Lecture No 52
Phase Equilibrium for Mass Transfer

Processes (Refer Slide Time: 0:39)

Well, so this is the last class we are going to have in the course of phase equilibrium
thermodynamics till now we have had a large amount of discussions on vapour liquid
equilibrium solution for different systems. Now for all these systems we found that vapour
liquid equilibrium phase diagrams and the equations for both ideal and non ideal solutions
they cover a wide range and this is primarily important for the design of distillation columns.

Now apart from distillation there are also a also a large number of other mass transfer operations

and I felt that a small or a brief discussion regarding the phase equilibrium relationships which

govern the other mass transfer operations, if they are not at least touched upon, the topic of

phase equilibrium thermodynamics remains unfinished or incomplete. So therefore in the last

class I have decided to touch upon the phase equilibrium relationships which are important for

other mass transfer operations the phase equilibrium relationships other than vapour liquid

equilibrium data which is important for distillation columns.

937
(Refer Slide Time: 1:46)

Now let us see what are the other mass transfer operations that usually a chemical
engineer deals with? Definitely distillation we know that the principle underlying
distillation or the mechanism of separation which is the used in the distillation
processes is that it exploits the difference in volatility of the 2 components to be
separated. Apart from distillation we have absorption and stripping, what happens
under this phase? There is a gas phase we bring it in contact with the liquid and the
liquid absorbs the gas from a mixture of gases or from a pure component gas.

The most common example is absorption of carbon dioxide or sulphur dioxide using
sodium hydro caustic soda or caustic potash solutions. Now in this case also we cannot
have the complete absorption or maybe suppose we have some amount of sulphur
dioxide or carbon dioxide, there is a limit to which this gas can be absorbed in the
sodium hydroxide or any other solvent that you use and the maximum amount of
absorption that can take place that depends upon the equilibrium concentration of
sulphur dioxide in the caustic soda solution or any other solvent that you’re going to use
and this is governed by the solubility of that particular gas in the solvent that we are
going to use.

Naturally since sulphur dioxide is more soluble in caustic soda solution as compared
to water. So definitely we will have a greater separation or a greater purification of
the gas phase if we use caustic soda solution in place of water. So this is governed
or the process of absorption is governed by the difference in or rather by the
solubility of the gaseous constituent in the contacting liquid phases.

938
Now in stripping, mass transfer occurs in the reverse direction, in this case mass
transfer occurs from the liquid to the gas phase or the vapour phase and in both
absorption and stripping, since it is a mass transfer between a gas phase and a
liquid phase both of them they depend upon solubility of the gaseous constituent in
the contacting liquid phase, accordingly the equilibrium relationship in this case is the
solubility of the particular gas in the liquid under constant temperature and pressure
conditions.

And naturally if you’re dealing with solubility of the gas it’s quite natural then we will be more
concerned about the mathematical equations which will be governing this is the ideal dilute
solution law, the Henry’s law in contrast to Raoult's law which used to govern vapour liquid
equilibrium for distillation cases. There is another mass transfer operation which we have
which is known as extraction, in this case what we have? We use a second liquid phase to
extract any particular component from a from one particular liquid phase for example
suppose we have acetic acid in toluene we would like to purify to toluene or would like to
remove acetic acid from toluene we use water we bring the 2 phases in contact and we try to
see rather we try to purify toluene from acetic acid to the maximum extent possible.

Now what is this maximum extent? Definitely it depends upon the relative solubility of
acetic acid in toluene and in water. So therefore in extraction we find that the mass
transfer between the 2 immiscible liquid phases it depends upon the preferential
solubility of the desired component which is to be separated in the 2 liquid phases.
Then we have another mass transfer operation which concerns solid liquid or solid
gas mass transfer that’s known as adsorption and the reverse is leaching.

939
(Refer Slide Time: 6:16)

In adsorption a solid phase it absorbs either a gas or a liquid phase and in this case
naturally the mass transfer is governed by fluid solid equilibria and in leaching the
mass transfer occurs in the reverse direction. Now for all these mass transfer
operations what do we need to do? First we need to create 2 phases and then we
need to have mass transfer between the 2 phases, how do we create the 2 phases?

For distillation we find that creation of the 2 phases is just by heat we have the
component in the liquid phase we apply heat we have a vapour phase naturally the
vapour phase is richer in the more volatile component the liquid phase is richer in the
less volatile component and we keep on contacting them and separating them and
again contacting them, in this process the liquid progressively becomes richer in the
less volatile component, the vapour becomes richer in the more volatile component
and in this way we can get a substantial good amount of separation.

In absorption, leaching in extraction we find that we need, in fact apart from

distillation for all other mass transfer operations we find that we need a third
component in order to create the same phase or rather in order to create the second
phase unlike distillation where the contacting vapour and liquid phases are derived
from the same source, for all other mass transfer operations a third phase is required
in order to affect separation. For example in absorption, leaching and extraction this
third phase is the liquid phase created by introduction of a liquid stream.

940
(Refer Slide Time: 8:30)

Liquid stream is often termed as the absorbent in absorption or it is termed as the

solvent in case of extraction, leaching and also in absorption we often call it as a solvent.
In stripping the third phase is an external vapour stream it is often referred to as an inert
stream and in adsorption it is the adsorbent contacting it is actually the solid phase
which creates the adsorption this is this is wrongly written this should be the solid phase
which creates the medium where the liquid or the gas is preferentially adsorbed.

(Refer Slide Time: 8:47)

Now in all these particular in all this all this mass transfer operations we find that for

separation of a binary mixture for distillation the minimum number of components is 2 and for

all other operations it is 3 and for all operations number of contacting phases is 2, what

941
are these 2 phases? For distillation or rectification it is the vapour and liquid phase of
the same component, in absorption and stripping it is a gas phase and a liquid
phase. In extraction it is the 2 immiscible liquid phases, in leaching it is a solid and a
liquid phase. In adsorption it is either a solid liquid or a solid gas phases.

(Refer Slide Time: 9:40)

Now for all these phases again the way I had told you the first thing which we need to do
is to create the 2 phase system. I have already discussed how we create the 2 phase
system, in distillation it is created by heat for all other cases it is created by a addition of
a third component then we have to effect mass transfer between the phases once the
mass transfer separation occurs when we try to separate the 2 phases.

(Refer Slide Time: 10:30)

942
Now this mass transfer is governed by phase equilibrium, we cannot have as much
mass transfer as we need and the mass transfer under constant temperature pressure
conditions it is limited by the phase equilibrium considerations and this phase equilibrium
is denoted in different ways for different mass transfer operations and just like we have
we have already discussed we can have graphical representation of phase diagrams
and we can also have mathematical equations for the phase diagrams.

Remember for all phase equilibrium problems the governing equation is the same,
what is the governing equation? For each component i, the chemical potential in
phase alpha and in phase beta have to be the same this alpha and beta they can be
maybe one is the gas phase the other is the liquid phase, both of them can be 2
different liquid phases, one can be a solid phase other can be a liquid or a gas phase
whatever be the case the condition of thermodynamic equilibrium remains the same
it is the equality of chemical potentials of each component in the 2 phases.

And therefore how do we be develop mathematical equations? For developing

mathematical equations we try to express the chemical potential of each component
i in the 2 phases in terms of the compositions of component i in the corresponding
phases and those particular equation is when they are equated we get mass or
rather we get phase equilibrium relationships.

(Refer Slide Time: 11:50)

943
(Refer Slide Time: 12:05)

(Refer Slide Time: 13:01)

Now tell me for absorption what is the standard equation? For absorption as I have
mentioned the ideal dilute solution or in other words Henry’s law is the governing
equation in case of absorption. For adsorption if we see for adsorption the basic
equation the simplest equation it is of the Henry’s law type where the equation is given
as q equals to A into C star. Now the first question is what q is and what is C star?

So well, q it is the naturally in the, for adsorption it is the solid phase as well as a liquid or gas

phase. So q it is a amount of solute adsorbed per unit mass of the adsorbent that is the amount

of sol solute which is adsorbed it can be a liquid phase or it can be a gas phase which is

adsorbed per unit mass of the adsorbent and C star it is the equilibrium concentration of the

944
solute phase the C star it is the equilibrium concentration of the solute in the mixture,
that means in the gaseous mixture if it is a gaseous mixture then a conventional way
is to express the concentration in terms of partial pressure of the gas.

(Refer Slide Time: 13:30)

So therefore usually if it is a gaseous mixture then we prefer to replace C star with Pi

star and if it is a liquid mixture then we express it in terms of concentration where the
concentration can be either in terms of weight per unit volume or weight per unit
weight or maybe moles per unit volume etc.

Now in this case we find for absorption while the Henry’s law was the most common
equation or it is the ideal equation which is used we have already discussed Henry’s
law and in this particular case the I will just write down the equation for convenience
one, it is the Pi star equals to Hixi which you already know.

Well, there is one thing which I should be mentioning before I proceed further that is the
different unit in which the concentrations are mentioned for the different mass transfer
operations. Let us see the different concentrations before I go to the other mathematical
equations. We find that while we were discussing vapour liquid equilibrium the mole
fraction was the typical concentration ratio typical concentration unit that we were using.

We were using the mole fraction in the vapour phase and the mole fraction in the liquid

phase, the mole fraction in the vapour phase was denoted by y the mole fraction in the liquid

phase was denoted by x and usually we had Txy diagram, Pxy diagrams and xy diagrams.

Often for

945
absorption we use the mole ratio but mole fractions but more common is the usage
of mole ratios.

(Refer Slide Time: 15:18)

(Refer Slide Time: 15:40)

What is the mole ratio? As I have already mentioned in this in this table it is the
moles of species i divided by the moles of species free i, isn’t it? So therefore if I
have a total solution where x is the mole fraction, naturally 1 minus x is the mole
fraction of solute free system and x is the mole fraction of solute.

So in this particular case the mole ratio which is denoted by capital X or capital Y, naturally

capital Y refers to the vapour phase and capital X refers to the liquid phase the given as x by

1 minus x

946
and Y is given by y by 1 minus y the reason for often preferring your mole ratios to
mole fractions are that these often lead to linear plots Y versus X often leads to
linear plot as compared to small y versus small x and once more I will try to
emphasize that when we are dealing with gaseous components then very frequently
the concentration unit which we prefer is the partial pressure.

So therefore the partial pressure in the vapour phase or the gas phase is often
preferred for absorption, stripping and adsorption as well as leaching and when we
are dealing with liquid phases say in liquid-liquid extraction or in adsorption of a
liquid by a solid under that condition usually we prefer to use concentration or any
particular property related to concentration and here I would just like to mention that
usually the concentration and the solid phase is denoted as small q and the
concentration in the liquid phase is denoted by either a small c or a capital C.

(Refer Slide Time: 18:06)

So therefore the first thing which is which is important for you to remember are the different
mass transfer operations and for the different mass transfer operations are concerned with
and the second thing is for all these mass transfer operations the separation is guided by
phase equilibrium considerations and these phase equilibrium considerations are obtained
from phase diagrams or from mathematical relations relating the different equilibrium
properties and these particle phase diagrams they are denoted either in terms of mole
fractions like we have already seen in distillation or they’re denoted this is one in terms of
mole fraction or they are denoted in mole ratios the advantages immediately evident while
we have a curved xy relationship we have a linear mole ratio relationship.

947
They can also be denoted by partial pressure as a function of mole fraction this is
usually the case for absorption and stripping where the partial pressure of the gas
over the solution is plotted with the equilibrium concentration of the gas in the liquid
solution. Here I think this has been plotted as a function of temperature and as we
know gas solubility decreases with temperature as a result of which the isotherms
they become steeper as we go to higher temperatures.

And here this is the typical XY diagram which we have already dealt, this is distillation,
this is absorption and this is this is also absorption and this is a typical isotherm for
adsorption. Adsorption we plot or rather we construct the phase diagram with the fraction
or the amount adsorbed in by the solid phase and the equilibrium concentration of the
adsorbate in the remaining gas phase or in the remaining liquid phase.

If it is a liquid phase I repeat, we denote the coordinates with concentration and if it is a

gas phase we denote it with partial pressure and this is a typical phase equilibrium
diagram for liquid-liquid extraction, we will be discussing or rather I will be giving some
time on it because it is different as we see, why is it different let us understand? When
we are dealing with the other mass transfer operations, what do we see?

We usually consider that there is only one component which is being transferred
from phase a to say phase b, in distillation it’s either from liquid to vapour or vapour
to liquid. In absorption it is from vapour to liquid, stripping liquid to vapour.
Adsorption it's from liquid or gas to solid, leaching solid to liquid or gas but when we
are considering liquid-liquid extraction, what happens?

We find that there is one particular liquid for a component is dissolved we bring
another particular liquid and although we consider that the 2 liquids they are
completely immiscible in practice we hardly get any liquid where which are
completely immiscible with one another. So basically in practice what happens? The
solute which has to be extracted suppose we have A is one particular solvent where
the component C is dissolved we bring B to it such that it extracts C from A.

Now C therefore it dissolves in A, C it dissolves in B and even if A and B they have got very

less mutual solubility there will be some amount of solubility between A and B. So therefore

there has to be equilibrium between AC, BC and AB in order to show all these equilibrium in

948
a single curve usually we take the help of a Ternary equilibrium plot which I will be
explaining after I discuss the other parts.

(Refer Slide Time: 22:05)

And so therefore this is a typical solubility curve for gases in water here we find that in
this particular case the coordinates are partial pressure of solute in gas as a function of
the mole fraction of the solute in liquid and we find that the steepness of the curves that
we get, they denote the extent of solubility. For example we find that there is a the partial
pressure is very high at a low mole fraction for the case of sulphur dioxide which shows
that sulphur dioxide under this particular condition is relatively insoluble in water.

While on the other hand the partial pressure is much less at higher mole fraction for
HCl solution which shows that HCl is much more soluble in water. So therefore the
steepness of the curve gives you an idea of the solubility of that particular
component and from there we know that whether the absorption with this particular
solute is going to be feasible or not. If the solubility is very low then there is no point
in using this particular solvent for extracting or absorbing this particular gas.

949
(Refer Slide Time: 23:05)

(Refer Slide Time: 23:14)

Now let us go for adsorption, adsorption as I have said it can either be a q versus C
as have shown here this is for adsorption of a liquid. We can also have in terms of
partial pressure as we have shown here. Now from this particular curve it is quite
evident that adsorption isotherms they are called adsorption isotherms because
mostly they are plotted at constant pressure they are not always concave upwards
we can have a wide variety of adsorption isotherms as we can see here.

And something very unique to adsorption which I should be mentioning here is that for all other

mass transfer processes see we can extract the phase equilibrium data either while we are

increasing the concentration or while we are decreasing the concentration and we find that

950
usually for all the cases we get the same equilibrium concentration. For example in
case of absorption, say carbon dioxide being absorbed in water, okay.

We can either start with a highly saturated value of carbon dioxide in water we can keep on
changing the pressure so that the solubility keeps on decreasing and we can record the
solubility as a function of pressure or a function of temperature or else we can start with pure
water we can keep on dissolving carbon dioxide and increasing the temperature or the
pressure and finding out the equilibrium concentration, for both the cases usually we get the
same equilibrium concentrations but this is not the case for adsorption.

(Refer Slide Time: 24:51)

If we generate the adsorption isotherm during adsorption and during desorption usually we
find that the 2 curves they do not coincide with one another as a result of which it exhibits
the phenomena of hysteresis which is mostly exhibited by adsorption and not by other mass
transfer processes and during this exhibition of this phenomena of hysteresis which is
usually exhibited over a particular range of concentration we always find that the desorption
equilibrium pressure is always lower than the adsorption equilibrium pressure.

951
(Refer Slide Time: 25:33)

(Refer Slide Time: 25:56)

Well, as far as the mathematical equations are concerned I have already mentioned
for absorption it is the Henry’s law and for adsorption the simplest equation is
something like the Henry’s law it denotes a linear relationship between the amount
which is adsorbed per unit mass of adsorbent and equilibrium concentration in the
liquid phase or the or the equilibrium partial pressure in the vapour phase.

Apart from that we find that several other equations are available to predict the adsorption
isotherms under different conditions, the equations and the special features are mentioned
here it’s not important for you to remember all these equations it is just important for you to
remember that since absorption it so very erratic as I’ve shown you the isotherm as have

952
shown you the phenomena of hysteresis therefore several equations are needed in
order to predict adsorption under different conditions.

(Refer Slide Time: 27:20)

Well, the last which I would like to discuss with you is liquid-liquid extraction. In this
case as I have already mentioned that what happens when we are dealing with liquid
vapour cases? The gas phase is always miscible, liquid phase it may be miscible it
may not be miscible, so therefore usually a two-dimensional plot of the 2 equilibrium
concentrations under different conditions of temperature and pressure are sufficient
to generate an equilibrium plot.

But in liquid-liquid equilibrium as I have mentioned or taught while ago, we find that they
have mutual solubility’s among the different components, let us take a concrete example say
we have A where C is dissolved and we add up B in order to and here what we do? We add
up B in order to extract C from A, we know that C is soluble in A and C is soluble in B, right?
And also A is soluble to small extent in B and B is soluble to a small extent in A.

Now how to represent such a type of phase diagram? What we do? We usually
represent them on an equilateral triangle where the 3 apexes of the triangle refer to pure
components, so this apex it refers to pure A, this is pure C and this is pure B and all the
sides of the triangles they refer to binary solutions this denotes solution of C in A, this
denotes solution of C in B, this denotes the liquid-liquid mixture of A and B.

Along this particular site, the relative proportions of A and C is denoted by the particular point

where we select for example a point here it has got a small amount of A, a large amount

953
of C where as the point here is a small amount of C and a large amount of A. Now
we know C, C and A form a completely miscible system C and B form a completely
miscible system, A and B their solubility’s are very low in one another. So therefore
they form 2 immiscible systems which are denoted by point L and point K.

If the systems are highly immiscible point L and point K they shift more and more towards A
and towards B. So therefore this particular range it shows the extent of miscibility of A and B
and we find that between L and K there is a dome, what does it represent? This particular
side from L to P it denotes that A and B they are mutually insoluble and with the addition of
C the change in the mutual solubilities of A and B are denoted by this particular dome.

So within this region, this region it is heterogeneous region comprising of 2 phases,

outside this region it is a single phase liquid solution and solubility of C in A and the
solubility of C in B both of them they meet at a point P which is known as the plait
point and within this particular dome any particular point say point M it comprises of
2 liquid phases one liquid phase comprises of solution of C in A and the other liquid
phase comprises of a solution of C in B, the compositions of the 2 phases are
denoted by the Tie lines which pass through M and intersects this particular dome.

So therefore at M in reality we have a solution of A and C corresponding to

composition R and the solution of C in B corresponding to a composition E and we
find that all these Tie lines they are usually not parallel to one another they are
usually not horizontal they gradually change their slope but the sign of the slope
more or less remains the same there are few cases where we see that the sign of
the slope changes and we find that the slope gradually from negative slope it
becomes horizontal and then it becomes slightly on the other side such systems are
known as Solutropic system but the systems are not very common in one another.

So therefore we find that, in this particular case if we find the solution of C in B is

much richer as compared to the solution of C in A this is also denoted by this
particular phase diagrams which automatically suggests that B has got a higher or
rather C has got a higher affinity for B and therefore B can serve as a proper solvent
for extracting C from A and again another thing I would like to mention that any point
inside the triangle we can know the proportions of A, B and C here, how can we
know it? We know that sum of the perpendicular dropped from this point on the 3
sides they equal to the sum of the altitude, so what do we do?

954
We assume that the altitude is 100 percent, so therefore these 3 they sum up to 100
percent and the perpendicular distance dropped from M on AC it represents the mole
fraction or the fraction of B, the perpendicular which has been dropped from M on
BC represents the mole fraction of A and the perpendicular of M from AB represents
the mole fraction of C. So accordingly by this process we can find out for any
particular point inside the triangle the composition of the liquid mixture corresponding
to that point when it is inside the triangle it is a it comprises of all the 3 components,
when it is along the sides of the triangle it comprises of 2 components and when it
forms any particular apex of the triangle it refers to that particular pure component.

So therefore this completes our discussions on phase equilibrium thermodynamics, we have

not only dealt with vapour liquid equilibrium but also other equilibrium considerations which
are which are important for other mass transfer operations which we will be covering in the
course of your chemical engineering curriculum. From here we could understand how phase
equilibrium is important for all mass transfer operations and in what way phase equilibrium
relationships or phase equilibrium diagrams should be represented for different mass
transfer operations and it’s also important to remember that the units in which we express
the phase diagrams they are also different for different systems that we consider.

Usually mole fractions are used or preferred, mole ratios are preferred to obtain
linear phase equilibrium curve and partial pressure is a much more relevant unit
when gas phase is concerned just because it is easier to measure partial pressure or
the pressure as compared to concentration, for any liquid phase involved or for any
solid phase involved it is the concentration which is important. Usually concentration
is denoted by C in the liquid phase and by Q in the solid phase.

And it is also important to remember that in adsorption equilibrium isotherms they

are often marked by the phenomena of hysteresis and therefore adsorption
isotherms they’re comparatively difficult to obtain and they are mainly based on
equilibrium experimental data and several curve fitting of experimental data as have
been proposed by the different adsorption isotherms that we have discussed and
such ternary liquid such ternary equilibrium plots they are not only common in liquid-
liquid equilibrium data they can also be common for adsorption or other mass
transfer operations when more than one particular component gets transferred

955
or in other words when all the components involved with one another gets
transferred into one another but usually ternary plot is much more relevant for liquid-
liquid extraction processes. So with this I wish you good luck and I really hope you
have enjoyed journey of phase equilibrium thermodynamics. We have been solving
your doubts and queries and we are there for addressing further doubts that you
have before your exams. So good luck and good day to all of you

956
THIS BOOK IS NOT FOR SALE
NOR COMMERCIAL USE

PH : (044) 2257 5905/08 nptel.ac.in swayam.gov.in