fntroduction To

Melallurgical
Thermodynamics

R. H. Tupkary
Ex Prof. & Head Dept. of Met. Egg.
V. Regional College of Engineering
NAGPUR.

Itr Pupir.sners ir-agpur t+o ozz

Published by
Mrs. Mbha B. Saoji
SUMATI,41, Atre Lay-out, Nagpur - 4AO22
(Ph:pp. -0712-234f,73)

Legal Warning

This book as a whole or part thereof can not be reproduced in any forrn
whatso e\,er, or translated, except for review or criticism in scientific literature,
without the written permisssion of the author and the publisher

. FIRST EDITION - 1995

erice f $.15q/9
Printed at - Surya Offset '
Ranrdaspeth, 28 farm land
Nagpur - 440 010
 DEDICATED TO

All those scientists and technologists who are

striving hard to raise the energy efficiencies of various
englneering and technological processes in lndia to bring
these up to the best standard in the world.
 Preface

Thermodynamics istaught at under graduate level in the degree course

of metallurgical engineering all over lndia and yet no appropriate book is
available for the beginners to understand this most abstract and conceptual
subject. The available books are rather too elaborate and orer exhaustive and
an average student finds it difficult to separate out the required material for his
purpose. The non-availability of an appropriate book becomes an handicap in
understanding this abstract subject. lt may be one of the reasons why this
subject is dreaded mosi by the students and the net result is that the subject,
which ought to be understood properly for understanding other subjects at
higher lewls in metallurgical cunicula, is appreciated by the students only
mechanically for the purpose of solving numericals alone.
Having taught the subject at the beginners level for over a decade these
difiiculties of students became more and more apparent. The desire to write an
introductory text book on thermodynamics for metallurgical engineering stu-
dents became intense but availability of leisurely time was rather remote. This
became available only while canvalescing after an accident in which both tibia
and fibula were fractured.
I hope the students will like this new addition afterthe earliertwo - steel
making and iron making - equally well. Even if the students do not find if so
useful as the earlier two books, the pleasure of preparing the manuscript of the
book on thermodynamics always made me forget the pains of a fractured leg
during the period of convalqscence. And now when the book is being published,
the unpleasant memories of the accident if not fully, at least pontially have
been wiped out.
I take this as if desired providentially.

Nagpur
Makar Sankranti P.. H. Tbpkary
1 995
 Contents

l. lrtuoduction - 1
Scope, Concepb and terrre,
2. Ffustlawof Tbemod5mamics ' 15
hrbrtr.al Energr, E*raby ard theirdelvatives
3. Secondlewof ThemodSnramics - 38
Enhropy ard ib Derivatives

4 Thermod5nramic Fotentiatrs - 53
F\ee Ene1ry ard F,quilibrium Crfufu
5. ThndlawofTbermodJmamics' 70
Estimatbnof Enbopy
6. ZerothlawofTbermodynamics - 76
Phase Sbbility
7. ChmicalEquilibda- 83
Fauilibrium Constant
8. Phase Equilftria in Single C.,qmponent System - 9l
Clausius - Clapeyron F4uation
9. Sohrtion I- 100
Gibbs -Drhem Equatio n anil Activily Laws

10. sohrtiontr- 116

MDdng and E><c e s s F\rrrc tio ns and,Re guhr So lutb n

11. Sohrtionltr 125

Applicatbns of Gihbs - Duhem F.quation
12. nnagrams
EquililsriainPhase -133
13. Hecfuocbemicalhocesses - lM
Emf I\lbasuremenb
14. Ftee hergy- Tbmperature Relationships ' 151
Eilingham Diagrams
15. IGnetics of Reactiore' 180
Process rates

Bbbg:aphy I
Symtiob dnd Abbreviatio ns n
[*lex Ily
 Acknowledgement

The author is indeed indebted to many in the preparation and publication ol

this book in a very short span of time. I am grateful to all my earlier colleagues and
staff members of the Department of Metallurgical Engineering, VRCE, for their
constant encouragement in writting this book. Special thanks are due to Shri. A. P.
Patil, lecturer, for his untiring efforts in making me conversant with computer-com-
posing of the initial manuscript of the book.
I have no words to express my gratitude to Shri. Sandeep Gadge for his
uncomplaining help in the preparation of the final manuscript on D.T.P. of the Shri.
Narkesary Prakashan Ltd,, Nagpur. I am equally thankfulto Shri. Ajay Dhakras of
M/s. Surya Offset in getting this book printed in almost record time.
I have deep sense of appreciation of my wife Mrs. Madhuri and chlldren,
Vidyesh and Shailesh, without whose suggestions and constructive eritiiism this
book would not harle been in the present elegant forrn.
The scientific tradition as exemplified by the folloWing quartet

r rE[$lfli[E[rE[ zfftlq[: cfulffi: ggffigT

T{dr qRE rrqrfr +treFilFt: g*: I

cim'Sir -Rerq sn+r$ {dE: .

rHr *'+-6{rqAq F* qqrgwrqq' I ir
t

always inspires me to contribute my liftle might in regaining that scientific glory of our
nation. May this piece of scientific work be my humble offering at the alter of our
beloved motherland.

R. Fl. Tupkary
 CHAPTER

lntroduction
scope, coneePts and terms

1.1 l-llsroRlcAL

1.1.1 Energy in Historical Perspective

available i'e'
The stone-age man sustained himself by eating whaterrerwas naturally
He must have
green fruits, roots, dead animals, etc" and perhaps resided in natural caves'
felt the warmth of the bright sun withottt knowing that it was due to the thermal energy
same
associated with the sun rays. He must have accidentallydiscoveredfireandfeltthe
gradually understood
warnrth that he experienced in sun-rays. With the passage of time he
dried leaves and chips
the need of cooking his food, aibeit crudely, may be on fire lit by burning
of trees. slowly some crude implements made out of branches of trees or pieces of stones
to build crude habitat made
might have been developed by him and similarly he appreciated
out of similar naturatly occurring materials like bushes and trees'
He was moving around in search of food thereby doing
some labour for sustaining
some work' The human energy
himself out of natural instinct. ln modern terms he was doing
illustrated here by
was being used to do work. This is true of all beings of animate world,
being is the energy that is
the example of hurnan being. The food consumed by human
work' A dead man does not
converted through biological and physiological processes into
do such work because he does not consune food. ln modern
days the food consumed by
food he consumes' A robust
a human being is assessed in terms of calories equivalent of
manconsumesmorefoodancJ,atleast,iscapableofdoingmorework'
As the human civilization prcgressed from stone-age into the bronze-age
man

appreciated and developed the art and science of extraction

of copper and tin from their

minerals, therby producing alloy bronze and its conversion into

useful implements for
been used in this extraction
sustaining his civilization. The chemical energy of coal must have
This is really the beginning of use
operation without understanding it as is understood today.
and tin minerals, into useful
of energy for chanEing the existing materialforrns, i.e. copper
implements for better human life.
use in the development
The same is the story for all materials that slowly came into
to
of hunran civilization since the stone- age. The bronze-age progressed into iron-age
produce a much wider range and better quality materials for the
use of mankind' This
uttimately brought in the industrial rewlution and progressed
further into its rnodem form
through the enormous engineering and technological activities.
Alltbe modern manufacturing activities are nothing but changes brought about in the
existing chemical, physical and mechanicalforms into new forms. For example, iron ore is
reduced by carbon to produce iron and then converted into steel. Steel is processed to produce
varieties of forms and shapes. A tree is finally conrerted into a chair or a table. A bamboo
is converted into a paper which is then made into a notebook. All these are changes
deliberately broughtt into the existing material forns for the benefit of mankind. Without these
manufacturing activities the present day human civilization just could not exist.
ln fact the history of human civilization is nothing but progressive increase in anpunt,
quality and cornplexity of materialsmadeavailablefortheuse of mankind. This was all
possible because industrial revolution could obtain energy on an increasingly larger scale,
hitherto unknown, for the benefit of rnankind. No wonder the quality of human civilization is
assessed in terms of per capita' consumption of various materials, in particular the basic
materials like steel, aluminium, copper, etc. or per capita consumption of energy like
electrical energy.
Energy was available to the stone-age man as is available at least in the naturally
occurring forms. He however could not use it because he could not harness it as is possible
today. lndustrial revolution is a 'revolution' in terms of h4rnessing the available energy.
The steam engine is the device that was invented first to convert essentially the chemical
energy of coal, by first converting it into thermal energy, and then into mechanical energy and
thereby started off the industrial revolution. Consequently the railway train, textile mills,
tool manufacture and so on got off on hitherto unknown larger scale.
Soon chemical energy could be converted into electrical energy in thermal power
plants and manufacturing activities carried out on increasingly larger scale using electrical
power. Electrical energy could be generated by constructing dam, on rirrcrs. The potential
energy of water , so stored, was subsequently converted into electrical energy through
hydel generators. Similarly chemical and electrical energy was interconverted as per
need, through electrochemical cells to produce one from the other. This finally culminated into
the development of almost infinte source of energy through conversion of mass into energy
through nuclear reators, which is nothing but a miniature form of celestial star, a perenial
souce of energy on earth.
lf one looks at the history of human civilization from the point of vieur of energy
r.rtilization, few facts come out quite vividly.
1. The human energy is limited and hence the work that can be carried out manually is
limited. lt is estimated that a, human being can offer energy equivalent of about
1lB horsepower.
2. Externalenergy is required to obtain new rnaterials and in their new forms.
3. Energy is required to produce materials in their proportion, quality, complexity, etc.
4. Use of more and better materials is rasponsible for the developmwnt of hunran
civilization starting from the stone-age upto the modern super industrial civilization.
5. lndustrial revolution is nothing but the development of means of harnessing energy for
the benefit of mankind.
 ln energy terms therefore, the history of human civilization is nothing but
development of means of harnessing energy in ever increasing proportion, from and.in its
varied fornrs, for the benefrt of mankind.
The present day civilization sustains itself on its tremendous capacity to bring about
changes in existing material forrns. A change may occur as a natural one, for example,
burning of coal, accompanied by evolution of energy. However most changes are to be
brought
about by supply of energy from outside. These are unnatural changes typically illustrated
by the example of extraction of metals from its rninerals. The necess'rty to evaluate the
actual energy requirements for such changes and the progressiw decrease in the
energy consumption by way of the improrements in such activities has been one of the
landmark dev'dopment in human civilization'
1.1.2 Thermodynamics in Historical Perspective
The establishrnent of industrial revnlution on a firm footing by the eighteenth
century, as was the resutt of human ingenuity, was also the eause for further development
of human ingenuity, quite naturally, compelled these 'revolutionaries' - the men of
science, to understand the scientific principles underlying the means adopted for har-
nessing th6 therrnal energy readily available from burning of chemicalfuels'
The early
energy
approach was essentially related to the steam engine- a device that converted therrnal
thermal
into mechanical energy. This eventually crystalised intothe study of this dynamism
of

energy into what is now known as the 'science of thermodynamics'. For

quite sometimes
even after the cornrnencement of industrial revrlution thermal energy was the only energy
of
form available to be harnessed for useful industrial activities.This is the origin of science
thermodynamics.
With the progress of industrial revolution on the one hand and development
of
science as a whole on the other, led the scientists to discover ser,eral other forms of
energies like electrical, electrochernical, magnetic, nuclear and so on' The harnessing
of

these energy forms inrolved the development of means of interconversion of these

energies. The science of thermodynamiqs had to deal with each one of these intercon-
version ofenergies, as it appeared on the scientific horizon. The traditional name howerer
continued to be used for all such studies although the interconversion
ai times did not at
all involve thermal energy as such. Strictly speaking therefore the science of interconversion
cover allforrns
of energies need to be named anew as something like 'energodynamics'to
of energies, but the same traditional name thermodynamics continues
to be referred to the
study of interconversion of energies.
years but
The science of thermodynamics really developed over the lastfew hendred
more so in the last century, The formulation of Nerns't Heat Theorem in
1906

completed the formulation of 'elassical thermodynamics' i.e' the laws of thermody-

namics and related mathematical deductions. lt should be remembered that the
desire

and the necessity to assess theefficiencyofsteamengines,theninuseforquitesonretime'

more than anything else, led to the derrelopnrent of classicalthermodynamics'
Hence
engine or
the related mathernatical deductions are often illustrated with reference to a steam
in its idealised form as an idealgas engine'
1.1.3 Scope of Metallurgical Therrnodynarnics
The science of thernnodynamics here deals with only conventional forms of
energies like, electrical, mechanical, chemical, etc. The non-conventional energy like
nuclear energy related to atomic and sub-atomic particle forms has to be dealt with
separately because in that case all matter would have to be considered as forms of energy
as per the famous Einstein equation E = mC2 . The nuclear thermodynarnics is altogather
different branch and in order to distinguish the general thermodynamics the terms chemical
or metallurgical or mechanical thermodynamics have been introduced lately. Here the subject
matter of discussion is chemical and/or metallurgical thermodynamics alone. The systems
under discussion will be one of finite dimensions consisting of a large number of particles, i.e.
macro systems.
.I"2 THERMODYNAMICS AND RATE OF CHANGE

The knowledge of thermodynamic science may indicate the possibility of the change
but it does not indicate in any way the time required to bring about the change i.e. the rate
of the change is not predictable by thermodynamics. The rates of changes are studied
under a separate brach of science known as science of 'kineties' which helps to assess the
time required f6r a certain change to occur vis-a-vis the physical configuration of the system'
This can best be illustrated with the help of the following example :
The loss in potential energy or more appropriately the exehange of energy as the
water flows under gravity is the same irrespective of whether the outlet is say 10 mrn dia,
25mmdia orintheextremeacapillary.Butthetimerequired for a ceftainamountofwater
to flow down varies considerably in these cases.
lri other words the rate of exchange of energy is different in these three cases'
More appropriately the rate of thermodynamic change is different in these three cases and
it depends on the physical nature of the system.
1.3 EI\IERGY
is the abilityto do work, useful or otherwise. This is too
What is afterallenergy? lt
mechanical an answer. lfsulphuricacidandwater are mixed, even gently without much
mechanicalwork, heat is produced. Where was this thermal energy residing earlier? in the
acid or in the water? Similarly if silicon is dissolved in iron huge amount of heat is produced.
Where was it earlier that it is evolved on dissolution?
Having once established through the famous Einstein Equation E = mC2 that
even mass is a condensed form of energy, the whole universe is nothing but energy in its
various forms, like mass, thermal, kinetic, chemical, rnagnetic, etc. On terestrial level also
it is true to the extent lhat everything is in terrns of energy. Quite obviously the energy
associated with di{ferent material forms is different and therefore when one form changes to
another the energy eitlrer gets liberated or is absorbed in one or the other forms of energy.
As a corrollary energy may be used to bring about changes in material forms, Natural
changes in material forms are accompanied by evoltttion of energy, e'9. burning of
coal, whereas unnatural changes have to be brought-in by investing energy. Natural
changes are nol many. Even many such have a history such that earlier it had been
unnaturally brought into the forrn, like water stored in overhead tank, that now it flcnars down
as natural change. Unnatural changes in material systems are necessary to produce so
many materials for sustaining human civllization. Therefore nothing will change unnatu-
rally from as-is-where unless some energy is invested to bring about that change'
Energy is therefore capaclty to bring about ehanges in the existing nnaterial
forms as per requirements. This definition is much broader than the mere ability to do
mechanical work. Some material forms possess so rnuch energy that given a chance they
will undergo changes on their ourn, liberate excess energy and take a more stable, lo+v energy
level, material forrn in consequence.
Understandably such material forms are unstable because of higher energy associ-
ated with them. Ely and large naturally occurring materialforrns are not at high eneregy level
and are processed on industrial level to produce a variety of material forms required
for human civilization. These manurfacturing activities therefore require energy' The
industrial manufacturing activities are such that more than the theoretical energy is always
required to carry out these industrial activities. This gires the energy efficiency of the
manufacturing operation. This energy efficiency is one of the most irnportant parameter that
affects manqfacturing economy and is under constant survillence for improvement. The
thermodynamic study is relevrant from this point of view'
1.3.1 Energy ProPerties
The reader upto this stage is familiar with mass, pressure, temperature and volume
as these are far less abstract icjeas. One can feel or almost 'see' these p;'operties.
Although energy exchanges during heating or cooling or phase changes can be assessed
in terms of these simple properties, these properties by themselves are not adequate enough
to assess energy exchanges during most of the changes. One of the simple reason for such
an inadequacy is that these properties are not by thernselves energy paranreters as such.
There must be sorne energy parameters or properties of maierials such that during any
change, and change they will, so that the energy exchanges would be assessible direcily
in terms of such energy pararneters.
The thermodynamic science has evolved several such properties of materials
which are energy parameters and by mere algebric summation of their values before and
after the change can rer,eal the possible energy exchanges in the changes. The following
examples can be cited as illustratlons.
1. A system takes some heat and in return does some work. Where is the balance energy
? The theimodynamic science calculates this in terrns of changs in 'internal energy'
which is a energy Parameter.
2. A liquid is converted into vapour at the boiling point without changing the pressure
and the temperaturel The latent heat is provided foi'the change. The question is
where and in what form the energy of vaporisation exists after vaporisation. The
tliernlodynamic science evaluates this in terms of a property called - 'entropy'of a
system and which is a energy parameter'
3. A metaland oxygen combines to form the oxide. ln what form the energy is lost from
the materials ? The thermodynamic science assesses it in terms o{ either 'enthalpy
 or free energy' which are yet another energy parameters.
4. When electrolysis of a solution or an electrochemical reaction takes place it is
thermodynamic science that can assess the energy exchange in terrrs of enthaly
and free energy parameters of the chemical reaction taking place.
1.3.2 Thermodynamic Laws
The observations on changes taking place in this universe have led to evolve
sorne definite facts which were formulated in due course of time into specific statements
knownas theThreeLawsofThermodynamics.TheZeroth Law carne uplater. Although
there is no direct proof to prove the validity of these thermodynamic laws but their
applications to various changes taking place in this universe do yield correct results. These
latrvs are applicable to finite changes in materials of finite dimensions consisting of large
number of particles. These laws can not be applied to changes in materials of infinite
dimensions such as on an universal scale or equally to micro-systems of just few atoms
or molecules.
1.3.3 Fornns of Energies
Since the thermodynamics deals with energies in all its forms it would be worth-while
to know the possible forms of energies in material aggregatas.
The concept of energy is foremost in modern science and it is the basic consideration
in manufacturing processes from the point of view of process economy. Yet the concept of
energy is not obvious. The concept of kinetic energy became apparent first, potential energy
inalienably follored. ln 1851 Lord Kelvin defined energy as, "The eilergy of a material
system ls the sum expressed in mechanical units of work, of all the effects which
are produced outside the system when the system is made to pass in any manner from
the state ln which it happens to be to a certain arbitrarily fixed initial, i.e 'standard
state'.". lt became obvious that there is nothing like absolute energy but only relative
energy and that only energy changes can be measured. From the point of view of chemlcal
and metallurgical engineers kinetic energy is seldom of interest. The total energy devoid of
kinetic energy is called 'internal eneregy'.
Energy has also been defined as capacity to do work. Work is considered to be
done if a body moves under the influence of force applied on it. Therefore
w=fxd
where f is force in dynes and d is the distance of movement. Work is measured in terns of
erg in C.G.S. and Joule in S.l. units.
1 Joule = 107 ergs
ln otherwords what is capable of doing work must possess energy. Energy manifests
itself in variety of forms like thermal (heat), mechanical (in tern-rs of movements), electrical
(in terms of flow of electrons), chemical (in terms of bonds), nuclear (in terms of mass) and
so on. Except the nuclear all other forms of energies and their interconversion and
conversion to work specifically shall be deait with in the text to follorro.
Mechanical;
i) kinetic energy = 112 x m€15S x velocity
 ii) potential energy = lrl€lss x height x acceleration component
iii) congifurational = integralforce x distance over energy certain limits

Thermal :-
Heat exchanged = rrass x specific heat x temperature difference

Electrical :-
Electricalenergy = cuff€ftt x time x potentialdifference'
Chemical:-
Chemical energy = nurnber of chemical bonds x bond strength
't.3.4 Sources of EnergY
Although theoretically errcrything in this universe is energy, practically the resources
of energy are limited. The naturally occurring materials which on mere initiation undergo
changes into materialforms, and thereby liberate energy during the natural change, are one
of the chief sources of energy on the earth. The coals in all their forns, mineral oils ancj natural
gas on burning give thermal energy which is either used directly or converted into electrical
or mechanical energy. This is the chief source of energy and these materlals are classed
as chemical fuels. Fr.lel is considered to be synonymous for source of energy' Water stored,
naturaly or artificially by dams at high levels has potential energy which can be convefied
into electrical energy by hydel generation and hence is a source of energy.
These are conrrentional sources of energy. With the rise of modern civilization into a
enorrnous manufacturing activity the energy requirement is ever increasing. The natural
sourcs of energy are limited and hence attention is focussed onto alternate sourees of
energy.
ln this category cornas utilization of heat energy of sun-rays or kinetic energy of
winds and tides, by converting them into suitable energy forms preferably electrical,
mechanicalorthermal. The bio-gas generation by disintegration of vegetables and bio-mass
also comes in this category.
The nuclear energy available from fissionable materials like uranium was considered
to be a very good potential source of energy on earth but the hazards associated with its
waste disposaland the necassity of high safety standards has posed great hurdles in its full
development as a source of energy. The initialeuphoria in this field in the fiftees has
now dampenedconsiderablY
The advantages of chemical fuels as a source of energy is that it can be stored and
carried conwniently and useid as and when required . On the contrary electrical energy can
be conveyed over hundreds of kilorneters within no time but can not be stored readily. lt must
be used when available. Energy can be stored through chemical cells but the investment
required is rather too large.
1.4 TI-IERMODYNAMIC SYSTEM
It would be worth while to define certain tern'rs before understanding the thermody-
namic laws. Since the laws of thermodynamics are applicable to matedd aggre-
gates offinitedimension,awell defined finiteaggregate of materials has tobe visualised
or in reality - constructed for the application of these laws and their derivatives for
precise deductions. Such an aggregate of the material or a portion of this infinite universe,
with real or imaginary dimensions, is referred to as a 'thermodynamic systefii', quite
often for the sake of brevi$ only a 'system' . ln a dynamic bulk a portion of it can
be imagined with finite imaginary boundaries as a system. The space or materials
around this therrnodynarnic system under consideration ls generally referred to as
'surroundings'. The system may exchange matter or energy or both with the surroundings
during the change. The thermodynamic systens can be classified based on its interaction
with the surroundings or on its orrun material distribution or on its o\,vn composition as shown
in Fig.1.1.

THEROMODYNAMIC SYSTEMS

IN TERMS OF ITS BASED ON

INTERACT]ONS MATERIAL COMPOSITION
THE SURROUNDINGS DISTRIBUTION

ISOLATED- OPEN BOTH CLOSED SINGLE MULTI-COMPONENT

gCHANG. MATTER AND CAN Q,A COMPONENT
ES NEITHER
MATTER
ENERGY
EXCI.IANGE
E(CHA.
NGE
Xu
Ei
fi O
H@
de
f,ur
83
z=
uJo-
9q
r--l @o a
NOR
ENERGY
WITH THE
OCCURS EN E RGY
BUT NOT
MATTEB
fr$
:E>
Xe
=a
@
3
2
l)
Eq
ri
tr)o
^ft;
o

32iH
:)

U.r
/F
!=
U^ qlH6
SURROUND
o *u) oX(E
INGS
o
5xg
irJv aoi!
= == J!I

Fig. 1.1 - Classification of tkermodynamic systems.

Atherrnodynamicsystemmust, of necessity, be stablewithrespecttoitschemistry

during its study. lf inherently the system is undergoing continuously some chemicalchange
it can not be considered as a thermodynamic system. For example a live animate body
like tree or a human being are not systems because they are continuosly unclergoing some
changes - the life sustaining processes. Similarf a closed circuit Daniel cellcan not
be a system because the anode and the cathode chambers are continuously undergoing
chemical changes. On the other hand almost all of the inanimate aggregates of well
defined bounddries can be called thermodynamic systerns as their chemistry is fixed (systens
in which extremely slcryrr chemical processes are going on are exceptional). There can be
sorne peculiar unstable chemicals which may not fall under the category of systems but
these are exceptions. ln short there must be complete-internal- chenoical-stabilty for
anything to be called as a thermodynamic system.
 1.4.1 State of System
The system has to be described by some experirnentally determinable paranreters
so that it can lead to its complete description and is understood by one and all as that
particularsystem. For exampleforasinglecomponent homogeneous systerntherefore only
two variables are to be fixed out of the three viz. pressure, rolurne and temperature for
its complete description. Such variables are known as 'state 'variables. The minimum
number of variables required to describe the state of the systenn are called 'independent'
state variables. lt is also referred to as 'state properties'.
The state of the system can also be described in terrns of energies associated with
it' The algebraic surnmation of the energy parameters of the systern in two different states
gives the energy exchanged during the change of the system from the initialto the final
state.

T.5 EXTENSIVE AND INTENSIVE PROPERTIES

lf a state variable, whether dependent or independent, is a function of the mass of
the system it is known as 'extensive'state property e.g. volume, energy properties, etc. lt
only means that these properties get added up in proportion to thelr mass or the size of the
system and are expressed per unit rnass like per gram mole. lf hower,er the function is
independent of the rnass or size of the system it is called 'intensive' state property like
density, temperature, pressure, conductivity, etc. These are not additive properties. For
example when one gram of a substance at T degrees absolute is added to say a ton of similar
material at the same temperature the resultant temperature is still T degrees only.
Product of an intensive and an extensir,re properties is also an extensi\,e state variable
whereas the rario of two extensive properties yiels an intensire propefi. The values of
extensive properties, expressed per unit quantity of the system or substance, become
intensive variables like specific volume, specific heat, partial properties, etc.

1.6 EQUATION OF STATE

The state of ihe system can be described in the form of some mathematical
equation involving some state variables.
This analytical {orm as applicable to the system under consideration is known
as 'equation of state'. For exarnple the pressure, volume, temperature and mass of an
ideal gas are interrelated by an expression as
PV = nRT (1.1)
where n is the nurnber of moles and R is the universal gas constant. This is the equation
of state of ideal gas. Therefore if any two variables are fixed the third is automatically fixed
for a giwn mass of ideal gas. The abow relationship is an exprassion which correlates
the pressure, ternperature and volume of an ideal gas. ln fact this is found to be true in
case of solids as well as liquids though the exact form of this relationship may not
be known. The same can be described in a generalised functionalform as
F (P,VT) - 0 (1.21
It only means that there exists a function for every substance in any form which correlates
pressure, volume and temperature. lt is not even remotely implied that it must be known. lt
ryray
or it nray not be known.
There can be sereral such relationships correlating the available state propedies
of a particular system and all such relationships shali be called as equations of state for
that particular system. These are useful relationships to calculate the required
state
properties' knowing sone others.

1.7 THEREMODYNAMIC PROCESS

A material system may undergo a change, under externally or internally irnposed
constraints, in terms of their state variables, from the existing one to sorne different
variables. such a change a'thermodynamic process''
in the state of the system is known as
physical one of breaking
lf a ten meter long rod is cut into two pieces the change is merely a
it into two pieces. But if in the process of breaking, the energy required to break the rod is
taken into consideration then the process of breaking is considered to be
a thernnodynamic

process. similarly if iron combines with oxygen to form its oxide, it is a simple chemical
process but if the energy exchanged in this process is taken into account
then this process
process takes place
of oxidation is a thermodynamic process' The way the thernndynamic
,process path'. There are different types of thermodynamic processes
is known as its
named accordingly and are
depending upon the conditions imposed on the system and are
shown in Fig. 1.2.

lsoBARlc
Fis.12 -
Sc he matic tion of
re Pre s e nta

g pressure vs. volurrte in vqri-

tr G
E ous
f o tkermodYnamic Proc-
|o
rt (, €.ss'es like isabaric,
lt o
g
G
o. kolhermal, isochoric, adi-
abatic and polytroPic-

VOLIJME

lsothermal - ln which temperature remains constant and is mathematically

de-
jacket
scribed as dT = 0. lt is imagined that the system is in contact
with a thermal
maintained at the required temperature'
imagined that
lsobaric - ln which pressure does not change and hence dP=o' lt is

10
the procass is carried out in open atmosphere or in a vessel where the pressure is maintained
constant.
lsochoric - ln which volume does not change and hence dV = 0. lt neans that the
process is carriedout in a vesel of known volunre.
Adiabatic - ln which no exchange of heat occurs with thesurroundings and hence
dq = 0 and is described as PV as constant.
It is assumed that the process is insulated from the surroundings perfectly. The F or
V may change as per the given equgtion.
Polytropic - Those processes which obey equation P\d is constant whereXis
any positive number between one and y but not equal to either of them. ln this all variables
can change but since any two can determine the third therefore, only two vary and the
popular form of this equatlon therefore takes the form involving P and V only,

1.8 PATH ^AND STATE FUNCT1ONS

It is easier to visualise that the change in pressure, volurne dr temperature during a

thermodynamic process is entirely related to the original state of the system and the final
state to which it has changed to. lt does not depend on which way or by what process path
the change has actualiy occurred. For example if a gas is expanded from say a urlume of
10 meter cube to 15 meter cube by an adibatic or isothermal chahge, the change in
volume is obtained by algebraic summation of the two end states and it is 5 meter cube
in this case. The same is true of temperature and pressure. The property whose change
depends on only the initial and the finalstates of the system and not on the path adopted
tobringaboutthe change,iscalleda'statefunetion'. Mathematicallythereforeiftheproperty
is a state function then in a cyclic process, when the system undergoes a change but
returns back to its original state, the net change in the state property or function is zero, lf r
is the state function then
fdr = 0 (1.3)
and if say 1 is not a state function then f dy need not necessarily be equalto zero, where
g indicate the integral over the entire cyclic chanEe. As against this there are properties
such that the change in them in a process depends on the path followed for this change.
It is interesting to note that energy is compounded of two terms nanely, heat and work effects
produced outside the system, neither of which are uniquely related to the state of the system.
Let us consider a gas being compressed or expanded from state A to state B as
shown in Fig. 1.3. lt can be changed from state Ato state B i.e, from configuration Pr Vr to
Pz Vz by any of the courses 1;2, or 3. The work done in such a case shall be the area
under the relevrant curve from Ato B, and it shall be integral PV. ln course 'l the process
is carried out isothermally. ln course 2 it is first by cooling follcnrued by heating-and in
course 3 it is first by heating followed by cooling. The work done in all these cases is
different though the initial and final states are exactly the same, because of different paths
follovrred in these cases. Therefore the work done by making the system change from
state A to B is not a function of states A and B but it definitely depends upon the path
adopted for the change, $imilarly the ability to absorb or desorb heat i.e. heat

11
 in
lrt
c Fig. 1.3 - Schematic P-V dingrant
rtt showing the expawion and contrac-
o
lg1
aa
lion of a gus between stateA depicted
o,
by Pt Vt and stcte B depicted by p2
02
Vz.

V1 v2
-'+
*Vb'"frfr?E"
exchanged in the process depends on the path followed. Hence work done w or heat
exchanged q are not state functions but path functions.
The usefulness of state functions is also obMous from the above example
because mere algebric summation of the values of the state functions gives the change in
the state but in this instant case of assessing the work and heat exchange effects such
a summation does not lead to the required value as these are not state functions. Whether
a state function or not can be mathematically assessed as follcnrus. For an
any function is
ideal gas T = PV/R wherein P and V are taken as independent variables and T as
dependent variable. The above can be expressed in its total differential form as

dr = (#) dv . (#) .' (1 4)

generalised form if a system has two independent variables say x and y and any third function
z can be expressed in its total differential form as ;
dz = Mdx+ Ndy (1.5)
where M and N may be functions of x and y hence function z shall be a state variable if and
only if
AM AN
Ay Ax
(1.6)

Whether a function is a state function or not can thus be established by relating it with
two other known state functions through the above relationship and if it cornes out to be true
it shall be a state function.

1.9 THERMODYNAMIC EQUILTBRIUM

This is very important concept for understanding the usefulness of thermody-
namic science and its applications. This can be readily understood with the help of common
examples. Let us imagine a inflated tube of an automobile vehicle. lf it is punctured in
an open atmosphere the pressurised air from the tube leaks out through the puncture like

12
a jet till thereno extra pressure left inside the tube. What will happen if the same inffated
is
pressure ? The
tube is punctured in a chamber which has also been inflated to the same
are equal l.e' since
tube will not get deflated. since pressure inside the tube and outside
mechanical potentials on either side of the tube wall are the same' there exists
a'pressure
or mechanieal equilibrium' in the system - the tube and its surrounding chamber'
florvs
Similarly if two objects at different temperatures are physically ioined heat
lf the two were at the
from the one at higher temperature to the one at lovrrer ternperature.
same temperature no heat flow shall occur. lt is then said to be in 'thermal
equilibrium''
lf two phases are in contact such that a chemicalreaction takes place between
rates of forward and the
thern, but the two phases can be present in such a way that the
called as in 'chemical
backward reactions are exactly maiched then the system can be
same in that case'
equilibrium. The chemical potentiais in both of these phases must thebe
,thermodynamic equilibrium' is thus that situation where the system is
Complete
The
with respect to all such potentials like mechanical, therma!' chemical' etc'
in equilibrium
term equilibrium is applied to a system and not to a process'
'1.9.1 Paltial Equilibrium

systern can be in a state of partiai equilibrium. lt only means that the

two phases
A
involved in a system are in equilibrium with respect to any
one of the above mentioned
parameters, thermal, chemical, etc. or a combination of these but not
all of them at a time'
say a mixture of nitrogen and
if a system contains a semi- permeable membrane separating
hydrogen gases on one side and only hydrogen on the other,
it is likely that the partial
pr"."rru" of hydrogen on either side may be equal because of the membrane being
of nitrogen being
permeable to hydrogen but the total pressures may be different because
impermeable through the membrane. lt may take ages before
the system has total
pressures equal on both the sides of the membrane. Strictly speaking
the system is not truely
using
a thermodynamic system and hence not amenable for thermodynamic treatment
on either side the
of
state variables because of the continuously changing composition
membrane. Yet within a reasonable time
'scale the systern can be considered to be a
system and treated accordingly. The equilibrium in such a case
is partialequilibrium
und be so ireated with negligible error inrrnlved in the calculations'
"un
1.10 REVERSIBLE AND IRREVERSIBLE PROCESSES
Therrnodynarnic processes can also be subdivided as reversible
and irre-
inrrclving a system but not to a
versible processes.These terms are applied to a process
system as such.
change in the
A process that can be reversed in its clirection by an infinitesimal
driving force i.e. change in one or more of the state variables
is said to be a'reversible
pressure of the
process' A classical example of this is the gas cylinder and a piston' lf the
piston' the gas
gas is say P atmospheres and if (P + dP) is exerted from outside on the
insicie the cylinder shall be compressed. However if
the external pressure is (P-dP) then
the gas shall expand' A classical chemical reaction like
A+B-C+ D
'13
 if can be made to go in the backward or forward direction by changing
one or more of the
variables, then it is a reversible reaction. A Daniel cell has
op"n circuit emf of 0.76 votts
and if the circuit is closed then curent flows in the direction "n
sf zinc to copper in the solution
and copper to zinc in the electronic conductor, lf an opposite external
emf is applied the
current slowly decreases and is zero when exactly the
same value of emf is superimposed
on the cell in opposite direction. Any further increase
now wollld make the current florru
from copper to zinc in solution i.e. reverse. Such a process
can be made to go in the desired
direction by changing the superimposed emf. This is a typicar
reversibre process.
on the contrary a match stick when it burns, the process can
not be reversed by
changing one or the other parameters. once burnt, the
stick can not be reproduced by
reversing the process' This is typically a 'irreversible process'.other
examples are mixing
of two gases' mixing of two liquids to form a sotution
or flow of electric current through a
resistor and so on. Alr natural procqsses are most often
irrerersibre changes.
lf the case of a gas-cylinder and a piston is taken as an
experirnentattool, it can be
shown that the work done in expanding the gas very slovrrly
through a wrlume change V1
to Vz is giwn by

w1-2 = fu' Pdv (1.4

"v,
and that to get the system back to its original position is

wz-r = [:, p dV = I^Vz pav (1.8)

rVt
lf the process is carried out not so slovvly then it can be.proved that the
work done
in similar expansion and contraction is less than that done when
/ the process is capied out
slorrly. Hence
work (rerrersible) work(irreversible)
This is so because in reality the pressure difference driving
the system to do work is
maximum in rewrsible state.
Although this is a simplified picture under isothermal conditions applicable to ideal
gas it holds true for any working substance gas, solid or liquid under any conditions.
1.11 CONCLUDING REMARKS
The enormous industrial rnanufacturing activities required to sustain the pre-
sent-day civilization requires equally enormous energy. The energy
resources are limited. lt
is therefore necessary to understand the scientific principles
underlying energy require-
ntents in these manufacturing activitias so that the minimum
energy inplt n12y bJachieveo.
The thermodynamic laws and their applications contribute towards
this understanding.The
basic terms such as energy' system, process, state and path properties,
reversible and
irreversible rocqsses, extensive and intensive properties and
so on are required to under-
stand these thermodynamic laws and these are therefore covered
in the present chapter and
the details of the thermodynamic raws are covered in the foilornring
chapters.

14
 CHAPTER 2

First law of thermodynamics

internal energy, enthalpy
and their derivatives

2.1 HTSTORTCAL

As seen earlier, from the definition, the measurable energy of any system is
dependent on a arbitrarily chosen standard state, There is nothing like measurable absolute
energy of a system, it is all relative. Thus only energy differences can be measured. A
system may have huge amount of energy but in a given surrounding it will be possible to
give out only a certain anrount of energy and that will be the energy exchange possible
and itwillbe measuredonly in terms of theenergyso released.lnthe presentdiscussion
and as is of no consequence, the kinetic energy and the mass as a form of energy in
nuclear terms, are omitted, being constant in any process, unless otherwise stated.
It was Runrford who first appreciated the conversion of work into heat in qualitative
manner in 1798. This was more properly explained by H.Davy in 1799 by showing that ice
melted when two ice blocks were rubbed against each other under vacuum. lt was howerrer
Joule who established through his famous experiment, in an adiabatic enclosure, the
quantitative equivalence of work and heat as
work
:'-"-= J, w
heat Joule or = (2.l',)
q
lf both work and heat are expressed in the same units then for a cyclic process

It means ,nrl;rJin ,
the Jorm of work gets transformed lnto anotn" ,.r::'?
energy as heat fully. This formbd the basis of the law of 'conservation of energy' which
is the progenitor of the first law of thermodynamics. lt also means that it is just not
possible to produce energy of any kind without dispensing an equivalent amount of
energy of some other form. lt is therefore impossible to construct a'perpetual machine'
which will do work perpetually without putting energy into it in any other form.

2.2 FORMUUTION OF FTRST I-.AW OF THERMODYNAMICS

ln a gravitationalfield work done by a body as a result of its rnovements depends on
its initial and final positions and not nn the path taken by the body between these two
positions. Similarly the work done by a charged particle under the influence of electric field
 depends on the potential difference between the two points, between which
the particle
mo\'€s, rather than on the path taken. Thus the work done, in the abor,re, by a
body in its
morrements from state A to state B depends on the functions of states A and
B and not
on the path the body takes for its movements.
ln otherwords the work done by a body in an adiabatic enclosure when moved from
its initial to final state involves the same amount of work and which is independent
of the path
followed for the change. The change in the internal state of the systern, as a consequence
ofwhichtheworkisdone, hastobedefinedandis called the'internalenergy'ofihesystem
denoted in the literature by either letters 'U' or'E'. Here it will be referred to as ,U,.
ln an adiabatic system the work done by the system on the surroundings.shall be
equal to the change in the internal energy of the system as
AU = w (2.S)
Similarly if a system while doing work on the
surroundings simultaneously
accepts heat from the surroundings then the work done shall be equal to the change
in the
internal energy of the system as well as the surroundings. The change in the internal
energy
of the surrounding shall be equal to the heat exchanged. Hence

Q = AU' (surr) (2.4)

lf the system and the resenroir supplying heat are considered as a single system
and if this is isolated adiabatically, ihen the work done by this combined system equals
the
change in the internalenergy of the combined system, or
w = (AU'_AU) (2.51
putting Eq.(2.4) into this gives
w = _ AU + q (2.6)
or on an infinitesimalform
dU = Aq -0w (2.6)
This isnothingbutmathematicalstatementofthe'First Law of Thermodynamics'.
ln a generalised form it can be said that if dg amount of heat is supplied to a system and if it
does work 6w on to the surroundings the change in the internal energy of the system is
given by Eq.(2.6).

2.9 INTERNAL ENERGY AS A STATE FUNCTION.

Let us go back to Fig.1.2. The system happens to be in a state A and undergoes a
change to state B. Firstly it follows the path 1 and second time it follovvs the path 3. ln both
the cases the system returns to its oiginalstate via path 2. ln these two independent cyclic
processes the net change in energy is zero as the system returns back to its original state.
This has to be. The heat exchanged and work done in each cycle will have to be equal.
Even from Eq.(2.6) if energy change is zero q has to be equal to w.
Therefore heat exchanged and work done in each cycle can be equated as in Eq.
(2.21as
w+ w1 + q1
clz (2.4
and t4l3+ w qb+ q2 (2.8)
16
 q2-W = u,h-Cn {2.e)
and g2-w = w-q (2.10)
and hence for that matter for any path follcnared for a ryclic change with state A and state B
as initial and finalstates
q W = constant
or when expressed in infinitesimal form as
Aq - 0w = con-stant (2.11)
Therefore for any change the actualvalues of w and g shall depend upon the path follored
but the difference shall always be the sarne €rs long as states A and B as initial and final
states remain the same. ln other words the differnece (g - wl shall not depend on the
path, although individually both of them depend on the path folloved, but on the states A
and B only and, therefore (q - ui shall be a'gtate function'. For U to be a state function
the values (q - vt has to be zero for a cyclic process and which is the case in Eqs.(2.7)
and (2.8). Internal energy ls therefore a state propefi.
It reveals that the system is capable of absorbing heat from the sunounding and
doing work on to the surrounding such that Eq.(2.6) holds true. lt is quite possible that the
system rnay, €ts a whole, resuh in positive or negative change in U depending upon the
actual values of heat g received and the amount of work w done.
Hence sum of all forns of energies exchanged by a systemwith the sunoundings is
equal lo the change in its intemal energy. lt can also be stated as that 'energy san not be
created nor can it be destroyed'. This is one of the statetrpnts of the first larrr of
thermodynanics. lt can also be stated as 'tfte total energy of a systecr rnd it* suffound-
lngs shall be constant and that it may change lts lorms or rnay exchange between
themselves'.
The system is therefore capable of accepting energy in one form and deliver it in the
sarne or different form and in the same or different amount. This is possible because the
system is capable of acting as a'reservoir' of energy. The stored energy is nothing but
the intemal energy ol the system referred above and il increases or decreases depending
upon its interaction with the sunounding or changes within itself.
The internal energy consists of all $pes of energy forms contained by the system
including mass if nuclear reactions are under consideration. ,As per Kelvin's definition thle
presence o{ any particular form of energy will be seen only as a manifestation that the
system produces, during its interaction with the sunoundirqs. For e:<arnple if nuclear
reactions are noi being considdred mass wlll not manlfest itrsetf m energy. Sirnihrly if
electrical efiects are not considered electricalform of internal energy will not rnanifest. Jn
chenical and metallurgical processes the thernral and mechanical energie alone manifest
and therefore internal energy in those forns alone shalf be manifest and would be
considered hereafter, unless otherwise stated specilically since chemical and rnetalfurgi-
cal thernrodynamic-s is being considered here.
So the reader b conr,ersant with the terrns,volune, pressure and temperature as
frar
state properties and mw internal energy U has been added to this llst ol state properties
to describe the energy content of the system. The question is whether there is any
17
correlation between the directly measurable state propertiers like P,V and T and the newly
coined property U ?
lf the system expands by a volunre change dV against pressure P then the
mechanical work done by the system on to the sunoundings isJ fOV and the Eq. (2.6) shall
be modified as : dU - dg - PdV - vl (2.',t2)
where PdV is the mechanicalwork and d is all otherforns of work like electrical, magnetic,
surface, etc. ln normal chemical and metallurgical systens w'is almqst always zero because
the system does only mechanicalwork and therefore henceforth w' is assurned to be zero,
unless otherurise stated specifically. Therefore
dU - dq -PdV (2.13)
It is desirable to have equation involving U the internal energy in the form where all quantities
are exact differentials, i.e. in terms of state properties. But this is the best first law of
thermodynamics can lead to. This can be cast in the desired form only with the help of second
lar of thermodynamias.
2.4 INTERNAL ENERGY IN TERMS OF PARTIAL DERIVATIVES
It has already been Sated that for a homogeneous system consisting of a given
amount of substance of fixed composition the equation of state of that system can be
described as F (RVT) 0= (2.14')
and since out of RV and T only two are independent variables, the third gets automatically
fixed and it can be described as
P=F(V.T) ; V- F(RT) and l= FMP) (2.15)
The internal energy U being a state function will have to be a function of P, V and T and that
is to say =
U F {RVT) (2.16)
This also therefore can be stated as (like in Eq.2.15)
lf = F(P,V)t U= F[I,P) and ll = FMT) (2.1n
and fiom the theorem of partial derivatives dU can be expressed as

du = (#)" dP . (,ry), dv (2.18)

du = (,+), dP . (#), dr (2.1e)

du = (,ry), dv . (#)" dr (2.20)

and from these under the conditions of isothermal, isobaric or isochoric processes the
dependenry of U on the state variables can be established. These equations in which
internal energy is expressed as a function of independent state variables are called 'caloric
equatlons' of state as these allorrr evaluation of thermal capacities or heat capacities. For
a single component system the caloric equation can be expressed also as
F (U'T,V) = 0 (2.21)
and for mufti-component systern€ itwill be
F (U,-T,Vn1 ,nz,Il3,.....ni ) =0 I e.zt^l
18
2.5 HEATCAPACITY
ln any process, but more so in chemicaland metallurgical processe, materials get
heated or cooled and therefore it is necessary to know the amount of heat required to heat
or as a corrollary amount of heat liberated, in cooling a material, o\,er a certain temperature
range. Different materials require different annunts of heat to get heated through the same
temperature rise. This is because the materials have difierent heat capacities. 'This is so
because of the variation in the crystal structure of nnterials and their related parameters.
The heat capacity of a substance is the amount of heat required to raise its
temperature by one degree. ln a limiting case the same can be exprassed as
aq,
vc -= aT e.z2)
The term specific heat refers to the heat capacity of unit mass of the material under
consideration i.e. it is that heat required to raise the temperature of a mass of a gram of
substance through one degree. Since g depends on path the value of heat capacity may
also depend on path, if expressed in the form mentioned as above. ln an adiabatic process
when g is zero the factor q \ dT becomes zero, which is absurd. lt is necessary
therefore to express heat capacity in terrns of state functions.
As internal energy is related to g and as it is also related to state functions,the
heat capacity, through g could be related to state functions. lt should also be noted that the
heat capacity depends on conditions of constancy of volume or pressure, particularly in
the case of gasas, perhaps less so but definitely as well in the case of liquids and remotely
so for solids as well. Hence two different heat capacities have been estimated, one as Cv
as heat capacity at constant volunp and the other Cp as that at constant pressure.
2.5.1 Specific Heat at Constant Volume
From first law of thermodynamics as in Eq.{2'13)
dU = dq- PdV
or 0q= dU+ PdV
substituting in this the value of dU as given in Eq. (2'20)

o q = (,j#), ou . (#)" dr + pdv

= (#)u o'- Pr dV (2.231

'(#),+
when dV = 0 under constant volume conditions

aq = (#)" dr

q_ - /!l/\
_
a
dr - /!_u\
lorju
or !_e
ar lurju
AU
-]-= (2.241
tr'V =
dt
Thus heat capacity under eonstant volume is nothing but rate of change of internal
energy with temperature. The same thing can be arrived at in a differentway as follors : -
19
 Under the conditions of dV 0 no mechanical work is done by the system and
=
hence w = PdV = 0 and hence from the first lal equation
(d q)y = (du)v

or (*",)" = (#)"
or Cv = fH as shown in Eq' (2'2a)
\ / i
Ahhough g is a path fuhcticin heat exchanged under constant volume
--^ conditions
-^-r!^

ls a state functton and equals change in internal energy'

2.5.2 Speclfic Heat at Constant Pressure

lf pressure is maintained constant during change of a system from initial state (i) to
final st€rte (f the work done will be

w= lrou
- Ji '-- =Pf
Ji
dv =PM-Vi)
The first larr equation as in Eq.(2'6) can be rewritten as
Ur-Ui = dq- PM-Vi)
or on rearanging these ternrs
(Ur+ PVr)-Ui + PVi) = Q

The expression (u+ PV) is referred to as 'heat content' or'enthalpy' o{ the

system and is ofien designated by the letter H as

fl= U+ PV {12.251
(2.26)
orthus H/ H, = AH = dg
-
bY dTwillgive
dividing

fry) r/t or cp=#

=f+) (2.24
lu
r/, \d
The heat capaclty at constant pressure ls nothing but rate of change of
enthalpy with temPeraturc'
it in Eq' (2'13) as:
The sarne can be derived using Eq.(2.19) and substituting
dq= du+ Pdv
= PdV .
(#r), dP + (#), o'
as dP= o ds - Pd/- f+)r/p
dr
\d
dividing by dT gires
f4{) = c}D - - /d v\ /alfrJ'
u\

beingexactdifierentials
\d,/
dU= dU and dT=
l*,l'- dT

and hence cp = e'28\

[*#-],
substituting H lor U ( + PV) in the above
20
 cP = (.+), as in Eq.(2.27)

Altematively the sarne can be derived as: From the mathematical

definition of enthalpy f{= U+ PV
difierentiating gives dH=dU+PdV+VdP
and since from first law dq=du+Pdv
hence dH= dg+ VdP
and at constant prassure since dP = 0
dH= dq
or AH = dg as arrived in Eq.(2.26)
and by dividing this by dT Eq. {2.24 as obtained.
Since enthalpy is a state function the heat exchanged by a system at constant
pressure becomes a path independent property. ln other words it is a state proper$'
Also by definition itsetf H depends on allthose parameters which are by thenrsblves state
properties and it itsetf, in the derived form, is a state properS. Since enthalpy is a state
property it can be expressed as a function of any two variables out of the three RV and T
and thus expressing it as a function of P and T as
H(RT)=0 or H=F(P'T) (2.2e)

2.6 INTERNAL ENERGY vs ENTI'IALPY

One can readily understand the term lnternal energy as allthat energy stored in
the system. But it is rather difficuh to understand the term enthalpy because it is defined
by mathematical equation. Therefore what does enthalpy mean? This can be best
illustrated by the follcnruing example :

Let us imagine the process of calcination of caleiurn carbonate Eiven as:

CaC0s= Ca0+ COz (2.30)
Heat is supplied to break the chemical bonds between CaO and COz . Assume for the tine
being that both CaO and COz are solids. Then the invested bond energy would be internal
energy of the process of calcination. ln this case the oniginal system of CaCOs with say
conditions as P,V and T rernain practically the sarne after calcination. ltwould arnount to
merely a solid/solid phase change.
However if COz is allcnroed to form as gas in this process then on breakirtg of
onegrammoleof CaCOe nearly 22.4lilres of CQ atSTPwillbe forrned as shown in
Fig. 2.1. There is a tremendous increase in the volume of the system on calcinaiion. The
system must do work of expansion to attain the nesultant expected system. The process
is supposed to be carried out at one atnrcspheric pressure in norrnal atr;lospnere. lf the
system has to attain a volume expansion of the order of 22.4 litnes at STP t['ben it should
first attain a positive pressure against the ambient atmospheric pres$ure and then
expand. ln this process of expansion, the system will do work equivalent of J PdV as
mechanical work on to the surrounding. F{Bnce in addition to the requlitement of energy
for breaking the chemical bond of CaO - eOa , additionalenergy equivalent of JeaV witt

21
 Fig.2'1 -

Work done bY the Process of

I
i:)l
,Tr calcinatian of CaCOz un'

V@LUfl4€
l0 voluME' '22'4tttet der arnbient Pressures of 1
I
rM io
l-l amos and 10-3 atnos 4t two
I
,l
6ry5 6Ts4O€i tPRE9CURE
oF ,e%
16*eono" i N
OFr
ir^..",
iCOl G'Ag different calcination tem-
".1:--- lot I I eFrpru perstures. Note the volutne
:i: -L i-='t- r-,---I-- change after the reactinn-

g6.MoLe CoeOg T G. MOLE O*CO9

HSAT.€D 7o
o
9Oo c HEATED To "
660-c

and this
this process making a total of u + PdV energy
have to be supplied to the systern for internal energy U'
on calcination, quite difierent from the
is the enthalpy change of the system
Forchemicalprocesseswherethereisnoperceptiblechangeinvolumeasin
2CaOs. SiO2 = 2CaO + SiOz
U and H are almost the same' The difference
tneJeUV term is practically absent and hence
gaseous
component in the system'lf there is no
in U and H arises because of gaseous will be
component and only condensed
phases are piesent as in the equation above'there
condensed phase changes to another condense<l
no difference between U and H. When
or can be negtgcteo orJeav term shall
btt
phase, the change in volurrre is negligible
tor Jeov arises only because of gaseouii
treated as zero. The perceptible value H'
hence ih" diff"r"nce in values of U and
component inrnlved in the reaction and,
from this may have enthalpy alsc '
A system thus has internal energy and as distinct
enthalpy is often more relevant 6nrj
ln chemical ano mJallurgical world therefore the
Even if the term Jnov isabsentitisworthwhilir
distinctly different from internal energy.
torefertoHandforgetabouttherealityasittakesintoaccountwhatitshould.
Theusageofenthalpychangesasameasuringparameterisobviouslymoti:
appropriatetotakeintoaccounttheworktermoftheprocessratherthanmereinternrll energy in chemic :tl
thus very popular rather than internal
energy. The usage of enthalpy is of her'
:[he term ,heat oi reaction' is restricted to the exchange
and metallurgical world.
does not take into account the mechanicalwo
l'
during a chemical reaction and which
term associated with it as explained above'

2.7 |NTERREI-AT|ONSHIP of Cp and Cv

Fromthefirstlawolthermodynamicsatconstantpressuredq=dU+PdV
given in Eq'(2'20)
and substituting the value ol dU as
gives dq = (ut#), ou . (ruY)"'
ot + PdV

22
and dividing by dT gives

# = (#), (,+) . (#)"

assuming constant pressure conditions
/e1\ = cp and f-a u\
lfrh \fr,,|u
= ./v
it will reduce down to

= (#), . e]
[(,+),
For an of internal energy with volurne at constant temperature
ideal gas the rate of change
is nil. That is for an idealgasbecause of its non-interactive nature.

l-l =U
la
v/r
Hence cp
- cv = (#), ,
(2.31)
For idealgas PV = RT
or ylT = R/p
(when V changes with 1 obviously P gets fixed up)
and therefore = R/P and substuting
dvidT in Eq. (2.3.1) gives
Cp-Cv= R (2'321

2.8 CONSEQUENCES OF FIRST LAW OF THERMODYNAMICS

ln the light of first law of thermodynamics various
thermodynamic processes can be studied to derir,e additionalrelationships. The system
studied is ideal gas and for which
PV = RT and (AU/OV) = Q

2.8.1 lsothermal Process

The process is carried out when dT =0. The heat supplied to the system in this case
must fully come out in the form of work done on to the surrounding. Hence
Aq = 0w or dU =0 (2.33)
It is also true that in this case

6q= 0w= { pdv=/v'Tou'=*t{] +=*t{] drnV

Similarly (2.33a)
' AH= aU+ /ev
= AU + (PrVr - PiVi)

23
 = AU + R(Tr - Tt) (2.341
Since there is no tenrperature change
Tf - Ti = 0 and combining with Eq. (2.33)
Al'{ = 0 (2.34a1

Hence in an isothermal process both internalenergy and enthalpy rernain unchanged

and work is done by the system equivalent to the heat input. Therefore now for an isothermal
proqess not only dT O but dU = 0 and dH
= =
0 and reversible work done by the
system is
q - !t/ = RT ln M/Vi) (2.35)

2.82 lsochorlc Procesr

Under constant volume conditions the system does no mechanical work on the
surrounding and henceJPdV equals zero. The first law Eq' (2.13) gets reduced to
dU= dq (2.36)
and also 65=dU + PdV=-dq (2sn
ln otherwords the heat absorbed or rejected bythe system during the process equals
the change in the internal energy or enthalpy of the system. Therefore for a constant volume
process dV = 0 and also dU = dH = dq
2.8.g Adiabatic Process
There is no exchange of heat between the system and the surrounding i.e'
dq = 0. Hence first lar takes the form
dU=-dw= PdV (2.38)
dU = CvdT = - PdV (2.3e)
For an ideal gas
Also since by definition fl= U + PV
therefore dFl= dU+ PdV + VdP
and combining with Eq. (2.38) dH = VdP
That the system does work at the expense of its fi/n energy and therefore temperature of
the system shall fall. Using these relationships for a adiabatic process the interrelation of
P and V or P and T can be found as
PV = Constantand tt(?) =
constant
Where y is constant. The reader will recall now how the relationship for adiabatic process
was giwn earlier in Chapter 1, Section (1.7). Expressing dU in terms of partial derivatives
as in Eq.(2.20)
du = (#)rou - (#)"ot = Pdv asds - o

and since (au/dTlv = Cv and (au/aVr = 0

for an ideal gas and also assuming Cv to be constant over a certain temperature range it
leads to Cv dT = - PdV
Adiabatic work done can be obtained by integration of Cv dT over Ti initial and Tt s
finaltemperature as
Adiabatic work done = -Ci/(Tr-Ti )
(2.4o1

24
andsince dU = -w= -PdV
(2.41\
adiabatic energy change = Cv (Tt - Ti)
2..8.4 PolYtroPic Processes
Since the polytropic processes obey P\f = constant, the final expression of dU
and dH accompanying the polytropic processes is quite complex because the mechanical
work done isJnV and notJPdV. Since both P and V change during the process, the
work done by the system will have to be calculated taking changes in both into
account
level.
and this is quite complex and therefore, not introduced here at the beginner's
2.8.5 Throttling and Free Expanslon
This is a irreversible process carried out adiabatically as there is no chance for
heat exchange to occur in such a instantaneous process of drop in pressure and free
expansion in volume'
since dq = 0, du = - dw and hence the system performs work at the expense
if there is a iet
of its oarn energy and therby the temperature of the system shall fall. Therefore
the jet is lcruver than
coming from a nozzle in an open atmosphere the fluid temperature in
the temperature of the fluid fed to the jet.
2.8.6 Deviation of Real Gas from ldeality
The ideal gas equationcorrelatingRV and Tisapplicableforidealgases'lnreality
the actual volume of gas is not the apparent one but, that minus the
actual volume
occupied by the gas particles (molecules). Hence clausius modified itto.
P (V b) = - RT e'42)
where
,b' represents the volume occupied by the molecular particles of the gas as
such. The real behaviour is further deviated from ideality because of molecular interaction
and hence van der Waal further nrodified it to :
/p-4) x N-b) = Rr P'42a1
\ v'l
This is known as van der Watl's Equation and it comes more closer to real
obserwd behaviour of gases. This is a cubic equation for volurne with
three alternative
values.Whenallthethree values are real,gas can be liquified. The lowesttemperature
gas will not
at which this is possible is called the 'criticaltemperature' and abow which
be liquified. The relevant criticalvolume and critical pressure are the ones when liquifaction

willbe possible.
Although the modified forrn of gas equation to correlate the obsenred behaviour
of real gasets, is notdirect consequence of the first law of thermodynamics' it is explained
a
here to understand limitations of ideal gas and its use in deriving
certain expressions as
in Eq.(2.32) in the form of derivations from first larv of thermodynamics.
2.9 THERMOCHEMISTRY
occur with
ln thefieldof chemicaland metallurgical processeschemical reactions
are nothing but making
or without mechanical work involvement. chemical reactions
in the form of
or breaking of chemical bonds (in sonre both nny occur)' Energy exists
25
these bonds (in terms of bond strength) and hence energy, most often heat, is required to
breakthese bondsand conversely it is releasedif bonds are formed. Henceinchemical
processes heat may be released or absorbed depending upon whether chemical bonds are
formed or broken. lf both of these, making and breaking, occurtogether, then the net algebric
sum of the bond energies shall be released or absorbed, as the case may be. ln a
heterogeneous process in which gaseous phase is inrrclred the workterm may be substan-
tial. Whereas in liquid- liquid or liquid-solid or solid-solid heterogeneous systens the
volume change, ewn though finite, may be negligibly small to contribute any worthwhile
work factbr in the process. ln systems containing a gas phase, if a perceptible volume change
occurs in the process the work term may be dominant.
Besides the direct chemical reactions, energy exchanges are also associated with
phase changes that may occur in the process e.g. solid to liquid or liquid to gas or vice
versa, in any of the phases involved in the process, Phase transformations in solid state
likd order-disorder or austenite-martensite transformations are also accompanied by heat
exchangesbutto asrnall extent. The processofformation of solutions or de-formation
of solutions can contribute a substantial part of heat exchanges occurring in a chemical
process.
The field of thermochemistry is to study all the above mentioned heat exchanges
during chemical or physico-chemical processes. Calorimeter is often used to study
these in a laboratory.
2.g.1 Hess's law - Corrollary of First l-aw
As the heat exchanged in a process is seen to be a path dependent property, the
heat exchanged in chemical or metalfurgical processes might also be enoneously vieured
as path dependent property. These processes are studied under either isobaric or
isochoric conditions and as shown in Eq'(2'2al and (2.271 x:
aU = dg (for isochoric) and dH - aq (for isobaric)
The heat changes in chemical processes.are not path dependent but are state
functions only. This fact was realised,though not quite in the way as is understood today,
by G.H.Hess in 1840. He suggested that the total heat exchanged in a chemical process
ls the same lrrespective of whether the process occurs ln a direct single stage
or through multiple stages. This is now known as Hess's Law.
It is now considered as a corrollary of first law of thernndynamics. lf Hess's law
does not hold good then in a cyclic process when the foarard process is carried out in say
directly one stage and the reverse in multistages then it will result in a net gain or loss of
energy. This ought to be zero in a cyclic process and hence both should neutralise each
otherinsuchawaythat the net amount is zero as per the first latnr ofthermodynamics.
Hence the Hess's lzul is a corrollary of the first larrv of thernodynamics.
The larrv can be explained by an illustration such as reduction of iron oxide by
carbon or carbon-monoxide gas. The reaction can be carried out in one stage as
' 3Fez Os + 3C = 2Fe + 3CO (2.4?)
or alternatively in stages as

26
 3FezOs+ C = 2FesOn+ CO 12.44\
2FesO++ 2C = 6FeO+ 2CO (2.45l.
and 6FeO + 6C = 6Fe + 6CO (2.461
The total enthalphy change in both of these routes is exactly the same. This can be proved
thus:
The enthalpy change in reaction (2.4i]) i.e. single stage is gi\,€n by

AH1z.+s1 = 2Hre 3Hco - 3Hc- Hr".o.

+ (2An
and enthalpy change in reactions Q'a l b (2.46) are
AH1z.aal = 2HFeooq + Hco - 3H5er9. - H6 (2.48)

IHp.+sy - + 2Hco - 2Hresg. - 2Hc

6Hneo (2.4e1

AHlz.my= 6Hre + 6l-lco - 6Hreo - 6Hc (2.50)

which on addition gives the same equation as inEq. Q'al
Example : Calculate the enthalpy of forrnation of WC from the follcnrving data at 25 C:

(1) wC + z| oz - Wos + Coz

AHo = -285800 caVg mole

(21 C+ Oa=COz
AHo = -940950 cali g mole

(3) w +'tloz - 'z

wos
AHo = -195700cailgmole
Solution : WC is formed from C and Was
W+ C- WC
and is obtained by addition of equations 2 and 3 and subtracting from this, equation 1
and

the algebric summation of the AHo values give the required enthalpy change as

AHo= AHz+693 -AHr

(-285800)
: -;ff' ;r]ffl"*
2.10 HEAT CAPACITY CALCULATIONS

Heat capacity for any substance is not a fixed value. For the same material it changes
with temperature and the way it changes with temperature varies considerably from
sub-

stance to substance. The actual relationship has to be established experimentally. Howewl

Fol
certain empirical rules have been suggested and are used for approximate calculations'
mono-atomic gases at high temperatures, the specific heat GV or Cp approaches a
value o1
312R. For diatomic gases ihe is close to 5/2R and for solid materials Dulong and Petil
"a*e
suggested that the molal heat capacity is equal to 3R i.e. 3 x 1.987 = 5'96 cal/de'
gree. This statement is known as Dulong and Petit'l-aw (1319). Einstein developed the
n*"rory theory for calculation of heat capacity based on quantum theory with
certain assumptions regarding the nature of crystal structure. This is known as 'Einstein's
Theory ol Specific Heat of sotids'. This was found to be reasonably valid for high
temperature ranges but it failed to explain the observed values of specific heats
at lcnn
27
temperatures. This has thus been modified by Debye, that belor a certain temperature,
known as 'Debye's characteristic temperature' and , which varies from substance to
substance, the specific heat is given by
cp or cv = . 234R 1T/0o)3 = 464.5 F/0o)3 (2.s1)
or Cp or Gv cr l" (2.s2)
where 0 o is the Debye's characteristictemperature. This is also known as Debye's fp Uaw
for lorry temperature heat capacities. The obsened values of specific heats of various
materials at lorv temperatures show good agreenrent with the above equation.
Since the variation of specific heat of any substance with temperature is a complex
relationship, after exhaustive experimentation empiricalrelationships have been developed
for a number of elements and compounds in the form
Cp=a + bT *"1."_, (2.s3)
and Cv=a'+b'T+c" (2.54)
where a,b,c,a',.b'.andc'areall constants characteristic of thesubstance over a certain
temperature range. lf data are not sufficient the final term may be omitted. Simplification
is also possible by trsing mean specific heat over a certain temperature range as defined as

(or9p)
Cp or Cv (meanvralues) =
{'w (2.55)
(Tr -Tz)
Often tables are available giMng the values of constants a, b and c for a large number of
substances,
2.10.1 Kopp Rule
Where specific heat values of compounds are not known from experimentation it
can be approximately evaluated by the use of 'Kopp Rule'suggested in 1864. lt says that
the specific heat of a compound A B shall approxinntely be given by a formula as :
Molar Sp. Heat of A" By = rsp.heat of A + y.sp. heat of B per g atom (2.56)
This is fairly reliable for ionic compounds wherein the individual elements obey
Dulong and Petit's law.
This is extended to solutions, metallic as well as non- metallic, to work out approxi-
mate values of specific heats frorn the specific heat values of individual components.
The specific heat of solutions is equal to the weighted sum of the heat capacities of the
components. This can be explained by two examples as:
Example 2.n
A certain slag contains 40oA CaO, 35% SiOz and 25"/o Alz Os
lf the specific heats given are:
CaO - 0.01031 cal/gl0o/mole
Si0z - 0.01498 --do---
Alz 0s - 0.01906 --do----
then the weighted sum of these values shall be
= 0.01031 x 0.40 + 0.01498 x 0.35 + 0.01906 x 0.25

28
 = 0.0095 cayg/Oo /rnole for the slag under consideration'
Example 2.2
lf the sp.heat of copper is 0.096671 and that of silwr is
0.05582 cal/g/oo, what will
be the nalue of sp.heat of an alloy of 60% Cu and 4Oo/"
Ag?
The weighted sum shall be

= 0.096671 x 0.6 + 0'05582 x 0'4

= O.O77322 caVglOo

2.10.2 Varlatlon of Enthalpy with Temperaturc

The heat content or enthalpy required to heat a substance over a certain tenperature
of direct relevance in chen$cal or
range or that which would be etolved on cooling is
metallurgical processes. The basic equation of enthalpy
as it varies with temperature is
Jz
(A H)P = Jr, CP dT

ArelationshipasshowninFig2.2canbedrawnfor^Hasitvarieswith
temperature. The change in heating the material from Tt to T2 temperatures
"ninapy and Tz ' A varieties of such
shall be equal to the area under the cune between Tt
depending upon their nature'
relationships would be observed for different substances

AH = Hz - H1 = [';, (a + b'T + c'T-z) dT

=(rz-rr)a+(rz-rr) brz+ c (+ +)
(2s4
related to cp the
Knowing the values of Tr and Tz and a, b, and c as constants
like absolute enthalpy a relative
enthalpy cnange can be calculated. since there is nothing
H298o k as the base levelfor
value is only calculated. lt is therefore custornery to assign
as
comparison and actualvalues of enthalpy changes
Hzs'ok- Hzssok (2.58)
tabular form for ready
are estimated from the specific heat data and are compiled in
in the form of HzsEok data'
reference for various pure elements and compounds
Nornally reactions are carried out with the reactants being
at room temperature
temperature and then the
to start with. The reactants may be heated to sorne elevated
reaction occurs to produce the products'
ForareactionsaY A+ B - AB .-,-
the value of A Hzssox values of
the enthalpy change at eggo K can be evaluated knowing
reactants and Products as
A Hzsao - A Hzeeo (AB) - A Hzsso G) - A Hzgeo
(D (2'59)

tnageneratbedwaythevahrotAHzgsoK
canbe evaluated from-theknowledgeofAH2g8ok and
values of specific heats of reactants and products
inrolwd in the reaction knowing the
Kirdhofi's larl as explained in the lolloling section'

29
I Fis.22'
Schematic v ariation of e nthalP
o-
of soliy' lrquid, and gas Phases

F for different materials'

zUJ

al'
-1

tt TEMPERATURE
T7

-
Example z 2.3 Formetallothermicreductionof solid oides to
be afeasiblepropo-
sition the enthalpy change required is around 600 cal/g
weight of the system'
Establishwhichoneofthefollowingreactionisaviableproposition?
1. SPbO + 2Al = Al2 03 + 3Pb
. 2. 2PbO + Si = SiO2 + 2Pb
3. Fez Os + 2Al = 2Fe + Alz Os
4. 3ZnO + 2N = 37rr + Alz Os
5. 2MnO + Si = SiOz + 2Mn
Given for the solid reactants and products :

2PbO + 02 = 2PbO AHo = - 104500 cal/g/mole

4/3Al + Oz = 2l3Alz Oe AHo ' = - 256600 --do-

4l3Fe + 02 = 2l3Fez Os AHo = - 196000 --do-
27:rr + Q = 2ZnO AHo =-83000 --do-
2Mn + oz= 2MnO AHo =-92000 -do--
Si + Oz = SiO2 AHo = - 209000 --do-
therefore
2PbO+ 4l3Al = 2lgAlzOs + 2Pb

from Hess's law

AHo = -152100
30
 AHo= -152100
3PbO+ 2Al= AlzOs+ 3Pb
AHo = - 228150
sp.heat/gofsystem= - ry= -315cal
where 723 isthe sum total of gram atoms of products or reactants'
-iherefore even if pure oxide was reduced it would not be possible to carry this
reduction. tn fact in practice the oxide will be impure, that is additional
dead weight will
be there, and the specific heat will still further decrease
and rnake it non-practical'
lnthe same way the others can be calculated andassqssedtofindwhichoneis
a feasible reaction. lt will be observed that the reduction of iron oxide by aluminium will be
a \Ery {avourable one and that is the thermit process of welding.
Example : 2.4
iron is oxidised by
Calculate the temperature rise when 0.01% P dissolved in molten
process to make liquid calcium
supply of oxygen and solid lime at 27o C during a steelmaking
phosphate slag at 17500 K by a reaction :
4[P]tzsm + 5{02}gmo+ 6 <CaO>3ooo= 2(3Ca0'P2Os)tzsoo
Given:
1. {Pz} = zLPi AHr = - 58500 cal
2. alslPzl + {Oz} = 215 <PzOs> AHz = - 15160 cal
3. {Pz 05 } = < Pz Os> AHs = - 3600 cal
4.6/5<CaO> + 2l5{-PzOs) = 215<CaO'PzOs> AHe = -405300ca1
5. mean sp.heat of olrygen gas, solid cao and molten iron are 0'255, 0'1 and 0'1
caildegree/g resPectivelY'
6 Heat of fusion of phosphate = + 4000 caU mole
Solution:TheprocessistobecarriedoutasshowninFig.2.3.
lary as follows to obtain
The final reaction can be built up in stages using Hess's
the overallenthalPY change :

2Fl = 1ez) AHr = + 58500 cals

4lillPl = 215 {PzI AHr = } 58500 x 215 = 234o,0

215{PzI + {Oz}= 215<PzOs> AHz= -15160ca1s

45[P] + {Oz}300o = 215< Pz Os> AH = AHr + AHz

= 8340 cals
add to this a H of the reaction 4 and add solid cao to it to give
a/s[P] + {Oz} + 215 <CaO> = 2/5 < CaO'Pz Os>
AH = 8340-405300 x 215

= - 153380 cals

31
 g{E} * {QIoc*= t(80') t(Co0}*t
lu
tI6
fl;
tfli la
l2
It
-
(t
3
F

J"ll
rlG
{ll
'['
ili +
U
t
*'
$tPl {Q}.ro'* tt Eq) * t(Coo)oo*= € (gCaO'E$r)
+AHt
&6C0.0'EG)

Fig. 23 - The physbal mode of the reactinn of oxidatbn of phosphorus ftom

molten iron to give molten phosphate.

Algebrically account for heating of reactants o4ygen gas and solid lime to 17500 K
and for nrelting of calcium phosphate.
These are : {Oz} smo = Cp x dT x mass
= {Oz}rzso AH
= 0.255 x 1450 x 32
= 11840 cal
6/5<CaO>smo = 6/5<CaO>l7sooo AH= 0.1 x 1450x 56x 6/5
= + 9744 cals
and ?5<CaO.PzOs> = 215 (3CaO.PzOs) AH = 21000 x Us
, = 1600 cals
All these will have to be met from the exothermic heat to obtain the liquid product phosphate.
The net enthalpy change is therefore
= -153380 + 11840 + 9744 + 1600

= - 130636 ca|s
This enthalpy is erolved when (4/5) x 31 = 24.8 g of P is oxidised in mohen iron
togetequivalentmolten phosphate. Whenl00gof meltistakenand0.0l gof Pis oxidised
adiabatically the temperature rise shall be
AH 1306
Cp' Q.l x 1@x 4hx
Vl - rn^ AL ., 4a
mass 31

= 5.2o C
2.10.3 Enthalpy Change during Phase Change
When phase changes like solid to liquid or liquid to gas or vice versa or order-
disorder changes or solid state phase changes like austenite to ferrite andlor cementite
occur, energy exchange is a must and hence the enthalpy changes in such processes are
given by the corresponding latent heats of such processes.

32
 During heating of a materiat (or eren cooling in rewrse) from say Tr to Tz
temperature if a phase change occurs in between then the enthalpy change will be given
by
Process enthalpy change AHoir' _ t ) = 4:' ce dr(phase 1) + +
^Hf
- f;, cpdr(phase2) (2.60)
where A Hlo is the enthalpy of the phase change taking place at temperature Tf which is in
between the temperatures Tl and Te . This equation shall get extended if more then one
such phase changes occur in the over all process by the addition of similar terms in
between the temperatures Tr and Tz . For example if manganese is heated fiom b00o
C to say 12500 C, the overall enthalpy change shall include the enthalpy changes
accompanying the phase changes for the intermediate phase changes such as Mn (cr) to
Mn (F) at7zoo C, Mn(p) to Mn (y) at 11000 c and Mn (y) to Mn (6) at 11360 c and
therefore the enthalpy equation shall take form as belcnru ;
aH = I Cp(Mnc,) dT+ AH(.r-F) +JCe(MnB) dT+ AH(p-y) +
u3-993o K 9930 K 993 -13230 K 13zgo K
+/Ce(Mnr)dT + AH(y-6) + f Ce(Mno) dT (2.61)
1373-14090 K 1409.oK 1409-15230 K
This evaluation of enthalpy as shown mathematically abore can also be depicted
schenratically as shown in Fig.2.4. lt would be clear from both that this evaluation of
enthalpy necassarily requires the knowledge of Cp values of all such substances
(crystalforms) inwlred in the overall process. lt is also clearthat if enthalpy is known at sorne
temperature then iis value at some higher temperature can be found out as

AHrr=Hrl+ cpdr
t, (2.62)
and with appropriate changes in sign in this equation it can also be determined at a lqryer
temperature as well. lt means knowing the value at some temperature on the graph in Flg.2.2
the remaining can be added to or subtracted from it.
2.11 KTRCHOFF,S LAW

This is the law which helps to evaluate enthalpy of a reaction at a temperature

when il is known at some other temperature. The variation of enthalpies of reactants
and products with temperature for a certain reaction are shown in Fig. 2.5. lnitially the
reactants are at state A at temperature T; (initial) and the final products are at state F at
tenperature n (final). The reaction can be carried out by the route ABCDF or ABEDF
wherein the reaction occurs qt temperature Tr and Tz respectively. ln either case the
enthalpy change is the sanre.
The total enthalpy change can be evaluated by knowing the enthalpy of heating from
AtoB,thenenthalpyof reaction attemperature Tr and the enthalpyof heating of products
to temperature Tf . This requires the knowledge of Ce values for reactants and products
and also enthalpy change --f the reaction at Tr . Alternatively the sarne process can be
carried out by heating the reactants to temperature Tz followed by reaction at temperature

33
 u--+ Dhase change ^S+d Mn f+d Mn

\;:+= i;i;; --; il; : : = ------ -- ='n;7 t4 og"

I
Hr*H4'change f
ph"ose ts
P*:f, Mn
qt
H l____- -:-- ---7 rgrso

o-
J

F | --;l ^*{
--/o
i
I

t
I

f-*'-lr--
i
z I

I
t! I

I pnase-change
I

ac+F P1'" il i
I
I

I {l^^,. I

Ht
'l o-T, )*ffinn' i

I
I

t
i
I

J-
ly" lr
2AO 400 60 o 800 1000 1200 1400 1600
9K
1800
TEMER ATURE --+
changes on heatingrnfinganese aver 4
Fig.2.4 - Representation of enthalpy
changes in rnanganese'
rcmryrature range involving phase
for
This also needs the knowledge of ce values
Tz and then heating of the products to Tt Tz ' ln either
change of the reaction at temperature
reactants and products and ine enthalpy final states are the same
is the same s]nce the initial and the
case the overall enthalpy change is zero and therefore
process BCDEB the net enthirlpy change
in both the cases. For a cyclic
Jz * cp (react) dr
dn1 + J- CP(Prod)dT-
^
Hr, ftt
D-E E-B B-B
B-C O+ D

(2.63)
or A Hra - a Hr' = f:, I Gp (prod) - Cp (reast)ldT

as [Cp (prod)
'- Cp (react)1 = A Ce
if it is shown
(2.641
aHr2 _ orr,= 1], lOp dr

Knowingthevaluesofspecificheatsforreactantsandproductsandenthalpy
changeatsornetemperature[Ir)theenthalpychangeatsomeothertemperature(Tz)can
becalculatedfromtheaboveequation.rn""itentnapyvalueisknownateitherTlorTz
the change in the enthalpy
the above equation' Alteinatively
the other can be calculated from
and Tz can be calculated as
values between temperatures Tt
u
 E --t"'

I
I

5
a
F
zrrt

T at T
TEI(PERATURE
-- into accountJhe changes in
Fig.25 - Enthalp change during a reaction taking
enthalpics of the reactants and the prodrcts with tetnperature.

aHr2-AH1 = aAH=/' ACPdT

/q#)
o' = Acp (2.65)
I E
of constant pressure
The subscript p is to indicate that this is valid under conditions
as perthe original equation used in thederMation' ThisbthestatenrentojKirchoff'slaw'
Itmeans thatthe raieof changeof enthalpyof aprocessorareactionwlthtemperatur€
of producis and
ls given by the difference of the specific heats at constant Pressure
the reactants. The same c€rn be arrived at as

cp (prod) /aH\ = [#,), and cp (react) = (9*),

by subtracting
0H proa :-d Hre*t = cp(prod) cp(reaa)
AT
and since the left hand side is the A H of the reaction

(H), -rcp as shown in Eq. (2.65)

lf the reaction is carried out under adaibatic conditions

either by route ABGDF or ABEDF the
totalenthalpy change will be given by either

35
 + AH1 +
f" Cp dT
{ ,CedT
roactants reaction Producl*
E
Jo cPdr + aHr - { cpdr
reectants reaction Products
the
The temperature rise of the product can thus be calculated from these knowing
either ABCDE
mass of the reactants and the products. Basically the reaction process is
or ABEDF and this includes the heating and reaction steps. Given
the required details

of any such process the remaining unknown can be found out'

Example: 2.5
Calculate the enthalpy of vaporisation of Mn at 19000 K (sub- boiling
point temp)
and of fusion at 12000 K(sub-melting point) when these values are known at their normal
liquid and gaseous
boiling and melting points and when the specific heat values of solid,
Mn are known.
Solution :
The process has to be imagined to occur as shown in Fig'2'6' The overall
enthalpy change will be
200
J 580 -1

=* |
Jt2oo
Cp dT(solid) + AHr (150!0 K) + r1580
I Cp dT(liq) - AHf (12ooo K)

which is equalto zero for a cylic process'

Hence

AHr (12@o )K = Jlzm

-1560
ce dT(solid) + A Hr (15600 O -fi: Gp dT {liq)

(1200oK) can be
All the values on the right side are known as given and hence AHr
give the value of AHv (1900oR
calculated. Similar exercise carried out for vaporisation can
-r'
2.T{ ADIABATI C FI.AM E TE MPE RATU R E

vo.- lt is often advisable to know the maximum temperature of the flame

that can be
achieved theoretically when fuel is burned. This is possible to
be calculated assuming
conditions of burning. lt means that whatever heat is liberated
in the combustion
adiabatic
ofwhateverconstituents (C,H,etc.)thatarepresentshallbeavailabletoraisethetemperature
values for CO' COz ' HzO'
of the products that make the flame' lt also requires that the Cp
is the enthalpy of the overall
etc. be known for that temperature range. lf it is assumed that AH
process
Reactants at Tt .* burning "* Products at Tz
andif C + Oz = COz
and 2H + oz = HzO
from the composition of
are the combustion reactions then the products can be ascertained
the fuel and amount of air and or olygen supplied for its burning.
And
Jr -Cp (Hzo)
aH = f "
J2g8"K [Ce (coz)'nco, + 'nxro +
CP (oz).no, + CP (N2)'nNJ dT

36
 S ol Ldn/n ( rs Bo"x, #ffH0fo* L iluid
(
Mn

A I cl
rd l{
F tl- J F
fi c, o n
I' u o
U
+

(soooK) PHASC CHANqE

Soltd.'Mn _AH4 (raoo-n)
PROCess
S?ART
a,oil*.{l
HERE

Fig.2.6 - Imaginary scheme of carrying out the process to

obtqin the desired end prodrct

The excess oxygen term will appear only if excess oxygen or air is supplied and
similarly nitrogen term willappear only if
air is supplied for combustion and in proportion to
7g:21. The above will have to be integrated to obtain the unknown T as the adiabatic
llanne temperature.

2.13 CONCLUDIT.IG REMARKS

1. Energy cannot be created nor can it be destroyed, it changes only its form'
2, Energy can be interconverted.
g. Although the first law as such does not impose any limitation on such interconr,ersion
and,as is observed in practice, that mechanical work can be fully convrented to heat
energy, yet out of all conceivable changes, only some actually occur in practice in fully
converting the form, whereas many others do not do so.
4. There is a certain tendency for the conversion of energy to occur in sone prefer-
ential direction rather than for either direction to be equally possible. This limitation
shall be dealt with in the lollowing chapter.
This anormly, though not conceived in the first law of thermodynamics, can bei
explained only in terms of the second law of thermodynamics. This will be discussed in
the follo,ving chapter.

37
CHAPTER 3

ondLawOfThermod namlcs
entropy and its derivatives

3.1 INTRODUCTTON
unless acted upon
A thermodynamic system in equilibrium shall stay in equilibrium
equilibrium pasition' However if the
by external constraints to make it move away from the
system is away irom equilibriurn it willtend to achieve
the equilibrium state' systens' which
are away from equilibriurn, upon initiation, shall move
towards equilibrium and such processes
processes' Examples are
are referred to as 'natural' or'spontaneous' or'irreversible'
heat flow from a hotter to a polder body (in
'contact with) or depressurisation of an inflated
tube in a low pressure sunoundings or a free fall
under gravity and so on' The
state is possible because in the
spontaneous change lrom an existing state to an equilibrium
potential and which is the driving force
existing state the system happens !o be at higher
thermalpotential and which
forthe changeto occur. Highertemperature istherefore a higher
higherpressureisa higher
makesheatflowfrom higher to a lowertemperature. sirnilarly
mechanical Potentialand so on.
reverse direction in the
lf a system is in equilibrium and if it is to be moved in the
or non'sporilaneous processes' There-
above examples, it would be termed as 'unnatural'
in the form of mctor-pump
fore water cannot be raised to an overhead tank unless energy,
heatwillnotflowfrorna colder
setisprovidedtothesystem from thesurroundings. similarly
energy as
to a hotter body unless aided by the surroundings in the form of compressor processes
observed but in chemical
in a refrigerator. These are the examples that can be
it is not readily possible to assess as to what is a
naiural process or an unnaturalprocess'
state of the system'
This can only be evaluated fr6m the knowledge of equilibrium
The knowledge of equilibrium co/coz ratio in contact with Fe or FeO
can only
of iron oxide by providing
guide us as to how to prevent oxidation of iron or effect reduction
a suitableco/cozgas mixture as surroundings. Thefirst law ofthermodynamiasmerely
lt also
direction in which this will occur'
states the inter-convlrsion of heat and work but not the
potential to make a process take
does not give any indication of internal energy being a
of driving the process in
place. The second law is able to devise- a potential term capable
the brm of a force'
energy is fully
The lirst law is based on the essential observation that mechanical
it passible to carry out this process
converted to heat(from Joule's adiabatic experimen$. ls
 energy?' This was of
in the re\€rse direction i.e. heat to be converted fully to mechanical
in the early days and also to
direct relevance to understand the working of steam engines
the performance of steam
improre upon their efficiencies. These efforts to appreciate
engines in the 18th and 19th century ultimately led
to the formulation of second
lar of therrnodYnarnics.
study an ldeal gas engine producing work by taking heat
It would be worthwhile to
from a reservoir. Fig. 3.1 shows the P-V relationship
of such an engine' lt takes heat
A to state B doing mechanical
from the surrounding and expands isothernnlly from state
to
worxequatof pV.lnordertorun it as anengine this must be done repetitively
The systern therefore must return
produce work by taking heat from the surroundings'
back to A, its originalstatg.This is not po!9ible
ory 4 adiabatic comPression.

A
t
el
I'OTHgiFIAL
zo ,{ t
I AT TE&iP :Tt I
ld gE
E l. tt sl
f kc'
io. 0(
o
(r) 3n
ld 6l'
co- <ct lrl
0(
I3O'THERIfiAL o-
compn?gstoN

E- VOLUME+
(b)
(a)
related to a gas
Fig. 3.1 - Typbat P-V diagrams in cyclic processes
engine ; (a) shows a three stage cycle and
(b) shows a four stage cycle'

processes, one of which may be isotherrnal, are needed

to revert
At lea$ two reversible
in Fig'3'1 as processes BC and
the gas system to its original state A. These are shown
cune AB' The work
cA. Both of these will have to be on the lcnryer side of the isothermal
c*cF PV to leave out net
IJ6 PV anO rc f"^ PV together must be and is less than lo
nnechanicalwork done by the system in a cyclic
process"
the gas
ln such a ryclic process involving minimum three difierent processes
and expands to do work' on rerrerse side
takes heat at a temperature Tr from the surrounding
it has to isothermally compress first, desorbing
the equivalent heat to the surrounding' and
then to adiabatically it further to attain the original state A. This necessitates that
"o*pio" at-af?fperaturelower
the systemmust.gire out heat duringthe isothermal compression
work by expansion' The adiabatic comprassion
than at which it earlier accepted it for doing the
the original state A ls achieved' This whole
shall raise the temperature to Tt again when
39
process, for conversion of heat to work necessarily produces a permanent change in the
second surrounding reservoir by way of release of heat into it. This inevitable tnansfer of heat
to the surrounding at a lower temperature, which is a permanent change in the second
surrounding, to carry out the conversion of heat to work is called by sone as 'comlpensa-
tion'. Hence heat can not be converted to work without such compensation. dt also means
that the entire heat that is absorbed initially is not converted into usefulwork but a part of it
is rejected to the second reservoir as of necessi$. ln other words 'ln a cye8lc process it is
lust not possible to convert all heat into mechanical work, although 8t is possible to
conveil all work into heat'. lnanon-cyclicprocessitmaybepossibleto convert all heat
into work but in a cyclic process it is not so. The important congideration is therefore
that in a cyclic process all heat can not be converted into work fully.
lf heat is to flow from a relatirely colder to a hotter body, it is only possible by attaching
the colder body to a gas engine which will receive heat from the colder body by way of its
expansion and by adiabatic compression raise its ternperature to that higher than the hotter
body and thus desorb heat taken from colder body to a hotter body. The work done by he
gas engine as surrounding is the required compensation.
The second lary assuned the form in the early days in the wonds of Carnot as'it is
not possible to construct a cyclic devlce to convert heat into work fully on its own'.
The same thing was put by Kelvin as 'it is iust not possible to con\rert heat into work
ln a cyclie process without compensation'. A later version of second law by Clausius
states that,'in a cyclic process it is imposible to transfer heat frorn a colder to a
hotter body without producing permanent changes or cornpernsation in the surround-
tng'. Max Plank suggested that'it is irnpossible to construct a eyclic device extracting
heat from a reservoir and produce thermal effects elsewhere' (compensation).
The first larrrr states that a system is a energy reservoir and hence can accept energy
in any form from the surroundings and give it out in the san€ or other energy forms in the
same or different anpunt. ln the above example the system ( in the form of ideal gas engine
and piston) accepts heat fiom surroundings but does not convert it all to mechanicalwork.
A part of this ene!'gy is converted to work and a part is parted off to another system at a lower
temperature. ln other words in a closed cycle operation converting heat into work a part
of the heat input has to be rejected to a colder reservoir. The efficiency of this engine is
given by net work done on to the sunounding
heat supplied to the system

3.2 CARNOT CYCLE

ln order to work out the efficiency of a steam engine Carnot in 1824 imagined a
cyclic operation consisting of two isothermal and two adiabatic, each with a expansion
and a compression, thus totalling four stages in a certain way. This is.shown in Fig.3.2 and
is known as'Carnot Cycle'. The operation of such an idealised gas engine can be imagined
to start from any point. Let us say it starts from A. The engine with ideal gas as the working
substance is initially at A at pressure Pl , volume Vr and temprature T1 conditions. lt is
put in contact with athermostat at temperatureTt andisallorvedto expand, takingheat
$ and doing work lfi on the surrounding. lt thus reaches B at corresponding state properties

N
 A (PrvrTr)
I9OTHERMAL EXNilSION
llr r -ot
B (itrr)
I .AOIA6ATsE
Fig.3.2 -
rc ADIAlATIC
E ErsNaroN Typical
)U, -arr1 Carnot cycle
an Q,t o operation.
br
0t
o.
(qy+rr)O
+ol I C(tYr
ISOTHSRMAL G€f{PtcSlror{

voLUME ._F
,Vz and Tr remains the same, at the end of the isothermal process of expansion.
as Pz
Now the thermostat is removed and it is insulated and allowed to expand further
adiabatically to conditions Ps , Vs and Tz doing further work W on the surounding and
thereby reaching position C. Now Tz has to be less than Tr because of loss of adiabatic
work from thesystemtothesurrounding. lt is now again put incontactwith athermostat
at temperature Tz and isothermally compressed putting in work vra to achieve conditions
Pc ,Yq and Tz at position D. ln this it gives out heat q to the reservoir because of work
r,16 being put into it from the surrounding for compressing it from position C to D. The
thermastat is now renpved at D and system is insulated and further compressed putting
we into the system from the surrounding which would raise the temperature from Tz to
Tr and pressure and volume to the original Pr and Vr and therby completing the cycle.
On the whole the situation is as follows :
The gas engine takes heat qr at Tr temperature and does work wt and rejects heat
A at temperature Tz while receiving work vE . The two adiabatic processes, one of
expan6ion and the other of compression nulli! each other as far as the net change is
concemed. Also since it is closed cycle the net work done and the net heat absorbed are
equal and hence net energy exchange is zero.
Therefore the total work done can be equated to the total heat exchanged as
folkxvs:
lsothermal work done during Vr to Vz and during Ve to V+ equals the heat
exchanged during the same period as q and gz . Hence as per Eq.(2.33) in Section (2.8.1)
the two isothermal work terms can be written as
RTr ln(Vzffr) + RTz ln(V+rus qz) = q1 + (3.1)
ln order to obtain maximum work all operations are carried out rewrsibly.

41
 From first law putting g = 0 for an adiabatic process
gives dU = -g/ = -PdV
and since dU = CvdT and CvdT = - PdV
for an idealgas Cv dT =-
(RTM dV
and rearranging the terns it becornes
dV Cy dT
v=- R'.7
integrating between the limits Tr and Te when Vz expands to Vs

ln (Vz A/s ) =
_RCv . ln (Tz /Tr ) (3.2)

Cv
Vz = /l-l Tz\
-E- V+
Vs \ Tt/ Vn
Vs
hence on reananging Y
Vr = Va
(3.3)

Hence Eq. (3.1) reduces to

RTr ln (Vz l/e) RTz ln (Vz l/r )
- = q1 + A (3,4)
or itmeans R (Tr - Tz ) ln (Vz A/r q1 +) = clz

and since
RTr ln(Vz lr'r ) = gt or Qt lT'r = R ln (Vz l/r )
therefore (Tr - Tz) (qr/Tr) = q1 + q2 (3.5)

or
Tr-Tz_q1-q2 (3.6)
Ir ql
that is 1 -[fafrr) = 1+ (qtlql
or V*!-o
1z rr
(3.7)

lnotherwords 0 ry =o as all the processes

in Carnot cycle are carried out rewrsibly.
The cyclic summation of qnvft being zero holds true for any substance in the
Carnot cycle engine and also for its any state like solid; liquid or gas as there is no
condition-precedent like this in arriving at the above relationship. Hence the property
ch.vft is a fundamental property and being zero in a cyclic process, it must be a state
function and not a path function. The grev , othenrise refened to as path function, denotes
here only the actual anpunt of heat exchanged and the temperature at which it is
exchanged. As was observed earlier while deducing the relationshigs lor Cv and dU and,
Cp and dH that,the term qldT is also equi\ralent to dU under constant volume conditions
and to dH under constant pressure condition and thereby become state functions, so also
here, the term grevlT is a state function and is designated by the term 'entropy' and often
denoted by the letter'S' and hence,
r$dS= 0 or dS= drev) or AS _ qqev)
-dT (3.8)
T
42
 efficiency of the camot engine
The equation (3.6) in fact rneans the thernrodynamic
in the instant case. The total work done
by the engine operating as rnentioned above is
cn + q2 and is divided by the heat input
into the system as q1 ' This fractor is nothing but
on the two temperature limits between which
efficienry of the engine. lt oepenos solely
efficiency is therefore independent of
the engine operates. The thermodynamic engine
the working substance and depends only on the two temperatures' since Ta cannot be
zerocentpercentconversionofheatintoworkisjustl:'tnoss|ble.lnotherwordsallheat
heat
larr is-thus proved'Theconversionof
cannot befullyconvertedintowork. Thesecond
at lwertemperature
into work isaccompanied by the pernanentchangeinthereservoir
whereinapartoftheheatsuppliedbythemainhightemperatureresenoir,isdumped,
and which is the required compensation'

3.3 ENTROPY CHANGE IN NATURAL PROCESSES

Allnaturalprocessesorirrewrsibleprocessesorthosewhicharearuayfrom
equilibrium, when initiated they all result
in increase in entropy as will be obvious from
the folorirring examPles :
reservoir to a lcrt,trer
Let us imagine that heat flovrrs froma highertemperatureTl
temperature,o"*oi,Tzthroughaconductorundersteadystateconditions.The
the reservoirs as surround-
in this instant case and
conductor be considered as the system
ing. under steady state conditions there
will be temperature gradient Tr to Tz across the
There is no change in the conductor as
a system during the transfer of heat
conductor.
fromTtreservoirtoTzreservoir'TheconductorreceivesdlrentropyfromreserroiratTr
-fz
Tz ' Since Tt ' >
and gives out entropy qflz to reseryoir at

+ '+ i,-r,...^r heat is

and gives out more' since the flor of
and hence the system receives less entropy entropy'
a natural process creates or generates
a natural process the conclusion is that
a hotter body is not possible as per
The re\€rse process of heat flovrr from a colder to just like that
therefore entropy can not be decreased
second laal of thermodynamics and it
of an inflated tube can be cited' on puncturing'
without compensation. Another example gets
gets deflated. The net change is increase
of entropy' Also the kinetic energy of a body
generated' lt means
decreased, in frl"""" by frictionalforces' wherein heat is
" ""iur.r of iceand
weight falls undergravitysayinamixture
increaseof entropy. similarty if sorne fall and the impact' lt
potentilt energy released by the
water some ice is melted by the
meansentropyoficeandwaterhasincreased'Thesameargumentcanthusbeputfor
allnaturalprocessesthatentropyincreaseswhennaturalprocessesoccur.
ltisdifficulttoobviouslyproveitinnaturalchemicalprocessesbutitisso.
conditions' lf the cycle is carried out
The carnot cycle operates under reversible
the work done in each
under identical state positions but, under irreversible conditions then
stageshallbelessthanthatdoneunderreversibleconditions'ltmeansthatthework and
exchanged but less than heat exchanged
done shall not be equal to the heat terms in
invotveo in a carnot cycle the two work
hence for a irneversible natural process
Eq.(3.1)'
43
 Rr, rn (#) . Rrz rn (":) < (qr + qz ) (3.e)

On processing these as was done earlier while deriving the Camot rycle this inequality
therefore flnally leads to

o>i(+eg > o (3.10)

for Carnot cycle involving irreversible processes. For any isolated system therefore
AS = or > 0
depending upon whether the process is reversible or irreversible (natural).
This equation is the expression of universal la,rv of increase of entropy which
states the entropy of a system in isolation can never decrease'. lt means the isolated
system has to go towards npre randomness or chaos.
lf the argument is carried further and if the process is to be made to go like melting
of ice or lifting of water into a overhead tank or reduction of oxides of baser elements and
so on, then the entropy change is less than zero or dS < 0. ln all the above discussion
the entropy change means total entropy of the system and the surroundings put together.
These processes are all unnatural and hence it can be deducted that the entropy decreases
in unnatural processes.
For any given conditions of an isolated system there is only one and one true
equilibrium state of the system and therefore the entropy law can be stated as that 'the
conditions of a system in isolation can not be varied in such a way as to decrease
the entropy'.
lf a solid is melted into a liquid by supply of heat at the same temperature in the form
of latent heat of melting then where is this energy ? The temperature, pressure are the sane
and the volurne is practically the same. For sure the minor change in volume accompanying
the melting process can not account for the huge amount of latent heat supplied as work
done. The internal energy is changed because of supply of heat and one may say that
dU = g (latentheat)
-but such an equation hardly conveys anything beyond theappreciation of firstlaw of
thermodynamics. The Carnot cycle states that the latent heat has resulted in increase in
the entropy of the system equivalent to
Hfo
Latent heat of fusion
Melting temperature = MP =+ = ds (3.11)

Physically the rigid atomic arrangernent of solid has, on melting, become more flexible with
more mobility for atoms/molecules, etc. ls entropy therefore a measure of this mobility ?
Thesaneargumentholds trueforliquidtovapour.transition atthe same tempera-
ture. Now the volume change and consequent mechanical work are not equivalent to
latent heat of vaporisation. This is again a case of increase of entropy as
Latent heat of vaporisation _ Hv _ g _ de
boiling point - B.P. - T (3.12)

The randomness of atoms/nplecules in vapour is far more than that in liquid. Therefore is

4
entropy a measure of randomness again ? The same argument can be put forth for mere
heating as well.
Surely entropy is therefore a measure of randomness of atorns or molecules in
the system. ln solid state, phase changes like crystal changes or order-disorder transfor-
mations and thelike occur, then the entropy change as a measure of randorness is more
apparent.
Since heat suply or heat evolution is an inelienable part of chemical and metallur-
gical processes, the associated entropy change in the sys-tem, €us a consequence, is
additional and more important parameter for study of energy changes during such
processes.
This can best be illustrated fromthe example of order- disorder change. The two
structures one as ordered and. the other disordered are shown in Fig. 3.3. These.two
arrangernents must have two different energies associated with them since the pres-

ORDER€D DISORDERED

Fig.33 - The ordered and the disordered arrangements in a-solid crys*I lattice.
The dnrk and the hollow circles represent two different types of atoms-

sure, volume and the temperature are all the sarne in both the cases. ln changing from one
form to another sorne heat say g will have to be exchanged at say a temperature f K. This
is based on the basic fundamental that two different systems with similar chemistry must
have two difierent energies associated with them. lf the associated energies were
exactly the same thenthereisnoreasonforthemtobetwo difierent systems becauseP,V
and T are all alike. The two therefore differ in only one form and that is the configurational
arrangerpnt of the atoms in them. This energy difference can only be described in terms
of its entropy alone which is different for different configurations and is described as qf[ i.e.
dS, amount of energy.

3.4 COMBINED STATEMENT OF THE FIRSTAND THE SECOND tAW

lf a system is capable of doing only mechanicdl work the first lanv equation can be
put as given earlier as Eq.(2.6) as dU=dg-PdV
and although it is not exact differential equation it is all that first law can do. This equation
holds true for a reversible process. The second law for a reversible process gives

45
 dS=df or g=TdS
The two lavrrs therefore can lead to
= TdS - PdV or 169 = dU
6t-,l
+ PdV (3'13)
lawsof thermodynamics'
This isthecombinedstatementof thelirstandthesecond
Thiscombinedstatementincludesonlythoseterrrswhicharestatefunctionsandhencejs
and the entropy' This relation is
exact differential equation in terms of the internal energy
therefore independent of path and depends
on state properties only'
For irreversible processes second lar gives
ds>0 (3.14)
or therefore TdS>dU+ PdV
On combining the Eqs (3.13) and (3'14)
gives ri.1')
TdS > dU + Pdv \\
used forfudher deductions'Areversible
This is themqstusefulformandismostextensively
hencethe equalitysigninthe above equation
process isveryvery close toequilibrium and
sign in one djrection indicates spontane-
indicates equilibrium relationship. The inequality
.ous or natural process as shown abow. By the same logic if
(3'16)
TdS < dU + PdV
indicatesunnaturalprocessandenergywillhavetobeinvestedtomakeitgo.
3.5VARIATIoNoFENTRoPYWITHTEMPERATURE(CoNSTANTvoLUME)
Thisdependenryofentropyontemperaturecanbeevaluatedunderconstant
volume conditions as follols :
Second lal gives dS = EdI
putting vlaue of dqfrom first lar equation
ds = # - i'ou
(2.20) and using
putting value of dU in terms of partial derivatives as given in Eq'
dU=CpdT
(3.17)
leadsto os=Sar+
['. (#),
1
+
at constant volume dV = 0,
Cv
it reduces to (dS)v = T
dT=CvdlnT
Jz (3.18)
and a finite change (AS) = Jr, . CvdlnT

or Sr, = Sr, * Jrt cv d ln T (3'19)

Tr ' As a corrollary it can be deduced that

:
where s1 is entropy at some reference temperature
do withheat exchangedbythesystem'
By definitionitself entropyhassomethingto
change has to be zero'
Hence il aq = 0 then the entropy
AtconstantrplumeconditionswhendV=0thefirstlaralequation
 dU = Ag - PdV reduces to dU= dg= TdS
or /du\
=t (3.20)
[fitju
or if dq is zero, there is no entropy change in the system then
dU=-PdV
or /du\
l,*,l.=-P (3.21)

ln other words there can be conditions of constant entropy or so to say adiabatic

condltions when the lnternal energy change ls equivalent to mechanlcal work
exchange.

3.6 VARIATION OF ENTROPY WITH TEMPERATURE (CONSTANT PRESSURE)

Now (ds)p = aqT by definition
putting dg = dU + pdV = d(U + pV) = 611

(ds)p= (+)"= *(#; o'= ?o'

or for a finite change (AS)p = -f:lr Cp d tn T
or at sorne ternperature T2
Jz
Sr.= Sr,+ l- CPdlnT
'lr

sr2=sr'-4, ?.ot (3.221

Exarnple 3.1 : Calculate the entropy change when supercooled aluminium freezes at B00o K
alongwith that in surroundings at the same temperature.
Solutlon : The process can be imagined to occur as shown in chapter 2 as Fig. 2.6 The
solution of this lequires the sp.heats of solid and liquid aluminium and Hr value for
aluminium.The AH values are thus known and from this the entropy changes at various stages
can be evaluated and then algebrically summated to get the entropy change in the system
and the surroundings separately.

s.7 VARTATION OF ENTROPYW|TH VOLUME (TSOTHERMAL)

The mathematical proof for evaluating the variation of entropy with volume under
isothermal conditions is considered unnecessary at this stage and hence only the finalform
is given belor ;
lf the therrnal volume coefficient of expansion is denoted as

o= g (,+), (3.23)

and similady the compressibility as

p=-1r rry) (3.24',)

Iar/"
then dS = (3.25)
ffou
47
 As= (3.26)
t{ dv

alternatively Svz = Sv,. * JVr

^Vz
[, dV P.zn
the solution of this equation requires the knowledge of entropy value at sorne known
volume as itwill not be zero

3.8 VARIATION OF ENTROPY WITH PRESSURE

It can also be shown that
ds=-(#),0' (3,28)

or dS=-crVdP (3.2e)

3.9 THERMODYNAMIC EOUATION OF STATE

It is possible to show that
P=r(,+)"-(#), (3.30)

and also v - '(#'),. (#), (3.31)

which correlate P,V and T and hence are referred to as additional thermodynamic equations
of state.
The mathematical derivations are omitted here as these are considered unneces-
sary at the introductory level of the subiect.
3.10 INTER RELATION OF CP AND Cv

It can be shown that Cp -- = (3.32)

"u asf described
in the preVous section.
where cr and p have the usual meaning
This relation is useful particularly because cr and p are often known and it is rather
easy to experimentally determine the values of Cp and hence knowing these Cv can be
evaluated and likqivise the experimental data of cv and ce n|ay also be verified.
3.11 ENAOPY CHANGES WITH PHASE TRANSFORMATION
The evaluation of changes in entropy values associated with phase transforma-
tions requires the knowledge of heat exchanges associated with these processes like,
latent heat of fusion or vapourisation or sublimation, crystal transformation and so on.

Melting : AS=+=#
Vapourisation : AS=+=#
Hs
SolidStateGhanges : AS = temp of transformation

4
 ln generalised way the entropy change in heating solid iron from say 5O0o C
a
to 16000 C can Ue evaluated by a complex equation involving all possible changes from 50Oo
to 16000 C. lron undergoes transformation €xs ct to p at 10330 K, p to y at 11830 K, y to 6 at
16790 Kand mettingatlSl2o K.Theoverall equationgiving thetotal entropychangein
this process is as follcnrs :

4:' ?ot . nffij) - J:'; {ftc

Cp (F)
srazsoK = T
ur *
lz
* '1673
J''*
gry or * Al{1;Q r*
r 1ozg
J8O9
J.,.r. P**
. ffi; +-) dr (3.33)

Or in a generalised form thG total entropy of a substance at . a certain higher temPerature T

can be evaluated from a equation of the type :

As=Aszss"*fn
'J2ga 9+Odr*+*f
I lr ,T,
9#E)or* andsoon
I
(3.34)

till the temperature under consideration is reached.This is graphically shown in Fig. 3.4. in
a schematic manner and the solution of which needs the knowledge of entropy value

22.60

p.zz 8y;;1;
-bE r: I
I
3o I
I
= I I

ils *.gs 6J3 -- I'Y o'rai

-t slreg
I
t
I
fa
(-) tt."o1'j: o.'F o
lcrlo!5 r t
r i I
lr
t ll
I
I
rl
iro
o {. tt I
t
I
I
I
g Ir
F l1 I
z 55
-+ ----
al
It
I

t! lt
1

,l I
r1 I
ll I
6 ll
tl
t

I
o I
irosfi 1185 Ol
I L6 73., $Lf
298 50 0 1000
oK + 1500
fEMPERATUR-E
oC
on heating iron from 25 to above its melting
@ o29eoK
point. TheS is nken as 6-5 calldeglmole. Note the absence of lilrcar
relati.onship between temperature artd the entropy values although it has
been asswned to be so in plotting this relationship'

49
at 2980K(25o C). These values areavailable asASzssoK (liketheearlierASzs"inconpriled
form).
Example:3.2
Calculate the change in entropy when 2 Kg of solid silver at 4O0o C is added to
already molten 5 Kg of silrrer at 1100o C.Find out the entropy change in the process.
Given : CP (solid)= 5 cailoo /mole
Ce (liq) = 7'3 --do-
Ht (Ag) = 2690 cafmole at 9600 C (m.p.)
Solutlon :
The solution of this requires that first the final ternperature of the total mass of
seven Kg of silr,er is determined. For this the process of mixing of the two has to be imagined
as shown in Fig. 3.5.. lt means that first (AHr + AHz + AHg ) shall be released to the
silver comes down to 4000 C temperature. Then the

t s KG _A3(
rF g 960'c
-F 7
z J
u, o
o
()

Fig, 35 - Shows the imaginary route of carrying out the

thermodynamic process to abtain desired result.
mixing is carried out without any enthalpy change. The mass is now 7 K!. Whatever enthalpy
was earlier giren to the surrounding is now pumped back to the system. Since it is not certain
as to what extent the heating will take place, the heating is carried oui in stages. First the
mass is heated to the melting point 9600 C. lf the available enthalpy is not enough even to
heat upto this temperature, to whatewr temperature it is capable of heating shall be
the final temperature. But if the enthalpy is surplus then the remaining is used to melt the
silver. Depending upon the surplus the equivalent silverwill be melted while the other shall
be still sslid. ln that case the finaltemperature shall be the nnlting point 9600 C'

50
 But even after melting allthe mass of silver, enthalpy is still left
then it will be used to
these calculations require the
superheat silver and the final temperature can be calculated. All
of mefting and which are given'
specific heat values of solid and liquid silver and its latent heat
point then the amount
Once the finaltenperature is found out and il it is the melting
of solid left will be calculated. Then the entropy changes can be evaluated knowing
the

enthalpy changes in each stage. Their algebric sunmation

shall give the required value'

i.rz ENTRopy cHANGE Assocl,ATED wlrH PRocEssEs

For a reversible process the system receives heat from its surrounding and gairs
entropy and if it looses heat to the sunoundings it looses entropy, amounting to the
heat excfranged divided by the tenrperature. ln this process it mr'st be
noted that the
surrounding also takes p.* in the process in terrns of either
loss or gain of the sarne amount
obeyed' The net sum of the
of entropy as the case tnay be for the first lar has to be
therefore always zero 6 :
entropy cfiange of the system and the surrounding is
AS (system) + AS (surrounding) = 0
(3'35)

for a reversible Process.

lf the process is irreversible or natural then in the light of Eq.(3.10)
AS {sYstem) + AS (surrounding > 0 (3.36)

and conversely if the reversible process is carried out in a sYstem isolated from the
surroundings then
> (3'34
AS (sYstem) 0
is no heat exchange no entroPY change
Also if the process is at equilibrium then since there
will occur and hence
(3.38)
AS {sYstem) = 0
the entropy change is given
Now. if a system undegoes a change from Ato B then
AS= Sa-SR % an,
assuming certain restrictions of conditions'
lf sa sa then as 0 or it means tfiat the system A
= - B is at equilibrium and will not
=
undergo a change.
spontaneously or in a natural
lf se > sR then As > 0 and the procFs occurs
way from Ato B'
Attematively if sl > sg then As < and the process can not occur unless helped
0
by external energy source. All these above equalities or inequalities are for the total values
the second
conditions therefore
of entropies i;e. systqm pltts surroundings. Under certain judge the
a criterion to
lar of thernndynamics provides in the iorm of entropy change,
equilibrium or non-equilibrium position in a process'

3.13 ENTROPY CHANGES IN CHEMICAL PROCESSES

entropy being extensive
Like enthalpy change as predicted by the Hess s lar,
energy state function, snalt Jso be subiect to
Hess's larru' Therefore entropy change in
the change'
a gii"n pro"""s is the sarne irrespective of the path follorued in bringing about
i.e, in one step or in multiple steps'
EntropyofanysubstancecanbeevaluatedasfromEq.(3'22).as

51
 fz Cp'dT
I
^S-- JT., T
The change in entropy is often related to datum level of AS2eso values which
have been established for a large number of elements and compounds and hence are
available in tabular forms.

Hence Asr = Aszsao -

4, *.ot (3.40)

Therefore entropy change in a generalised reaction

A+B=C+D
shallbe
AS8go" = AS(C) zge" + AS(D) 2eeo - AS(A) 2e6" - AS(B) ass"
lf the reaction is carried out at any other elevated temperature then, as was done
while deriVng the Kirchofi's law for evaluation of enthalpy, the process can be imagined to
take place at 298 K and the products raised to reach reaction temperature T or, reactants
raised to temperature T and then reaction carried out. Therefore to the above if the entropy
of heating is taken into acount the entropy of reaction can be calculated.
Therefore ASrr=lsr,*.f'f ot (3.41)

where ACp has the same meaning as given in Kirchoff's law.

Example 3.3 : Calculate the entropy change for the reaction
FezOs+3C=ZFe+3CO
at 2980 K. Giren : The standard values of entropies at 29Bo K for various compounds are :

Fe O 21.4 cal/O /mole

C 49.5 --do-
Fe 32.63 --do-
CO 47.3 --do--
At 2980 K AS = 2Sre + 3Sco - 3Sc - SFe, os
= 65.26 + 141.9 - 148.5 - 21.4
= 37.26 cayOo /rnole

3.14 CONCLUDING REMARKS :

It is not possible to con\,ert all heat into work without bringing some permanent
change in the surroundings. Because of exchange of heat the system undergoes change in
entropy - a measure of randomnbss of the system. Natural procssses occur with increase in
entropy. The second law proMdes method of calculating only difierence in entropy of a
system. The entropy change is an indication of direction of the change as to wether it is
natural, unnatural or is at equilibrium. ls this the only parameterthat indicates the direction
of a change? The answer is giren in the next chapter.

52
 CHAPTER 4

Thermod namic Potentials

free enerry and equilibrium crieteria

4.1 INTRODUCTION
A system by itseff, in isolated state, or in contact with surrou rcings shall stay in
equilirbium unless acted upon it by some constraints, But if a systemtends to mcve, as a result
of being not in equilibrium or as a result of external constraints, there must be a driving force
making the system morrc from within itself or underthe applied constraints. This driving force
is referred to as thermodynamic potential driving the system to change to a new state. ls this
driving force or potentialthe same under all conditions of the system or itvaries from condition
to condition ol the system ? ln this context as heat flows from higher to a lower temperature
the 1ow is possible due to the higher thermal potential in the form of higher ternperature.
Similarly the higher pressure is a driving mechanical potential. Also higher electricalvoltage
is driving the electrons under the influence of electrical potential as voltage' These are
more
tangible examples and can be understood more readily. However the question remains as to
to
what is that potential, forcing a thermodynamic process' wherein energy exchange occurs,
take place ? lt is easy to imagine that a system with higher associated energy will be relativeS
when
unstable as compared to the one wherein the associated energy is lower' ln otherwords
a natural change occurs the system moves from a higher to a lower energy level or mo\,es
from a higher to a lower thermodynamic potential. lt also means that energy is a
potential
potentialg
driving the system for change to occur. lt also means that for equilibrium to exists the
of all the systems or sub-systems within it must be at the same potential.

So far the internal energy, enthalpy and entropy have been eurlved as energy
paranreters but it is notyetknown as to whetherthese actsingly orin combination as potentials
in driving a particular Process ?
ln deriving various relationships from second law of thermodynamics entropy was
seen to suggest that for a natural process it is more than zero l.e. + ve and therefore for a
unnatural process it is less than zero i.e. - ve. Therefore for equilibrium to exist the entropy
shall have to be the same for all the systems and sub-systems. existing together. But it must
be noted here that for a natural change the system ntorrqs from a lower to a higher entropy
level and hence entropy can not be considered as driving force bringing about the change'
Can intemal energy and enthalpy quali! as potentials capable of driving the process towards
equilibrium, if already at higher levels ? lf the internal energy and the enthalpy are potential
terms then changes in them will have to be zero at equilibrium and these will har,e to decrease
for natural changes or irreversible processes and increase for unnatural processes' Let us
therefore see if criteria can be erolved mathematically to eraluate what constitutes as
thermodynamic potential or potentials and if so under what conditions ?
4.1.1 Potential Under Constant Volume and Constant Entropy Conditions
These are peculiar conditions in as far as eonstant entropy is concerned, since
constant volurne processes are well known. The constancy of entropy i.e. isentropic condition
isthatwheredS=0i.e.afi=0andsinceTisgenerallynotzero(excepttheextrernecase
of absolute zero state) therefore dg will have to be zero. ln other words the process is
possible under
adiabatic and no exchange of heat between the system and its surrounding is
of
the situation under consideration. The combined statement of the first and the second larrs
thermodynamics as stated earlier in Eq'(3'13) that
4g=Tds-Pdv
and putting dS = 0 and dV = 0 it reduces to
dU=0'
where dU = 0 for reversible or equilibrium process and dU < 0 for natural or
irreversible process and du > 0 for an unnatural process. This is based on the earlier
obseruation that dS > 0 for natural and dS < 0 lor unnatural processes'
The equilibrium for such a process under eonstant volume and constant entro'py
conditions shallbe defined as dV = 0, dS = 0, Q = 0 and dU = 0
a higher to a
and for a naturat procqss the internal energy will decrease i.e. will change from
loryer level and for aunnatural process just the reverse'
The internal energy thus fully qualifies to be called a potential under constant volurne
under the
and entropy conditions. lt also means that the equilibrium shall be characterised,
therefore
constant volume and entropy conditions, by highest possible entropy' As a eorrollary
expressed as
the intema! energy must be afunction of entropy and volune and cAn be
(4.1)
u = F (S,V)
which on partial difierentiation shall give
ou = (#)u ds + (#), ou
(4.2)

The intemal energy is also given by the combined statement of the first and
second
lar,rrsas
'on
dU = TdS - PdV
comparison of Eq. (4.2) with this leads to

(#)"=rand -P {4.3)
ffi,=
T
ln other words by specifying U as a function of S and V it is possible to evaluate
of any other two
and P and thereby describing the system fully. By exgressing u as a function
variable it will not be possible to describe all the variables, as is possible, under the
speclfication of U, as a function of .s and v. For exarnple by stating say
u = F (qv)
s and T of the system. The reader
it will not be possible to determine the rernaining properties
should now understand why such peculiar conditions were assumed
in the beginning in
deriving the internal energy as a potential term'
u
 Now that since internal encrgy as one of the forns of energies of a system can be
related to all the vadables of the systern, any change in the propertis of the system can be
directly conelated to the energy exchange and that is what is required.
From the combined staternent of the first and secondlaurs as dU = TdS - PdV it
Le readily inferred that in a reversible isochoric process change in intemalenergy is given
"an
by Jcu rJos= = oq 14.41

or by the reversible heat exchange. Also in areversibie adiabatic process i.e. when
ag = TdS = 0 or under isentropic conditions the change in internal energy is equal and opposite
of reversible work done on the system as

Jau=-eJov (4.5)

4.1.2 Potential Under Constant Pressure and Constant Entropy Conditions

Now therefore it should be clear that these conditions are so chosen such that the
complete systemis describable.
The conditions here are that dP = 0 and dS = 0'
Now H=U+ PV
therefore on dffierenciation
dH=dU+PdV+ VdP
by putting in first larrr and second law equations
dH=g+VdP
(4.6)
= TdS + VdP
Atequilibriurnwhen dP = 0 and dS = 0
dH=0
ln a naturalprocess since T dS > 0
dH > O for a natural Process
and since TdS < 0 for an unnatural process
dl-t > 0 for an unnatural Process
So therefore on the whole
dH >=< 0
for unnatural, equilibrium and natural processes as the case may be. Therefore the term
enthalpy qualifies as being calted a potential term under congtant pressure and entropy
conditions. The enthalpy can therefore be described mathematically as
H=F (P,S)
On partial differentiation the enthalpy change can be described as
g.n
dH = (#r)-o'. (,s),.'
Comparing this with already derir,ed equation (4.6)

(#u), =' and (#). =u (4,8)

Again by expressing H as a function of P and S for any system the remaining variables T and
V can be readily found out and thus the system gets completely described.
It is also clear from the equation (4.6) that for isobaric.processes_the change in
enthalpyofaprocessisnothingbutretrersibleheatexchangesince dH = TdS = dgandfor
adiabatic processes when g = TdS = 0 the change in enthalpy will be J VOe over the
pressure limits of initial and final state.
4.1.3 Potential Under Constant Volume and ConstantTemperature Conditions
These are useful conditions under which many practical processes are canied out in
the form of a fixed enclosure and a fixed temperature. These conditions mathematically are
dV=0 and dT=0
andas TdS = dU + PdV for a reversible isochoric process
therefore (dU - TdS) = 0
or d(U-TS) =0
Let us define a new mathematicalterm as
A=U_TS
then dA = 0 for a reversible Process,
dA < 0 foranaturalProcess
and dA > 0 for a unnatural Process'

This newly coined function A as defined by (U - TS) qualifies for being referred to
as a potential term under constant volume and constant temperature conditions. Being
a function of variables which are all state functions 'A must also be a state function.
Expressingasbefore A = F(T,V)
and by partial differentiation

(,+)" dr + (,t),*
oo = (4.e)

Now fi= U-TS

therefore dA = dU - TdS - SdT
and since TdS = dU + PdV
therefore dA = - SdT - PdV (4'10)
and comparing the two leads to
f4\ ---s and ($) =-p (4.11)
lur/u \av/.
As A varies with T and V and if this is known then S and P of the system can be
evaluated to describe the system fully.
This function Awas first defined by Helmholtz and therefore is known as Helmholtz's
Free Energy. From the definition of A it can be written after difierentiation as
64 = dU - TdS (at constant temperature)
and for a reversible process clrev = TdS from second laal
and therefore Qrev = dU - dA
According to the first law, dU = grev - wrev
and hence ftev = dU + wtev
comparing the two equations leads to
- dA - wrev
or dA=-wrev
Change in A during a process is given by
M = - Wrev'
 That the change in Helmholtz free energy in a procass is equal to the amount of
work
done by the system on to thesurroundings or is equal to the amount of work the system
is
work possible). The function A is therefore
capable of doing (being rewrsible it is the maximum
.work function'. lf the system is capable of doing mechanical work alone then
also known as
The loss in A
the loss in mechanicalwork is equal to the change in Helmholtz's free energy.
as a result of giren isothermal
is equal to the maximum work the system is capable of doing
change only'
change and that being maximum, will have to be achier,ed through reversible
From dA = - PdV - SdT as in Eq'(a'10)
it is clear that for isothermal processes
J on = - e.J ov g'12')

or the change in free energy is given by maximum work the system is

capable of doing on the
surroundings by way of rolume expansion. Also for isochoric processes
Jan=-s.Jar the system'
and free energy change will be in terms of temperature variation of

4.1.4 potential Under Constant Pressure and Constant Temperature Conditions

metallurgical
These are byfarthe mcst commonly adopted conditions in chemicaland
engineering Practices.
It means dP=0 and dT=0.
written as
The combined statement of tlre first and second lzurs of thermodynamias can be
d(TS) = dU + d(PV) (4.13)

or d(U + PV-TS) = 0
Now let us define a new term mathenratically as
(4.14)
G=(U+PV-TS)
and therefore obviously dG = 0 for reversible or process at equilibrium
and since dS > 0 for natural Process
U + PV-TS < O for natural process and hence
dG < 0 for natural Process'
Conwrsely and obviouslY
dG > 0for an unnatural Process
for being called as
The property or function G as defined above thus qualifies
potential term under constant pressure and constant temperature conditions'
The
hence G also is a state
iunction G is a variable of functions which are all state functions and
function G is referred to as 'Gibb's Free Energy'as distinct from the
Helmholtz
function. This
free energy.
Now G =U+PV-TS
or putting value of { =U+PV (4.15)
AG =H_TS
or for a finite change AG =AH-TAS
or dG =dH_TdS-SdT
and at constant Pressure dP =0
dH =9=TdS
and hence dG =-SdT

57
 (#),--s (4.16)

Again 66 = dH - TdS - SdT

on putting = dU + PdV
611
66 = dU + PdV + VdP - TdS - SdT
putting in this dU + PdV - TdS
dG=TdS+ VdP-TdS-SdT"
or =VdP- SdT {4.17)
at constant temPerature when dT = 0
dG = vdP
/dG)
or l*Jr
= V' t4.18)

since G is a state function any change in G can be represented as

6Q=AU-TAS+ A(PV)
- wrev + A(PV)
or for small changes
dG=-wrev* d(PV) (4.1e)

Therefore the change in Gibb's"free energy is equal to negative reversible work

for
obtainable from the thermodynamic process (as is Helmholtz free energy) but corrected
the
change in volume during the process i.e. corrected for mechanical work done during
process. ln the definition itself it is so as would be obvious as
A=U_TS
and G=U -TS + PV (thePVtermisthecorrectionfactoQ'
By adopting G as a function of T and P as
G=F (P,T) (4.20)

on partial differentiation
dG = (#),.r. (#), o,
and since dG = - S dT + V dP as in Eq' (4'1f
on comparison

f*'l
\drip
=-s and f#) =V
\"-/r
(4.211

Since all are exact differentials dG in place of dG makes no change and the same
applies for others.
two
Thus by knowing G as a. function of T and P it is possible to establish other
variables v2. S and V and obtainthe description of the system fully'
Example 4.1 : Calculate the free energy change of-a reaction
<Si> + {Oz} - <SiOz>
at a temperature 9000 K
Solution: AGo$oo=AHogmo-1ASosoo,

='[AHozsa - Jfl:
299
Ace drl - T[asozsa +
 9m Acq
* Jr* J: drl
(reaaans}
where AC = 2 CP Froduc{s) X Cp
directly given or indirectly
inforrnation is required either as
Therefore to solve this tne tottoaring
calculable. _r^ ^-roroducts in the range of 29go - g00o K.
(1) Cp values of individual reactants and
(2)AHoandASovalueslorallreactantsandproductsal2gs:..Kasstandardvalues'
be sotved and the XGosoo can
be obtained'
*qtratlon
llthese are available,n*it " above "tl
U, H, A and G and their Extensive Natul
4.1.5 Functions
Allthesearemathematicallydefinedfunctionsandareinterrelatedtooneanotherand
toP,VTandSas
H=U_PV
A=U_TS
G=U-TS+'PV=A+ PV
=U+PV-TS=H-TS
Alltheseareusefultoevaluateenergyexchangesinthernrodynamicprocesses'Since
energyofasystemcanberelatedtoth*uu,i'bl""ofthesystem'anychangeintheproperties
o{thesystemcandirectlybecorrelatedtotheenergyexchangeandthatiswhatisrequired'
ltshouldalsobenotedthatoutofalltheseenergyparametersonlyGissuchthatitcanbe
assessedintermsofallvariablessuchasP,V,T-andSatatime.ThefunctionAcanalsobe
similarlycorrelatedas-A=9-PV-TS,butitisnotexactdifferentialequation'The
functionsUandHarerelatabletoonlysomeoftrresefourvariables.Gisthereforebyfarthe
mostexhaustiveenergytermthatcanaccountforvariedvariationsinasystem.Nowonderit
in chemical and metallurgical
procbsses
is used extensirely tolsess the energy
"t"n"ng"u
whereinnotonlyP,VandTchangebutalsothechemistry.Allthesefourpropertiesare
process involving chernical and/or
physical
and are additive' no'
extensive state properties "ny the free energies ol the two
f*" energy is given uy the difference in
changes tnu cr,angJ
'; other after-th" ;;";;; has occ.urred. Therefore 6Q =
AH TAs -
states, one before state' so
"no,n"u""n shown to b" m"."rred with reference to a reference
sl.*
and since H and like H and s at the
asslsea in reference to such a reference state' Hence
G also has to ue
referencelevel - ..^ +^ao
AGo=AHo-TAS(
SimilarlytheHelmholtzlreeenergyAalsocanbgexpressedwithareferenceleve].
trom Systems
4.1.6 Availability of WorUEnergy
Howmuchenergyandinwhateverform,isextractablefromasystem?ltcannotbe in a well defined
but it is fixed once. to, il from a well defined system
any amount "no from the mean sea leve
surroundings. Let us say that water is *rJ ", " tvel :l ll
jneters

electric is equivalent of (Mr - Mz

andthehydro-electricgeneratorislocatedatMzmetersfromm.s.l.Theenergythatisavailable l

to be converteJ to ?:**t
in the form of potentiai energy
meters.FromthesamedamltgeneratorsarelocatedatMgwhichislessthanMzoratM.t
bvelwhichismorethanM2leveltheatlailablepotentialfromtheSarTleresenrciriseithel
59
increased (at Ms ) or decreased (if at Ma ). This is also obvious fiom the definition of energy
that it is not absolute but equivalent of effects it produces on the surroundings. lt is measured
as a difference and upto the ler,el to which it can be lowered. lt can not be lowered below that
of the surroundings.
Now let us irnagine a system at pressure Ps and temperature Ts in contact with
surroundings at P and T. Let us also assume that the system exchanges heat and work with
the surroundings. Let us assume that g amount of heat is given by the surroundings to the
system and
therefore q q = or < (TsAS)
Now since AU = g - v/ = or < (Ts AS - Ps AV) where work done is only
mechanical and on the surroundings only. By transposing the terms
AU-TsAS+PsAV=oI( 0 (4.22)
The Gibb's free energy has been defined as above on the left hand side and hence from the
above AG = or < 0
This is the energy freely available in the form of heat and mechanicalwork from thesystemorits
,avaitability' (disregarding energy changes in all other forms). lt may be surprising to see that
the energy available is a function of Ts and Ps and not P and T i.e. the functions of the
surrounding conditions and not of the system itself'
The inequality sign indicates non-reversible condition and the equality sign the
reversible condition when maximum work will be carried out. Hence taking the equality for the
change
AG= AU-T"AS+ PsAV (4.231
putting AU= q-w = TAS-PAV intheabove
AG = TAS - TsAS - PDV + PsAV
= (T - T") AS - (P - Ps) AV (4.24),
The term -TAS is the heat extracted from the system and can be gainfully converted to work
through a Carnot engine operating between T and Ts temperatures.The work done by the
Carnot engine would then be - (T - T=)AS. Similarly if the system is allcnrued to expand as
AVthe mechanicalwork done by the system shall be effectively (P - Ps ) AV . Therefore the
above equation (4.24) shows the free energy or the maximum work (energy in generalised
term) available from the sYstem.
lf p = ps and T = Ts then no net work is available as AG = 0. The system is then in
equilibrium with the surrounding and hence no exchange of energy is possible.
The larger are (P - Ps ) and (T - Ts ) gaps the larger will be the the energy available
from the system. ln chemical reactions when chemical bonds change, the energy exchange
is taken care of bY the term - TAS.
It is thus clear that maximum bnergy available from a system is not absolute but it is
with reference to the conditions of surroundings. ln a gir,en surroundings AG gives the
maximum exchange possible and hence very aptly named as free energy of the system'
ln the above example of hydroelectric generation the lewl of the dam is the surround-
ing, source of energy, and the level of the hydro-electric generator as the system.
lf the temperature and the pressure of the surroundings are fixed the availability of a
given system is fixed.

60
4.2 PARTAL PROPERTIES AND CHEMICAL POTENTIAL
It is easy to appreciate extensive properties like volume, internal energy, enthalpy,
entropy, free energy, etc. for pure substances in terms of unit mass or per mass mole (as molal
propefi). But when the substance is in dissolved form r.e. in an homogeneous multicomponent
system, what is its molal extensive property ? The answer is not simple. This can be understood
by taking an example of volume for the purpose of illustration. This is chosen because it can
almost be 'seen'and thus readily understood.\Mten in solution the individual component
occupies the total volurne of the solution as its ovrrn volume irrespective of its concentration in
solution. Can it then be said that the molalvolume of the component in solution is equalto the
solution volurne ? lf this is assumed then the molal volume is practically the same for all
concentrations of all the components in solution. This is absurd'
Similarly it is rather difficult to visualise the potentialterm driving a chemical reaction
between phases wherein more than one components are inrCIlred. The question as to which
way a component would move between the two phases under a given set of conditions needs
to be answered in the same way as answered above while deriving conditions for determining
the potential term. Under the characteristic conditions of constancy of S and V internal energy
acts as the potential driving the process.Under these constancy conditions does U become
the potential term driving the chemical component from one to the other phase? Same
argument can be made for H, A and G values also.The answer is not direct and is interlinked
with the molal internal energy of that component in the multicomponent system. The molal
property of a component in a multicomponent system thus needs to be properly understood.
The problem is solred indirectly interms of thermodynamic concepts by coining a new
term called 'padial quantity or property'. This is applicable to extensiw properties only.
Let us imagine the components, solute and solvent, initially in separate vessels as
shown in Fig 4.1(a). Each is one g. mole in mass. Let us imagine that the solute is transfened
tothe solwnt but somehow kept separate as shown in Fig.4.1.(b). ln the next stepthe partition
is renrcved and the two are allowed to get dissolved to form a homogeneous system - the
solution. lt is almost rare that the solution volume will be equal to the algebric sum of the
componentvolumes. ln reality the solution volume is either less or more than the algebric sum
of their molal component volumes as shown in Fig. a.1Q). lqw let us imagine that whatever

H Fig.4
The process of
.1 -
mixing of two
components to form a solu'
tion with some imaginary

H (ol
61
steps.
be the individual sum total of the atomic or molar volunres of the components in solution, these
are somehow isolated in the form of two layers, the algebric sum being equal to that of the
system tolume. Whaterrer is the individual component molalvolume, the molalvolume in the
system is either less or rrore than that value when in pure forrn. This change is due to the
interaction of the components when present together.The interaction originates from their
different chemical nature,electronic structure and so on'
Although the abore picture can not be construed in reality it can be evaluated by
mathematical formulations as follors :

While deriving the thermodynamic potentialterrs it has already been shown that
U = F (S,V); H = F(S,P);A = F(T'V) ; and G =. F(T'P)
These are valid for systems of fixed composition and when characteristic two of the
four variables P,VT and S are fixed.
For example at constant T and P the system attains equilibrium at a minirnum value
of G. But if composition is changed keeping T and P the same, being an erdensive property
the value of G gets changed and hence the system no longer remains in equilibrium. The sarne
.arguments can be made for U, H and Avalues as per above equations. Hence strictly speaking
if chemistry of the system is variable the above equations get modified as :

U'=.F(S,V,nr,112,h3,....r...ni) (4'25)

H' = F(S, P,nr,Il2,Il3,.. ! ! r..' ni) (4.26)

A'= F (T,Vnr,ll2,ll3'...!'..'ni ) $.2n

G' = F (T, P,nr, Il2, ll3'., r..... ni) (4.28)

The prime on these functions is to indicate the overall property of the solution as
distinct lrom that related to only one mole or atom. The folloning is illustrated for G only but
the sarne treatrnent is valid for other properties like U, H and A'
Differentiation of Eq. (4.28) gives:

dG' = (#)*", .. n
"""3,...
- /aG\ .dnr+ /aG\ .dnz +
lFjr*nz,n3,..... ni l*,,|rr,",,"3,...,. ni

- /aG\
|.fr,lrr",,n2,..
dn' (4.2s1
.. .n,-,
lf composition is fixed, except the first two terns on the right hand side all other vanish' This
equation can be simplified further as

dG' = - sdr + VdP - (fir0",,"".....n,-'' dn' (4.30)

*
ln this the last terms mean the rate of change of values of G' with changes in

concetration of one component at a time. How is it possible to change concentration of one

component in a system and yet not change that of the others in the system ?

62
 This difficulty is orrercorne by irnagining a system large enough such that by addition
of one mole of a component I would not materially alter the composition of the other
conponents.
The change AGr accompanying this one mole addition is thus the molal G value of
as
that component i in the rnulticomponent system. lt is expressed in the differentialform
/aG\ (4.31)
l,6,lr.",,""..... ni-1
as the molal property of component i in the system consisting of several other
conponents.
o{ten denoted by putting a bar
This is known as 'partial ProPefty of that component'. This is
on the symbolas for exampleG called as G-bar or bar-G. Therefore
(4'32)
PartialmolalinternalenergyU = f+)
\d^i]s,vn' nr,... .. ni-i
1aH\ (4.33)
Partial molalenthalPY H =
lfrJr,*n,,",,n3. .... ni-l

Partial rnolal Helmhol? A= (u*),,n",",, (4.34)

n-1
free energy

and the partial nrolalGibb's free energy is as shown in Eq' (a'31)

The total extensive property of a homogeneous system, comprised of more
than one
therby making
component, consists of partial parts contributed by each one of the component,
contribution of a component to the overall total
the total property of the system. This individual
for total mass of one
extensive property is called its 'partial property' and when expressed a
expresed
gram mole or atom in the system becomes its 'partial molal property', and when
is its 'partial molar property'.The
for a fraction of the total one gram mole or atom of solution
depends upon
partial propefi is ordinarily differentfrom its pure phase property . The variation
the interaction between dissimilar components in the system.
AtconstantTandPthefreeenergyofasolutionwouldbe:

dG'=
T
i l,#)
\""'/n1,n4n3.....ni -l
= )aGi
1
{4.35)

lf the
For a given composition with minimum G' the system shall be in equilibrium'
the factual values
concentration of one component is changed, obviously depending upon
given by the equation above' The system shall gain
either G' will decrease or increase as
or with another
equilibrium by allorting exchange of that component with the surroundings
ln other words the value
sub-system till again thl original equilibrium composition is achieved'
otei i. e. partiaimolarfree enprgy is acting as a potential in driving that chemical component
in a certain direction to achieve equilibrium. The partial molar free energy
Gi is thus the
potential in terms of chemical composition of the system. This is therefore designated as
ichemical potential' and expressed commonly by the letter p in the literature. By arguirq
similarty for U, H and A it can be stated as :

63
 /au\ _
JUL = - /!H\
l6jr,nn, n,,n,.. ....i -r lanijr,r"r,hz,ng.....hi -r
/aA\ /aG\
= l-l {4.36)
lfrJrnn,,n.,n.. . . .. ni -r \dn'/tRn,,nan3..... ni -r

or p = 1Ui) or F'i) or tA or (Gi) t $sn

However each term shall act as chemical potential term under the conditions of
constancy under which it acts as the thermodynamlc potential term. lt must therefore be
understood that U is a characteristic function of independent variables S and V and composi-
tion and likewise others as discussed earlier.These can be written as :
I

dU=TdS-PdV+ pi .dni (4.3e)

1

T
dH=TdS+VdP+ 2.. U.i .dni (4.3e)
1

dA=-SdT-PdV +) pi .dni (4.40)

1

dG=-SdT+VdP+ ) Fi .dni (4.4',t1

1

Now since from first law dU=g-w

from Eq. (2.12) = TdS-PdV-v/
therefore dU = TdS - PdV - [chemicalworkintheprocess(r+/)] (4.421

ComparisonofEqs(4.38)and(a'a2)girethatchemicalworkis=>
1

it means that migration of chemical species over the available coordinate when multiplied by
the driving force that is the chemical potential, pi in this case gives the chemical work done
in the process by the migration of the chemical species.
Partial molar energies as shown above are the driving chemical potentials. lf a two,
phase multicomponent system exists, then the p;ghasel) = p;(phasen) for equilibrium to exist
forcomponent r'. ln case these are not similarthen i will migrate from higherto a lorrer potential
side.
Like all other potential terins the chemical potentials of a component in different
phases in equilibrium or between system and its suirroundings or between sub-system of a
system will have to be equal for the component to stay in equilibrium. lf il is unequal in that
case the higher chemical potential shall drive the component from higher to a lower potential
zone.
Likewise complete chemicalequilibrium shall be possible only if chemical potentials
for all components are correspondingly the same in all zones of the system. Transfer of
chenrical species from a zone of a higher potential to a lorer potential shall be a natural
process and vice versa as a unnatural process.
Most of the chenical and metallurgical processes are carried out under constant
tenperature and pressure and therefore partial molar free energy 6i is commonly used as
chemical potential term to €Fsess chemical equilibria.
4.2.1 Maxwell's Relations
We have derived
dU = TdS - PdV as in Eq.(3.13)
dH = TdS + VdP as in Eq.(4.6)
dA = - SdT - PdV as in Eq.(a.10)
dG = - SdT + VdP as in Eq.(4.17)
All these are exact differential equations.
From the propefi of exact differentials the follaling relationships can be stated :

(#), = - (g)" (4.431

(,+). = - (,+), (4.44)

(*^), = - (#)" (4.45)

/€\ = _ rcy\ (4.461

\aP/, \ar/,
These are popularly known as 'Maxwell's relations and are usefulfor calculations of
thermodynamic pararneters. There are four variables viz. RVT and S and the variation of one
with the other is related to the variation of the third with the fourth variable under certain
conditions, in the serelationships. Thermodynamic Mnemonlc square is available to anive
atthese relationships by a thumb rule. These equations apply to any system which is
determined by two parameters and the rernaining two are determinable, even othenrise as
seen before.
Also from the above stated relationships for dU, dH, dA and dG it can be derived
directly for fixed chemical composition the folloling relationships:

r- (#)"=(#), (4.44

p=-(#).= - (#), (4.48)

V= (#), (#), (4.4e)

and s-- (,+)" /!G\

- |.rtJ' (4.so)

65
4.g GIBB'S HELMHOL]Z EOUATION
It is known that G=H-TS
putting values of S as in Eq'(4'16)
G=H.t(**),

dividinsbyl€ +-+ (#)r= +

e +[+ (#), I=+
or [%(9o1,
=-+ or ## = H

[a (^qp] = aH
(4.51)
or
La(yn JP

[a (^ArDl - ^u (4.52)
SimilarlY it can be derived that
La(1^) I

TheEqs.(4'51)and(4.52)areknownasGibb,sHelmholtzEquationslorfreeenergiesas
they vary with temPeratures'

4.4 FUGACITY
system doing only mechanicalwork
It has ben shown earlier that for a
66=VdP-SdT
and at constant temperature when
dT = 0
dG = vdP
For one mole ol an idealgas
V = RT/p (4'53)
therefore dG=nf'Oplp=RT'dlnP' component across
can be in equilibrium with only one
It can be imagined that a gas mixture this system
to that component only' Hence in
a diaphramwhich is permlable
(mixture side)
dG (pure component side) = dGi
p"tti"r pressure of that component on the mixture
or therfore dG = RT ;';p
*n"'" 'p i" tn"
side. r LL^ ^. pressures as
limits of G and the corresponding
This can be integrated'over known (4'54)
' Gz Gt = RT ln (Pz
- /Pr)
Pr will be one atmosphere
taken as agreed standard in which
and if conditions for G are to be
then ln Pz
(4'55)
G2 = Go + RT
will take the form as
For an ideal mixture of gases this (4'56)
Gi = GP + RT ln Pi

66
it can also be imagined that an ideal gas mixture can be in partial equilibrium with only one
component across a diaphram which is permeable to that component only. Hence in this
system
dG bue comporcntside) = dG (mirctrre side) g.5n
or therefore dGi = RT ]n pi where pi is the total pressure on pure side or p on the
mixture side. On integration it again leads to
G = GP + RT ln pi (4.58)

Strictly speaking this equation is applicable to ideal gases or their mixtures only. The
free energy-pressure relationship is linear in that case. But for the real gases as pressure rises
the integration of this equation does not remain linear and poses sonre difficulties. lewls, the
larnous thernrcdynamics scientist introduced a new term called 'fugaclty'and mathematically
defined it as dG; = RTdlnt (4.5e)
On integration RT d ln f = G + I where I is the integration constant and it is so
chosen that f=p €tS P+0
Mathematically the functions U, A and H can also be correlated in a similar way as:
Since 611 = TdS + VdP as in Eq. (4.6)
or at constantentropy = VdP
and since V = RT/P = RT d lnP = RT d ln f (4.60)
As dA = - SdT - PdV as in Eq.(a.10)
at constant temperature = - PdV
= - RT(dVM = RTd lnV
therefore n=l"t
JVi
dlnV (4.61)

As dU = TdS - VdP as in Eq.(3.13)

= - VdP under constant entropy
= - RT(dP/P) = - RTdlnP
= - RTd lnf (4.62)
The introduction of such a function ensures linear relationship between allthese extensive
properties, U, H, A and G and logarithm of fugacity in any state of any gas.
For an ideal gas, fugaclty is equal to pressure at all prassures. Similarly for gas mixture
the fugacity of each component equals partial pressure, as the total or also partial pressure
tends to be zero. The fugacity of a component in an ideal gas mixture is numerically equal to
the partial pressure of that component.
For an ideal or real gas component the pressure is mean of the fugacity and the
pressure it wouldhave if it had behaved as an ideal gas.
Therefore P+0 then f=P
and P+0then flP=1
lf a gas mixture is separated by a diaphram trom one of the constituent gas on the other side
and depending upon the fugacity values ol that component on either side of the diaphram the
diffusion of that component acrqss the diaphram willbccur if
RTd lnP(pure) r RTd lnPi (4.ffi)
RT d ln f (pure) * RT d ln ft (4.e)

67
 as is obvious like mechanical equilibrium
and in the direction from higher to loiver fugacity side
!n terms of Pressures
of the component in a system'
Therefore fugacir.y represents the escaping tendency
Fugacity values can be calculated frcm above relationships using equation of state and
compressibilitydataanditsvariationwithtemperatureandpressure'
Fugacityisalsoapotentialtermlikeanyothertermdefinedearlierindicatingnatural
process, unnatural process and equilibrium state'
very lor and hence can be
solids and liquids do have vapour pressures which are
with solid/liquid as the case may be
equated to as the fugacities in gas phase in equilibrium
fugacities'
and hence by equilibrium considerations solids and liquids do have

4.5 ACTlVlw
with the surroundings' The
Activity of a component is defined as its ability to interact
This ability is maximum when fugacity is
activities of pure substances is therefore maximum'
of fugacity of the substance in
maximum. The activity has therefore been defined as the ratio
thestateinwhichitexiststothefugacityinitsstandardstateas
t=Ft Of -fi
af,=
f,
(4.65)

and the i th component' The

where f and fl are the fugacities of pure component
standard state is commonly pure liquid or solid
at one atmosphere pressure and at the
atmospheric pressure at the temperature under
tempertaure under consideration or gas at one
consideration. Obviously only one, solid or liquid
or gas phase' is stable at the given
The fugacity in the standard state is
temperature, excepting the transition temperature.
activity is unity for pure substances' Since
nornrally described uyin" symbol f and ltence
take pure substances alone as standard states'
standard state is arbitr4ry it is not necessary to
ltiscommonpracticetotakelwt%dilsolutionasstandardstatedealingwithmetallicsolutions
With the abow
a = (flP) or ar= (t I GP)
andhence f =f'a or t = aGP
and since f and to are fxed and constant, lherefop
(4.66)
RTdlnf = RTdlna = dq
{4.6n
and RTdlnt = RTdlna = dGi
The similarity in activity and fugacity is obvious'
lntegrating from the standard state €lat = 1 at f = f
or != GP to anY arbitrary state
(4.68)
RT lna = G - Go
(4.6e)
and RT lna = Gi - GP

and on differentiatino
,,- - Art.- ..o\
RTdlns=d(G-Go)
(4.70)

RT d ln a;= d(Gi - G1
$'711
and
is almost proportional to fugacity and the
This activity is also a potentialterm since it
be a potentialterm' lt indicates that if activities
latter being potentialterm so activity also should
68
on either side of an interface of two phases in a system are the sar€, then the system shall
be in equilibrium. lf the two are not same on either side of the interface then the component
with higher acJivtty shall'hare a tehdency to migrate to the region of lower activity zone across
the boundary, if physically otherwise possible.

4.6 CONCLUSTONS
The parameters like intemal energy, enthalpy,work function and Gibb's free energy
have been the potential terms under their characteristic conditions. These energy properties
are extensive properties. Potentials are intensive properties and the difierentiation of these
two must be properly understood. The properties like temperature, pressure and chemical
potential are intensiw properties and are potential terns of one kind or the other and which
make the heat flow (underthermal potential gradient), make work exchange possible (under
pressure gradient) and make chemicalspecias mow (under chemical potential gradient). The
values of the potential terns decide whether the system is at equilibrium or is poised for a
spontaneous change or whether energy will have to be supplied to make the system move in
the desired direction.
The terms fugacity and activity are directly related to each other. lf say two immiscible
liquids like a slag and a metal or like oil and water are in equilibrium with respect to certain
componentthe activity of that component in both the phases must be the same. ln otherwords
the fugacities also must be the same and since and which in turn means that their chemical
potentials in the two phases must also be the same. Therefore the terms fugacity, activity,
partial energy properties under the characteristic conditions and finally the chemical potential
means the same potential term, only in different forms, that dictate chernical equilibria in
thermodynamic systems.

69
CHAPTER

HERMODYNAMICS
estimation of entropy-

5.1 ZERO ENERGY LEVEL

energy and enthalpy of a
The first law allows measurnent of changes in intemal
not in absolute terms but only
system when therrnodynamic processes occur, though
energy or enthalpy are purely arbitrary' Both
relatively and hence the zero lewls for internal
oftheseareenergyparameters.Theentropydefinedbythesecondlawisa]soaenergy
in a thermody-
p.r.*". fne seJonO law allows measurement of only entropy changes from the second
namic process but the entropy value in absolute terns
is not determinable

law.
The Carnot cycle leads to a relationship
_tz
ftt = gz lqt
and Tz are fixed when q and
which states that the difference in temperaturqs Tl
respectively' The range of
q2 amounts of heats are then exchanged with the surroundings
the nature of the system in the carnot
temperature scale orrer Tr and T2 is independentof
by adiusting the realvalues of
cycle. ltcan €rswett iefixed as on Oo to 1000 Celciusscale
lower and higher ranges of the same
qr and qz . TIre same scale could then be extended to
Celcius scale. From this it comes out to be '27go
Cas zero degrees abso.lute. or 0o K' As
energy and entropyreduce
both energy andentropy decrease with fallingtemperaturewill
to zero at absolute zero temperature ?
Allpuresubstancescannotbeassignedazeroenergyvalueatabsolutezero
and
because then the heat of formation of compounds will have to be zero at absolute zero
fact that though internal energy
which is just not the case. This is also obvious from the
to zero at absolute zero temperature'
changes with temperature it can not be extrapolated
Energies at higher temperatures can be estimated
experimentally fromJCe OT orJCv dT
these have not been
and these can be extrapolated. to absolute zero temperature' But
internal energy to even
obsenedtobezero at0oK. Thus anyassignment olzerovalueof
elenents at absolute zero temperature is purely arbitrary.
The integration of Gibbs-Helmholtz equations
d (AAfD = -AUfl" dT
and d(AG/T) = - dT
^Hff
gives separately + =-f;, Sor*ru (5.1)

and ST Jrr 4or*r,

=-f'" f
(5.2)

where Iv and Ip are integration constants indicating the constant volume or the
constant pressure conditions of the systens. These constants Iv and Ip €n not be deter-
mined from first or the second laws of thermodYiramics and empirical methods alone becorne
necessary for their evaluations.

5.2 ZERO ENTROPY LEVEL

The entropy variation with temperature has already been shorrln earlier in terns of
the specific heats as
dT
dSy = Cv
T
dT
and dSp = Cp
T
and on integration leads to

Srv Q
= eOV -d (s.3)
"uS
Srp = Sop -{ "rS
(5.4)

where Sgy and 36" are integration constants or values of entropy at absolute zero
law
temperature under constant volume and pressure conditions respec'tively. Again second
by itsetf can not solve this for evaluation of Sq, and 36"values except by direct experimental
work.
Nernstwho after analysing the data of variation of AG and AH with temperature'
tt was
par$cularly at lower temperature region, found that aG/T and aH/T tends to zero as the
temperature approaches zero value as schematically shown in Fig 5.1.

Fig. 5.1 - Showing schematically the

observations of Nerns't about vari-
ation of Mf and LG of reactions with
o temperatures.

tG

TEMPERATURE +

71
 d (4G)
0
a(AH) _
AT dT-v ^
T+O J+O
slnce
#=-as, AS= 0
T+O
(5,5)

.Nernst generalised these findings in the form of a statement that for all reactions
ln which substances are ln condensed state AS is zero at the absolute zero tempera-
ture. This is known as Nernst's Heat Theorem and is one of the early statements of the
third lar of thermodynamics. Hence if reaction occurs as
A+B= AB
the entropy change of the reaction is given by
AS(reaction) = ASAB - ASA - ASe
and if the entropies of elements are assigned zero values as per the aborre theorem then the
entropy of the compound AB is also zero at the asbsolute zerc temperature. This is in fact
what the Nernst Heat Theorem states.
The combined statement of the first and the second laws of thermodynamics is
dG= vdP_sdT
and this can take the form
dAG=AVdP-ASdT
or H= ^v#r ^s
at constant pressure (H), = - AS = 0 (s.6)

T+o
Similady 6Q=AH-TAS
differentiating partially with respect to temperature at constant pressure

("P),=(H),-'(H) -* (5.7)

putting values of

therefore (5.8)

Hence ACp = o al T= 0 unless t (#F), is infinite and which is not

at all so.
According to Debye's law
Cp or Cv c T3 or Cp or Cv = kf3 where k is constant
henceat T=0 Cp or Cv=0
Hence the specific heats are zero at absolute zero temperature and hence the isothermal
processes like change in pressure and/or volume, since without any heat exchanges as
Cp and Cv are zero, therefore occur without variation in entropies.
72
or
(H),=o (5.e)

and (u#)"=oas r+o (5.10)

or in other words the entropy attains a value of zero at absolute zero temperature.

5.3 THIRD I,AW OF THERMODYNAMICS

The third law may now be stated as that the 'entropy of any homogeneous
substance which is in complete internalequilibrium may be taken as zero at zero
degree absolute temperaturc^ Complete internalequilibriurn means that unique lorest
energy state which is obtained in the crystals when cooled under perfect equlibrium
conditions, such that the atoms occupy uniquely ordered arrangement and hence the uniquely
lcwest energy state.
There is first and the second larrus of thermodynamics but
no theoretical proof for the
for the third lavrr proof can be given in terms of quantum statistics.
It should be noted that, similar to internal energy and the enthalpy, the entropy of
elements at the uniquely loruast energy level is zero at absolute zero temperature, is
again arbitrary. This is however fruitful because this also, as a consequence, leads to that
entropy of a compound at absolute zero temperature is also zero, provided that the
compound is in complete internalequilibrium.
It is rather difficutt to achieve the state of complete internal equilibrium and hence
atons in compounds are not that orderly arranged and hence may not have zero but some
entropy which might cause error in calculations.

5.4 EVA!-UATION OF ENTROPY AT HIGHER TEPERATURES

The variation of entropy of a substance with temperature has been shown in Section
3.5 and 3.6 as

at constant volume Sr2 = Sr', * -f lr . Cv. dl nT and, (5.11)

at constant pressure Sr, = 91, . I';, Cp . d lnT (5.12)

lf this integration is carried from 0o to a temperature f K then using third lzur these equations
would reduce to :

at constant volume sT2 =."0f, cv dln T (5.13)

and constant pressure S-, = Cp dln T (5.14)

d
lf this is split into two different temperature zones one lover tha4the Debye's
characteristic temperature 0o (which is generally around 15o K) and the other abow this
temperature then.

73
 ^9o
65r, = J6 ?o'.{" ?0, (5.15)

ptltting value of Cp = k 1€ as per Debye's 1€ law

then lsr, = /oS ot - l]j ?o'

=t kf ot.{" ?o'
= {. td" * d' ?o'
as the first term vanishes at T = Oo K

aSlr= Jcrtrol . tt;+ dT (5.16)

orwherethefirstterm on the right hand side is the heatcapacityatDebye'scharacteristic

temperature. This can be solr,ed graphically by plotting either Cp /T vs T or Cp vs. ln T.
Theareabetweenthepoints 0oand Tz under the curve bound by thetemperature
axis shall bd the value of the second term on the right hand side and when 1/3.Gp16oy is
added tn it shall give the entropy change required. This presumes that upto temperature T2
no phase change occurs. lf phase change occurs in between, before Tz then the equation will
have to be extended as in Eq.(3.41).
o
ln a generalised form the values of Szga K are known from the values of Cp
between the temperature range 0o - 2980 K and are available in tabular form for many
substances and hence entropy of a substance at a temperature T may be evaluated as :

S?= S2sa"+fr = Sggeo*dru" Cv dlnT (5.17)

J 298"
"u.S
s?=s3se.*1t
" "rS = s2sao-C*", cpdtnT (5.18)

the superscript indicates standard states.lf there is a phase change in between then the
entropy change will be given by as discussed in Section (3.11).
Example : Calculate the standard molar entropy of zinc at 3000 C assuming that it follops
Debye's equation below 25o K. Assume average mean specific heat for solid zinc as 0.36
caUg/0o K and that at 25o K is 0.091 cal/0o /gram.

sotution: ASsoooc=Asgzs"k = f;' As+ f::' cp.drnT

{
a
Cetzso) '+ Cplave) x [ln 573 - ln 25]

0.03 + 1.931 = 1.961 cal/0o /gram

1.961 x 65.5 = 132 caVOo /g.mole
It is generally not so accurate to take mean value of specific heat like this over a
temperature range. ln fact Cp value in this temperature range must be evaluated by
experiment and then the integral 25-5730 K is evaluated by plotting Cp vs. ln T and the area

74
between the limits 25o and 5730 K shall be taken as the value of that integral. This is more
precise.

5.5 CONSEOUENCES OF THIRD IAW

The coefficient of thermal expansion and compression tends to be zero as tem-
perature approaches absolute zero. This will be obvious from Ma,xwell's relations. That is
1 /av\
"=VIFj,
substituting Eq.(4.46) in this gives

cr= -(r5)r|=o asStendstozeroas T+ 0 (5.1e)

* = 1P /.48\
" (ar/v
substituting Eq.(a.a5) in this gives

U=(#)rf =o asstendstozeroas J+o (5.20)

It is also true that the third law itself suggests impossibility of attaining absolute
zero temperature. This can be prowd thus :
Let us imagine a Carnot cycle which is operating between 0o and f X. tne entropy
change during such a cycle is zero and hence the entropy change of the cycle :

fr***=o
The term go ff at zero absolute is zero and hence Qr/Tr must also be zero. But this
can not be so because it is a finite quantity and hence the only alternative conclusion
possible is that Oo K is not attainable. Therefore it has been stated that 'no definite serie
of processes could lead to achieve absolute zero temperature'. The third law therefore
finds a way out to estimate absolute values of entropy based on quanturr statistics.
to the entropy of uniquely stable arrangement of
The assignment of a zero value
elements is nothing but convention. The assignment of zero value to the entropy of
uniquely stable compounds is a consequence of the above convention and is stated as the
third laru of thermodynamics.

5.6 CONCLUDINGREMARKS:
Third lanl finds a way out to estimate entropy of substances at absolut e zero
degree and ihereby completes the nrethod of calculating the entropy of substances. This
ultimately is required to work out energy exchanges during thermodynamic processes.

75
CHAPTER 6

ZEROTH LAW OF THERMODYNAMICS

phase stability

6.1 INTRODUCTION
lf a system is fully isolated by an adiabatic wall and left to itself, it attains a stable state
such that no further change is perceptible, no matter how long one waits, then the system is
said to be in equilibrium or has attained equilibrium. Left to itself it shall continue to be in the
state of equilibrium and shall change from the equilibrium state only if externally acted upon
by some constraint. A system may also be in equilibrium with its surroundings in the sarne
way. lf a system consists of only one phase, say liquid or solid or gas then left to itself, being
a homogeneous phase it is normally in the state of equilibrium except some exceptional
circurnstances. This is what is known as 'stable equilibrium'. This is best illustrated by an
example of a ball resting in a valley. A system in stable equilibrium shall get displaced from its
equilibrium state on application of an external constraint, but as soon as the external constraint
is removed the system shall gain its original stable state.
lf such displacement, under external constraint makes it stable in some other state
then it is said to be in 'metastable equitibrium'. The best example to illustrate this meta-sta-
bili$ is the ball resting in a small depression at the top of a hill. The other befter examples of
metastability are the case of a supercooled liquid or a rnixture of hydrogen and oxygen. A
human body is most stable in horizontal lying position. Standing vertical, though appears quite
stable, yet it is only meta- stability. ln all these cases the system shall move to a stabler form
if giwn a chance by initiating the displacement.
On the contrary on displacement, if the system changes to a displaced state, it is said
to be in 'neutral equilibrium'. The example for its illustration is a ball resting on a flat plane
surface.
ln the chemical and metalurgicalarena quite often it is necessary to know whether a
certain phase is stable or not and if it is unstable, what is the stable phase under the given
constraints. Similarly it is more often required to deal with two or more phases containing
several components in contact with each other. lt is also then necessary to know whether
these phases are in stable equilibrium or not and if not, then what is the stable configuration?
The former part is dealt with by ascertaining the conditions for phase or phases stability and
the later part by knowing generally the chemical equilibria between the phases involved. The
cases like two-phase-equilibria in a single component system like liquid-vapour or liquid-solid
 can be dealt with in terns of phase stability of an individual phase or phases inwlved in the
system.
The thermodynamic studies are confined to stable equilibria only. lt has already been
pointed out that the system and its sunoundings can be in a state of mechanical equilibrium,
if pressures in both are equal or in thermal equilibrium if both are at the same temperatures.
Similarly they can be in a state chemical equilibrium if the chemical potentials are the same
in both. The sane holds true within the system itself vis-a-vis the different phases involved or
vis-a-vis the different sub- systems within the system itseff under consideration.
The concept of mechanical stability is a static one as will be obvious from the example
of overhead tank. On opening the tap at the lower lerrel, water florus down and stays at that
lovrrer level, r.e. it has acquired a static stability at the lower lewl.

ln chemistry howeverthe concept of stability is a dynamic one. ln ternr-s of the theory

of reaction rates, the system acquires stability only when the rates of forward and that of the
baclcrard reaction exactly match to have no net effect whatsoer,er. The two equal and
opposite rates match but do not reduce the system to no activity level. This is why it is a
dynamic concept.

6.2 ZEROTH TAW OF THERMODYNAMICS

This lantr is more basic than even the first lary, although it was discovered only after
the first and the second lalvs were formulated and, formalised only after the enunciation of
the third larr of thermodynamias. ln stead of giving it the number fourth it has been preferred
to call it zeroth lanrr thereby indicating its more elementary nature than even the first law. This
lavv can be appreciated eren without appreciating the idea of temperature derirred from the
second lalr.
Two basic properties, particularly of fluids and in that also, more easily appreciated
in the case of gaseous phase, are 'pressure'and 'volume'. A cylinder and a piston can be
imagined to contain a certain amount of gas. The volume and pressure of this gas can be
adjusted by the movements of the piston. Let the values of pressure and volume for a certain
gas in this rylinder be Pr and Vr . The gas is thus defined in terms of pr and Vr . lf another
gas at volurne V2 is brought in contact With this gas through a diathernral wall the two can be
brought to equilibrium by adjusting the pressure of the second gas P2. The two gas systens
may have the same hotness or coldness or fixed so by placing them in a certain oven or a
refrigerator. Out of ihe four variables three may be taken arbitrarily but the fourth will have to
be fixed, in relation to these three for equilibrium to reach. lt means that if gas one at pr Vr
is placed in a orren and gas two of Vz is brought in contact in the same oven with diathermal
walls between the two , the Pz will have to be adjusted to bring the two gas systems in
equilibrium.
Let us inngine a third gas which is brought in equilibrium with the first one in place
of the second with Ps and Vs values such that Pr Vr and Pe Vs are now in equilibrium in the
oven or the refrigerator. Now from the pair 'one-three' the first system is removed and the
second is brought in, then the pair'two-three' must also be in equilibrium, as is obMous and
obsened in practice. The same need not be true with only gases but shall be true with any
other bodies.
n
 This observation is stated in the form of 'zeroth law of thermodynamics,
as :
'lf of three bodies say A, B and C, B and C aro separately
in equilibrium with A then,
B and C are also in equilibrium with one another'.
The converse is also true in that if three bodies are in contact, each
to each and are
all in equilibrium together, then any two taken separately are also in equilibrium
with one
another.
This leads to the fact that the function of P and V for a certain amount
of gas is the
empirical parameter hotness or coldness of the system and is denoted
by the palmeter T
in the form of an equation as
F (P,V) = T
This then would be the equation of state of that fluid. lt is this
function of the fluids,
which is nothing but temperature, which has the property of taking the
same value for fluids
in equilibrium with one-another. The fluid is thus uniquely defined by
any two of the three
variables P, V and T.
The original zeroth law statement therefore reads that of three bodies
A,B and C, if
A-B and A-C are in thermal equilibrium meaning thereby that at the same
temperature, then
B- C are also in thermal equilibrium, meaning thereby that, they
are at the sarne temperature.
Conversely this is possible for fluids only if their temperatures are equal,
that is the product
PV for each one of the three fluids A, B and C are equal.
Similarly if Vr Tr of A are in equilibrium separately with Vz Tr of B and Vs Tr
of C
then Vz Tr of B is in equilibrium with Vg Tt of C as they mean same pressure
for A, B and C;
which shallbe the case of mechanicalequilibrium.
Equally true will be the case of chemical potentials. lf chemical potentials
of A and B
are equal and that of Aand C are also equalthen they are in equilibrium
and hence B and C
also shall be in equlibrium from the chemical point of view. Since thermodynamic
equilibrium
means equilibriumwith respectto all possible potentials, thermal, mechanical
, chemical, etc.,
the generalised statement as above has been made as the zeroth lar statement
for an overall
equilibrium.
It may also be stated that if three or more systems are in contact, each with
each,
and are in equilibrium with respect to any one or more of the thermodynamic potentials,
then
any two taken together shall also be in equilirbiumwith respect to that potential.

6.3 Phase stability

A substance can exist in either a solid, liquid or a gas form depending upon
the
conditions of pressure, temperature and volume. The knowledge of conditions
under which
a phase is stable i'e. in equilibrium is important in systematic thernndynamic
analysis. This
will also give rise to the conditions under which two or more phases eists
together in
equilibrium like at melting and boiling points. From the point of view of
consideration of
thermodynamic potentials, of all the possibilities, that phase which has the lowest
associated
potential shall be the stable one relative to the others, lf the phase
has a higher potentiat, it
shall change over to the one having lcnrver potential.
 Single component phase is stable in its existing form if and only if the rate of change
of entropy with its own mass is negative r.e.
(!e\ . o
\d"/
This is obvious because if this is positive then this will be a natural change possible
and hence phase will be unstable.
Other conditions are that the specific heat values Cp and Cv must be positiVe and
the compressibility p must always be positive r.e.
p>o or _+(r+),,,
Free energy must vary with chemical concentration in a certain direction. lt means
activity must vary positively with concentration. lf this variation changes or reverses phase is
unstable..
The problem of phase stability can best be understood from the example of liquifica-
tion of a gas by compression coupled with lorrrrering of the temperature. The Fig. 6.1 shows
the P-V plots for a gas at various temperatures. lt is seen that as the temperature is lowered

Fig,6.1 -
I
ul Pressure-volume rela-
c
3 tionshipfor air as affected
o
o by temperature to show
LI
c
a.
the conditions for ltquifi-
cation of air.

STAELE
PHAgE
VOLUME
the P-V isotherm ->
is lcnruere.d in position until a temperature Tc - called critical temperature
is reached, belcnrv which the gas liquifies. lt means the gas stability is lost and liquid is formed
for a certain range of P and V values at a temperature lower than the Tc . fre gas is stable
in the range A-C and B-F but not in between in the range C-F.
As the gas is cooled (AV/AP)Ihas a + ve value of compressibility upto temperature
Tc at which a slight inflextion occurs in the P-V isotherm. Belonv Tc at D and E the slope is
zero and the gas becomes unstable. A small range before this stage is the metastable alea
 C-D and E-F. ln the range D-E the compressibility is negative and hence unstable for gas
phase to exist and hence the change to a more stabler phase that can eist. i.e. liquification
takes place.
Phase stability will be lost and phase transition shall take place if the free energy
variation shows a discontinuity with temperature. ln other words a stable phase region shows
continuous variation of free energy with temperature. The first derivative of free energy as
derived from the combined statement of the first and the second law as dG = Vdp
- SdT
and hence

(#), =-g' (#),=' and

lffi] =
"
that is S,V and H vary continuously for the phase stability. Any discontinuity is an indication
of phase lnstability. Similady the second derivatives pzehf ) and the third (a 2C/at€ ) atso
be varying continuously with temperature for phase stability. Phase stability will be lost if heat
input increases the temperature in a two phase stable system like solid-liquid or liquid-gas.
ln magnetic transformation S,V and H vs.T are continuous curves but the second
derivatives of free energy vs.temperature is a discontinuous one.
Order-disorder transformation shall be possible if S,V and H show discontinuity in their
variation with temperature. These changes do cause abrupt change in sp.heat, coefficient of
thermal expansion and compressibility. These are second order manifestations of free energy
variation with temperature.

6.4 TWO PHASE STABILITY/ EQUILIBRIA

It is a common knowledge that at the melting point both solid and liquid and at the
boiling point both the liquid and the gas co-exist. This is because the two phases solid and
liquid in one case and the liquid and the gas in the other case are in thermal equilibrium. This
is because the two phases in contact have the same thermodynamic potential at such critical
temperatures. The change in free energy when one component gets transferred from one to
the other phase is zero at such critical temperatures.
Since AG = AH - TAS, at equilibrium AH = - TAS. The latent heat is enthalpy at
constant pressure and temperature and the internal energy at constant temperature and
volume. The heat input is latent heat and matches - TAS and hence AG = 0. The lntegral

/eOV should be zero for such changes. ln Fig.6.1 the two thatched areas are equal and
opposite and hence the total effeCiveJPdV is zero when both liquid and gas coexist together.
It can be mathematically proved that for two phases to be in equilibrium like liquid and solid,
in any system, the temperature, the pressure and the free energy of both the phases will hare
to be in equilibrium and this will be possible ff and only if the chemical potentials of all the
components i.e. partial molal free energy values of all the components, though different values
for different components, are equal for each component in both the phases.

9t4.1 Phase Rule

-/ A phase is a homogeneous part of the system u",ith definite boundarie and interfaces
with other phases or sub- systems. The number of components in a system is the srnallest
number of substances in terns of which the composition of each phase in a system can be
defined. The degrees of freedon are nothing but number of independent intensive variables
to be controlled externally for describing the system completely like pressure, volurne and
temperature. The phase rule states that for any system
F=C-P+2 (6.1)
or F+P=C+2 {6.21
where F is the degrees of freedom, C the number of components involr,ed and P the nunrber
of phases present. For systems nornnlly encountered in the metallurgical area the pressure
is often equalto atmospheric pressure and hence is fixed at one atnnsphere and the systerns
are condensed phases and therefore the phase rule takes the form as
F=C-P+l (6.9)
because one degree of fieedom in the form of pressure is not arailable.
6.4.2 P-G-T Relationships vs Phase Stabitity
On the whole in a closed one-component system at a certain temperature T and
pressure R the equilibriurn phase or phases in the system shall be that or those having
minimum value of free energy. The equilibrium thus can be evaluated by knowing the variation
of G at constant T and P. But if any of these two also vary then through the relationship of
G-PatconstantforG-TatconstantPorasG-P-Trelationships,asallthethreevary,
may be useful. Phase stabilities can also be appreciated from P-T diagrams but G-R G-Tand
G-P-T diagrarns are more appropriate.
The P-V relationship for a single component system as shown in Fig.6 2. lt is clear
that at one atmosphere pressure upto melting point Tu solid is stable, above boiling point Te

I
I
I
I
I I ut9
il I
I
a

I
I
I
,

EA TEMPERATURE
E
T
Blx
TEMPE

F ig. 6 2 - Pressure-temperature relat Fig.63 - The free energt-tempera-

iols hips to indicate stabiJity regbns ture relationship to show relative
of soltd, Iiquin and vapour phase in stabilitics of phases (shown by
a single comprent system (shown solid lirrcs) and their rnetastable
by soM lines) along-wilh rtte trvtas- regions (shown by dotted line$.
tability regiotts (sho*n hy dotted litn).
81
gaseous phase is stable and in bef,ween Tu and Ta liquid is stable. At T both solid and liqukl
and at Ta both liquid and gas phases co-exist. As pressure is lowered the boiling point
decreases rapidly and the nntting point also decreases but rather slorrly. Consequentty at a
certain ternper&r;s Ts all solid, liquid and gas phases co-exist. The pressure at which this
happens is much behw one atnnsphere. At temperatures belovrr Ts therefore solid and gas
can co-exist together and hence sublirnation is possible but above Ts solid will hbre to change
to liquid and then to gas on increasing the temperature.
The sarne conclusion can be drawn from the G-T relationship for sold-liquid-gas
phases br a single cornponent system. The fiee energy values are calculated for all phases
as they vary with ternperatures and at different pressures. The minimum value of G indicates
the stability of that phase under the conditions under consideration. This is strown in F19.6.3.
It clearly shorrrs that at lov pressures vapour is more stable than liquid and hence direct
sublimation rather than melting takes place. At high pressures the temperature rise shall result
in solid melting to liquid and then converting into vapour. The situation is clarified further by
plotting G-T diagram as shorn in F1g.6.4. Below melting point Tu stable phase is solid and
abore it liquid. The unstable phases are shown by dotted lines. Similarly if the AG of solid to
liquid phase change is plotted againsttenperature the situation is shorrn in Fi9.6.5. lt clearly
indicates that AG is negative for this only above Tu temperature indicating its possibility and
bdor Tu it is positive and hence not possible.

--_J1og,a
rg
c.
a
z
q
d
arl
ul
d
IL
TI

itr,
orN*.,

TEMPER,ATURE + TE'{PERATURE;+

Fig.6.4 - Free energt temPrature Fig.65 - Tlrc free energl ehange as a

relatiarahip to indicate Phase furrction of temperature when plwse f

t
snbility itf solid and liquid phases. change occurs.

6.5 CONCLUDING REMARKS

Although obviouS stillthe equilibrium between various systems or subsystens within
a system has been established by the zeroth law of therrnodynarnics. The stabilitis
of phases

".nu"evaluatedknowingtheirP-T,G-T,G-PorG-P-Trelationship'Thisisuseful
from the point of view of their uses in practice. A phase which is not stable can pose
problens
il it is to be used in practice over a certain duration.
 CHAPTER

CHEMICAL EQUILIBRIA
equilibrium constant

7.1 INTRODUCTION

Jhe modern day civilization can sustain itsetf by conrerting the naturally occurring
rnaterials into a series of products with different chemistry. Eren the already changed chemistry
is changed further several tinre to obtain the desired chen{stry of the products. Literally
therefore there are innurnerable chemical reactions that are canied out in actual practice,
particularly in the field of metallurgical and chenrical engineedng. Although these chemical
processes are carried out in such a way that the reaction proceeds in fqward direction but the
knowledge of equilibrium conditions of such chemical processes is a must to work out the
conditions for rnaking the reactions go in the desired direction. lt is tfrus the prinrary objective
olthernndynanjc study to firs* establish the equilibrium conditims of chemical processes ol
interest. lt is from this pc$nt of view tnt this topic has been dealt with independenily in this
chapter.

7.2 THERI'ODYi'IAMIC EOUIUBRIUM CONSTANT

Let us consider a general chemical reaction at constant prqs$rre and tenperature
as aA+bB+....... =cC+dD+...... (7.11
where capital letters are used to denote the reactants or products and the srnall letters
are used to describe molecutes inrolved in the balanced form of the reaction. The free energy
change may be wdfien as
AG = 2G (products) - )G (reactants)
= (cGc + dGo + ....) - (aGn + bGa + ....) (7.21
where indMdual free energies are described in terns of the respective G values. lf the
reactants and the products are assurned to exist in their standard states, then the free energy
change rnay bewdtten with suffixe of standard state on each fiee energy value as
o
-
AG = (cGE + dGB + ......) (aGl + bGB + "..) (7.s)
Subtracting Eq.(7.3) from(7.2), standard and non-standard state equations of free enerry
changes and, rearranging the terns suitably, leads to

-
AG AGo = [c(Gc cB) + d(Go Gb + .,..1 ta(Gn -Gl )
- - -
+ b(Gs _ cg) + ........1 (t.41
Substituting s derived earlier G-Go = RT In a in the above gives
AG - AGo = [c RTln ac + d RT ln ao + ....] - [a RT ln a,q +
 + b RT ln ae + .,....1 (7.5)

on rearranging becomes

aG-^co=Rrrn 4aA'a8.
"l ... (7.6)

lf the reaction (7.1) is at equilibrium r.e. if the reactants and the products are
in thermody-
and that case the parameter, which is of constant value
namic equilibrium with each other in
is denoted as :
. .ln .8. :- .''=
-..- "S. K F,N
al.aB....
of the reaction'
The term K by definition is called the thermodynamic equilibrium constant
of K remains
The activities of individual reactants and products may change but the value
constant at a fixed temperature for a giwn reaction'
AGo RT ln K (7'8)
Hence AG - =
and if the reaction is at equilibrium then the free energy change is zero i'e'
AG = 0 and
tn"r"tor" AGo = - RT ln K (7'9)

Putting value of R as 1.987 cal/deg/mole and changing to common logarithm

AGo = - 4.575 T log K
(7'10)
hence K is also a
It is thus clear that the free energy is a tunction of temperature and
the
in eraaluating
function of temperature. This is one of the most commonly used relationship
would occur in
chemical equilibria. This evaluation indicates whether a particular reaction
practice r.e. whether there exists a tendency in the system to move in one
or the other
explained by two illustrations as follcnrs
direction or is at equilibrium. This is best
:

Example 7.1 : Will magnesia crucible stay in snape in a furnace operating at 16000 C
under a vacuum of 10a atmosphere ? Given is the free energy for
the reaction
2{Mg} + {oz} = 2< MgO>
is AGo=-358700+102.6Tca1/g'mole
Solution : The value of AG at 1600 + 273 = 1873 K is given as
AGrezso = - 166500 cal/g mole

stnce r = Pfig
^&g--
. Po,

AGrezso = - RT ln K = - 4.575x 1873,on P[ts' Poz

={_
putting aMso = 1 for solid MgO,the abore reduces to
166500 = + 7600lo9 p2Ms po. or log p2rug Poa = - 21'91
- existing are magne-
ln the ambient atnnsphere under vacuum of lStatmos.the components
vacuum pressure' The
sium vapour and oxygen gas. Each at best can be equal to the
proportion of magnesium uupou;. and oxygen in the gas phase
is not known and there
is no alternative but to assurne that these are in proportion to 1 po. = 2
ptrg
moles and hence
dissociation
aftersubstitutingthisinthe above givespo2 + 2pr'rs = vacuumpressure'Butif
gas phase and hence only oxygen
does not occur then no magnesium will be there in the
that dissociation doas
partial pressure shall be equalto the vacuum pressure' First assuming
notoccurthen- 166500 = + 7600logpo, or 21'90 or ps2 =lO'"'Under
logPoz = -
the equilibrium partial
this situation sincethe ambient partialpressure of oxygen is morethan
84
 pressure of oxygen the reaction shall har,e a tendency to morre
tornrards formation of MgO,
that is magnesia shall stay in shape. lf now dissociation is considered
to occur then product
Por shall be equalto 1/3 of the vacuurn pressure which will still be less than when the
dissociation does not occur and hence chances of dissociation of MgO
are quite rennte. ln
other words Mgo has equilibrium oxygen potential equal to rnuch less
than the vacuum
pressure at 16000 C and therefsre MgO will not dissociate under
the conditions of vacuum
and is thus stable under the given conditions, r.e. Mgo shall stay in shape.
Example 7 '2 : Will it be possible to anneal copper ingot at B00o b before
rolling under a gas
mifiure of Hz lHz O equalto 102.
Given 4cu + oz = pcuzo aGo = - 7g7oa+39.3T cailgmore
and 2Hz + Oz = ZHzO AGo = - 118000 + 26.T5T _do_
Solution : Forareaction 4Cu + 2HzO = ZCuzO + ZHz
Hess's law gives AGo = - 39300 + 12.SS Tcal/g.nnle
orfor 2Cu + HzO = CuzO + Hz AGo - _ 1gl50 + 6.27 T
and AG?ozso =- 1242J caVg mole
The equilibrium constant of this reaction is
acuzo'
* - PH3
a6u' PHao
and assuming the activities of solid Cu and CuzO as unity the value of K reduces to
K = pHzl pH.o
and since AG = - RT ln K = - 4.STSTlogK
putting values for B00o C it gives
- 12423 = - 4908 log Ps, / p*.o
or tog pH2 lpu"o = ffi = 2,53
i.e pHzl gn o = 192'se
This gas mixture is richer in hydrogen than that when the ratio is 102 . Hence the tendency
for
the reaction of oxidation of copper is to go towards lefi. The atmosphere in the ingot
annealing
furnace is reducing towards copper and hence it will be possible to anneal .oppu; in
the given
case without the danger of its oxidation.

7.9 TYPES OF REACTIONS

Chemical reactions occur either in a single phase as homogeneous reactions or

between two or more phases as heterogeneous reactions. The homogeneous reactions
can
best be illustrated as :
a) double decomposition of acid and base in an aqueous solution like
HCI+NaOH=NaCl+HzO (7.11)
or as
b) reaction of two gases like 2Hz + 02 = 2HrO (7.121
The equilibrium constant of reaction (7.12) is

K= pfrro lp\" . po,

which is characterised in such a way that for a fixed temperature and partial pressure
of
oxygen the ratio Hz I Hz} is fixed. ln other words the equilibrium constant of such reactions

85
 and
shall be proportional to sucfr ratios or it will be equalto fixed ratio for a fixed temperature
fixed oxygen partial Pressures.
phases like:
The heterogeneous reactions essentially occur between two
i) solidsolid <ZnA>+<c>=Zn+CO
iD solid-liquid lMnl + <SiO> = [Si] + (MnO]
iii) liquid-liqukt tsl + tMnl + (CaO) = (CaS) + (MnO)
iv) solid-gas <FezOg> + {CO} = <FesO+> + {CQ}
v) liquidgas 2[9+{Q}=2{co}
activities of
The evaluation of equilibrium con$ant of such reactions need the
condensed phase, each
reactants and those of the products. For the cornponents in a
the contacting gas phase'
cornponent will be in equilibrium with its o\/n vapour pressure in
with their fugacities
The vapour pressurs shall be extremely lor and hence could
be equated
without inrolving any significant enor. These fugacities, being
so low will be substantially
and pressure ranges. Therefore the ac'tivities of
constant even owr a significant temperature
to their fugacitie in gaseous phase'
the component, in the condensed state are taken as equal
purposes, considering the
and shall be constant and hence are taken as unity for usual
activiUes shall be- different from
cornponents in pure state. lf these are not in pure state the
generally on the larrer side and
unity. Sin$larly the actual values of vapour pressures are
are taken ln the lorm of
hence are equated to their fugacities and therefore the actlvlties
thelr Partial Pressures.
in molten iron interacting
lf the component is not in pure form, like the one dissolved
of activities is a must
with components dissolved in molten slag or gas then the knowledge
for evaluating the equilibrium cons*ant. Alternatively the ralue
of K can be obtained from
can be worked out when others
the free energy data or knowing K the actvity of a cornponent
are known.
PERATU RE
7 {V ARIATION OF EOU ILIB RI U M CON STANT WITH TEM
K depends on ttre
J" Since the Eq.(7.g) implicitty means that the AGo and therefore
etablishing the value
for
tenperature, its variation with tenperature will have to be evaluated
is knoyvn'
of K at a temperature when its ralue at some other temperature
The combined form of first and the second of thernndynamics giws
larrus

variation of G with T in the form of an equation as dG =

VdP - SdT
are in their standard .
This may be applied to when the reactants and the products both
statesforageneralisedreactionasstatedearlierinEq.(7'1)as
dG8=VtOp-S8Ur
dGB=vBcP-sSor
andsoon dcl= VldP - SRdT
dGB=VBOP_SBdT
(7.13)
and so on
.:-^ each
multiplying nna by
^^ah one rrrr rho r
the corresponding number of atoms or rnolecules inrrulved
inthereaction(asdepictedbysmallletters},theorrcrallprocesscanbedescribedas
85
d(cG8 + dGB + ..... -aGl-bGB....; = (cVB + dVB + ....-aVl-bW "..) dP
- lcs8 + dsB .... - asl - bsB ....) dr {7.141
The three expressions in the brackets mean the overall lree energy change,
volume change and entropy change in the overall process. Therefore
daGo = A\P dP - ASo dT (7.15)
assurning one- atnnsphere pressure as_ standard conditions and putting
dP=0 AGo=-ASodT
from definitional form AGo = AHo - T ASo

therefore rco-
A9-T
AHo- AGo (7.16)

Putting this value in the aboie

d ,'=-
AGo
aHo - aco
v.1n
This is one of the lorms of Gibb's-Helmhottz equatlon valid at constant pressure. lt can be
also written as
TdAGo -.dTdT = -
- AGo AHo dr
dividing by T (the integrating factor being tff ) gives
daco _ _ AHodT (7.18)
T-T2
d(AG"4) _ _ AHo
or --dT- =-F (7.1e)

or d(+,g?r
o(trr = AHo v.2al

and since - RTlnK,tnus $

AG = = - RlnK

d(Acfn=-RdlnK Rd lnK AHo

or or
dT=T2
(7.211

dlnK =-- AHo dlnK =- AHo

(7.221
therefore
dr Rf n d(w
This is known as van't Hoff equation relating variation of K with T. Tlre elfect of
T on K depends on the value of AHo lor the reaction. lf the reaction is endothermic
(i.e. AHo = + ve) then K increases with increasing tenrperature and if it is exotheinib
(i.e. AHo - - ve) then K decreases with increasing ternperature. lntegratio,n rof.this
requires knorruledge of variation of H with T and which depends on ACp values of the
reaction. lf AHo is independent of temperature then K varies linearty with ,inrerce of
tenperature (1fD. With increase in temperature the reaction will mow in such a way as
to nulify the external constrhint as per the le-Chatelier's principle.
Exanrple 7.3 : The equilibrium constants of the reaction
<NiO> + {CO} = <Ni> + {COz}
at g00oK is 4.53 x 10-1 andat1100oK is 3.13 x ioo .

Find the enthalpy change of this reaction.

Solution : Using Eq (29
AGgooo = 4.575 - x 900 x 4'53 x 10-1

87
 = - 1865caVg/mole
smilarly AGrrooo = - 1575 - do -
* = - s Hence - s - = - 1.4sca/smte
Since G = H-TSthereforeHgooo=-1865 + 900x (-1.45)
=#
Enthatpy is senerallytaken as ;Jr"-i#:l:;0"-rure and hence catcutated
at any one temperature.

7.4 VARIATION OF EOUILIBRIUM CONSTANTWITH PRESSURE

As per Le-Chatelier's principle the reaction inwlving one or more gaseous compo-
nentslphases will only be affected by the variation of pressure and the effect will be such
that increasing pressure will tend to move the equilibrium in such a way as to nullify the
change in pressure as the externalconstraint. Any reaction which results in pressure increase
by way of fornration of gaseous component from condensed phases shall tend to move in
fonivard direction upon decrease in pressure and any reaction which results in pressure
decrease by way of formation of condensed phase from out of one or more gaseous reactants
shall tend to more in forward direction upon increasing pressure and vice-a-versa. For both
the situations, for example for a reaction of dissociation of calcium carbonate as

CaCOg= CaO+COz n=- acuo'Pco,

dOaCOg

with increasing pressure the forward reaction will not be fawured and hence K shall
increase or reverse reaction will be fawured and vice-versa. lf the pressure is decreased
by employing vacuum the process of calcination shall take place readily ,that is, at a loruer
temperature as has earlier been shown in Fig. 2.'t. ln general therefore if a reaction
aA+bB=cC+dD
represents the reaction involving gaseous components then

K=p8'p$
pft' pB
Will be independent of overall pressure if
c+d-a-b=0
then Kp = Kx
and lf this is not so pc+ b 'a -b

where Kx is the equilibrium constant expressed in terms of activities as mole

fractions. This is of direct importance in reactions involving condensed and gaseous phases
where c+d-a-b*0.
The folowing example shall illustrate the use of this more vividly.
Forareaction 2Hz + Qz = 2Hz O
KP = ( vl"o)r(ofr..pa) =
tr# Noa.P
t
/ Nxro\z 1 1 NHro\2
=ln,rr,l N"r'P =lN"',,| p",

88
or poa = No, . P =
[#)' *F or Kp Poz= (#)'
Also from Eq. (7.8) rd = - 6*pon . - expRr
At a giwn temperature Kp is therefore known in terns of AG of the reaction.
Therefore for a certain partial pressure of oxygen or equilibrium oxidising potentialthe volume
ratio of HzlHz O can be obtained
Similarly equilibrium CO/CO2 can also be calculated for reaction like
2CO + Oz = ZCAz .
Example 7.4 : The fiee energy of the reaction :
< CaCOg> = < CaO> + {COz }
is AG = + 40250 - 34.4f cal/g mole
What is the highest temperature at which this reaction will not occur in open atmosphere.
Also find out the temperature if the reaction was carried out under avacuum of 10'5 atmos.
Solution : The equilibrium constant of this reaction is
K=
8<CaO> .F.COzr/a<CaCOs>
Since both CaO and CaCOs are pure solid phases their activities are taken as unity and hence
K = Pocoa' = 1 atmasphere'
Since AG = - RTInK = -4.575 T log K = 0 = 40250 - 34.47
or 34.4T=40250 or T=1170oK=897oC.
It means till nearly 90Oo C is reached, the pressure of COz shall not exceed one
atmasphere and hence calcination will not take place tillthis temperature (see Section 2.6)
in open atmosphere. Under a vacuum of 10-s atmos K will be equal to this pressure and
hence 40250 - 34.4 T = -4.575 Tlog 10-5
orthe required temperature shall be 7030 K. The decrease in ambient pressure has reduced
the calcination temperature.

7.5 FREE ENERGY. TEMPERATURE RETATIONSHTPS

It has been shown that G = H - TS and H and S both change with temperature as
shown in Section (2.11.2iland (3.6) Therefore G also must vary with temperature. Since
both H and S are not measured in absolute terms but only relatirely with some reference
state, G also is measurable only in terms of reference state or, only changes are measurable'
Hence AG = AH - T AS and if all the reactants and products are in their standard states
then AGo=aHo-Taso
for a process AFlo = X Ho (products) - : Ho (reactants)
and ASo = X So (proaucs) - E So (reactants)
therefore AGo= [X Ho (products) - : Ho (reactants)] -
- T I: So (produas) - : So (reactants)l
= AHo (reaction) - TASo (reaction)
be evaluated has been
The way in which enthalpy and entropy can dealt with in Section
(2.11) andsection,r,r,i;l"l%""i:'_"1:T"j:", A + B = AB
this can be illustrated by an example:
Example 7.5 : Calculate the standard free energy of the
reaction [FeSJ + <C aO> = (FeO) + (CaS)
89
from the follo*'ing data :

(1) 2(Ca) + {Oz} = 2<CaO> AGol = -307100 - 51.28 T

(2) {Sz} + 2(Ca) = 2<CaS> AGo2 = -283100 - 42.93 T
(3) {Sz} + 2[Fe] = [FeSl AGos = -64080 - 19.04 T
({ 2[Fe] + {Cz} = 2(FeO) AGo4 = -1257N - 30.69 T
(5) <CaS> = (CaS) AGos - + 19000 + 9.0 T
Solution : The required reaction can be worked out using Hess's lav by algebric summation
of the given reactions and in turn it also gives the fiee energy change of the reaction
by the conesponding summation of the fiee energy values given as:
Add (-reaction 3) + (-+eaction 1) + 2 (reaction 5) + (reaction 2) + (reaction 4)
to give the required reaction.
(the - ve sign of reaction means the reaction in reverse)
The free energy change is thus :
= -AGos - AGor + 2AGos + AGoe + AGoe = 380 + 14.7A T cailg.mol.
The enthalpy AHo of chemical reaction is substantially a constant value without
variation with temperature. But ASo does change with temperature and also with volurne
considerably. lf these are plotted vs. temperature for a reaction the net effect of H and S can
be graphically evaluated to work out the variation of G vs. T. The G vs. T relationships are
varied depending upon the nature of the phases invofued and volurne changes accompanying
the process. These shall be discussed later in Chapter 14. As per third lary, elements do
not hare enthalpy or entropy at Oo K and therefore the free energy of compound is taken
as free energy of formation of compound from respective elements.
lnotherwordsforareaction A + B = AB AGo=GoRa-GoA-GoB
= GoAB since Gol and Gog are zero.
ln orderto simplify the matter Gole (2980 K) are determined using specific heat data
and Kirchofi's Law and are plotted in tabular forns so that AGo for any other reaction at any
higher temperature can be evaluated as follqrys ;
Forareaction MnO + C = Mn + CO
AdbE = Gtoo.o + GRtno.o - GF,morruo - GEr*o

= Gtoanro - Gf,tnroo

The AGor value at any higher temperature can be evaluated using the enthalpy and
entropy data and Kirchofi's Law.

7.6 CONCLUDING REMARKS

The ewluation of thermodynamic equilibrium constant is a major achievernent in

understanding the chemical equilibria since its knowledge gives the equilibrium positim
which can be juxtaposed with the real life situation of the reaction to work out, if any, the
desired direction in which the reaction has to be rnade to more. lt also gives the parameters
which could be used to make the reac{ion mow in the desired direction. lt also gives the
limitations of making a reaction move in the desired direction as reaction can rrpve only upto
equilibrium, and that too, often not even upto the equilibrium state but nu.rch ahead of it.

90
 CHAPTER

PHASE EQUILIBRIA INII

SI NG LE COM PON ENT SYSTE[V

clausius-clapeyron eqnatior r"

8.1 TNTRODUCTTON

ln practice metals are often heated to high temperatures for their processing in
either solid or liquid state. As temperature rises the vapour pressure of metals in gas phase
increases and may become substantial. The equilibrium vapour pressures of metals are of
direct interest. Firstly if metals are to be kept at high temperatures for a prolonged period il:
can lead to a substantial metal loss via vapour forrnation. For example in LD steelmakinll
process almost one percent iron is lost in vapour form which not only is a net lqss but poseri
pollution hazards. lf such losses are to be minimised then conditions will have to be generata i

to make vaporisation rninimum. This requires the knowledge of equilibrium vapour presiurr r

of metals or broadly, the thernpdynamics of single component multi-phase system is requiret I

o\€r a temperature and prasbure ranges of interest. Similarly for dealing with alloys tht,
thermodynamics of mufti-component, multi phase system o\,er a temperature and pressu'
ranges shall be of interst.
The melting point and boiling point by implication means those temperatures wher
transitions solid-liquid and liquid-gas occur when the ambient pressure is one atrpsphere
When pressure is difierent, the m.p. or b. p. do change. The equilibrium solid-liquid, liquid-gat
or solid-gas, as a function of temperature and pressure therefore needs to be established. The
equilibrium phase i.e. stable phase under a girren set of conditions can be ascertained fronr
the P-T, P-G, G=T or P-G-T diagrans for such pure components as described in the
previous chapter. The lines separating the two adjoining phases in such diagrams are
univariant i.e. with one degree of freedom and needs to be expressed mathematically tc
explain the equilibria involved. These are nothing but the interrelationships of two-phase.
single-conponent system with ternperature and pressure .

8"2 EQUILIBRIUM IN SOLID-LIOUID SYSTEM

Let us imagine a pure substance system at its melting point T when both solid anc
liquid coexists in equilibrium at a pressure P. lt also means that their free energis are equa
as Gl=Gs
Let us irnagine that pressure changes by an infinitesirnal ammount dP and as a result the
temperature changes by an amount dT to establish a new equilibrium and in that an
infinitesimaly small amount of solid or liquid may change orer to the other phase. lf the
associated free energy changes in solid and liquid phases are dGs and dGr and if the system
is isolated then these two free energy changes must match as
dQ = dGs (8.1)
The vapour prqssure of both under the given conditions are expected to be very small
and hence their fugacities can be equated with vapour pressures considering the behaviour
as like ideal gases. ln that case if dP is the change in the equilibrium vapour pressure of this
system the combined statement of the first and the second law can be applied to both solid
and liquid phases as : dGl = Vr dP - Sr dT (8,2)
dG5 = VsdP - S"dT (8.3)
It is also certain therefore that
VrdF-SidT=VsdP-SsdT (8.4)
or (Vr - V') dP = (Sr - Ss) dT
/dP\ _- /sr- ss\ _- /Afu_r.)\ (8.5)
lot/"' lu,- v"f lavr,-"il
At equilibrium AG = 0 and therefore AH = 16 g.
thus /dP\ _
l *,J., = TAll
^H (8.6)

where AH is the latent heat of mefting per g.mole and AV is the difference in molar
volumes of liquid and solid phases. This is known as "Clapeyron Equation" interrelating the
variations in temperature and pressure for maintaining the equilibriurn between the two
phases.
lf AV" -1 is positive , as is normally the case, and if AH" - 1 is a positive quantity as is

normal, *" (*+).. will be a positive quantity i.e. as pressure increases the transition

temperature should also increase.ln other words the equilibrium melting point shall
increase with pressure and vice versa. However for ice-water system AV. - r is negative
and hence is negative and which is fully exploited for the game of skating on ice
(:+)",
wherein the skate pressure increase shall decrease the m.p. of ice and hence help maintain
ice without melting for skating purposes, i.e. skating is readily possible.

8.3 EQUILIBRIA RELATED TO LIOUID.VAPOUR SYSTEM

ln the previous section solid to liquid change was logically analysed to arrire at some
mathematical relations. The same logic can be applied here to analyse the liquid-vapour
change. The same thing need not be repeated here but the equations shall differ only with
respect to change frorn solid to liquid and from liquid to gas phase as follows:
Therefore as in Eqs. (8.2) to (8.4)
(Vv-VtdP=(Sv-StdT (8.4
dP _ AS(v*tt
dT - AV(v- /r
andsince Vv >> Vr (difference of liquid and equivalent vapour volumes)
therefore AV = Vv - Vr = Vv

92
 / dP\ AHv
or ldrl =
TVv
{8.8)
\ /eq
as usualy the vapour pressures are very lcrrv, it can be assumed to behar,re ideally then vapour
pressure will be y= RT/P as per ideal gas equation and hene
/!ly) -= Pv' AI{v where Pv is equilibrium 'apour pressure;
lorl"o Rf
or /dpv\ = A4 . -'or or dlnP dr
AHv
= *T' (B.e)
l,*,J,, Rr'
'This equation is one of the most famous and useful equations dealing witn vapour pressures
of
and is known as 'Clausius-Ctapeyron' equation. lf AHv is assumed to be independent
temperature then on integration it leads to

ln Pv = Ary * I where I is the integration constant (8.10)

RT
positive
which correlates the vapour pressure with temperature' Since R,T and AH are all
quantities vapour pressure of any substance increases with temperature and vice versa:
3270 C are
Example g,1 : The densities of solid and liquid lead at its normal melting point of
The heat of fusion of lead is 1190 caUg at its
found to be 11 .05 and 10.645 g/c.c. respectively.
pressures practically remains the same.
melting point. prove that the m.p. of lead at erren high
to derivatives
Solution : The basis of solving this is to correlate the volume change on melting
of pressure and temPerature'
The gram atomic volumes of solid and liquid lead can be calculated as

solidlead = **# = ffi = 1e'8oecc

and similarly for liquid lead is = 19'445 cc
dp 1190 * 20_7 ,,, ,,_, _ 6443.5cay0olcc
=
dr=@--va
changingfromlcal - 41.193ccatmos = -2657?atmos/0o
or dT/dP = 1126572 = 376 x 10'7 0o /atmos
When P = 1 atmos m.P' of lead is 3270 C,
when P = 100 thendP = l0OandthereforedT = 3'7 x 10-3 oC
1 00 atmosphere pressure
Hence the melting point of lead shall increase because of application
and hence it practically remains unchanged'
UV onV O.OO37. 6 and which is negligibly small
l*rrnpr" 8.2 : Vapour pressure oiliquid zinc above its melting point of 42oo c is giren by :

logPzn=ry+ 5.5
1400 cailg atom'
The heat of fusion of zinc is 1600 cafig atom, for abow the m'p' and lt is
solid zinc below the m'p'
belcrrry its m.p. Derive a formula for vapour pressure over
Solution : For temperature above m'p' 4200 C
dP$ql oHY
P -= RTz = d rn P(,iq)
and for temperature belcnl its m'p'
dP(sol) AHs
P - = d ln P(sol)
RT2

93
at the m.p. of 420 C both are equal and eliminating the natural logadthm

log pzn (sorirr) = ,on pzr

ffi {rq}

and putting values as given = + 4.9

8.3.1 Vapour Pressure . Temperature =P Relationshlps i

As AH is expressed in terms of ACp values and since Cp is expressed in a gener-

alised form as:
CP = a + bT + clacaVrnole/Oo
the enthalpy of vapodsation is expressed as :

AHy=AHo+aT+pf+l]3+ (8.11)
where AHo , cr, p, 1, etc. are all constants for a given substance. This leads to

rnP= -#*ffmt**t* *f +G (8.12)

or in a nnre simplified form

lnP = f - AbgT + C (8.13)

lf log p are plotted vs inverse of tenperature then straight line plots o\ er a certain tenperature
range ?s shown in Flg. 8.1 are obsened experimentally. lf such an experimental data are
available AHv br substances can be calculated by evaluating the slope of these plots.

8.4 EOUILIBRTA OF SUBLIMAIION

The abore equations can be suitably modified to develop the equation forthe process
of sublimation r.e. the solid-vapour equilibda as:
/ dPv\ hrAHs (8.14)
l*J..= *rc
where AHs is the enthalpy change of the sublimation process

or idlnPv\=-DAHs (8.15)
t-aT-/ Rf

tn py = - ff - c fintegration constant) (8.16)

At the b.p. or the sublirnation point the value of Pv is one atrnosphere and if AHs value is
known then the integration con$ant can be evaluated and it is used for other calculations
thereafter.

8.5 VAN'T HOFF EQUATION AND EQUILIBRA

The equilibria so far studied and the equations dereloped to establish the equilibdum
relationships of vapour piessure with temperature can also be derir,ed from well known van't
Hoffs equation as follonts :
the equilibria under study can be described as :
liquid +
vaPour (gas Phase)

94
 A(n

uf
o H
d -25
\rt
o
6 3\
o z.\
J z\ ul\
$r\ a\
F\ o\
o\
('\ o\
z\
I\ t\
a t2 B\
l- \t
\o
\o
5
TE MPER.ATU RE
Fig.8.1 - The variatbn of vapour pressure with temperature as lag p vs. 1/7.
The near parallel lines and inverse slope are.characteristics of this
diagram.
and the equilibrium constant K = pv/a where E is the activity of the component in liquid phase
and since it is pure liquid phase therefore E is taken as unity,and hence K = Pv. Putting this
value in the van't Hoff's equation giws
dlnpl _ _ AHv
d(n - R
ssuning Al-& to be constantwith tenperature then
AHv
ln p - - fr7 +I (integration condanQ as before'

thisis the sanre equatinn as in Eq.(8.10) derived earlier.

Similarty the sane equation can also be derived from Mar<well's relations given earlier in
Eq.(4.4i1).
 8.6 TROUTON'S RULE
It has been observed from experimental data that the values of AHv varies linearly
with boiling points and that the plots obey broadly the retationship

+!Y
IB.P.
= 21 cat/mote/oo (8.14
This is shown in Fig. 8.2. This was first noted by Trouton and hence is known as 'Trouton's

.t
o
E STOPF r 2l CaLlolt. 5,7
Fig. 8.2 - VariatiCIn of latent lrcats of vapor-
3 isation of substances as a functinn of their
Y

normal boiling points to indicate the va-

lidity of Trouton's RuIe.

BOILIN6 POINTS 9K

Rule' stating that the entropy of vaporisation of any liquid is approximately equal to 21
cailmole/Oo K. Therefore at the boiling point this is the measure of increase in randomness on
vaporisation and which is approximately the same for many substances including both the
metals and their compounds
This is useful because the value of AHv can be evaluated for a substance approxi-
mately at the boiling point using this rule. For example at Py = 1, AHv = 21 x Ts.p. and
therefore on substituting these values in Eq, (8.1 6) gives the integration constant C = 1 0.57.
Using this value of C now vapour pressure of any liquid can be calculated at any other
temperature.
Example 8.3 : The normal boiling point of copper is 28680 K. Galculate the wpour pressure
of copper at 1100o K. \{hat are the assumptions made in solving this?
Solution : The vapour pressure is one atmosphere at norrnal boiling point Hence
ln 1 =
1.e#*s68 . c
assuming that the obeys Trouton's rule AHv = 21 x 2968 cal
"yri"*
The value of integration constant is.thus C = 10.56B
The vapour pressuie at 1100o K will be
21 x 2868
log p, =
1.987 x 1100
+ 10.568 or pv = 4.6 x 10-7 atniospheres.

8.7 RAMSAY.YOUNG RULE

It has been observed that for two chemically similar liquids like same family of organic
groups or inorganic liquids like oxides, sulphides, etc. there exists an approximate relation-
ship of \rapour pressure and temperature. This can best be illustrated as follows :

96
 Let two such similar liquids 1 and 2 have same vapour pressures at temperatures
Tl
and Tz respectivelY and therefore
lnpr = lnpz
or AHvr' AHv(zt
=ff+cz
*.iy+C1
or +=tHffi-(cr-c')**
similarly if fi and fz areanother such two temperatures at which the vapour pressures of both
are Pv and hence it reduces to

* = #ffi - (cr -c')#*

subtracting one from the other gives
11 Ii - (cl-cr)
i i, AHvtz) ft-fz\
- \'
and therefore
experimentally it has been found that the right hand side is almost zero
n-t'l (8.18)
Tz- Tz
temperature at whieh the chemically similar liquids have the same vapour
or the ratio of
to as
pressure is a constant. This is known as 'Ramsay - Young Rule'. This is also referred
;Duhring,s Rule'. Conversely 'the ratio of boiling points of two ehemically similar liquids
should have the same value.
8.8 RTCHARD'S RULE
It was observed from experimental data that the heat of fusion
for various substances
observed as
varied approximately linearly with their melting points. The slope was
:

(8.1e)
{xM
IM.P
= z - 4 caVmole/oo
and since entropy
This is shown in Fig.8.3. This is equalto entropy change on melting
materials on melting is
is a measure of randomness of materials, this randomness forvarious
almqst the same. This should be obvious'

8.9 EQUILIBRIUM PHASE TRANSFORMATIONS IN SOLID STATE

gas phases can
The equations derived so far for equilibria between solid, liquid and
transforrna-
be suitably modffied for change in phase in solid state like austenite-martensite
equilibrium in the gas phase' lf
tions. These phases shall have their ovvn vapour pressures in
the solid phases are designated by numbers 1 and 2 then :
/ dP\
Sz - 51
ASlz-r1 (8.20)
=
l*,|". Ul*
= Av(z-1)

97
 / dP\ AH(r-z) ...
where aHlr -zy is the enthalpy of transforrnation and
[ * j". = ffffi
V1z-r; is the molar volume change accompanied in the change.

+
I
I

!r gt opi. afatlcK M.A

Fig. 8.3 - Vari"arton of latent heats of
:,
d
melting of various substances as a
!J
v2 function of their metting points to show
r the validity of Richard's Rule.
<l
I

TEMPERATURS OK +
MELT'T'A POINT.s

8.10 VARIATION OF VAPOUR PRESSURE WITH TOTAL PRESSURE

Let us irnagine a liquid in equilibrium with its vapour at ambient pressure P at a
iemperature T. lf P is one atnbsphere then Pv shall be the true vapour pressure of the liquid
at tenperature T. Let us imagine that the system is flooded with an inert gas to increase the
total pressure of the gas phase in equilibrium with the liquid. What will be the vapour pressure
of the liquid at the same temperature? Will it be less or more? lt is bound to increase with
increase in total pressure and decrease if the total prassure decreases, like application of
vacuum as per the le Chatelials principle.
Let us therefore imagine that the system liquid and vapour are at equilibriurn at
temperature T and pres$lre P when p is the true vapour pressure.Let trs also imagine that
rpw the total pressure increases by dP and conespondingly the vapour pressure increases
by dpv
Using the combined statement of the first and the second larrs the change in the free
energy of each of the condensed and the gaseous phases as a result of this infinitesimal
change is given by as follors:
forgaspnase dGg = VsdP - SsdT
and for condensed phase dcc = Vc dP - Sc dT
considering isothermal conditions
dQ = Vg ' dPv 'where Pv is vapour pressure
and dcs = Vc dPr where Pr is the total pressure
ln an isolated system like this therefore the two free energy changes shall be equal and hence
dFv _ Vc
(8.21)
dP1 - Vg
it reduces to

since Vg is quite sanilland.can be assumed to behave ideally,

therefore Vs = RT/Pv
or putting the value of Vs gi\,w :# = *?
onrearranging H =Soe, or dlnPv =ff ae, 18.221

or otr a ln Pv =
S o oe, (8'23)

thus, as the total pressure increases the vapour pressure also increases and what wc
qualitatively suggested eariier is prowd here. ln the Eq.(8.22) above the A ln Pv is the
difference in vapour pressure at any total pressure Pt and that at any other reference state
(nornnlly one atnns) on log scale and A Pr is the difference between total presure ard
reference state pressure (one atnns)'
Example 8.4 : Calculate the vapour pressure of zinc under 10 atmos pressure at 7000 K. The
norral b.p. of zinc is 9000 C and density of zinc is 6.5 g/cc. At,wt'o{ zinc is 65.
Solution : Assuming Trouton's rule
AHs (zinc) = (900 + 2731 x 21 = 24633 caVg'mole
taking lnP = - AHS/RT +- 10.57 (wherc 10.57 is the inbgralion constant)

putting the values ln P = 12'4 x 10-2 atmos.

= 12'4 -x '10-2 atmos'
This is true vapour pressure and hence ln p,,
Molarvolumeof zincis 65/6.5 = 10cc = 1 x 10-2 litres
APt = Pr - Pv = 10-- 12'4 x 1A-2

Putting values in Eq. (8.22)

sives ArnPv=|--fl#)
= 0.001616 or 16.16 x 10
Since AlnPv=lnPv-lnP or lnPv=AlnPv+ lnP

or tn pv = f, r t ro-:trf_rll-to-')l
zoo ,._.. x
+ (re.+ 1o_2)
l--T082 " )
= 0.1256
or pV = 1.13 atnros under a total pressure of 10 atnnsphere at 7000 K.

8.11 CONCLUDING REI\'[ARKS

- lt is seen that it is easier to develop mathenratical equations to establish the equilibrium
relationships of phase equilibria in a single component multi-phase system and from these to
evaluate the vapour pressures of substances as desired. The equilibria in multi-cornponent
systens have not been derived to awid elaboration and complexity.

99
CHAPTER 9

SOLUTIONS - I
Gibb's-Duhent equation and activity laws

9.1 INTRODUCllON
Strictly speaking pure element is only a imaginary concept for it does not exist in
reality. Even the so-called 'spec pure' (spectrographic standards) elements do contain
impurities, may be of the order of parts per million or parts per billion or perhaps parts per
trillion, But the levels of these impurity;contents are such that these can be
disregarded for practical utilitarian purposes. However the fact remains that eren the
so-called pure elements are nothingbut'mixtures' and since these are homogeneous single
phase materials, are referred to as solutidns. ln practice therefore we deal with
solutions alone, whether liquid, solid or gaseous, may be dilute or very very dilute or otherwise.
The simplest possible solution is a gas mixture beause all gases together form a
single homogeneous gas phase at all concentrations of the components. Forrnation of
liquid solutions from two liquids, or liquid and solid, or gas and liquid and so on are
innumerable. Formation of solid solutions are also not at allexceptionaland many usefulalloys
or ceramic materials fall in this category.
The composition of a solution is expressed in weight percentages or, atom or mole
percentages or, fractions and, partial pressures, particularly for gaseous solutions.
number of atoms i
(atom%)i =
total number of atoms in solutions
ni
n1+ n2+ n3...+ ni
The individual number of atoms can be worked out by dividing the weight percentages by
their respective atomic weights. Hence
(atom %)i _ (wt %)i/(at.wt)i
- (,ut"/o)t /(aLwt)1 + (vtt"/o)z/ (at.vfi)z + . .. twtl"lrZ(afwtli
The molar percentage can be obtained similarly by taking into account moles in
place of atorns and molecular weights in place of atomic weights in the above expression.
A solution may consists of two or more components of which the largest in concen-
tration is referred to as solvent and the smaller ones as solutes. A multicomponent solution
can bedescribedas consisting of atoms or molecules of components 1, 2,3,.... i as
(nt + n2 + n3 + . + ni ) total atoms or moles. lf the solution contains only two
 ternary and so on. The minor uncontrollable
components then it is termed binary, if three as
a large number of components are there like
impurities contained therein are disregarded. lf
solution'
in iron and steel then it is a multicomponent
AccordingtoDalton.slawofpartialpressuresforagasmixture
(9'1)
pi =Ni P
of the component in solution and P
where pi is the partial pressure and Ni atomic fraction
thermodynamic properties of a solution are
is the total pressure of the gas mixture. The
components because of interaction of the
most often quite different from those of their
need to be studied precisely' The thermo-
component atoms/molecules and hence lhese
dynamic properties of components in solution is best expressed in tern's of their partial
properties as explained earlierin section (4.2). By
taking free energy property forthe purpose
of illustration it has been shown
that G' = f+)
\oniJr,Rn, nr,n.,.. ni- 1

and the total property G'of the solution can

be obtained bY summation as
(e.2)
(G)'rp = )i n' 'n,

6 Ni = ----
n1+ n2+.+r--
n3...=- ni
i_ (e.3)
thenmolar G=>; Gi .Ni =GtNt+GzNz+""'+GiNi
Similarlytheotherextensivepropertiesofthesolutioncanbeexpressedas:
i_ (e.4)
Molar volumu Vi .Ni = Vr Nt + Vz Nz + " " ' ViNl
).'
":
Molar enthalpy )' Fi 'Ni = Hr Nr + Hz Nz + ""' Hi Ni (e,5)

": I
(e.6)
entropy Si 'Ni = St Nt + Sz Nz + ""' Sit{l
Molar )'
i_ (e.7)
MolarHelmholtz'Lt1 Ai'Ni = Ar Nr + -fu ltz + ""' AiNl
free enqgv and so on'
,/
3.{ ctaB's DU H EM EQUATIoN pressure and ternperature gets reduced
The differential equation (4'Q-0)' at constant

to dG' =
ffi)ro*".,.. n,'on' i (#)r,*"1,n3, ", n
" dnz +

/aG'\ (e.Bi
* + "" + "on'
lffijr*"1,n2,. " ni'dns 'r#)rr", ", ". n, '

= 6r dnt * f,2dnz a f,3dns + """ * fr; dn' (e.e)

Similarly the total property of a solution is

101
 (9'10)
G'=nr6r+nGz+-ngGs+"''+niGi .'
or dG' = tn1ae1 --gOGz + ngdGg + nidGii + (Gr dnr +
a {2dnz * Q3dns + ""+ Gidni)
(9'10a)

On combining E3.(9.9) and (9.10a) leads to' .: (9"11)

ildGr+nzdGz+nsdGs+""+nid6i =0+ n2 + n3 + ni)
i.e. (nl ' ' '+
dividing by total numUeiof atoms/molq:in the solution
il; "-' r.rr-oe.r 1 [2dGe + NsdG 3 + NidGi =
""'+ 0 (9'12)

tni" i" knorrn as Gibb's Duhem equation'

1 and 2 then
lf the solution is of only two coFnponents nanrely
Nrdbr+NzdGz - o
(9'13)
dG 1 (9'14)
or Nr dG.r = -'ru, oG-, = - (1-N1) dG 2 = (1-Nz)
all other extensive properties can be
By the same treatment Gibb's Duhem equation for

;"H;.:::,:;",n, Nrdul + Nzdlz + Nsdus"" Nidui =0 (e'15)

forentropy Nri-qr.f.fzC$z+NedSs....NidSi =0 (9'16)

forworktunction hit;At*rutdAa+Nsd!3""NidAi = 0 (9'11

forvolume Ni OV1 +Nz*!z+NsdVe"'"NidVi = * (9'18)

forenthalpy ildgt*rutOtrz+NedFis....NidFii - 0 (9"19)

and so on. ln fact the Gibbis Duhem equation

lor !:inary can be stated for any evtensive molat
propefi Z as:
Nl dZr + NzdZz = A {9.20}

FROM MOLAR OUANT1T1ES

9.3 DETERMINATTON OF PARTIAL MOL'AR QUANTITIES
OF SOLUTION
ThemolargeneralextensivepropertyZatatwocomponentsolutionisgiven
Ers z'; it?t * NzZz {g'21i

differentiation giws
67 = Nt Z-; + N2 {Zz + Zl dNr + Ze dt'{z {s.22\
solution as given in Eq.(9.20) when added to
Gibb's Duhem equation for a two component
this leads to (e.23)
67 =7t dNr + ;2::Si-:
Nol in a binary solution Nt + ire - i
hence dNl+dNz = 0 or dNt=-ditia
by Nr /rtNz gives
muttiplying both sides ol the equation {9'23}
(e.24)
N1 7t+NtZz
ffi=-N1
from equation (9'21) Nr Zr = Z - NzZz
substituting this value in Eq. (d'2+1 iives
(e.2s)
Nt
ffi - - z + NzZz + NrZz
= -. 2 + Zz (Nr + Nzi
(e.26)
--Z+Zz
Putting Nr = (1 -Nz)
dz p.2n
2r=z+{1 -Nz) etNe

102
and simihr$
dz
Zt=Z+(1 -Nr)--_ (e.28)
dNr
For a multi-component system it should be

2.r = z + (1 - 1,-1.,y /i{; (e.2s)

\dNtl,u,*"...*,
It only means that if total molar property of a solution, as it varies with composition of one
component, is knorun, then the prtial molar prooerty ef other components can be found one
by one using the above relationship"
I
9.3.1 Graphical Method
The variation of molar extensiw property with conposition of a solution is
shown in Fig. 9.1. Graphically from such a graph the partial nolar property of a component
1 and 2 can be obtained by drawing a tangent on the curve at the relernant point indicating

Fig 9.i - lhrintbn nf any molm

extensive properg of a binary solu-
Fi tian as a function of composition
Ti
>i
2
lt

I
i
i
-,._" J

the composition at which the partial pr;iperty is required. The tangent is shown in Fig.9.1
at point A when the conceniraiio* i: flr . lt strikes the Zaxis at E at 100% Nr and at F at
100%Nz composition.Theint*rcepts at E and F give the_values of the conesponding
partial properties. The partialproperties are shou asZr and Zz as DE and HF respectively"
Graphically Zt=DE=DB-BE=q4-gg
GA = Z as per the graph
and BH =.BA
E; = i1 - Nri stope of the graph ar A
, i1 -Nri tr
Naut Fr=DF,-z-.. ij N1)
*Z (e"30)

and sirnilarlv 2z = HF = I - {t Nr}

aff; (e.31)

103
Thtrs it pro\€s that the tangent.drawn on the curve at the relevant composition point, g1e
intercept on the ordinate gives the partial molan quantities.
Also at Nr = 1,Nz = 0 and Z =2t (e.32)
or it is the sarne point where curve cuts the ordinate i.e. at C.
Similarlyat Nz = 1, Nl = 0 and Z =22 at point J (e"33)
9.32 Additional Relationships of Partial Properties
For a solution of large amount G' = H' - TS' (e.34)
on differentiation
1aG'\ ont 1aH'\
= |.frjrrn2,n3,...
|.6,lr.".,".,... n, ni

-r /39\ te.35)
ldnt/Tp,nr,nr,.. n,
Itmeansthat Gr = Fir - TSr (Q.86)
It also means that the relationships valid for free energy of a solution is also valict, in the same
fonnat, for iqldividual components irithe form of the relevant partial molar properties. Similarly
therefore
Ht = Ur - PVr (e,34
Ar = Ur - TSr (e.38)
ln the same way the cornbined staternent of the first and the second laws of therrnodynarnics
for the solution can be expressed as
dG'=V'dP-S'dT (e.3e)
or similarly dE = VidP - SidT (e.40)
and all such relatlonships shall be valid for solution as well
like f-aG'\ v' (e.41)
l,F)r,^',,",.. . ni
and fH
\ /l'nt,na,.'.nr
=-s' (e.42t

Arso /aB')
r aP l- - - Vi
=-f te.€)
\ /l,nl,nz,...ni
and 1@)
larl- =-Si (e.44)
\ /l,nl,na...ni

simirarrysince thus
ffi],:* l+ff],-- (e.45)

9.4 IDEAL SOLUTION

9.4.1 Raoult's l-aw

ln the early days the study of solutions was lirnited to the measurements of vapour
preessures of solutiens at various concentrations sf the dissolred cornponent in binary
solution systerns. These ryleasurernerits were neadily feasibte by actual experimental
vtork. The observations so made were codified byRaoultintheformof astaternent, linnited

't04
to birnry sokJtians ody. 'The bwerlng of the vapo{rr Pressune of a solve!* due to tle
addifion of solute is equal to the mole fraction of the solute in the solution," suggested
Raoult.
lf pr and pz are partial pressures of components 1 and 2 in a binary sclution at a
certain temperature and if pf and pE are the vapour pressures of pure cornponents 1 and 2
then
tr9- p. A=t-Nz=Nr
".-=NzOf 1-1=N.
71
or
P?
(e.46)
P?

and ll:-P'=ry1 or t-!=rur or \=t-Nr=Ne P.47)

P2
where Nr and Nz are the mole fractions of the components 1 and 2 respecti'iely and that
Nr + Na = 1. lnageneralisedwaytherefore
Pl
- Pf or since pP is fixed at a certaln
trti =

temeprature and pressure pi c, Ni (9'48)

or vapour pressure is directly proportional to its mole fraction. lf the vapour pressures
are low, these can be approimately equated with respective fugacities and hence

Ni = +=
PPf,1=3i
(9'49)

This mathernatical statemenrt is known as Raoult's Law and is often stated in the form that
.activity of a solute !n a solution is equal to its atom or mole fraction'. This is shown
in Fig g.2 by the diagonal line as the Raouft's law line. The activity of a component is equal
to itj actual atonn/mole fraction in the ideal solution and hence Raouhian behaviour is
considered as an idealsolution behaviour'
Those solutions which obey Flaoult's law are therefore known as idealsolutions and
thisholdstrueat all temperatunesandpressuresatwhichthesolution is capable ofexisting.
The solution tends to behave ideally at loru vapour prqssures because the vapour then tends
to behave as an ideal gas'
ln reality howewrthe solutions hardly obey Raoult's law orrerthe entire composition
rance.Fiaoult's law is seen toholdtrueonlyif theconcentrationofthesolvent is too large

d
Fig. 9.2 - The activity-composition
I relatbnship to show Raoultian and
F Henrian behavioux The deviation
l--
() from ideality is also shown.

o'6
ATOM FRAGTION

105
and obviarsf frat of the solvent too small. ln other words it holds true in dilute soltdions
only. ln rnn{iltrte solutions the behaviour deviates considerably from the ideality in both
pcitive and negative direction. ln.the deviated form the activity is either more than (+ ve
d€vidion) or less than (- ve deviation) the ideal values. The negative deviation is shown in
pracfbe by many solutions like Fe-Si, Fe-R Fe-S, Cu-Sn, Al-Zn, etc. The positir,e deviation
b strotn by relatively fewer systems and is typicaly illustrated'by the Fe-Cu system.
Near ideal behaviour is shown by Fe-Mn, Fe-Ni, Fe-Co, ionic melts con-
tahing gommon ion like Feo-Mno, FeS-Feo, FezSioa -MgSio+ and the like.
9.42 Process of Mixlng and Connponent lnteraction
ln pure state atoms are surrounded on all its sides by like atoms. Thd process of
mixing of components changes the atomiclmolecular configuration such that now unlike
donrs are available to sugound it and such number of unlike atoms surrounding the
dom increases on increasing the concentration of the solute in the solution. This gives rise
to short range as well as long range ordering even in liquid solutions in certain preferential
way. This short or long range ordering basically arises due to different atomic structures
of the component atrnos and it nanifests as interaction of the components in solution, The
solution formation is a natural process at least to some extent of concentrations of the
conponents. Hgnce the components have natural tendency to form short Qr long range
ordering between unlike atoms.
Almost the entire volume together of the components is available for component
atonrs/molecules to move. ln other words thb entropy of the components on solution
formation positivelY increases.
Allthis leads to changes in the energies of the constituent particles on forming
a solution and hence energies will generally be liberated or absorbed depending upon the
above changes with reference to the characieristics of the component atorns inrrclved in
forming the solution. These therrnodynamic changes by no means are insignificant and
treating the solution simply as a mechanica! mixture of the components can lead to
Yery erroneous deductions.
9.4.3 Solution Models
ln a solution the activity of a component should ideally be reduced down to ils own
oropcrtion i.e. atonVmole fraction. But the real behaviour is quite different, Solutions exhibit
wide ranging variations in the activities of the component which in turn means equally wide
ranging thermodynamic properties as has been experimentally observed' There is there-
lore necessity to explain these obsenations in ternrs of some models based on

irteraction of component atoms/molecules in solution'

Three main theories have.been suggested to e:<plain this namely {1} ti:e classical
thermodynamic model assuming formation of cne ci rnore c'f complexes, {2} ti;* statistical
thermodynamic model assuming free vclume for vi'nration *f each particle lrl a soiution and,
i3)thequasi-chemicaltheorywlrichtakes!nr';aecc)''|r!rlih*--pref+r'e"'li-'distribrr-;tionerr'
corrconent particles in terms of bonCs in :. sliutici:

",c,
 Various assumptions have been made in developing these models' Atoms are
atonrs
supposed to have a discrete size, Valency and electronegativity. \Mren two or more
atoms in
torm a solution, the differences in these three properties between the connponent
solution giw 1se to certain effects. The difference in sizes of atoms give rise to
size factor
effect manifesting as strain energy. The valency difference gires rise to coulombic effect-
which is sornething akin to effect of groups of atons in the periodic table. The
electrone-
ln
gativity difference gives rise to chemical bond forrnation. All these result ln changes
energy on solution formation'
9.4.4 Quasi-chemical Explanatlon of Activities
of
Without going into the intricacies of these theories the qualitative behaviour
solutions can be reaoity understood from the quasi-chemical theory ol
solution' Hence it
belorru. The narne itself indicates that it is something like but not
quite in terns
is described
of chemistry and chernical bonds of the components in solution.
of
ln condensed forrn, liquid or solid, atonrs do link to each other through formation
in working out the
chemical bonds. The quasi-chemical approach relies rnore on these bonds
energy of solution as distinct from their components. This is a limiteci approach as the
atonrs coming together
energy of solution shall be in terms of all effects of the two types of
over a
and not merely in terms of their chemical bondings. The bondings actually extend
distance i.e. long-range ordering will have to be considered' The
quasi-chemicaltheory
does not incorporate ihis comptication either. Nevertheless it does
qualitatively explain the
behaviour of solution well and therefore needs to be understood'
in a class
This behaviour can be understood from an illustration. Let us inugine that
he can do so
room where several students are seated and if any one wants to run arruay
on him and stay
with the same vigour if the rest of the students do not exert any influence
try to prevent him frorn
neutral to,rlards him. lf one or more do interact with him and, say,
running away, he nny not be able to run away with his orun vigour
for it is now reduced
tendency" lf the other
depending upon the extent of force (bonds) exerted on his escaping
to run auray with added activfu
students help him to run airuay by way of push he will be able
o'nore than his own. The exact inffuence of the forces exerted by others on him will be a
on him and on others or
conplex one depending upon the forces exerted by individuals
amongst thenselves'
lrr asolution since the component atonE aredifferenttheydohave forcesof
different electronic structures'
attraction or repulsion amongst themselves, because of their
This is knoryn as interaction amongst themselves'
lnabinarysolutionsayo!componentsAandBtheinteractionspossibleareA.B,
or unlike atoms. lf the
A-A and B-8. This can be summed up as lnteraction in like atoms
both components may
attraction in like atoms is more than that in unlike atoms then
dcminate
positivb deviation from ideality. On the contrary if attraction of unlike atons
""nili
then both may exhibit negative deviation from ideality. This attraction
reduces 'the escaping
shall be reduced to the
tenrJency of the components to the surroundings and the
activity
shall be seen in the activity'
extent this attraction dominates and hence negatiw deviation

107
 The srtx$rrr beco{rFs little more conplex if
a third elerrerd say c b & b
the tlinary solution. The eartier interaction of A-A, ts-B
and A-B are now influenced by
the interadion A-c, B-c and c-c and whaterer are
the net final fnteractive forces, to that
erient, the final ideal behaviour shall be affected and
accordingly the positive or the
negative behaviour shall be observed. The situation
becomes far more complex if more
than three components are invorved Le. inmurti-componentsystems.
The Raoultian behaviour willbe exhibited by'the components
if the atons are
similar in nature so that the interactions of like and unlike
atoms are correspgndingly almost
similar as if in their pure form. Strictly speaking such a
situation is non existant because no
two different atoms are alike. However if they are
the examples of which
almost alike,
are quoted earlier, the behaviour observed shall be practically ideal.
such cases are
obviously far.
on the other extreme if two different atoms are so widely
different that strong
interaction exists and in consequence intermetallic compounds
like Ax By tends to form.
Equally true is that if the unlike atom interaction is too
feeble and hence like atom attraction
dominates then the solution may exihibit miscibility :-:ap as is seen typically in silicate
metts.
9.4.5 ActivityCoefficient
From the foregone it is quite clear that the ac:ivity of
a component can be equated to
its atonVnple fraction if ideal in its behaviour, but if there is a deviation then what is the
actual
activity at a given concentration?
in idealsolution ai=Ni
the non-ideal behaviour is covered mathematically as

where v is the multipv,tq?lilL.,XnJLn9" the atom fraction Ni into the rear varue
This muttiplying factor is known as 'activity coefficient'. The ", "l?;lil
activity coefficient is not a fixed
value, it varies with concentration as will be obvious from
Fig. g.2. Most often dilute solutions
are under consideration and wherein the non-rdeal behaviour conforms
to a straight line
behaviour indicating that activity coefficient y has a fixed value
for a given binary solution. This
activity coefficient is often reffered to as yf to indicate that
it is valid for dilute concentrations
of solute only and hence is distinctly different from y for
any other higher concentrations.
Activity coefficient has a value at unity when ideal behaviour
is observed. For
negative deMation, fwill be less than unig while for positive
deviation, fwirr nefr*at*r tnan
unrty.

9.4.6 Henry's l-aw

The behaviour of a solute. crt dilute concentrations was found to obey

E- Ni * ,P
traiqNi
and was ndieed and stated and henice is known after its
observer as ,Henry,s
sld $e aetivity of a compoll@B1r-: ln dilute solution is proportional to
l:w,. j?j]
lt

fractbn as distinct from Raouit's Law. ln reality it has been observed that for
its atom/nrole
dilute solutions,
ilhe solute obeys Henry's law whereas the solvent obeys Raoult'larrv
(being at very high
108
 far too apart from each other and hence negligible interaction occurs to afiect
Raoultian
activity.
9.4.7 Standard States and their lnterretations
By definition of activity itself, it is obvious that it is a numerical fraction
to express the
interactive abiiity with respect to sorne standard interactive ability. The interactive
ability
is maximum when the material is in its purest form. This pure form can be solid, liquid
or
gaseous form of that substance, as the case may be under given conditions
of temperature
and pressure' This is purely arbitrary and without any scientific basis, though perfecly
alright' The activity of a component in such a pure form is assumed to be unity and
the rest
are expressed as its fraction when in impure form. The condition of the component
in
which form the activity is talien as unity is defined as the 'standard state' of that
component
for the purpose of calculation of its activities when in non-standard condition.
From the definition of standard state any other state of conr,enience, other than the
pure form can be arbitrarily taken as standard state. The Henry's larry offers
two such
alternative standard states. One of which is the infinitely dilute atom- fraction
standard
state- This is so chosen such that the Henrian activity approaches the value of atom
fraction at infinite dilution. lf the Henrian activity is depicted by 'h' (as distinct from ,a, used
to denote Raoultian activity) then
h= Ni as Ni ---' 0 (g.S2)
The Henry's law line as shown in Fig.9.3 is the ideal line for Henry's law and the vaiue of
unit activity is shown by yP on the y-axis. Actually the Raoultian activity
a=yfNi
where yf is the activity
coefficient, independent of concentration for dilute solutions. On
Henrian scale hi cr Ni
and hence hi = yP. Ni (e.53)
But yfl is the Henry's law coefficient. On assuming the dilute solution standard state for
Henry's law hi = Ni as yP = 1

z
:<
Fig.9.3 - Depictianof Henry's law scalevs.
c
zU Raoult's law scale to dffirentiate them
a
II
C
5
r\i
z
,.

r"rt %
.
r. too%
L

109
Bd f b the Henry's law coefficient. On assuming the dilute solution standard state for
Flenry's laru hi = Ni as Yl = 1
This 'rs one of the standard states on Henrian scale. lf there is a deviation from this behaviour
then mafrenratically it is equated as
6; = Nl . i t
where is the activit
coeficbnt on Henrian scale with relerence to infinite dilute solution standard state.
Abo therebre t"hi= r.t,
(e.54)

The intenelationship of these two standard states, nanely one as pure substance standard
state and the other as infinitely dilute solution standard state can be assssed
as

activity of i relative to pure substance standard state \lqfl -= t'

=-N' "f
acini
while dealing with chemical or
physico-chemical processes the energy ex-
pure f"e. at
change in the process would depend upon the initial state of the system, whether
free energy
unit activity or impur6 i.e. alalower activi$. Therefore it is necessary to know the
state and it is
change inrolred in transferring one mole of i from one to the other standard
given by

AGo - Gl (pure subsiance standard stab) GP (AiUC solution standard stab)

of i (pure substance standard stat-e)
AGo = RT ln' factivity of i (dilute solution standard state)
lactVity
= RT ln yP (9'55)

This is still far too complicated to deal with in practice for evaluating activitles or
energy exchanges.
ln a nrulti-cornponent system the Ni can be evaluated only after complete
chemical
analysis of the solution to its minute details including the small impurity concentrations'
This is rather too difficult. A practical way out is obtainable from Henry's
law by way of
assumption of one weight Percent concentration of a component as its another
standard state. lf this is assumed then 1wt% concentration will have unit
activity on
Henrian scale. Strictly speaking it is

hi = .(#j)rm*wr6*o
therefore at lolrler concentrations
hi = wt96 i as shown in Fig' 9'3'
The activity values are thus directly the values of chemicalanalyses
of components in
components otherthan the required
dilute ranjes obtained as weight percentages. No other
ones need be analysed in this case unnecessarily'
coefficient
lf the real behaviour deviates from this ideal Henrian behaviour, activi$
can be introduced as before as
fr; = [wt% i] x t
*- - hi (e.s6)
" wtToi
110
 The utility of this practical standard state lies in knowing the free energy associated
with the production of 1wt% dilute solution from a pure component. lt is obviously related
to the nature of the solwnt and will have to be assessed for each solvent separately. For
iron solutions it can be shown that the free energy for changing the component from pure
state to 1 vi"/" dilute solution in iron, using Mre and Mi as atomic weights of iron and
component i, is given by :

yf + RT ln [Mre /Mil
aGo = RTln
= (4.575 T log 0.5585yP)/Ui (e.54
The standard state is like a datum level for measuring the energy exchanges in
processes as absolute values are just not measurable. When reactants and products are not
in their pure forms any other datum lerel can be arbitrarily taken for the assessment of energy
exchanges" So long as all measurements are made withthis new reference state, problens
do not arise. But if some rneasurements are made with one standard state and the others
with other standard state then they can be transferred from one to the other standard state
by knowing the free energy change for transferring frorn one to the other standard state.
Example 9.1 : Molten silver was contained in a crucible of Sis Nl under nitrogen and
hydrogen gas mixture at 1400 C for the reaction
< Sis N+> = 3(Si) + 2{Nz}
and AGo = 209000 - 96.8 T cailg,rnole
At 4foA Nz equllibrir.lm Si content of silver was 0.238 %. Calculate the activity coefficient of
Si in Ag-Sialloy at 14000 C
Solution : The free energy change of the above reaction is
AG?loo = 209000 - 96.8 T
= 47054 cailglmole

since * = dl,', !fr'

?SiaNq
where Prua = 0.4 and dSiaNa = 1 as solid

therefore K = (wt.% Si; x ySi x O.4

using AG = - RT ln K
47054 = - 4.575 T {log ([wt.%Si] x ysi x 0.41]
putting values in the above gives
log ysi = 7.169 or ysi = 0.855
9.5 ACTIVITIES IN MULTICOMPONENT SOLU11ONS
-shown of a component is affected by the
It has been earlier that the activity
presence of other solute comporpnts in the same solution because of their mutual interac-
tions. These interactions are far too complicated for conect predictions but it is necessary
that knowing the activities of a component in some simple systems it should be possible to
mathematically predict that in a complex solution.
Let the activity coefficient of solute B in a binary solution A-B be given on Henrian

scale as
.
TB =
hg_B
[wto6 B]

111
 if a third element C is added to this solution then value of fB is altered to the extent
C has
interaction with A and B and hence
rr$-e-c- h0-8, fg {e.s8)
where € i" tn"t parameter or multiplying factor that corrects the binary activity into the ternary
one. ln a way it represents the effect of C on B,
or h6-B-c=1wt%81 t f$-B-c (e.5e)
where fd - B - c is th" activity coefficient of B in ternary solution.
Itmeans hB-t-"- twr%Bl* €x fg (e.60)
Sincefa ilself mayvarvwithconcentrationitistakenas fB astheactualvalueatthat
concentration.
Itfollowsthat f6-B-c= tB* fB
That is the binary activity coefficient when multiplied by the factor, which indicates
the influence of C on B, the ternary activity coefficient is obtained. Similarly in a
quarternary solution when D is added the activity is given as

n$-a-c =
[wt% B] x
fB ,, tB € ' (e.61)
provided the influence of D on C in the form of € is negligibly snnllto be neglected. This is
true in dilute solutions. ln a generalised form therefore
fe= fB'f8'fB.. 6 (s.62)
for a multi-component dilute solution (A- B - C- D ., . . i ).
Taking logarithms in a multi-component soultion
togb=logfB+ bgfS+ bgfB......+ togfb (e.63)
The experimental data on the influence of concentration of third element on the solute i.e. C
or D or....i on B in a solvent A, indicate that for dilute solutions the log fb u". wt% of the
third element, the infiuence is typically shown as in Fig. 9.r$. When the solution is dilute the
relationship is straight line and the.slope is given by

slope of the line = qrq

wPh i (9.64)
This slope has been designated as interaction coefficient or parameter and is expressed
as eB
Therefore fB = ,. wto/o C
log (9.65)
"8
where e$ indicates the interactive effect of c on cbncentration of B. Hence
hB = [wt% B] x fs
or = i* o/o'Bl x- # x fg , fB 6
(s.66)
on log scale
tog ha = logfB + bgf8 + log fB + . bg 6 + flog wt"/B]
= . (wt%B) + eB . $n"z"c) + eB . (wtoloD) +
"B
.... ub . (wt%i) + log wt% B (e.64

112
 ERRATA

L /L )v ,+ * +

Dorrgb A&aA
EN LA R-C' Eb

t2 5+5
o/o
A Poel E ugrte nr
Irig.g.4.SchematicvarirltioncfactivitycoefficientofsclluteBasi$.luencad
by the presence of third (solute) element i represented by Nos' I to 8'
 It is therefore possible to experimentally determine actiMty coefficients in
temary
systens and from them compute using the aboveequation, within reasonable accuracy,
the
actiMty of any component in a multi-component system. Although this is valid
for dilute
solutions the resuhs in the case of even non-dilute solutions have not been allthat
wrong.
Sonretimes eB is known but not ef . This can be found out using Wagner,s
approximation for dilute solutions as

eg Me = eB Mc (9.68)
where M indicates the atonic weights.
lf both are known their accuracy can be verified using the aborc equation.
Exanrple 9.2 : Calculate the activity of sulphur dissofued in moften iron under
the folloring
conditions.
% elernent in solution interaction parameter
s - 0.08 .E = - 0.03
Mn -
"$" = -
0.75 0.025
si - 0.30 tg = + 0.065
P - 0.35 = + 0.042
' Cr - 0.67
"5
e8' =,- o.oo5
Ni - 1.s0 e$i = - 0.002
c - 2.50 = + 0.110
Solution : The or,erall activity coefficient "8
of S in the complex iron nrelt shall be
togfs = togf8 + togf$n + tog€i + log€ * tog€, + bgf$i + bg€
= eE [%s] * "Sn. 1wt%Mn) * . 1wt%Si) * 1wt%p) +
"Ei "E.
+ e8' .ptzocr; + e$i . (wt%Ni) * (wt%c).
x "g.
= (-0.03 0.08) + (-0.025 x 0.75) + (0.065 x 0.30) +
(0.042 x 0,35) + (- 0.005 x 0.67) + (_ 0.002 x 1.S0) +
+ (0.11 x 2.50)
= - O.OO24 - 0.00018 + 0.0195 + 0.0147 _ 0.0033 _ 0.0030 +
+ 4.275A
= + 0.3004
Therefore fs = 0.5
Hence the activity of sulphur dissolved in iron is
€lg = wt% S x fs - 0.08 x 0.5 = 9.64.
9.6 SIEVERTS LAW
Gases dissolve in mohen metals readily. The gases like nitrogen, hydrogen, arnrnonia
and oxygen do not dissolve in metals in molecular forns like

{Hz} = [Hz] (e.6e)

These howeter dissolr,e by dissociation of their molecules as
113
 {Hz} = 2[Hl (e.70)
or lNHsl = [N] -3[H]
+ (e.71)
The dissolution behaviour of diatomic gases like Nz, Hz and Oz can be illustrated by the
exarple of hydrogen as shown in Eq.(9.70) above. The equilibrium constant of this reaction

shallbe *"= * (e.721

Assuming Henrian behaviour and equating fugacity with partial pressures

fwt%Hzl
K = *u - PHe
(e.73)

or the amount of hydrogen dissolved in the metal is given by

[wt%Hf = KH x PHa
or the arnount of hydrogen dissolved in the netal is proportional
lwt%Hl2 cr(pHJ (e.74)
Sinnilar equations can be developed for oxygen, nitrogen and such other diatomic
gases.ThiswasfirstobservedbySievertandhenceisknownas Sievert's Law which states
that 'the solubility of diatomic gases in metals is proportionaltothe under root of partial
pressure of the gas in the ambient atmosphere' and is stated as
for hydrogen twt % Hl = KH (Pu, )"' (e.75)
for nitrogen [wt %Nl = Kr'r (prur)tz (e.76)
for oxygen [wt % O] = Ko (po, )tz (e.7n
It is clear from the abow three equations (9"75) to (9.77) that if the metal phase is
brought in equilibrium with pure diatomic gas at one atmospheric pressure at a certain
temperature then the solubility will be equal to its characteristic constant as
I'ttl lo Hlt atms = Kg (e.78)

Now since solubility increases with temperature these characteristic constants are
expressed as a function of temperature as

logKn=-tT'' -1.246 (e.7e)

or tog Kx = - 'u+o - 1.68. (e.80)

and trom these the values of characteristic constant i.e. solubility values under pure gas
at one atnnsphere pressure at any temperature of interest can be calculated and from this in
tum the solubilities under any other conditions can be calculated.
Examples 9.3 : Molten alloy steelwas found to contain 10 ppm hydrogen when inside the
fumace where the ambient partial pressure of hydrogen was 0.2 atnns. The steel must contain
only 1 ppm before it solidifies. To what vacuum pressure it must be exposed before casting?
Solution : The Sievert's laru'gives
[rad %H]seel - Ku x pk"
and putting values 1 ppm =-0.0@1 wP/oHz
0.001 =Knx(0.2)18
or Ku=2.2x10-2
When 1 ppm is the limit 0.0001 = O.OZ2 x Pkz

114
or pk, = fqgj
I v,v1z. I
or P=2.06rt10-s dt.*.
The vacuum should be better than 4.5 x 10€ atmos.
Exanple 9.4 : Find outthe anpunt of nitrogen dissolved in steel melt in a Bqssemer process
of steelmaking in which air is blown for refining the bath. Later argon was bubbled through
the steel bath to reduce this nitrogen to 40 ppm for better solidification. Find out the amount
of argon bubbled?
Giren lof Kr.r =- # -'12$
Solution : Assuming the temperature of the steel at the end of refining as 16000 C. The value
of KH cornes out to be

tog Kn - - ---l!9-rc - 1.26 = - 1.346

KN = 0.045
% nitrogen dissolved in the steel bath will be that what is in equilibrium with air i.e. with
ptrt = 0.79
Therefore [%N] = 0.045 x 0]91/2 = 0.040 = 400 ppm
This is to be reduced down to 40 ppm. lt means that 360 ppm are to be reduced, or nitrogen
% be brought down from 0.04 to 0.004.
Let us find out the nitrogen partial pressure in a gas phase in equilibrium with 0.004% Nz in

X::k *,", ;.:;,Jffi[" ;ffij";":ff:-

PA=1-0'29=O'71
The anpunt of nitrogen remorred per tonne is
= 0.00036 T/ ton steel bath
0.00036 x 103
x 22.4at = 0.28 m3 at STP
28
Equivalent argon volume shall be = 71129 x 0.28 = 0.68 m3 at STP.
Therefore per ton of steel bath nearly 0.68 m3 argon gas will have to be bubbled to scavenge
nitrogen from the bath provided that errery time equilibrium is reached with the argon
bubble.
9.7 CONCLUDING REMARKS
Solutions are by tl'lemselves a category of materials possessing their orvn thermody-
namic properties as distinct from those of their constituents. lt is because of this reason that
a separate branch of solution thermodynamics has been evolved in the later half of the present
century to understand the behaviour of solutions. The basic nature of solutions in the form of
their models have been dealt with in this chapter and the finer properties shall be dealt with in
the next chapter.

115
 CHAPTER 10

SOLUTIONS I I
mixing and excess functions,
and regular solution

10.1 INTRODUCTION
lf pure silicon and silicon dissolved in iron, under similar conditions, are sepa-
rately oxidised by identicalsurroundings, willthe free energy change per mole of silicon
in both cases be the same? lf not why not? The answer is that the two values of AG can not
be the san€. lf these are not the same then the free energy associated with the reactants
i.e. silicon in two different forms, pure in onei'case and inpure in the other, must be dissimilar.
These two values have got to be different because these are different systems of the same
material, under similar conditions of pressure and temperature, and hence .associated
energy will have to be different. lf these were alike very anamolous situation can arise.
Fact is that these are not alike. Such situations are met almost always in metallurgical field.
Nowthereforethe difference in free energy changesassociatedwiththesetwoare
required to evaluate the net free energy change from pure to impure form. The essential
difference between these two processes is:
(1) Pure silicon € oxidation + silicon oxjd€
(2) Pure silicon + dissolr,ed in iron + oxidatiix + silicon oxide
The situation shall be complicated further if instead of formation of pure silica, a silicate is
formed. ln that case the proqNss (2) will be further modified by addition of the step
silicon oxide + reaction with base + silicate
at the end. ln process (2\ tree energy change associated with the formation of solution of
silicon in iron and that of silica in base matrix are additionally required in assessing
the overall change. '

The energy associated with pure element at any tempeature is merely enthalpy by
way of its having been heated above the absolute zero temperature. But as solution the free
energy of formation of the solution from the constituents is required. Pure elements do not
haw any free energy at absolute zero temperature because pure elements do not have any
enthalpy and entropy at absolute zero tempeature. The free energy of fornration of solution,
which is often not at all negligibly small, has to be determined. lt is often referred to as the
free energy of mixing or free energy of formation of solution and is generally expressed by
the syrrbol AGM , with M as the superscript to indicate the mixing nature of the process.
 1O.2 FREE ENERGY OF MIXNG
When to a large amount of solvent or to a solution, if a gram mole of another
cornponent is added, the free energy change of this process is the molar free energy of
mixing or integral free energy of mixing and is designated as :

AGM = G lsoturion; - )' (pure components)

"'
= (NrGr * NzGz +...+.NiGi) - (Nrc? + NzGS + ...+ Nic,!
= Nr (Gl-c?) * Nz16z-c$ + ... + NiGi-eP) (10.1)
where G i is the partial nplar free energy of component i and Gl is tne molar free energy of
the chosen standard form of the i component. The generalised entity @ i- Gf ) is known
as partial molar free energy of mixing or relative partial molar free energy and often denoted
UV GJU and hence Eq. (1 0. 1) reduces down to molar free energy of mixing as :

AcM = grGt + NzGl +....+ r.rielil) (10.2)

on the same lines all other extensive properties can be evaluated as :
molar entropy of mixing
asM = (N1 sl" + NzsI +...,+ rui-#) (10.3)
or molar enthalpy of mixing
AHM = (Nr HY + Nz HS + ., .. + ttiHlu) and so on. (10.4)
It can be mathematically proved that as the thermodynamic relationships hold true for pure
components, relationships of similar format hold true for mixing functions as well:
For total frce energy of solution
AGM= AHM-TASM (10.5i
ifr component
and partial nplar free energy of
AGJl'l= aHi -TaSi (10.6)
or aGlu = RT ln a (10.7)
The concept of a component in a solution having energy parameter 6 i different from Gf has
been explained in Chapter 4 in Fi9.4.1. This difference emanates from the fact that the
dissimilar atoms interact and alterthe short and long range ordering and hence, this change
in bond strength is seen in the form of absorption or release of energy during the
solution fiormation, i.e. the dissolution process. Obviously the Gf will haw to be different
from G i . That is why for dissolution reactions like e.g. silicon in iron
< Si> = [Si]rwez. (10.8)
AGo = -28000 + 5.54 T
that is AGrezsJx = - 17620 callg.mole
and for < C> = [C]rwtz (10.e)
AGo = + 5400 - 10.1 T and hence AG?ezsox = - 13500 cal/g.mole
It only means that the dissolution process would make them more stable in solution form by
liberating energy during the process of dissolution. lt would be interesting to note that
although the boiling points of both sulphur and phosphorus are too lorrr, a considerable
amount of both of these can remain in solution in molten iron even at such high tempera-

117
 changes associated with their
ures r around 16000 c. This is because the free energy
their strong interaction with eren molten
ctssolution procass are qulte negatire, because of
im, and hence the observed results'

10.3 IDEAL MIXNG OF GASES

space
Gases form solutions at concentrations because of free inter-molecular
all
some metals do form solutions orrer
availabletoaccommodate molecules of othergases'
over partial composition
range, both in liquid and solid state, and many
the entire composition
or molecules dominates over unlike ones the
range. \Mren the attraction between like atoms
rray be more'
chances of reaching solubility limit at certain concentration
Whentwoidealgassmix,becauseofeachbeinganassemblageofvolume]ess
between two ideal gas particles' conse-
non-interacting particles,lh"r" are no interactions
or enthalpy of mixing shall be available
quenily no bond changes, and hence no heat of mixing
is zero'
and hence the enthalpy change on mixing
that
In Sectlon (4'4) and Eq'(4'58) itwas shown

Gi = Gf + RT ln Pi
and since Pi = NiP therefore
(10.10)
6i = eP + RT ln Ni + RT ln P
Dividing by T and differentiatlng with respect
where P is the total pressure of the gas mixture.
to temPerature gives
a(GifD
- a(Gi^)
AT-AT
(10.11)

temperature vanish'
the latter two terns being independent of
From Gibb's-Helmhottz equation
dG/T) E arro -O(Gi/r) Hf or nfl = Fi (10'12)
-
aT T" aT =7
ln otherwords the pure phase enthalpy is
eqrialto partial molar enthalpy of the is component
to the total enthalpy of unmixed components'
or the entharpy of the gI" tiout" is'equal

tnotherword" aH* = )' n,H, - )' ni HP = o (10'13)

or there is no change in enthalpy on mixing

as qualitatively and logically indicated above'
Thefreeenergyofmixingofidealgaseswillhavetobeanegativevaluebecause
freedom to rnove
the process of mixing itsetf is a natural
pro""r*. Earlier the particles had
total
volumes, and now have randomness over
or randomness within their cnrn pure'state + ve
rernaining constant)' the as is definitely
volume of the solution (assuming pressure
quantity and hence AG is -ve quantity'
is giwn by
Free energy of mixing df idealgases
AGM =G lmkture; - G lunrvlixedcomponents)
il
= S'
lt1
niGi - )H1 niGf = T'
Lt1
niGi- cP)

(a.58) G i = Gf + RT ln AGM = ).l nr rn

$ (10'14)
Pr.rtring Eq. Pi
",
118
and if mixing is carried out at constant pressure then

AGM = ), RT ln Ni since f.fi =F (10.15)

"'
Therefore AGM liueatgases) = AHM - T ASM
andsinceAHM= 0 thus =-TASm
Alsotherefore ASM= -g=I )' ni RlnNi
Ltt
(10.16)

)' n; R ln pi
= 41 (10.14

then in stead of partiaipressuresfugacitieswillhaveto be taken

lf the gases are imperfect
into consideration and the deviation from abow relationships shall be evident.
10.4 IDEAL CONDENSED SOLUTION
It has been shown that for any soltltion
AGM = G(solution) - Glpurecomponents)

= )'
./41
(GP + RT ln pi) -.)iZ.r1 (cl + RT ln pf)

!' RT tn 4po =
= Zt1 T'
Lt1
RT ln ai as in Eq.(10.7)

rhererore on'==*ll,f ',rl"T'.",ir*'n';

; i.*-1i] ," "', (10.18)

By definition for an ideal solution obeying Raoult's law a = Ni as the different components
do not interact andthe Eq. (10.18) reducesto
AGM = RT(Nr ln Nr + Nz ln Nz + ...'.+ Ni ln Ni) (10.1e)

Now since - ASM " l#] r.e. Aifferentiating the Eq.(10.18) with respect to tern-

o"'"""j'::;
= R(Nrrnar+ Nzrnaz+,,...+ N rn a)+ Rr [-, 1op; -

N,{iF).. ;- f#*)I (10.20)

Now since = ldM + T

AHM ASr
and substituting in this the values from Eq. (10.18) and (10'20) givas
AHM = RT (Ntlnar + Nzlnaz + ......Nilna;) - RT(Nrlnat + Nzlnaz '.
.. Nirna)[*' ffi) . N, (sF) - *, ffl]
* Frrz ttottt

= -Rr2 [*, (3p) . N, (3F) . + - (iF)] t'orrr

since the first two terns fn 6e iignt hanb side 'cancel each other.

119
An ideal solution activity is equal to mole fraction and which is independent of temperature,
and hence all the terms.shall be zero and therefore
ffi '
aHM fideal) 0 = (10.23)
€rs w€ls obtaEned for idealgases earlier in Eq.(10.13). Obviously
ASM6a.al=-pl (Nrlnar+ Nzlna2+....+ N; lna) (1o.241

=-R(NrlnNr + NzlnNz+.... + Ni lnNi) (10.25)

and partial nnlar entropy of mixing in ideal solution is therefore
ASld(ioeat)=-RlnNi (10.26)
AGM liaeatl = - RT(NrlnNr + NzlnNz + ..,. + Ni ln Ni) (0.2n
Altematively the same is proved by Eq;(10.19). and hence partial molarfree energy of mixing
in ideatsolution is therefore
AGld tia".rl = -
FIT ln Nt (10.28)
On the same lines if (Gi - GP) are differentiated with respect to pressure, which is equal to
volume change, then it can be pror,ed that
Vi=Vf or Vf-Vi=O (10.2e)
that is the volume change on ideal solution formation is zero or
AVM 6aeat; = 0 (10.30)
lf at this stage Fig. 4.1 is recalled then, if the solution formed is ideal no volume change will
be observed. ,

10.5 REGUIAR SOLUTION

The concept of an ideal solution is only a concept and seen to be folloryed in practice
only exceptionally. The actual or reai solutions arelar out from ideality. While studying the
behaviours of a large number of solutions it was observed that although the enthalpies of
formations of solutions are widely different their entropy of solution forrnation are not all
that widely spread. ln fact in many cases it was observed that the entropies of formation
of solutions are just about the same as if they behaved as ideal solutions. This is
logical enough because the entropy is essentially a configurational property and depends
on freedom available for the particles for their movements. \Mren a solution is formed,
whether ideal or non- ideal, the freedom available is practically equal to the total volume of
solution available and it is almast the same for many solutions, since the volume changes
on solution formation being generally negligibly small. This is the reason why ASM fideaD is
seen to be nearly the same as ASM 1reat1.

Based on the above observations, Hildebrand and co-workels introduced the

concept of 'Regular Solution'as in way between the ideal and the real solutions. By
definition it means that'in regular solutions the entropy of fiormation is equal to as if it
was an ldeal solution'. Therefore by definition as derived earlier in Eq.(10.25)
ASM6egl = ASM(ideat) = -R(NrlnNr + NzlnNz + ....+ NilnNi) (10.31)
Hencepartialmolarentropyofmixingofregularsolutionis ASld - -Rlni (10.32)
holrever the AHM (reg) = whatever finite value the solution has:
Therefore AGM (reg) is given as

120
 AGMl,'sy = RT(N'lnar + Nalnaa +...... + Nilna) sameasinEq.(10'18).
Hence the partial molar free energy of mixing of a regular solution is
(10.33)
aelil {reg) = AGI\rr {real) = RTln a
combining these with the equation giving entropy change gives
AHM=AGM+TASM
=RTI(Nrlnar+ Nzlnaz+" ""+ Nilna)-
-RT(NrlnNt + NzlnNz + """ + NilnNi)l
(10.34)

andusing E=Ni xYi or lnl;=lnai -lnNi

iHMt"gl = RT(Nrlnlt + Nzln12 + '" '+ Nlln;yi) (10'35)
partial molar property is
Since this is the total soiution property i.e. integral propefi the
obviously giwn bY
(10,36)
lFld = RT ln yi
10.5.1 AtPha Function
it was
While analysing the results of activity coefficients of various binary solutions

found that the factor (1

,r'nt?,,, was substantially constant and was therefore called as ct
- Ni)
function. lt can thus be put as t .r =
1fty
and * z =
ffu
(10.33)

for a binary solution of components 1 and 2. lt was also obsened

that cr t= cr 2 and hence

described as onlY cr and

* - ln vt
=- Yz ln (10.38)
therefore -
(1 - Nr)2 1t- Nz)z
lnr,l _ NE (10.3e)
or
ln yz N?
putting Eq(10.36) in this gives
hYd- - Ne
- (10.40)
l" yr N?
- ^HY
AHlt
ln other words log-activity-coefficients are inversely proprotional to the square of their
partial molar enthalies of mixing'ln
atordmole fraction or also inversely proportionalto their
one of the above
fact it is a test, if experimental iata fit'in into a pattern where
a regular solution therefore
relationships are oUeyed, the solution is a regular solution' For
of components inrrclved but
the free energy is as for a real solution in terms of activities
enthalpyandentropyaredifferentasshownintheequationsabove.

Example 10.1 : An alloy of sn-Ag-Au was made by addition of Au and Ag of total 5 grans
Oo C to a 95 gr"* *olt"n Sri ai 2500 C' ln one
case an alloy Au-Ag (7O% Au and 30% Ag)
"t added whereas in the other a mechanical mixture of Au+Ag was added' The difference
was
in temperature drop of tin in these two cases was 0.5o
C' Find out the enthalpy of formation
temperature this value of enthalpy of mixing
of Au-Ag alloy at 70:30 composition. And to what
of Au-A! alloy refer to? ls it 0o , 2500 or any other temperature?
Given is-sp.heat of Sn-Au-Ag alloy = 0'03 caV0o/g alloy

121
Solution : The process of formation of the final alloy can be imagined to occur as
Process I Alloy (Au'Ag) + + Sn * alloY (Sn'Au-Ag)
Process II Mixture (Au+dg) + + Sn * alloY (Sn-Au'fu)
The difference in these two processes is only as if the alloy Au--Ag is de-formed into individual
elements Au and Ag then added.
The AHM of Au-Ag (70:30) shall thus be additionally required when alloy is added to make
up the final required Sn-Au-Ag alloy product. Conversely it can also be imagined that the
alloy Au-Ag is formed first from the mechanical mixture and then added to make both the
processes the same. ln that case the AHM of Au-Ag alloy shall be liberated and shall be
available for heating the final alloy. ln other words the finaltemperatures reached are not
the same. lf the finaltemperature reached is say f C in one process, which has to be belcw
2500C,thenintheotherprocessitwillbeT;0.5oC, The thermodynamic process offinal
alloy formation can be imagined to occur at 0o C in either case, Hent e
AH - AH (process ll) = AHM (atlo1{+-As}
(process D

Hence Cp dT(process,) -4. Cp dT(processll) = 6gM @[oyAu-As)

t
fl.t't cp dr(D -{ cpdr(ll) = 6sM @rovAur--Ag)
ln either case the net change shall be

AHM lattoyAu-Ag) = - C x 0,5 = - 0.5 x 0.03

= - 0.015 cal/ g wt of alloy Sn-Au-Ag
= 0.015 x 100
= - !.5 cal/100 g wt of alloy Sn-Au-Ag
This is because the total enthalpy change has occurred in 100 gratns of the final alloy. This
is also equal to the enthalpy change that has occurred in 5 grams of Au-Ag alloy formation
in the sum total process. The 5 g Au-Ag alloy has 3.5 g of Au and 1.5 g of Ag or it is
equlvalent
3.5*
to 1g7 1.5
g atoms of Au and Ag respectivelY.
10g
The totalg atoms in the Au-Ag alloy = A.A177 + 0.0138
or = 0.0315 g atom alloy
4760 cal/g mole alloy Au-Ag
0.0315 = -
AHM = -
=*=
This value of enthalpy of mixing referes to 0 C only because the process has been
carried out in such a way that its value at 0 C has come out as an answer. lt can not be at
any other temperature in this instant case.

10.6 EXCESS FUNCTIONS

The excess thermodynamic function has been defined as the difference between
thepropertyofrealsolutionand that if ithadbehavedasanidealsolution.Thus the general
extensive property zo" M H ,U, S, A, G, etc.)
Z'" = Z(real solutbn) Z(ideal soslution) (10.41)
-
lf free energy is taken as an illustration then
Gr = AGM lrealsol) - AGM (idealsol) {10.42)

122
Putting their values as calculated earlier gives
G*" = RT {Nr lnar + Nalnae + .......+ Nilnai) -
- RT{Nr lnNr + NzlnNz + .... + NilnNi)
=RT{Nrlnyr+NzlnTz+ +Nilnli) (10.€)

= RT Lr' Ni ln 1i (10.44)

The excess entropy 'i

can be obtained by differentiating the excess free nergy that is the aborre
equation with respect to temperature €ts AG: i dT and from this the excess enthalpy Ho can
be evaluated as before.
The excss property can be imagined in the form of volume from Fig. 4.1 in Chapter 4
as the difference between the dotted line and the solid line in (d).lt may be + ve or - ve as
the case may be. This excess consists of contribution from each individual component of the
solution as shown in Fig. a.1(c). This is known as excess partial molar component and is
denoted UV ff and can be written for fiee energy as

\:i =RTlnYi
=xs
h- r- (10'45)
For a solution therefore
Gxs = NrGf + NaG" + NsG" +.....+ NiGi'" (10.46)
The excess partialquantities are also interrelated as
Gi'" = nl'" - rsi'" fi0.42)
For a regular solution since
ffiSsl = RT {Nr lnyr + Nzlnyz + .....+ Nilnyi) asshown
in Eq. (10.35) and therefore
G"" = AHM lreg;
or Gi'" = AHlut"s) = RT ln yi (10.48)
Example 10.2 : Calculation of various properties of solution:
lf activities are known then various mixing functions can be calculated at any temperature
as follons:
Given : The activitias of Zn and Cd in an alloy at Nzlr = a'4 are 0'3 and 0.65 respectively
at 5o0o K.
lf Nzr = 0.4 then Ncd = 0.6 in a binary solution.
Hence I G$0" = RT (Nzn ln azn + Nca ln aca)
= 1.987 x 500 x 2.303 (0.4 log 0.3 + 0.65 log are 0.6)

= - 782 cal/g mole

lf the same solution was ideal then
lG$arl = FIT(Nzn.ln Nzn + Nca ln Nca)
-
668 ca/g mole
=
hence the excess can be calculated as

^c- = l1;_^il.,
= - 114 caUg mole
The partial properties can be determined as
123
 A G z. = RT ln aTtr = 1'987 x 500 x 2'303 x log 0'3
= -1196 caVg mole
and AGca = RT ln acd = 1'987 x 500 x 2.303 x log 0'6
= - 428 cal/g mole -
Now to e\raluate partial excess property of each component
AGzn (irleal) = RT ln Nzt=0'4 = - 910 cal
lGca (ideal) - -507
AGli = lG zr - AG a (iiear) = - 1196 + 910 = - 286 cal

AA{i = + 79cal
Now 0.4/0.3 = 1.33 and Y cd = 0.65/0.60 = 1.0.q
'GM =
yzn
*"""" f,"s) = - n1rutr.r ln Nzr + Nca ln Nca) = ASM (iuea1
- 1.987 x 2.3O3 x {0.4 log 0.4 + 0.6 log 0'6)
=_- 1.33 cal/g mole/Oo
Now AE zn = - RT ln Nzr = 1.987 x 2'303 x log 0'4
= - 1.82 cal/g/mole/oo
similarly AS 66 = - 1.01 -do--
Then I H$s can be calculated as
AHUg=lGHs+TASHs
= - 782 + 500(-1.33) = -1447 cal/gmole
knowing only one
It would thus be very clear that how the mixing functions can be evaluated
parameter like activity. conversely if mixing or excess properties are known then the activities
or the activity coefficients can be calculated'
10.7 CONCLUDINGREMARKS
are respectively
The free energy, enthalpy and entropy of formation of a real solution
by, in
givenbyEqs.(10.18),-lrO.eo1 and (10.22). Similartytheseforideatsolutionaregiven
the sanp order by Eqs. (10.27), (10.23) and (10.25). The same
for regular solution are
given,inthesameorder,byEqs.(10'18),(10'35)and(10.31)or(10.25).
The partial contribution of free energy of formation of ideal solution
is given by
ideal solution
Eq.(10.28) and of entropy byEq.(10.26);thepartialenthalpycontributionfor
offormation of regular solution is given by Eq'(10'18)' Similarly
being zero. The free
"n"rgy by Eq.(l0'31) t'e' Eq'
for regular solution enthalpy is given by Eq.(10.35) and entropy
(r o.zi) The partiil contribution in regular solution is
given as of freeenergy by Eq' (10'33) '
enthalpy by Eq.(10.36) and entropy.by Eq'(10'32)'
and partial part
The excess lree energy of formation of solution is given by Eq.(10'43)
by Eq.(10.a$.
and these
The energy changes on solution formation are by no means insignificant
energy changes taking in
arise lrom contributions of individual components because of the
when the behaviour differs
them individually, Such partial properties and excegs properties,
of solutions'
from the ideal one, do contribute to our understanding of the
thermodynamies

124
 CHAPTER 11

SOLUTIONS - Iil
Applications of Gibb's-Duhem equation

11.1 INTRODUCTION
The concept of activity of a component in a solution as influenced by the presence
of other constituents has been discussed in the previous chapter. lt has also been seen
that although the practical solution may be multi- cornponent, their properties
are mainly
estimated from the knowledge of coresponding binary and ternary solutions. lt is easier
to study and carry out measurements of thermodynamic parameters on binary and ternary
solutions. The most commonly measured property of a solution is the activ1y
of their
components. Several techniques are available like chemical measurernents, ernf measure-
ments, vapour pressure measurements and so on, to measure activi$ of a component
in a solution. ln a binary solution if activity of a component is measured then
mathemati-
cally that of the other component can be computed without carrying out the actual
measurements, Even if actual measurements are carried out even forthe second component
mathematical evaluation can be used as confirmatory test ori the actual measurements.
lf by measuring activity of one and detgrmining that of the other by mathematical
computation is possible the process can be repeated with more complex mathematics to
compute the activity of the third knowing those o(he two in a ternary solution. The proce.ss
can continue for higher order solutions. The accuracy of calculations however does decrease
with this progress. !

The activity is direcily related to energy as dd ; = RT ln a as per Eq.(4.69).

Therefore even energy properties can be found out for second component when that is
known for the first.
The methods of these calculations are described in this chapter.

11.2 DETERMINATION OF PARTIAL MOLAR PROPERTIES

The Gibb's Duhem equation for an extensire property of a binary solution of compo-
nents 1 and 2 is given by
Nr dZr + Nz d2 z = 0 where Z is any extensirre property. lt means that

d7t = -# oZ, (11.1)

This can be integrated provided the variation ol Ez with composition is known for component
2.ln a binary when concentration of one component is known 1 - Nr shall be the concentration
of the other component Nz . The integration can be carried out over the limits to be fixed.
Thenaturalchoiceisof lcnryerlimitasthatwhen Nz = 0 i.e. Nt = l ordZt= 0 and the
upper limit is that when Nr = Nt at which the dZ r is to be found out. The integration of
Eq"(11.1) thus gives

Jtlr - tJr ,flr - Nr

I x .dZt = -l"Nr - x dZz (11.2)
'Nt -
.
1 1

or Zt (Nr - Nr) Zr (Nr - 1) =-{,:i'

ffioz,
whenNr=12=Qand
*' Nz .-
therefore Zr (Nr - Nr) = - t'- r* azz (11.3)
'Nt = 1

This integral can be solved graphically to compute the value otZ t at Nr = Nr

concentra-
tion. This is illustrated below with the help of an example and the calculations involved therein.
Example 11.1 : Let us assume that the follorruing data were obtained for partial molar energy
of mixing of Ag-sialloys at 1150o C.
Nsi 1.0 0.9 0.8 0.7 0.6 0.5 4.4 0.3 0.2 0.1 00
6 si 0.0 90 12O 28O 400 5s0 800 1150 1400 1800 0 cal/mole.
Let us saythat partial molarfree eneregy of Ag at 0.2 atomfraction is required to be computed.
ln the equation (11 .2) or (11 .3) component 1 is Ag and 2 is Si. lt therefore reduces to :
Jllag = o'z J\las - o'2
fx
tNes-1 dGlo = | x
"Nes-l t
ffioGs,
or this will reduce down to

oGm=
- o.2 = - tNon-1
flon=o'z x
ffi cGsi
The available data will have to be processed further as

Nsi 1.0 0.9 0.8 o.7 0.6 0.5 0.4 0.3 0.2 0.1 00
Nns 0.0 0.1 o.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 00
-NsL
Neg
e.o 4.a 2.33 1.50 1.0 0.66 0.4! 0.25 0.11
Gsi 0 90 12O 280 400 550 800 1150 1400 1800 0 cal/mole
Nsi
A graph is plotted with G si on the x-axis and the ratio on the y-axis as shown in Fi9.11.1.
ffi
The area on the graph under the burve and bound by the x-axis between the limits Nns = 1 to
NAs = .02 is shown by thatched marks. The area can be worked out graphically but since
it extends in the form of a tail to - m i.e. asynrtotic form, it can not be determined accurately.
ln the present case it comes out to be 2118 caUmole and hence the partial molar free
energy of mixing of Ag at Nns = 0.2 is - 2118 cal/mole.
The values of GXs at various concentrations can be computed likewise for each
composition separately, by computing each time the area as above.

126
 Nsi SHADED AREA = dr= o.z --2118 cat
No*
4

300 600 900 , 1200 1500 t800 2t@

'-"+ a,G^.
-M Cat/G.Mo[e
Fig. 11.1 - Plot to evaluate the Gibb's-Duhem integral for evaluatian of
Partial molar ProPertY.
11.3 DETERMINATION OF AC'IIVITIES
ln the case of binary solutions it is experimentally readily possible to determine
activity of a component, particularly the solute, in a solution using a varieties of techniques
but it is rather difiicult to determine experimentally the activity of the solvent component at the
same composition.This difficulty is more pronounced in the case of dilute solutions' The
integration of Gibb's Duhem equation comes out handy to determine the activity of the other
component, usually the solvent, when that of the solute component is known by actual
measurements. lt has already been seen that activity is nothing but a function of partial molar
free energy and hence the method of integration of Gibb's Duhem equation in the previoris
section should be applicable for estimation of activity as well. ln this case the Gibb's Duhem
equation is to be transformed into an equation correlatingactivitiesof thetwocomponents
in a binary solution. By taking partial molarfree energy as the extensive property the Gibb's
Duhem equation can be written as :
NrdGl +'N2dGz = 0
Putting dG 1 = RT dln ai and dividing by RT
gives Nrdlnar+Nzdlnae=0 and (11.4)
eliminating natural logarithm
Nt dlog at + Nz dlog a2 = 0 and (11.5)

cor dlog a1 = T9 d log az (11.6)

- Nr
127
 lntegrating again between the limits Nr = .l at the lcnrer limit to N1 = Nr as the uper limit
gives
Jllr * Nr
Nz
J*, -., drogal = - 'Nr
ft-ru'
- 1 N1
q log a2 (11.4

or (d log a1)x, - Nr - (d tog a1)p, = 1 =- o,os *

dl:)' ffi
since when Nr = 1, €lt = 1 and hence log a1 = e
thus (d rog ar) = - aros*
d::)'
A graph plotted betwen Nz / Nr and log a and the area under
ff (11.8)

. the cune bound by the x-axis

over the range N1 = 1 to Nt = Nr is estimated to obtain the value
of log ar N, - n.,. tne
integration is carried out between limits upto different Nr concentrations. The.curve ap_
proaches at one end + infinity when Nr
= 0 to - infinity when Nz = 0 at the other end. The
difficulties encountered will be clear by taking a numerical example
and solving it as follorrys :
Example : Let us take the measurernents of actiMties of Ni in Fe-Ni alloys
atl6000 C made
by G'R'Befton and R.J.Fruehan (J.Phys.Chem.1967),which are shown in horizontal
'columns 1 and 2. Let
us try to carcurate the activity of iron at Nre
NNi 1.0 0.9 0.8 O.Z 0.6 O.S 0.4 0.g 0.2 0.1 = 0.9"
00
aNi1.00.890.7660'62l0'485o.g74o.283o'2o7o.137o.067
now calcuhtgd va[res with lhe same columns continrcd
NFe 0.0 0.1 O.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0O
Nrui
:-
NFe
9.0 4.O'.2.99 1.50 1.0 0.66
v'vv 0.4i| 0.25 0.11
v'D
log as - 0.0506 0.2076 0.4271 0.6940 "t.1739
0.1157 0.3142 0.5482 0.8664
ln order to solve the integralthe data are processed to' obtain horizontal
colums 3, 4 and F
and graph is plotted with log aNi on x-axis and Nrui / Nr" on the y_axis as
shown in Fig. 11.2.
The area underthe curve bound by x-axis between the limits Np;
= 1 to Nrui = 0.1
is estirnated as the value of the integral in eq. 1t t .8). lt is observed that the cune extends
to - infinity when Nrui = 1 and hence required area has to be calculated stricily
upto -
infinity limit. This is practically just not possible and hence the calculations
are in enor.
However with.all these errors the value of lob ape at Nre
= 0.2 is equal to the area shown
by thatchings and is equalto 0.85.

11.4 ESTIMATION OF AOTIVITY COEFFICIENT

lf activity can be intenelated through Gibb's - Duhem equation, activity coefficient
must also be similarly intenelated. As derived in the previous section in Eq.(11.a)
Nt dlnar + Nzdlnaz = 0
Since in a binary Nr + Nz = 1
hence dNr + dN2 = 0

alteringthefornssuitably Nt
# * ruz ff = O

means
it also Nr d ln Nr + Nt d ln Ne = 0
fromdefinition e = Niyi i.e. log a = log Ni + log y;

128
 I
a

7
(tR e SIIADED ARSA
LOG
a AGTIVITY OF Fe (o& N6ro€)r o.!5
-N*r
Nros

4
I

04 05 0'6 0.7 08
LOG O +

Fig. 112 - Graphical determination of activity using the

G ibb's - D uhe m inte gration
or log a- log Ni = log yi (11.e)
Subtracting Eq.(11.9) from Eq.(11.4) leads to
Nr dlogyr + Nz dlogy2 = 0 (11.10)

dtogyl = - #f dlos12 (11.11)

As before integrating between the limits Nr = 1 to N1 = Nr

J,lr - Nr - Nr
J.,lr Nz
I OlogYl = - |
JNr-l N1
ologY2 (11.12)
'Nt-l
(ros yr)Nr - Nr - (rog y r) = - d] :)' ff arog*
Sinceat Nr = 1, €t1 = 1 andhencef r = l andthereforelogyl = 0
Nz
therefore (tosy r)r.rr - Nr = - .(:: )' o logY2 (11.13)
NI
This can be solved graphically by plotting Nz /Nr vs. log T z and the area under the
curve bound by the x-axis between the limits Nr = 1 to Ni = Nr shalt give the value of
log y r and from which y 1 can be calculated at Nr = Nr . Such values at any other composition
can be calculated likewise. Knowing y r, the activity values can be calculated through the
relationship a = Ni . y i. The activity data of Ni in Fe-Ni system at 16000 C shown in the
previous section can be processed further with respect to Eq.(11.13) to understand the
method of this calculation more precisely.
129
Bringing back the data here again

Nui 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 O.2 0.1 00
aNi 1.0 O.B9 0.766 0.62 0.485 0.974 0.283 0.207 0.137 0.067
il*l 1"0 0.988 0.885 o.7N 0.69 0.67
0.95 0.808 0.707 0.68
,logyr* -0.0052 -0.053 -0.126 -0.161 -0.173
- a.azi - 0.09 - 0.150 - 0.167
Nre 0.0 0.1 A.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 m
x!i- 9.0 4.0 2.3s 1.50 1.0 0.66 0.4i| o.zs 0.11
NF€

The graph is plotted with these datawith log y r'l on x-axis and NNi/ Nre on the y-axis.
The graph ls shown Fig. 11.3. The activity coefficient of iron at 0.3 atom fnaction composition
is equivalent to the thatched area on the graph and which is the value of the integral as desired
by Eq"(11.13).The area is equivalent to log y re = 0.01186 and hence ] Fe = 1.31.

I fr (fftro'll = 9IADED
1fu = t€t

LOG tr{r

Fig,I1.3 - Graphical ileterminatian of activity coefficicnt using the

G ibb's- D uhem inte gration

130
ll.sESTIMATIONOFACTIVITYCOEFFICIENTUSINCALPI{AFL|NCTION
From the Previous section
Nr dloglt + Nzdlogl2 = 0 as inEq'(11'10)

-
dividing both bY dNz gires
ohfry-* m,ff= o (11.13)
N;

or dros1l = - # .# o*r (11'14)

As before this can be intelrated between the limits Nt = 1 to Nt = Nr gives

f*Nr x
JNr-1
d ros vr = -{l:i' . {e9J3
ffi 6p, (11'15)

By definition the term

ln Ni
- is the cr (alphd function for
(1 - Nr)z
binary solution and as has been obsened
lnvr ln
ano d2 = -:;
yz
at = -*'
N3 N,
.. 1r _ln ln yz (11 .16)
" = *i =
.
and u1= d2 = cL andhence Na

and by processing this equation it can be shown that

J'lr - Nr
logyi = * a Nt *t -{,-;' cr dNt (11'17)

{t'"kdNr.
tNe,.7 = SHADED AREA

U6.: O'OO2 ct Na'

t.q-
>lz
('lz
313
ll

using
Fig.11.4 - Graphical evaluation of activity coefrcient
alPha-furcian.
131
 there'foreif o' 'i'e' log y z /(1 - Nz) is plotted vs. Nr the area under
bound by x-axis between the limits Nr the curve
= 1 and Nr = Nr cen be obtained and if from this
computedvarue of -c Nr Nr issubtracted,
the varue of rogyl is obtained
There is no asymptotic shape of this curve and
exact area can be found out to get
the activitv coefiicient accuratery. This is shown
section for activity of nicker to work out
in Fig 1r.e usin!t;;';;;.;;';ilJo*""*
the activity coefiicient of iron at Npe 0.3.
value at NFe = 0.3 can be evaluated. = The .,

The value of the integralcan be assessed

from the graph in Fig. 11.4 and is shown
by shaded area' The above equation (11'1n
can now be solved to work out the value
rlt Nre = 0.'3 and from there the value of activity coefiicient of Fe works out to of log
at iron atom fraction of 0.3. be 0.002
Alternatively if activity coefficient of a component
in a solution is expressed by an
empirical equation then it can be proces'sed mathematically to get the activity
the second component. coefiicient of

11.6 CONCLUDING REMARKS

The Gibb's - Duhem equation is thus very usefur in evaruating activity, its
coefficient or any extensive property of
a component in solution when that of the other
compoenent is known over a range of
composition. Atthough it is tedious but
particulady when the measurmenets is usefut
for the other component
-"'F-rrYrrr lre
srv not accuratery rneasurable
by experimentation.

132
 CHAPTER L2

EQUILIBRIA IN PHASE DIAGRAMS

12.1 INTRODUCTION

ln a single component system i.e. in pure substances the phases that can exist under
appropriate conditions of temperature and pressure are the solid or liquid or gas forms of
the pure substance, For a given amount of pure substance if P and T are fixed the volume
is automatically fixed. Even if amount of substance is varied T and P are potential terms
and hence decide the equilibria lndependent of rnass of the substance. The areas where
each one of these phases exist are therefore available from T-P diagrams as shown in Fig .6"2"
The T-P relationships also indicate the conditions when solid- liquid, solid-gas and
liquid-gas shall co-exist. This is confirrned from G-T relationship in terms of associated
minimum free energy for the stable phase or phases as in Fig. 6.3, Although these
relationships do indicate phases in equilibrium in a single component system, these are
not called equilibrium diagrams.
Equilibriunn or phase diagrams however ryrean graphical replesentation of stable
equilibrium phase regions that exist with respect to temperature and cornposrtlon in a system
with minimurn of two components wherein at least one component is in a e,Pndensed form.
The pressure is fixed at one atmosphere normally in plotting these diagrarse fhe equilibrium
diagrams are often known in terms of the number of components inwlved irl the systern. lf
only two components are present, it is binary equilibrium or phase diagrarn, if three are
present it is ternary diagram and, if four, it is quarternary diagram and and so on. A binary
diagram can be depicted readily as atwo dimensionaldiagrarnwith composttion on thex-axis
and temperature on the y-axis. A ternary can be indicated in a trianguiar ir:rm as a two
dimensional representation of the three dimensional situation at fixed ternperature' The
quarternary forrn is still complicated. ln a more thara foun, that is truely nnuitE-eomponent
system, it is not at all possible to construct the diagrar"n as a lrariation of temperature
and composition and hence for a fixed compositioll the phases are ascertained or it is
reduced to a pseudo-ternary systern to be depicted in atriangular forrn,
The binary diagrams are by far the rnost populan useful, readily legible and available
forms of phase diagrams. As apre|irnanaryintroductlc;ntothesubjectof energy consid-
erations and' phase equilibria, the diseussion here is eonfined to binary phase diagrams
only.
 A binary system eren in a condensed form may exhibit, one or two liquid phases,
one or more solid sglution phases, one or more intermetalic or chemical compounds, one
or more mechanically mixed phases as eutectics, magnetic and non magnetic phases, and
so on. ln other words the diagram may depict regions of stabilities of several of such phasas.
The simplest diagram with one liquid and one solid solution, where complete solid and liquid
solubility between the two components exist, shall have the region of stabilities of liquid and
solid solution phases andseparatingthesetwophase regions, a area where both phases
co-exist ih equilibrium. There will have tc be a two phase region, separating two single phase
regions. When one phasechanges to the other on heating or cooling, the gradual changeover
through the two phase region is thus necessary and without this direct changeover from
one to the other phase is jurst not possible.
The equilibrium diagram of two completely immiscible components even !n liquid
state e.g. Pb-Fe, W-Pb etc. are of no interest because no,other phases, except the
two pure component phases, mechanically in contact wiih each other,'existthere. Therefore
phase diagrams where at least a partial liquid solubility of component, one in the other, is
possible, are ofrelevance.Alternatively at least complete liquid solubilitywithoutanysolid
solubility may also be relevant.
The phase diagrams are constructed using equilibrium measurements and sub-
sequent phq:e analyses to be carried out by either chemical, X-ray, metallography and such
othertechniques. Dilatornetric or differential graMmetric (D-G) or difierentialthermogravimetric
(D-T-G) techniques are used to know. thetemperaturesatwhichphasechangesoccur.
Those measurements are carried out overthe required temperature and composition ranges
to completii the diagram. These measurements are carried out reversibly to ensure equilib-
rium" The fact that these are equilibrium diagrams means that they indicate the equilibriurn
phases existing in a given tenperature-composition range under isobaric conditions with
respect to the composition. The second law clearly establishes that for a system of definite
composition at a fixed temperature and pressure, the equilibrium stable phase is the one
which possesses lowest free energy. Equally true would be that when two phases coexist in
this diagram the chemica!potentials, G i for a cornponent in the coexisting phases must be
the same. Alternatively the activities of the same component in the co-existing phases must
be the same since free energy and activity are directly related to each other. ln other words
the phase diagrams must show appropriate free energy composition and acltvity composition
relationships in conformity with the larars of thermodynamics.
The phase diagrans are evaluated by assuming cooling of certain composition of the
system from where it is fully liquid. As this cools no change will occur till the temperature
buches the liquidus boundary (above which only liquid exists). Every phase diagram
nodd pose this situation. At the liquidus temperature, since by its definition, that solid
phas€ shall seperate out from the liquid. The equilibrium inwlrrcd is
Liquid
and solld phase can ,be almost the pure component, a solid solution, an intermetallic
conpound or a mechanical mixture of components known as eutectic. This will continue
till all the liquid phase is transformed into solid. So the free energy of fornntion of one of

134
these solid phases, when minimurn give the clue regarding the seperate phase(s) or
equilibrium phase(s) and the temperature range o\,er which it(these) shall be stable.
Similafly the soiid phase so formed on further cooling can undergo phase changes as
1. Solid Solution *' two solids
{eutocbid change)
2. Solid Solution .' another solid solution
3. Solid solution € a liquid and a solid
As the above free energy or activity consideration shall apply to this and such other
equilibria, involved in the given corfrponent composition, as it cools to alrnost room
ternperature or even belor it if required. Any other ch'ange,s also can be treated likerryise.
ln order to correlate the phase diagrann withfreeenergies o{ component phasas
and the relevant activity of conponents a fsru simple diagranrs are considered bela'v.
412.2 FREE ENERGY OF FORMATION OF SOLU'NON
There are systens where complete solubility orer entire composition range in liquid
shall exist. But many ceramic systerns iike alkaline earth-metal oxide-silica systerns exhibit
only partial liquid solubility. There are many wher'e only one solid solutian exists over the
entire composition ranges like Cu-Ag, Cu-Ni, Pb-Au etc. But many others exhibit only
partial solid solubility at each end. ln the extreme the solid solubility is so negligibly smallthat
almost pure component seperate out without any solid solution formation at each end" The
pb-Sn is atypical system olthis type wherein the liquld freezes into afine mixture of Pb and
Sn as eutectic of a certain fixed composition and the excess of any component as a
seperate phase of that pure component.
The evaluation of a solution phase thr.rs start with deterrnination of free energyof
forrnation of a solution, initially in liquid state and then in solid solution form as well' The AGM,
AHM and ASM for the fornration of a solution from components 1 and 2 have already been
evaluated in Chapter 10 and can be reproduced
as aGueat =-TASiueat
= - RT[Nr lnNt + Nz lnNa]
since AHiaeal = 0 for idealsolution. However if solution is not ideal i.e. ai * Ni bd
a = y i Ni or hi = ii (wt%i) and
hence AiM *; = AHreal - T ASreal
' The actual value of AGM depends on whether AHM is -ve or + \€ i.e. exothermic or
endotherrnic. But when solution fornrs AHM will have to be -ve, being a natural
process.
Since the interaction amongst the components manifeqt as -ve or + ve values of AHM, the
the
actiVties of components shall show corresponding -ve or + ve deviation from ideality' lf all
infornration is known the AGM can be evaluated for any composition of solution at any
temperature. AGM is therefore a function of temperature and composition ( prassure 1
atrnos.

fixed).
The free energy of formation of a solution is shown in Fig. 12.1 under varieties of
conditions. The curve at 'A' is ideal solution fornration and 'B' is for the -ve and
'C' is for the
+ ve deviation from ideality. Depending upon the actual value of AHM , which will definitely

135
 Fig. 12.1 - Free energ) of formatian of
anidealsolution as afunctbn of com-
positian along-with those of real so-
lutions to indicate the positive and
ne gative dev iatbn ide ali 4t.
from

coMPostoN

be related to interaction of components the real aGM shall_acquire the value as

a resultant of
AHM and T ASM factors.
The fact that components form solution ,being a natural process, is itself an indication
of -re value of AGM for the solution. aHM also *itt nr* to be -ve as the natural process
is exothermic in nature. The ISM will however be a + ve quantity forthe same
reason. Another
interesting part is that the lGlv can be evaluated by drawing a tangent at the conesponding
point of composition on the AGM curve, and the intercepts of this tangent as shown
in Fig. 1 2. 1
onaxisatNr = 0 and Nz = O girethevaluesof n6lil andaGld.AtNr = 0 meansNz
= 1 and at N2 = 0 means Nr = 1, the slope will harre to be minus infinity and hence the
cune atthesetwo extremities mustbevertical since at Nr
= 1, AGl/| = 0 and
A@ = -o andatNz = 1AGI= 0andlG$r= - 6.
The -ve AGM value for any such solution is a sure indication of the stability of
liquidsolutionand upto whatewrextentitisso,thesolubilitylimitshall extend.The Fig.12.1
shows the state of affairs when liquid solubility is exhibited over entire range of composition.
No.other form of phase, like say a mixture of compeinents 1 and 2 shall be stable. Ewn if the
AGM of this mixture is eraluated its values would b* less -ve than the conesponding value
AGM for sotution as shown in Fig..12.1.
This can be proved from Fig 12.2 which i* reproduction of the Fig. 12.L for AGM as
a generalised case, not specifically ideal or olti*.r-wise. Let us imagine a mixture of two
solutions of composition 'A and 'B'with the cornsicsition average being formed at ,C,. The
AG''' will be the total of AG of these two cornponent solutions as giwn by point A and B
respectively' The sum total value shall be as depicted Uy C wnicfr is rnuch lorruer than (less
-rre) point 'D' which gives AGM of solution of A and B solutions. Hence solution D is
more

136
 stablethan mirdure represented by C'
This holds true for entire range and
therefore solution phase shall be more
stabler than mere mixture'
? Fig.12.3 gives the free energY of
I
formation of solid as well as liquid solu-
I tion over the entire cornposition range at
three different temperatures for a system
which exhibits solubility in both liquid
i5 and solid state over the entire composi-
tion range. The IGM for solid solution and
liquid solution at temperature Ts and Tl
make interesting reading. At temperature
Tl only liquid solution can exist in the
phase diagram. The values of AGU and
lG$ hare been plotted in Fig.1 2.3(b) . The
Fig.122 - Shows free energy of mixing I GU has allthrough values more nega-
as it varies wiih compositinn for a solution
tive than those of A G$ vdues for anY
solution. The phase
composition, meaning thereby that liquid solution is stabler than solid
diagram confirms this. Similarly the argument for temperature I Ts shall
indicate solid solution
axis cuts the
to be the stable phase as shown in (c). At temperature Ts-l the temperature
the diagram
liquidus at 'b' and solidus at'c' as shown in (a). The AGM vs. composition in

indicates that for composition left of b, liquid is stable and for

composition right of c, solid
the same and hence they
solution is stable. ln-between b and c, both AG$ and AGF are
of free energy
co-eist. The phases in the diagram are thus confirmed on the basis
considerations.

SOLUBILIW
12.3 FREE ENERGIES OF SYSTEMS WITH ONLY PARTAL SOLID
for such a
The system is typically shown in Fig.12.4. The AGM of solid solution
of formation of solid solution
system is like that in Fig 12.1 . Let us work out the free enelgies
the horizantal dotted line for the
and liquid solution at various temperatures, as depicted by
and solid solution p have been
corresponding values. The AGM for liquid, solid solution a .b,
between the compositions
olotted at temperature Ts-t in Fig. 12'4tb|' lt is apparent that
il#;;;;il ;;;id phase has the toryest vatue * compared to aGM for .' and p solid
seen in the phase diagram' Also
solutions and hence liquid phase is stable in this region as
the A G}l has lorarer value on the left of composition b as it is
this region where * seperates
out from the liquid solution. Similarry A G$ shall indicate
the precipitation of p solid solution
beyond right of comPosition c.
It is equally clear that at Ts-t- in between b and c that
AGM for .' - p solid solution
p, are not stable phases in
is more than that for liquid solution and hence o and individually

137
 lJ1 COMPOS|TloN tL Nt this region. As the temperature is lcnv-
*[ ered down belos Ts-l the miscibllity
gap decreases and eventually at
rr"rl! Te both o and p have AGM less than
that for liquid and hence are stable.
But for the region between 'g' and 'f ,
r3
both aG$ and AGff are same
whereas for composition less than
point'g' it ls AG$ . AGy and above
IAT TEMP
.li1
point'f AGy r IGS inOicating the cr
I
TL
I
and B slability ranges at either end of
I
the phase diagram. lf temperature
It I A
AGs
I
is further lcnrvered the overlapping
I
of a and p lines extend further thereby
decreasing the stability region of a
and p as independant phases. This is
AT TElvIP Tj I
obvious because of the slope of cr / (cr
s I + p) and F /{a + p) phases being
tg I
sloping towards lower solid solubility
f ranges.

TeM?l.rl 12.4" FREE ENERGY OF PURE

,{-T
EUTECTIC SYSTEM
I

Atypical pure eutectic system

is shown in Fi9,12.5 (a). Let us con-
sider the temperature which is be-
tween the m.p. of the two consituents
as shown by the line T*+1ry. lThe
Fig.123 - Free energ)-composition relation- conesponding free energy values,
ships for varbus phases at different tempera- AGU and AGU , are plotted in Fig
tures for binary systems exhibiting complete
12.5 (b). Since temperature is belcrut
solid soltbility. the m.p. of component 1 and since no
solid solution is formed at 'a', solid -1
must precipitate out bn further cooling. ln otherwords AG$trl must have a lower value than
f
AGY. a tangent is drawn at conesponding point 'a' on the free energy curve the intercept
will gire partial, in this case of pure component, free energy which is less than that of solution
as AGr = A6lnl . This is qhown in Fig.12.5(c).
At a larrer temperature Ts+1zy which is lcnarer than the lowest m.p. of the two
constituent, the situation as shown in Fig 12.5(b) will qualitatiwly exists at the other end
of the diagram as well.

138
 I
lltt
l".rl
I

F ig.I 2.4 - Free energ)-eornpositinn

'./ relatianships for various phases in

/r a binary phase diagramwith
partial solid solubility.
I"

:q

At temperature Te tne AC$ and AGr are overlapping on the left side of entectk
composition i.e.point'E', meaning thereby that component 1 and liquid coexist and on the
right similarly, component 2 and liquid solution coexist. lnterestingly the pure phase 1 k;
unstable on right of eutectic E composition and similarf phase 2 is unstable on the left ol'
E. lf situation at temperature Te and composition E is considered then AGU , and AGr shall
overlap on the left of E and AGU , and AGz shall overlap on the right of E to indicate thei
coexistence, which is the case, The AGr will be less -ve on the right of E and therefore beloit'
AGF cune and AG! willbe less -ve on the left of E, and therelore below AGU cune ar
is so obvious'in Fig.12.5(d)

12.5 LIMI]]NG CONDITIONS FOR MISCIBIUry GAP IN PHASE DIAGRAMS

When solubility exists in liquid state the AGf is always -ve, €ls it is a naturell
process. But if the constituent componehts show a +ve deviation fiom Roultian behaviour
then theAGU vduesbecomeless negative and if the+ ve deviation issufiicientlylargrr
then AGF rpy even have + ve values orer certain range of composition. lt means there is lit
139
 I t t
I
I rl"e I

2+E
I t ,
I l I
I I

Fig.125 - Free energ)-composition

relatiorchips at v arious temperatures
far a pure eutectic systefti vis-a-vis
the phase diagram.

I
I
Aor 'I
I

l-
,

tt FFu

tendency not to
form solution and exhibit liquid immiscibility in the region where lGf is
+ ve. The norma! convex downward peak of AGU ,when -ve, becomes convex upwards
as
shown in Fi9.12.6.

!t is clear that for solutions of compositions 1a and AGU are the same since
1u the
the tangent is sqme and hence must be coexstent in equilibrium. Thre AGU howewr
increases beyond 1a with composition and after attaining a peak in convex upward form
reduces to attain the maximum value at 'b' as shown in Fig 12,6(b). For any composition
of solution between the points 'a' and 'b'the composition can be imagined as a mixture of
two solutions, say for example as of 'd' and 'e' with the effective gross composition shown
by 'f'. The aGU for composition 'f is thus that given by point 'c'. But as a .mixture its real
value is depicted by 'f'. ln other words the system has tendency to exist as two liquids rather
than a hemogeneous solution having lG[ corresponding to point c because, free energy
corresponding to point f is lourerthan that for'c'. The stable configuration between compasition
of 'a" and 'b' are two liquids of cornposition 'a' and 'b,. on the left of ,a' and right of ,b'
homogeneous liquid solutions shallexist as stable phases.

14
 t
2\-
SF
-qr !>
7l .t EF
<t,
I .|Y'
.ot t
s<
)'r I

I * l--*-+cOMPLET€
i. --* - j - :lrlMlSClBlLlfY
o-1.
EI
tdl TENDENCY :
l-' TO LtbUrD :
tHMt{ctErur*

rThlO
Itrouto

lo ,? rb

Fig. 12.6 'The free energl-composition relationships

to explain the conditions when liquid immkcibility
becomes imminent.

The liquid misscibility gap is atways associated with a strong like-atomlmolecule

attraction or + ve deviation from Raoultian behaviour. Mathematically it can be deduced that
or O6/Ot,tr or O2g/ON? or O3G/Ctt? must all be zero for nriscibility to occur as composilion
changes from one end to the other in a binary solution. lt means the phase diagrans
musthave a liquidus, notvaryingintemperaturewith composition as shown in Fig 12.6
between compositions'a'and'b'. Also since activity is directly proportionalto free energy
the activity also should either show no + ve variation with composition or it should not
show any variation r.e invariance with composition. This is shown in Fi9.12.6(a) for

141
 (o)
Fig. 12.7 - Free energt-composition
re lationshftts shorv in g the s tabititie s
L2. G+2
of inte rme t allic c omp ounds.

:, i eic
!r-l
i :.. c clc,
96z- cailPogt?toN

compositions between 'a' a!.id 'b,.

That it should not show invense
deviation at some Stage as it varies
with composition goes without say-
ing.

12.6 FREEENERGYOFFORMA-
TION OFCOMPOUNDS !N PHASE
DIAGRAMS

lf the components are such

that their mutual attraction, in terns
of their electronic structure, is suffi-
ciently large then their cornbination
may resuh intheformation of acom.
pound often intermetallic at a cer-
tain fixed composition. A typical
diagrann representing this is shown
in Fi9.12.7. As per phase rule, on either side of the composition of these compound must
lie a two phase region.
The AGU at temperature Tr when erery thhg is liquid is shown in Fig.1 2,7(b) where
it is apparent that over the entire range of composition it is the liquid phase that is stabler
than solid.

At the m.p. of compound Ci.e. at T both the AGU and AGU hare sarne value and
hence both exist in equilibrium with each other. The composition range of stability of a
compound will be between the two entectic points Er and Ez ovenrvhich liquid phase shall
extend.Atastilllcnrer temperatureTs-r which is muchlowerthanm.p.of compound,itwill
seperate out in the range of composition a-b beyond which on either side is liquid phase
stable. ln other words the free energy graphs should show miscibility gap between

142
composition a-b. The tangent shall be cpmmon to AGU and A eB nut there will be two
solid-liquid equilibria on either side separated by a miscibility gap as discussed earlier.

12.7 CONCLUDINGREMARKS
It is therefore clear that the phase diagrams are nothing but the depiction of equilibria
between composition and temperature{pressure being kept constant) and the nature ard
stabilitim of various phases can be guessed from theirthermodynamic information like free
energy of formation of the phases or the component activities and so on.

1 4i|
CHAPTER 13

EL ECTROCH EM I CA L PROCESSES
ennf measurements

13.1 INTRODUCNON
It is a known fact that compounds on dissolution in aqueous or fused media
dissociate into respective ions. The conductivity of such a solution is due to the migration
of these constituent ions, ln such a solution if a metal (more precisely an element), whose
ions exist in the solution, is dipped, the systern now in the form of a metal in contact with its
orun ions, acquires a potential arising out of an equilibrium of a reaction like
M -r Mn* + ne (13.1)
between the metal and the solution. This system is known as a galvanic element or an
electrode. By virtue of the equilibrium of the above reaction, at the boundary of the rnetal and
the solution, the metal part acquires a potential characteristic of its orvn and, dependent on
concentration of ions (more precisely [s chemical activity) and the temperature. This is
known as the'electrode potential'. Two such elements or electrodes when joined as,
electronic to electronic and ionic to ionic conductor, becomes a galvanic cellwith potential
as net algebric sum of the two electrode potentials. Cell is formed because dissimilar
metal electrodes have different potentials. Electrochemical cells are thus of two types, v2.
concentration and galvanic cells. Two dissimilar metal electrodes when joined make a
galvanic cell and two similar metals but with different concentrations of the solutions
around them, when connected, form a concentration cell. Asolution in such cells is therefore
referred to as electrolyte. A typical example of galvanic cell is Daniel Cell with Cu/CuSO+ as
one electrode andZnlZnSOa as another.
Because the electrode potential is due to the reaction as mentioned above, the
overall cellreaction in DanielCell shall be
Zn+CuSOa=ZnSOe+Cu (13.2)
It means that Zn has a tendency to go in solution and eject Cu out of solution as stated above.
Likewise in a concentration cell the natural process is tlrat the nretal ion from electrolyte of
high concentration shall move to the lower concentration electrolyte.

13.2 CELL REACNON

ln either case of these two cells as described above, as the ions move over a finite
distance (greaterthan interatomic distance) electric current has to flow and thus the electrical
circuit needs to be completed. This is posible in both these cases as shown in Fig 13.11.
 ELECTRON'C CONDUCTOR

ELECTRODES

Fig. 13.1 - Galvanic cell showing

z the ion and electron flows.
) J
F F
ru
l!
I
ELEcrnoLYrts

oR BRtDGe
I yEssEt

The electrode reaction as in Eq.(13.'1) is an anodic dissolution reaction or is known

as anode-half-cell reaction generating electrons which are transfened through the electronic
conductor to the other electrode where the reaction
Nn* + ne N (13.3)
takes place resutting in deposition of N on the cathode or it is the cathode-half-cell reaction
which consunes electrons generated at the anode by the anode-half-cell reaction. The overall
cell reaction therefore shall be
M+Nn* + M*+ N (13.4)
There is a conrention of describing a electrochemical cell. The Daniel Cellcan be represented
6
zn I znso+ t*rt ll cusoo (sol I cu
(- ve) (+ ve)
in sucfr a way that positive charge mo\,es from left to right i.e. left hand electrode is negative
(anode) and right hand electrode is positive (cathode). The vertcal lines in the above
represenl the phase boundaries. The cellreaction is as shown in Eq.(13.2) and which is a
natural process. lt only means that Zn is at higher chemical potentialthan Cu or CuSO+
is at higher chenrjcal potential than ZnSOa and hence the above reaction is naturally possible.
Cosequenily the cell shows electro-motive-force, emf tor short, across the two electrodes.
The concentration cell can tlren be described as

zn I znso+ tsor) ll znsoo 1"orr I Zn

1Cr) 1Cz)
where concentration Cr < concentration G z. The natural process is that the higher
concentration (Cz) solution will decrease and the lo,rrer concentration Cr shall increase and
accordingly the current shall flovrr. ln this case by dissolving Zn on the left hand side the Cr
concentration of ZnSOa shall increase and by precipitating Zn out of ZnSOa from solution of
concentration Cz shall decrease.

145
 lf external enrf is superimposed on the cell to counteract the cell ernf then a
stage can reach when the superimposed ernf exactly matches the cell ernf in the
opposite direction, when no current shallflou in the cell circuit. There will be no net reaction
of migration of ions or electrons. This is the equilibrium state and all thermodynanic
considerations of the cell, since at equilibrium of under reversibld condition, can be applied
to such a cell system with no net current floru'
Example 13.1 : The electrode potentials for
co*2 + Co*3+e is - 1.82 volts
andfor Sn*2 + Sn*4 + 2e is - 0.15 volts.
Which one of these two will reduce the other? Also establish the cell that will be useful as a
source of ernf.
Answer : Subtract the haff cell or electrode-reaction of colbalt from that of tin. lt gives
Sn*2+2co*3=2iir-*2+Sn*4
the cell ernf would be the algebric sum of these two
t.e. emf =- 0.15 - (-1.82) = + 1'67
as will be shown in next section the free energy change of such a electrocemical prosses is
AG=-nEF=2x 123061 x 11'67
= -77023 cal/ g mole
(wherein the value of r tras been used as 23061 cal/v/g.equivabnt)
This is thus a feasible process and hence Sn2* shall reduce Cos* or reaction of tin
shall reduce reaction of cobalt. The cell therefore rnay be constructed as :
inertmetd lSn*z lSn*4 ions ll Co.s ions lCo*2 linertmetal
and that this
wherein the positively charged ions shall move from ldft to right as is required
cell shall have net ernf as 1'67 tlolts."

13.3 INTERCONVERSION OF FREE ENERGYAND ELECTRICALWORK

Now G=H_TS
=U+PV-TS
On difierentiation
66 = dU + PdV + VdP * TdS - SdT
where according to first law
dU=g-PdY-Aw'
and substituting in this the second law statement giws
69 = TdS - PdV - 0w'
therefore 6q = VdP - SdT - 3w'
when the cell works in open'normal atnnsphere at room temperature dp = 0anddt=
0and
(13.5)
therefore dG = dw'
i.e. electrical charge from one potential to
The electrical work done by the passage of current
another over a potential difference
w' = charge x potential difference(vottage)
one gram electron
As per Farraday's Laws for cell reaction when one gram atom is deposited,
or 96500 coulombs of charge flcnrs through, This is called Farraday and
is denoted bY 'r"
Ttre number of electrons involved in a cell reaction, if n, then nf is the total
electrical work

146
done by the cell through the cellraction and if the cellvoltage or ernf is z then the work done
or free energy change
for one mole is LG = - rYEf (13.6)
lf dn is the number moles of the ions passing through the cell for a spontan+ous celi
reaction then
AG=-ftEfdn (13.4
Now since AGo is the standard free energy change then conesponding z would be
EO
therefore - tEof
AG = (13.8)
substituting in AG - AGo = - RT ln K as in Eq. (7.8)

-JYET * nEfo = -RT ln K

E=Eo * 4 hK (13.e)

The sarne relationship can be defined by considering a general reaction as in

evolving the equilibrium constant wherein as in Eq (7.6).

= Rr'"**
^c-^co
tur - rvEo fo = RT ," "9 +
aR aB
(13.10)

where f, o is the cell enrf when reactants and products are in there standard states and then
AGo = nao f

Thus f,,=Eo+ RTln' a8 a$ (13.11)

aft aB
or = Ea + RTInK as in Eq. (13.9)

13.4 ACTIVITIES AND EMF

Nornrally it is not possible to have a cell reaction in which sewral reactants and
products will be constituting the cell reaction. These are vgry limited, often to one each and
therefore
?Product
E = Eo + RT 1n O (13.12)
reactants
This is also valid at only reasonable concentration and not at too loyv a concentration of either
the product or reactant, otheruise r might tend to becorne infinite.
The concentration cell is therefore simple enough to analyse. lt can be represented

Pure metal A electrolyte inpure metal A

with An* ions i.e, alloy
The cell reaction in this case is
A (prre) + A 6mpuc or a!oy)
an timPul.e)
AG = AGo + RT ln aA (pure)
(13.13)

'147
aA (pure) is one and when both are pure electrodes the cell enrf is zero and hence AGo =0
and it will reduce the above to

AG = RT ln as = - nAz- f or log an = ##
which is extensively used for estinration of activities of various molten alloys using emf
measurements.
Also activities of compounds in solution can also be found from galvanic cells. Let
us take cell as

Wwire lng lAgzO.BzOe.Sioz l02 (latrnosprerel I Ptwire

The net cell reaction is
4 (Ag) + {Oz} = (2Agz O)r (electo}yb rnelt)

E1 = Eo + # ln a2

similady at some other concentration of (Agz O) the enrf can be measured as

E1 _,-2= #," ffie)
out of these (Agz O)z can be taken as pure or melt saturated with Ag zO and therefore the
(Agz O)z shall be one and subsequently the activity of Ag zO at other concentration can be
found out. This technique in principle is used extensively for estimation of activities of oxide
components in molten oxide systens of interest in extractive metallurgy and in refractory
materials.
Example 13.2 : Forthe cell
[,to I ng--si I Bzog sioz - I
oz (r 'tno") pt I
(silica sat. melt)
the cell ernf was 1 .70 volts with Ag-Si was saturated with silicon. The cell emf was 1 .55 vohs
when Nis = 0.1 in the Ag-Si alloy at 11150o C. Also calculate the standard free energy of
formation of silica at 1100o C and acivity coefficient of silicon at N5; = 0.1
Sblution : The cell reaction is Si + Oz = SiOz where all are in their standard
states.Therefore
AGo = -n|r - 4 x fl.ZO t 123061 = -156814cailg mole
(wherein lhe vdrc of Fanaday constant is 23061 calfu/g equiv)
When silicon is not at unit activity then the cell reaction can be written as
[Si] +
{O2} = <SiOz>
and " K = €l ( SiOz> / alsq x p{or} and since the'activiV of silica is unity
because the electrolyte is saturated with silica and since oxygen is at one Stmosphere
therefore K= 1/atsil
AIso the non-standard free energy of formatip.q-of silica is
AG=-IrEf = 4x t{j.5*flx 123061 = -142978
Hence AG = AGo + FIT ln K
or -142978 = -156814 + RT'n =L
alsil

or Rf ln ! = '142978 + 156814
a[si]
or log a1si1 = 2.11 i.e a1si1 = 0.007
1zA
or ysi =alsil/Nsi =fffl=0.07
Example 13.3 : Calculate the {ree energy change at 8000 C for the reaction
FeO+Ni = NiO+ Fe,
if the cell emf for the cell

uo I uooz I electrolyte I r"o I r" is 17,8 mV

and for cell Mo I UoO, I electrolyte I f.fiO I rui is 28.4 mV both +r 800oC.
Solution : Cell reaction of tuo I naooz I electrolyte I f"O I r"
is Mo + 2FeO = MoOz + ?Fe
and for which the free energy change is, assuming FeO in pure form
t ."fu
a G% = - n Ef = -RT ln "rt'root
€lMo x aF.O

and for the cell n,lo I n,loo I electrolyte I ruiO I Ni

cellreaction is Mo + 2NiO = MoOz + 2Ni
,.-
and lGffri = - nEr = - RT In 9U9o'- =fr'
aMo x aNiO
For the reaction FeO + Ni = NiO + Fe

the free energy change lGPe-r.ri = - RT rn *9#

aNi x €lFeo

- 4 x 17'8 x F
Dividing AGFe-Ni = / AcRr, = 2ln +9:=
AGPe
aNi x ?Feo 4x 28.4x F
17 '8
Therefore aGo = - RT ln -
28.4
=- 650 cailg mole.

13.5 TEMPERATURE COEFFICIENT OF CELL EMF

Nowsince AG = nEf
-
dAG = - n f dE
/j4 G\ =-nr dE
*
t-ait;
and at constant Pressure

(#),=-^s
and since cells are generally operated at atmospheric pressure this condition holds true.
Therefore AS= ^t#
It means that the r values are to be measured at various temperatures of the cell and
the slopeof the plotof c vs. T givesthe entropy changeforthis reaction. Once AS is known
from knowledge of measurement of r vs. T, AH can be computed as
AH=AG+ TAS

--n?-y+nrT:f
'149
Once AG, AH, and AS are known along with the activities of the constituent of the electrode
then by using the following
6X =RTlnan = -nA E f
s[ = nn'(#),
HX = na r E tE -r l#)
l.dr io'
r

and exeess properties can also be calculated.

Exampte 13"4 : The enrf otthe cell co I coctz -Cuglz I cu is 1.2 vohs
at 0o d and 1.15 volts at 25o C. Calculate the AH of the cell reaction at 25o O.
Solution :'The anode reaction is Cd + CA*z + 2e
and the cathode reaction is Cu*Z+ 2e + Cu
The fiee energy change of the cell reaction is
AGzzsok- 2 x 1'2 x 23061 = -55346
=
nnd AGzgsok =-2x 1.15x 23061 =-53040
- 4%=#@ and after puttins the abor,e
AS =
# = g2.2caVg mole/0o
values =
Now since AH= AG-TAS
= -, 53040 - 298 x (-92'2\
= -25624caVg.mole
.!3"6 SOLID ELECTROLYTE CELLS
through
Certain solid solutions behare as ionic solutions since the current is carried
ThOz-€aO'
thern by coniCuction of ions. Some typical examples are ZtO -CaO(4%),
as ionic conductors above a certain
Thoz-yz os (7"a1 or ThQtanthana. These behar,e o these are oxide
o{ygen partial pressure which is fairly,lol like 1 0-1 atms at 1 0000 C. Since
can be and
cells made with special oxides as electrolytes at high temperatures, these
reactions at various
have been used for thermodynamic measurements of oxidation reduction
pressures of oxygen including vacuum'
The cell used are of the tYPes :-

< cD - < crz Qg> I zro, cao) I fot in Fe alloy

(a%

< cD - < crs 03> I z,o" - (7Y"Yz os) | tq in Fe alloy

for which the cell reaction is

2t
! <c> + [O] = - E
<CrzOe>

or E - E" = # ln ao since bith Cr and Crz Os are in

solid state.

150
 in Fe is replaced by orygen
Where rois calculated fiom the measurements of cellwhen [O]
knowingthefree energy of formation of
at one atmospheric pressure or can be.calculated
< Crz Oe> taking c as 23061 cayv' This
has been reduced to
13580 --10'oB E
log ao = 4'62 -

where c is expressed in rnilirlolts and T in

K
in Fe with any other elernent like C, P, Mn, S etc'
By employing the right hand electrode as [O]
the log {io = eL t z' i) been
can also be found out' The results'haw
and thus even the interaction parameters
tally with actual rneasured ones in most cases'
extreamly good in that the calculated ones
A cell of the tYPe

<Mo> I tMoo" lz'ot-c"o I truio'<Nio>

if set uP, means
afri
s, = Eo + RT ln luoo'
aNio'aMo
ChemicalpotentialofNi_Nio-ChemicalPotentialofMo.Mooz
EnE
Knowingchemicalpotentiall.e.partialmolarfree:n"'qlofoneoxidetheothercan
befoundoutbymeasuringthecellen/-x,.oralternativelycellenrfrisrelatedtothefree minus
sum of the free energies of products
energy of cell reaction anJwhich is the algebric oxide'
can be found out knowing AG of the other
reactants and f rom these the aG of one oxide

13.7 CONCLUDINGREMARKS
For example the conductance measure-
Electrochemical measurments are rnany. is
ments, ion transport estimation and so
on. But the electrochemical process or the change'
by the passage of current through the system'
only the change in chemistry accompanied
with this process and hence is of direct
The energy exchange is inalienably associated
interestinthermodynamicmeasurments'Thetechniqueoferrrfmeasurmenthas.been the
of thermodynamic parameters' in particular
extensively exploited ior the estimation
properties of the system in which the chemical
activities and therefiom the other energy
The use of solid electrolytes has
reactions occur inthe form of electrochemical'processes.
at fairly high temperatures'
opened up a large vista for measuremenis

151
CHAPTER \4

FREE ENERGY- TEM PERATU RE

RELATIONSHIPS
ellingham d.iagrams

14.1 tNlRODUCll0N
The free energy composition relationships were discussed in Chapter 12 to confirm
the nature of phase equilibria inwlved in phase diagrams. Often it is necessary to know
the free energy-temperature relationships, which are nothing but equilibria as affected by
temperature, to establish the temperature at which it is advisable to earry out a certain process,
particularly the chemical processess.
Free energy change in a process is an indication of whether the process is at
equilibrium, or it is natural or spontaneouly occurring or, it is an unnatural process. Since
most processe of interest are as such unnatural processes (see Section 1.1). the conditions,
particulady the temperature is of direct interest for its commercial exploitation. Free energy-
temperature-pressure relationships are of direct use in this regard.
Any material form, say for example iron oxide, as sorne oxide mineral of iron, needs
to be reduced to produce iron. lron oxide as such has a fairly loru value of free energy and
hence stable. Had it not been so iron would have been produced as a natural process,
which is not the case. Conditions can be altered suitably such that iron oxide becomes
unstable as compared to something else. For doing this pure iron oxide system is transformed
into an iron oxide plus carbon system and heated to a.suitable temperature (above 9000
C).Thefree energy ofthismixtureishigherthanthatof ironandwhatewrother products,
CO or CO2 produced at the higfr temperature. The system therefore undergoes a transfor-
mation and produces iron as desired. lt neans the reaction
FezOs + 3C = 2Fe + 3CO (14.1)
takesplace and the reduction is possible to produce iron.Knowledgeoffreeenergy-tern-
perature relationships is guidance to convert otherwise stable system into a unstable one
and get the desired results.
ln the reaction above, original iron oxide is destroyed, and a different one, oxide
of carbon isproduced.Thecondition oftemperature,inparticular,isfixedinsuch a waythat
the above reaction is possible or net free energy change is negative for the owrall reaction.
The above reaction consists of two reaction equilibria as
 ZFe+3l2Oz = FezOe (14.21

2C+Oz =zCO (14.3)

which are, manipulated suitably for reaction in Eq. (14.1)to occour.

With reference to above it can be said that if the relative stabilities of oides, as
to carry out
they vary with temperature, are known a suitable pair can be worked out as above
the desired process. The qualitative term stability is quantified as free energy
associated
formation of the oxide the
with these various ofdes. More negative is the free energy of
more stable is the oxide.
The same arguments can be made for any group of substances like sulphides'
sulphates, chlorides, silicates and so on, in which case one sulphide will
be destroyed to
produce another sulphide and thelike. lt means for example, that
Sbz Ss + 3Fe = 2Sb + 3FeS (14.4\

would be possible to produce antimony provided the free energy of formation

of iron sulphide
is more negative than that for antimony sulphide at some temperature at which the reaction
will have to carried out'
All chemical processes are qssentially making, breaking or both of chemical bonds.
ln the manufacturing processes chemical compounds are broken and new
ones are formed'
Being non- nuclear, the element remains the same before and after reaction but
the

compounds change from reactants to products'

The free energy ol compounds i.e. free energy of formation of compounds can
act as a stability scale for compounds and since free energy is a function of temperature
energy of formation
the stability shall also be subject to temperature as a variable. The free
of GO is more negative than that for iron oxide and hence reaction as in Eq.(14'1) is readily
possible.similar is the story of reaction in Eq' (14'4)'
The free energy value changes with associated interacting substances i'e'
with

the activity of the constituents in reactants and products as

G=Go + RTInK
where ( = proAuct of activitY of reactants
Hence the changeo*ifro* pure to impure form of reactants and products are going to
affect the free energy and hereby the relative stabilities.
All these tactors arequantitatively examined in the following in terms of free
energy-temperature relationship.

I4.2FREEENERGYoFREACTIoNSINVoLVINGoNLYGASES
difference {rom those
This group is separately classified because of its characteristic
typically represented by
reactions involving condensed phases' These reactions can be
(14.5)
ZHz+ Oz=2HzQ
(14.6)
2CO +.62 = ZCOz
(14.4
2SOz +'Oz = 2SOe
Let us consider the free energy of reaction (14'5)
AG = 2GHzo - ZGnz - Goz

153
This will be in equilibrium if AG = O
Therefore 2Gs.o = 2GHz * Goz
Putting Gi = GP + RT ln pi as in Eq.(4.56)
2Gffro + 2RT ln pH2o = 2GPr, + 2RT ln pxz + G8, + RT ln po2 (14.8)
where pHzo, pH and po2 are corresponding partialpressures of the components involved in
the reaction.
Since a G&a.st = zGR"o - 2G,ha - GBn therefore,
AGlrc.s) = 2RT ln Pxz+ RT ln ps, + 2RT ln pHeo (14.9)

= ZRr,^
#
= _ 2RT ln l$
rao
or 2los l$ = ;ffi (14.10)

where Kp is the equilibrium constant of the reaction when gases alone are involved. lt is
different from the equilibrium cbnstant Kx which takes into account condensed phases as
discussed in Sec 7,2.
For Reaction (1a,5) it has been estimated
aGfra.si = - 118000 - 26.757
therefore at 16000 C or 18730 K

2togKp,,ou,= * ffi * ffi = * 19.616

or log Ke = 9.808
lf it is assumed that the atmosphere required should have
Poz = 10{ atmos

then los Ke = e.aoo = **#

or Pn' 9.808 x 102 = 980
PHao
lf pH, is one atnnsphere then PHzo = 1/980 = say approirrately 1 mrn Hg
The temperature of water when its water Vapour has a pressure of 1 rnm Hg and if
hydrogen gas is bubbled through the water at this temperature then the resultinrg gas mixture
shall hare a ratio of px. /pn2o as desired above to be equivalent to partial pressure of orygen
equalto 10a atmosphere. Similarly for reaction (14.6)
AGo= -115200+ 41.507
2logKe=+;#-#
al1273o 2log Ke = 19.78 - 9.071 = 28'85
or log Ke = 14.42

= *
tog Ke
ffi p'$, -- l+.+z

lf partial pressure of oxygen is to be kept at Poz = 1A-2 atmos"

154
 Pco = 14.42 x 110 atmos. or Pco
= 142
PCOz PCOz
The CO and COz gases will have to be mixed in such a way as to obtain the aborc ratio to
obtain poz = 10-2 atmos. lt neans 142 ml of CO will haw to mixed with 1 nrl of COz to get
the required mi)*ure. Often Co2 is generated and passed over carbon at appropriate
temperature when COz + C = 2CO reaction shall qccur and the appropriate ratio of
CO/COz willbe obtained. The details shall be discussed later'
It is thus clear that if the sum total of powers to which each of the reactants and
products are to be raised are equal the equilibrium shall be independent of pressure but if
these two are'not equal, which shall often be the case, these equilibria inwlving onty
gaseous components shall be dependent on pressure. The system shall then be defined
at a given temperature and pressure and in that case at quilibrium the free energy change
willbe zero.
These reactions are often between two
elemental component gases giving rise
to a gaseous compound and often shall be pressure dependent, A reaction of the type
CO+HeO=COz+ Hz (14.11)
can be imagined and since

*' _ PCOex
pco x pH2o
PHz

appears as pressure independent. A clqse scrutiny of such a reaction shall reveal that this
reaction is a combination of reactions in Eqs. (14.5) and (14.6). lt only means that the two
shall be independently in equilibrium at'a fixed po, value. For the Eq. (1 a. 1) to be in equilibrium,
p6rforbothofthereactionsinEqs(14.5)and(14.6)shallhave to be equal. lnotherwords
the pressure is fixed for the equilibrium and that is what the K for (14.11) reveals.

1 4.2.1 Orygen Potentia l-Tem perature Relations hips

Eq. (1a.5) gives

AGo=-RrrnKp= Rrh[#5" q (14.12\

or RT ln poz= aco - 2RT ln fj}) (14.13)

\Pxzo,1
Forthe reaction (14.5) AGo = - 118000 + 26'75T caVmole
Therefore RTrn poz = - 118000 + zl.rs r - 2RT f*31 (1a.13a)
'" \PH,o/
For any fixed ratio of HpJHzO the RTln po2 can be correlated with temperature and
it shall be a straight line. fhe intercept of this line on y-axis Le. RT ln poz axis shall be
- 118000 cal. This is shown in Fig.l4.1 in the form of point'H', When Hz lHz O ratio is one
the RTln po. shall increase (move in less negative direction because of + 26'75 T factor)
'H' as
with temperature and hence the line is sloping upwards from the anchor point
temperature rises. lf Hzl1z 0 ratio increases beyond one, for ewry ratio the line will slope
further up but originating from 'H'. lt Hz lHz O is less than one the term 2RT ln (pn. /pHro ]

point l-
shall be + r,e and hence the RT ln po, will move to rnore -\,e side but every time the

155
 RATIO
lhttko rt. .6*rd dtc tr rp
t;c
x;5
c -l
o to
E
(il I
<t
r.d
H
d
5-: 50
lo
6

F
E

2t
tp
+TElrPcK 500 1000
to4
los9

Fig. 14.1 - Vari"att)n of RT in po, of the reactinn 2H + Oz = 2HzO with temperature.

shall remain the same for all such relationships and hence all lines for various Hzl4ze
ratios shall originate from point 'H'. For all RT ln po. has the same value as shown by point
H. As HzlHzO ratio decreases below one the lines shall rotate in clockwise and
when it
increases beyond one in anti-clockwise direction as shown in Fig 14,1 .
Additional scale of Hz/HzO ratio is superimposed on this diagram in the form of
additional y-axis.ln order to make the diagram a compact one since lhe y-axisextends on
either side of x-axis over a long range ,the y-axis of HzlHzO ratio is bent at right angles and
made parallel with x-axis. The lines emanating from the point H upto the vertical y-axis shall
cut across the bent y-axis and which is now marked by corresponding values of the gas ratio
ol HzlHzO. This obviates the need to have long y-axis and thereby the diagram becomes
more compact.
For the reaction in Eq. (14.6) the free energy change
AGo = - 135000 + 41.S7Tcallmole.
This can be processed as was done in the case of reaction (14.5) to get

Rr rn po. = - 135000 + 41 57 r- Rr ," (14.15)

hff) for fixed values of
and by the same logic RT ln p6, can be plotted against tempbratuie
COICOz ratios as shown in Fig.14.2.
The valueof RTlnpo2 at zero degree absolute is-135000ca| andisshownas
point'C'. Allthe lines forvarious CO/COz ratios shall originate from point 'C' and shall move

156
 J

-t€ -l+ -!l .b -t

to to to lo lo

lo
{
L'{
tp-
c
lo
t&

,o
io
t'tl
10
l+

#
to't

IEL
TEMP oK-5OO lroo t9

Fig.l42.VariationofRTlnpo,ofthereaction2Co+oz=2Cozwithtemperatute.
clockwise on decreasing the ratio belorr one and anti- clockwise beyond ratio one on
increasing the CO/COz ratio. ln this diagram the additional y-axis is of CO/COz ratio and
is shown in the same way as in Fig'14'1'
This is the standardised treatment and may apply to even reactions in Eq' (1a.7)
likewise. Sirnilarly for a reaction
ZHz + 32 = 2HzS
the RT ln p9, shall now be replaced by RT ln ps, and related to temperature at fixed ratios
of Hz /HzS. This is useful for sulphidation reactions. Erren a reaction of the type
ZHz+Cl2=1116S
could be amenable for chlorination procasses as a function of ternperature and Hz/HCl ratios'
So far,theratios ot{zlHz0, CO/COzand thelike areused towork outtheirvariation
as a function of temperature and a function RT ln po, . The requirement is however to know
the variation of RT ln poz 6 such as a function of temperature and partial prqssure of oxygen
itself.
This could be established using the relation given in Eq'(a'56) as
' G=Go+RTlnp
lf pi is one atmosphere of pure oxygen and assuming it to be the standard state for oxygen
then G - Go
confirming that it is the standard state and in equilibrium with one atmasphere of o{ygen'
which is obVous enough. The RT ln po2 = 0 shall be the correlation of RT ln po, with

157
tenperature at po, = 1 as shown in Fig 14.3 in the form of the horizontal line passing
through the point RT ln poa = 0 at any temperature.

a
ta
a
lo'

to^
!o
L
lo-t
-2
10

Fig. 143 - Variatian of RT ln P orwith temperature for various pressures of orygen.

a pressure less or more than one atnnsphere

lf the gas is now at then,
RT ln Poe'= G - Go
or in another words it will be the free energy change from standard state when p9, = 1

to a new state when po. r 1, i.e. for the reaction

Oxygen (one dtnns) = o)rygen (other than one atnns)
This free energy change varies with temperature for a fixed partial pressure of
oxygen as can be seen in Fig. 14.3. All these lines shall origanate from RT ln po. = O at
T= OoK shown as point'O'in Fig. 14.3. lf oxygen pressureis less than one atnns
(por<1)the lines shall moveintowardsmore-ve freeenergy and if olygen pressureis
more than one atmos (port t1 towards + ve free energy values. This is obvious because
pure oxygen, being at higher qhemical potential shall change to oxygen at le-ss than one
atnnsphere as a natural process or spontaneous change with -ve free energy change and
just the opposite when pe2 1 atmos.
1s8
 As before an additional y-axisaspg, is provided on the diagram as shown in Fi9.14.3
The equilibrium of reactions as in Eqs. (14.5) and (14.6) indicate that at a fixec
ternperatureif theCO/COz sr HzlHzOratio isfixedthenRTlnpoz i'e'poz assuchisfixec
and the system is invarient.
ln fact therefore co/coz , HzlHz 0 and po2 are one and the same thing
clear from the
expressed in different forms for a certain fixed temperature. This will be
folloriing.
LetusdrawafallinesofdifferentCQlCOzratiosoriginating frompointCanddifferenl
Fig.l5'4- Lel:
lines originating from point O such that they intersect each other as shown in
P%

u
o
E
.{.
d
U.
{

TE,{P oK 5oO
-

Fig.l4.4.Theinterrelationshipsofpo,,ColCozandHiJHzoratios
, und the method of determining these graphically'
the above mentioned intersectiort
us draw a line originating from point H and passing through
Let us say that Poz and co/coz lines intersect at a point shown as A' Let tri'
point A as T as shown' Therefo'tr
also work out the temperature corresponding to this
ratio to be read from relevantl
at temperature T the point A indicates certain fixed CO/COz
y-axis.
direction o-Ato lxr
The same point A also indicate a certain fixed value of p in the
read from aPProPriate Y- axis.
equilibritrr
lf alineisdrarynfrom H through point A it will read thecorresponding
H/H1O ratio on the conesponding y-axis'
159
Therefore at the temperature T in Fig.14.4,
CO/COz ratioof 10-2 = po, of 104 = HzlHzO ratioof 10-1'2
ln otherwords the orygen potentialcan also be depicted at any temperature by eorresponding
ratio of CO/COz ar HzlHz O.
So for a certain po, value at a certain temperature the equilibrium CO/CQ ratio can
be found out like this. Draw the relevant po. line originating from the point O. Draw a
vertical through the required temperature to intersect this line. Draw a line from point 'C'
through this intersection to read on COICQz ratio axis. The value obtained is required
COICQz ratio. lf in stead of through C, line is drawn through point H it will read the HelHzO
ratio on the HzlHzO axis in equilibrium with poz . This is shown in Fig.14.4 by the point B at
temperature of 9000 k when poz of 10{0 = COICOz ratio of 10-12 = Hz l1z O ratio of 1011. lt
is rather easier to obtain the equivalent CO/COz or Hz l{z O ratios for a given partial
pressure of o4ygen. All these indicate the oxygen potential of the system in various forms.

14.3 FREE ENERGY OF REACTIONS INVOLVING ONLY CONDENSED PHASES

This is other e>trerne when no gaseous component is involved in the process.

Such reactions can be exemplified as
2CaO + SiOz = 2CaO. SiOz (14.12)
Ale Oo + 2CaO = 2Ca0. Alz Og (14.13)
or e\€n carbide forming reactions from carbon and metal also falls in this category; and such
others which are more common in ceramic systems. The reactions of this type are
possible only because the two condensed phases harre opposite chemical characters.
Reactions of the type
CaO + MgO = solid or liquid solution
leads to solid or liquid solution and not any chemical compound . Out of the simple compound
types therefore acidic oxides like SiOz, PzOs and sornetimes Ala Oe form chemical corn-
pounds with alkali or alkaline earth metal ofdes as silicates, phosphate and aluminates.
The fiee energy change of these reactions are independent of pressures. These
however do change with temperature, as it should be. The free energy is given by
AGo= aHo-TASo
in which the enthalpy can be treated as substantially constant and is equal to the intercept of
the line G vs.T on the G axis at zero degree absolute. The entropy change in a reaction of
the type (14.12) shall be
AStt+.rzl = S!c"o.sio. - Slc,ro - SBio,
Since all are condensed phases the individual entropy values are small and are
often of the sanre magnitude'for many substances in the same physical state. Therefore
ASo is very smallfor such reactions. The net result is that the AGo vs T plot for such reactions
donotvarymuch with temperatureand hence areparalleltothetemperatureaxisas shown
in Fig. 14.5. These lines do change their slopes if the reactants or products are not in their
standardstates.This non-standard behaviour shall bediscussedinageneralisedway later.
These plots also change their slope when phase change occurs in either the reactants or
the products or both. This also will be discussed in a generalised way for all reactions later.

160
 o
o
f
;
(,d
!
g

ro00
TEMPEMTURE+
F ig. I 4 5 - F re e e nerg) - temp
e r a tilre re latio ns hip s

of reactians involving only condensed phases.

14.4 FREE ENERGY OF REACTIONS INVOLVING CONDENSED AND GASEOUS

PHASES

what will be the differance if condensed and gaseous components

are involved in
a reaction ? There is considerable difference between the earlier
two categories and this
category as far as the free energy of reactions are concerned.
By far this is the most widely
encountered practical system, because chemical changes often
involve a gaseous compo-
nent like oxygen, sulphur, halide, carbon dioxide, carbon monoxide,
and so on. ln fact in
the inorganic world these reagents dominate and hence becorne
a part of processes for
consideration. The free energiesof reactions involving any one of these shall be a function
of
pressire of such gaseous components. The free energy
does vary with the purity of the
condensed phase (s) in the reaction but this deviation shall
be discussed later. lntially only
pure condensed phases shall be deatt with, The equilibria
is typically described by realtions
of the type:
MetalM +egas=Meoxide
or Metal M + Sz Gas = MSz sulphide and so on.
Let us €lssurne that the temperature for consideration to be
belor the metting point
of both metal as well as the oxide and hence let us study the
heterogenous reaction of interest
exemplffied as
M (solid) + Oe (gas) = MOz (sotid) (14.14)

161
or as per the conrention used
<M> + {O2} = <MOz> (14.141
The temperatures are such that the partial pressures of M and MQ in gas phase
shall be extremely small and each oi this solid phase shall be in equilibrium with its o\ rn vapour
pressure. The free energy of solid phase is thus equal to the corresponding partial part in the
gase phase. Under the present conditions the partial pressure of solids in a gas phase vary
negligibly eren when pressure changes considerably. Hence the free energy of solid phase
can be conwniently assumed to be independant of pressure and thereby assumed to be
constant.The gas phase equilibrium can be written in the form of reaction as in Eq. (14.14)
and the solid statqs thernselves to be taken as standard states.
The fiee energy change of reaction (14.14) is thus
AGftr.rl; = AGftoz - AGRr. - AGU
=-Rrrnlila I
aM x po2l I
(14.15)

and AGo = - RT tn k = + ni ln 1ps,1 (14.16)

where K is the equilibrium constant of the reaction.
It also means that free energy can be expressed in terms of partial pressure o{ orygen as
AGo = RT ln po, fi4in
and therefore the y-axis plotted in Fig. (14.1), (14.2) and (14.3) as RT ln pe. is nothing but
the standard free energy change AGo of the reaction. Thus the RTln poz w.temperature
plots for metaUmetal oxide systems are nothing. but the free energyltenperature relation-
ships. Likewise the RT ln ps, shall be the free energy term for sulphides in such diagrams.
TheAGft+.r+listhusthestandardfreeenergychangeofthereactionasin Eq.(1a.1a).
Since K is a function of temperature, at any fixed temperature the equilibrium shall be
decided by the unique value of the partial pressure of o4ygen in contact with solid metal and
its oxide. From phase rule it can also be deduced that the equilibrium in reaction (14.14)
has only one degree of freedorn.
Therefore in a metal-metal oxide-oxygen system if partial pressure of orygen is less
than the equilibrium value, metal oxide will spontdneously reduce to make up the partial
pressure of oxygen to the equilibrium value and if the oxygen partial pressure is more than
the equilibrium, the excess oxygen will spontaneously react with metal to reduce itself to
theequilibriumvalue. Both solid phases, of course, must existtogetherforthistohappen.
The actual value of equilibrium p9, depends on the temperature and if the ternperature
varies itsvalue atsovaries accordingly. Similartype of equilibrium partial pressures of sulphur
in the case of sulphides, of COainthecaseofcarbonates,ofHzOinthecaseofhydroxides,
of Glz in the case of chlorides and so on shall be existing and could be erraluated,
ln all such equilibria therefore, if equilibrium partial pressure of orygen (in oxide
reaction) and the others in the conesponding reaciions, if plotted as it varies with
tenperature then the equilibrium gas pressure line shall divide the area in such a way that
the conpound plns the gas shall eist on one side and the metal ph.s gas shall exist on the
other side as shorn in Flg.l4.6.

162
 o

ZnO 02

or-40
tr A2
crl \
o

o'5 1'O t'ta*"ot, 2.4

INVERSE TEMPE RATURE

Fig. 14.6 - The varintian of equilibrimt arygen potential with temperature

orygerL
for a system containing zirc and. oxygen and" zinc oxi.de and
14.4.1 Varlatlon of Free Energy wlth Temperature
The free energy change in a process is given by

AG= AH-TAS
change
It means that there nny be a change in enthapy and entropy of the system leading to
in free energy of the system. Although AH varies with tenperature, the variation being nd
large, it can be treated as substantialty constant o\€r a hirty wide range of tenperature
without inwlr,ing much enor. The As horever changes with tenperature in a
pfo@ss'
partiorlarlywhen ga.seous phase is inwfued along with the condensed phases'
ln the generalised chemical process of oxidation of an elernent as depicted in
the enthaply
Eg.(14.14), if the reaction occurs as a natural process, may be after initiation,
change of this proess shallbe negative. Since one g . mole of o4ygen gOs disappears in

this process the entropy must decrease appreciably in this process.

TSese changes are qualitatively shown in Fig. 15.7 plotted under standard
conditims.
The plotting of such a plot needs the knowledge of enthalpy of this reaction and the entropy
change accompanylngine process. The slope of the AGo r,s. T plot is ASo . The
ralue of
- TAS; continuously increases with temperature. lf actual exanples are taken for calcualtions
all will slrorv simila'r
and the AGo values forvarious oxide formation reactions are plotted, they
originating from various lewls on the free energy axis equivalent to AHo of that
straight lines
reaction. For similar reactions of variotts oxides the entropy is given by
So(rrpu0 -. Sopxypnpcl
ASo = Soloritc; - (15'16)
possible 1n
and since So 1O**1 and So (mcbD are practically the same, within the accuracy
of
such catculations, the entropy change arlse predomlnantly due to dlsappearance
163
 tooo o
i+

Fig. 14.7 - The variation of free energl, enthalpy and entropy for the reaction
413 <Cr> + {Oz} , = 213 <Crzos> with temperature'
(LGo = - 178500 + 41'4 T) callgmole
one g mole of orygen gas and hence lt is negative. The entropy changes in various such
value. The AGo vs'T
metal oxidation processJs are expected to be substantially of the same
plots therefore slope upwards, towards less negative values of AGo ' lf the
fornrat of the
ieactions change from that in Eg' (14.14) to something like:
2M + Oz = 2MO (14'19)
4lg Al + Oz - ?3 AlzOs
(14'20)
or other'
even intersect each
their slope is likely to change and in that case thege lines may
There are exceptions to this general nature of slopes of these AGo vs'T plots in sorne
oxidation reactions. The case of
(14.21\
< C> + {Oz} = lCOz}
gives entroPY change as
lSf+.ztl=S8oz-Sc-So,
and
Since the gas entropy values are far more than that for the condensed Phases
since Scoa = 56, and Scoe >Sc or soa sc >
and therefore the resultant ASo = O for the above reaction'
dioxide gas has to
It only means that the AGo vs.T plot for the formation of carbon
be a nearly horizontal line as is the case. This is the exceptional behaviour
of c* coz
reaction.
ConverselY for a reaction

1g
 2C+{O2}=2tCO} (14.221
ASo=2Sco-2Sc-Soz
the resutt is increase in entropy, as volurne is nearly doubled, for the same pressure.
The slope of AG vs.T shall therefore be reverse of that of the earlier metal oxidation reaction.
This is in fact true. This is yet another exceptional behaviour of C *
CO reaction and slope
being in opposite direction, this line cuts across many other oxide lines and alters the relative
stabilities of the oxides vis-a-vis the carbon monoxide line and which is fully exploited in
extractiw metallurgy for metal extraction purposs"
14.4.2 Free Energy Change With Change of Phase.
So far the discussion centered around when both the reactant and the product
remained in soiid state. While probing the free energy temperature relationships the
temperature of interest may be high enough such that either the reactant or the product or
both melt or evaporate because the temperature under consideration transgresses their
melting or boiling points. The question arises as to whether the straight line relationship of
free energy and temperature continue to be the sarne or change at these temperatures when
phase changes occur? ln other words will it have the same slope or the slope changes?
Let us take the case first when the metal or the reactant first melts. The reaction
then shall be written as
(M)+{Oz}={MOz} (14.22)
The entropy of the reaction upto the temperature when all are solids has one value as shown
in Fig.14.4. Now the entropy change of this reactlon is different from that in Eq.(14.19) as
AS[ra.zzl = SMoe Su So, (14.231

lf the Eqs.(14.19) and (14.231are compared, earlier the entropies of solid metaland
solid oxide were equated, but now being different phases, are not that similar to be equated
to canceleach otherand hence Srr,t (liq) > Su (sotia).
Hence ASo = more negative ttan when both were solid,
and in that case if the free energy of Eqs.(14.19) and (14.23) are compared as
allsolid phases lG?= AHo-TAS?
when reactant is liquid AG8 = AHo - T ASa
wherein -TAS?<-TAS8
-AG?>-AG2
ln otherwords the plot shallslope further upwards or slope of this AG vs. T relationship
shallbe more as shown in Fig.l4.8 fromthe point of temperature as melting point.
Since the entropy of melting of many metals has fairly the same value the plots for
many metals may again be parallel originating from their respectiw melting points onwards.
lf the temperature of the process of oxidation is such that the reactant boils, the value
of - TASo shallbe still morenegative than that for liquid. ln that case the AGo vs.T plot shall
slope further upwards or rnore tcnrvards less negative value of AGo as shown in Fig'14.8.
Let us now consider that instead of reactant melting, product oxide melts first then
the entropy change shall be

165
 'Tat?)
('
d
td h(P)
zlll // Fig" 14.8 - Vari"atinn af free en'
ergt with temPerature as Phase
ad
ul ./ ,',
Tr94--' changes takz Place in reactnnts
tIr ,'
cndproducts.B is used to de-
og rnte boiling Point ntd M is
ttsed to derwte melting Pint
rll
The subsgriPt R is used for
reactant and P is used for Prod-

I uct.

TEMPE RATU RE
(14.24l.
ASo=Sogqridoxiic) -So(solilmcbl ) - So{st")
less negative than when both are solids
=
more negative than when both are
Therefore the free energy in this case shall be
oe more negative than when mefting of the
solids and as a result the slope of the line shali
dowrnalards when it boils' The combination
reactant occurs. Likewise these lines bend further
The actual values for a
oi allthese changes haw been shown Fig.1a.8 schematically'
values of various phases involved in
specific case can be evaluated knowing thl entropy
the reaction
Exanrple 14.1 : For three difierent forns of reactant metal M, solid, liquid and gas, being
oxidisedtoproducesolidmetaloxide,threedifferentvaluesoffreeenergychangeswere
obsened as :

(1) 2<M> + {o2} = 2<MO> AGo = -290400 + 46 T

tzl 2(M) + {Oti'= 2 <MO> AGo = - 3587s4 + 1027
(3) 2{M}+tbzi =2 <Mo> aGo=-298400+557
M?
specific state of
How to know which on" of thepe is for which
oi formation of soia oxide from solid and liquid M
should be
Solution : The lree temprature at
"n.tgy when equated shall give the
same at the m.p. ln othei words the two AGo
which AGo shall be zero.
are three equation-pairs narnely t-2, t-3 and 2-3
to obtain the required
Therefore there
tenperatures'
1-2 - when equated, T = 12200 K
gives
1--3 - When equated, gires T = 88Bo K
2-3 - when equated, gives T = 13050 K
166
The question still remairs, ev,en if 1-3 are solid-liquidtransition, beingtheloresttransition
tenprature, which one of the 1or 3 is for solid ? lt is clear that m.p must be 8880 K, and B.P
12znoKTheambigurtycanbesolvedbyevaluating values of AGo atanytemperaturebelow
88g Le. say8000K, in between 888and12200K i.e. tosayatl20ooKandabovetheb.p.
1220i.e. say at 12800 K.
The values are as follows:
Temperature reaction 1 reaction 2 reaction 3
800 - 253600 -277154 -25440
1200 -2544ffi - 256754 - 24i1400
1280 -231520 -228194 -288000
All reactions show negative free energy values of AGo and hence can be correlated
schematically as shown in Fig. 14.8(B). lt is logicalto expect that between land 3 at 8000 K
whatewr shows lass -ve AGo value shall be liquid as shown by supercooled dotted line'
Hence Eq 1 is for solid and 3 is for liquid. Obviously 2 is for gas. The proof is thatAGo ol
reaction 2 at 12800 K must be least - ve and which is the case.
The same can be argued in a different way as follows:
The entropy change in each one of these three cases has to be relatively calculated
as follors:
(1) WhensolidM ASo = S@xide) - Slo"vgengas) - S1maary

(2) WhenliquidM ASo = S(oxlie) - S(oxvsengas) - S(rn€bD

(3) When gas M ASo = S(oxae) - SFrygengas) - S(mctal)

The entropy change therefore shall be in increasing order in the order
solidM € liquidM + gasM
The entropy values are given in the three equatircns as a muttiplier of temperature T
Hence the equation of fiee energy lor reaction 1 is for solid, 3 is for liquid and 2 is for gas.

A-{ELLINcHAM DIAGRAMs
These are the free energy vs. temperature relationships which were plotted fol
various oxidation and sulphidation processes of several metals and gases by H'J.T.Elllngant
during the second world war for the first time and these later became known after him asi
Ellingham Dlagrams. Now such diagrans are available not just for oxides and sulphides bul:
for many other cornpounds like chlorides, carbides, nitrides, carbonates, hydroxides and sci'
on and also for many conplex compounds like silicates, sulphate, phosphates, and thelike
The Ellingham diagrars for simpler compounds are by far rnore useful and hence moret
popular. ln each one of this diagram the basic reaction is formation of the relevant compouncll
from respective elements, one of which shall be the anionic part like oxygen, sulphur, chlorine,
and so on. These diagrams are konwn by these anionic part as 0xide'Elllngham-Diagrams',
Sulphide- Ellingham-Diagrams, and so on'
lnitailly only oxide diagrams were prepared but soon the others were also plotta:il
because of their importance. The diagram plotted by Ellingham for oxides was recast blr
F.D. Richardson immediately after the world war second. lt includes not only the AGo Vs.'li'
relationship but also superimposed on this diagram are the additional axs of prs, , CO/COru
ratio, HalHzO ratio scales and relevant O, C and H points as shown in Figs' 14.1 , 14'2 ancll
l4.3.lnotherwordstheFig 14.1,14.2,and 14.3 are superimposed ontheAGo vs.Tdiagramr
167
The free energy is shown on y-axis but as per Eq:(14.17)
it is atso the sarne €n" RT rn psr2 .
It also means that the present Ellingham diagram for oxides includes
the free energy
temperature relationships for various metals and provides
means to obtain equillibrium
Co/coz ratio and equilibrium HzlHzQ ratio in contact with metal-metal oxide system
at any
temprature over the range shown in the diagram.
The diagram currently in use and re'ferred universally is shown
in Fig 14.9. The
characteristics of this diagram are discussed in detail in the follorrving.
These are :
1' Several metal'metal oxide systems of interest in chemical and metallur-
gical engineering are shown in this diagram.
2' The Zero free enerry trrcint is at the top and the y-axis is shown in the
form of, -ve AGo values.
3. This is a standard diagram meaning thereby that the reactants (metars)
end the products (oxides) are all in their standard statee or pure
etates,
i.e. et unit activities.
4' Alnost aII oxides have negative free energy of formation. The exceptions
being the nobre metars like platinum, gord and to some extent
silven
5. As the electrochemical series is scann€d from noble to baser
metals
downwards, nearly similar order, though not exacdy matching
the same
is observed, as the lines ere scanned from top to bottom in this
diagram.
The nobler metals are at the top and slowly change downwards
to less and
less noble to traser ones towards the lower side.
6. since the - ve AGo moves downwards the rower is the rine the lower ie
the free enerry of that oxide. In other words the lower oxides ere
rela-
tively more stabler than the ones above it. Obviously the nobler metal
oxides, which are most unstabre, are at the top and the stsbler oxides
of
the baser metals are at the bottom. This relative stability is true et
any
given temperature.
7' The carbon'cerbon monoxide line on this diagram is exeeptional
in thet it
slopes in the direction just reverse to almost all other oxides. In other
words the -AGo value for this reaction goes on increasing with temfoera-
ture. For all other oxides, except carbon dioxide, the - aGo go on
decreasing with temperature. The carbon-carbon monoxide line there-
fore cuts acroas many of these oxide lines and thereby reverse the
relative stabilt5r relationships with the eorreslrcnding oxide after the
intersection.
8. Any point on this diagram means certain temperature nnd certain
' value of AGo or RT In po, . rf this point
is joined to either .o, or .c, or .r1
points on the graph and extrapolated, ean give the equilibrium partiatr
presaure of oxygen or equilibrium ratios of CO/COz orlf,z/IflrzO in contact
with metals and their oxides. Therefore such equilibrium parameters

168
 o z
1;;;;"
20

gie d J
to'
4 6r +O1 z,2CryO
=C

at
o
E
C+qt rCoz
a
(,
v
Lzs
GI
o
.g
, L4o
fuofU
F
t
tl L6o
oct
tfo
I

2'20 tot

24
%o

t'1.?. ELEME{r -
N.?,
g.P. Q.,DE
EtgME{T
-27'
d
o 2o,o' 4@ 5@ Coo looo leoo illo0 leoo igoo
ook T5fiP-..+ oC Poz.1
CAtCo2

llr/HrO

Fig. 149 - Ellingham diagramfor oxide systems.

169
 at any temperature for any oxide csn be found out by ssaascfing the
point to the pivota 'O', 'C'or'H' as per requirements. This has alreadybeen
discuseed in Sec. Q,4.2.1\ and Fig. 14.4.
g. The lines on the diagram for most metals are nearly parallel as discussed
in s€c. 14.4.1. The slope of an individual line may alter on pslt'iang or
boiling of the reactant or product or solid etate phase changes therein ea
discussed in Section !5.4.2. This therefore may result in interse€tion of
aome lines snd thereby alter their relstive stabilities'
10. Free energy is an exteneive property and hence quantitative comparison
of, free energies of formation of oxides csn not be mode in two different
reactions involving different massea. Eroneous eonclusions may come
out if care ie not tsken in standsrdising the two reactigns for the eake of
eompanison. Free eneqgies sgn be compared ba.sed on one g mole of the
reaetant metsl or product oxide or o)rygen gas. There is no specific bar
'on sny of these bases or nny specific preferance on scientific prineiples.
The eonmon thing in all these reactions is orygen gas. This diagram is
widely used to asseas the relative possitrilities of varions oxidation reac-
tions with respect to one-anothen whether one particular element would
be in a position to reduce the oxide of a different metal under certain
eonditioneennbedecidedbywritingthetwooxidetionregctionsand
subtrgctingonefromtheothertoobtainthenetfreeenergychangefrom
thie diagrnrn. In the process of subtraction the oxygen part shall get
eliminated, if the base is oneg mole of o:rygen' etraightaway.If this is not
the base the two equations urill heve to be balaneed to get the same
result. It is therefore easier to have the free energy of reactions plotted
per g mole of oxygen. The resctiona on the diagram are written such thst
the reactant is one g mole of oxygen snd the equivalent metal and
oxide are a{iusted aecordinglY'
aa 2Mn + $2 = 2MnO
si+ 02=siQ
4/gAt + 02 = 2/3 Alz Oa
11. on the whole a huge smount of informstion is stored in this eompact
diagrem.

Theway this intormation has been exploited or is available for such use shall be
deah with in the sections to {ollcnrv.
Example 14.2 : Steel is made by preferential oxidation of impurities in iron of the
nominal-
is supplied
composition C-4yo, Si-1.5%, Mn-0.7/o,P-0.35% and balance iron' lf pure oxygen
to oxidise these find out the theoretical sequence of elimination of these impurities.
Giren : ' (1) 2<C> + {Oz} = 2{CO}
AGo = -58800 - 31.97

170
 (2) [Stl + {Oz} = <SiOz>
AGo = - 217N0 + 48.87
(3) 2[Mn] + {Oz} = <MnO>
AGo=-190800+ 39.257
(4) vs[P] + {Q} = 26 < Pz Os>
AGo=-93100+ 64.57
Solution : The {ree energy changes at 16000 C, the usual steelmaking tenrperature, are
found to be
AG8 = -118500 caUg.mole
AG$ =-125600 -do-
AGR/n = -117300 -do-
AGF= +277A0 -do-
Assuming the product oxide activities to be unity in all the cases and that the oxygen activity
shall be the sanp in allthe cases that element shall oxidise first whose equilibrium partial
pressure shall be the minimum" lt also means that the sequence of oxidation shall be in the
order first that element whose free energy change on oxidation is the minimum. Therefore the
strict sequence shallbe first silicon, then carbon, then manganese and lastly phospho-
rus. ln fact the first three are so close that they may oxidise simuftaneously and for
phsphorus oxidation to be possible the activity of the oxide will har,e to be made rnore
negative to make it possible.
*i'}.l=RMAL
t4.s"i DOSSOCIATION OF OXIDES
!i has already been noted that in a standard Ellingham diagram the oxide lines
slopes upwards from their plvots at AGo = AHo at zero degrees absolute. lt means that as
temperature rises the AG vaiues becorne less and less negative. A situation can arise when
the line intersectthe horizontal line at AGo = 0 t.e. Poz = 1 atrnos at sorne tenrperature. When
it does so it means that the metal and metal oxide are in equilibrium at one atmcspheric
pressureof oxygenandtheAGo = 0.ln reality in asystemopentoatmosphericairwherein
poz = 0.21 the non- standard AG = 0 position rnay be reached much earlier. Any further
rise in temperature would make the AGo now + ve, reaning thereby that oxidation ol metal
is no longer feasible. On the contrary oxide shatl dissociate into metal and axygen. ln fact this
is the ideal process of extraction of metal from its oxide merely by heating, provided it is
possible to separately collect the metal so reduced. The process can be typically illustrated
by the exarnple of silver oxide.
For 4 <Ag> + {O2} = 2<AgzO> (14.25)
*730CI + 15'76 T
and aGo = e at o' = -f.?*:
Hence above 4620 K silr,er oxide shall dissociate into silver and oxygen. This process can
be used for extraction of siiver from its oxlde merely by thernal dissociation. Unfortunately
silver is not available as oxide, at least not in naturai forrn, to make it a commercial
possibility.

17',|
 Theoretically allthese lines which originate from belor point O in the oxide Ellingham
diagram shall intersect the AGo = O line at some temperature thus thermally dissociate at
above that temperature where it intersects the p9, = 1 line. ln normal atmospheric air the
situation shall change accordingly. Normally these temperatures are fairly large and ewn if
the oxide dissociates at such high temperatures it will not be possible to collect it in pure
metallic form because of its highly reactive nature at that high temperature. The end result
may therefore be that the oxide does dissociate into metal and olygen at one moment and
immediately there after, while trying to separate the metal, it gets oxidised and producas
oxide again.
Similar considerations are possible for other diagrams as well. Mercury occurs in
nature as Hgz S (mineral cinnabar) and has only small -r,e value of AGo of formation. lf this
is heated around 5000 C it does dissociate into Hg and S, both gaseous forrns, and can be
condensed to recover Hg. This is used in commercial practice to win mercury from its
sulphide ores.
14.5.2 INTERSECTING LINES ON ELLINGHAM DIAGRAM.
Although by and large the lines on the oxide diagram belcnar the melting point of
both the metal or its oxide are parallelto each other, their slope undergoes changes at
the.melting or the boiling points of the metal or its oxide. Since the melting and the boiling
points for dlfferent metals are quite difierent, the change in slope in these lines takes place
at different temperatures. The slopes also change when phase transformations in solid state
occur. The net result can be that some lines may intersect eaih other at sorne temperatures.
The intersection changes the relative
positioning of these lines and
thereby its relative stability. lt
means that AGo values relatively
change their order on changes in the
phases of the reactants and the
products. The situation is described
6 qualitatively in Fi9.14.10. Let us as-
c sume that the two lines of oxidation of
ld
2 metalA and B intersect at tempera-
ld
!J ture f K as shown. Therefore belorrrr
ld the temperature 'T A-oxide is more
G
Ir stable, being lower, than B-oxide.
I
But after the intersection i,e. above
I temperature'T the situation is justthe
reverse. Hence belour 'T reaction of
reduction of A-oxide by B is not possi-
TEMPERATURE O11+ ble whereas above temperature'T it
' is readily feasible. Such situations can
Fig. 14.10 - Intersecting lines on the Ellingham be gainfully exploited to carry out
dingram indicating the reversal of relative certain reactions which otherwise
stabilitics after the temperaturc of intersectbn.
172
ar€ not possible below a certain temperature.
This is possible because, free energy consists of two parts, one enthalpy and thn
other entropy; enthalpy being substantially constant with temperature but entropy changintn
with temperature. The situation beconres so favourable after a certain temperature that thrlr
net resutt in these two different cases is rer,erse. This will be disussed further again later.
14.5.3 Carbothermic Reduction of Oxides
The efiect of intersection of lines on the oxide Ellingham diagram has just beeur
discussed. The C-CO line is such that it cuts across lines of many oxides of interest. Th+r
slope of C-CO line is reverse to those of other oxides and therefore it should be possible tn
reduce many oxides by carbon above the temperature at which the C-GO line intersectui
their oxide lines. lt is thus possible to reduce the oxides of copper, lead, iron, nickle, coball,,
molybdenum, tin, phosphorous, zinc, chromium, manganese; vanadium, tungsten, silicon,,
magnesium, aluminium, calcium and so on at increasing temperatures in that order usinl;;
carbon as a reducing igent. Carbon being rery cheap, in the form of either charcoal or cokrilr
is used in fully exploiting this technique for commercial production of these metals from theil'
oxides.
Metals like aluminium or titanium can not be extracted by carbon reduction becausrrr
these metals react with carbon and the product obtained is in their carbide rather than thrr
metal form. ln some cases the temperature required for reduction is so high that it is not irr
cornmercial proposition against other cheaper alternatives.
These reduction processes have to be carried out at high temperatures and hencril
gffi '"",T:-1tr""::'.i:ff ffi:J,"fl1"J[;i..:rnwhichsoridsiricawourduereouJerr
by carbon to produce liquid silicon.

Given 2<C>+{Oz}=2{CO}
AGo = -53400 - 41.9 T cafg.mole
< Si> = (Si)
AGo = 10050 --do-
<Si> + {Oz} = <SiOz>
AGo = -208000 + 43.3 T ..do..

Solution : From Hess's law for the reaction

< SiO> + 2<C> = (Si) + 2{CO}
AG = + 208000 - 43.3 T+ 10050 - 53400 - 41.9 T
= + 162600 - 85.2 T
This reaction shall be in equilibrium when AGo = 0 and therefore this will happen a l:

T= 16350 Cwhich is the minimumtemperature atwhich carbonwill be ableto reduce solirll

silica.

173
14.5.4 Metallothermic Reduetion of Oxldes
The fact that oxide whose line is belour is more stabler than that whose line is
relatively abo\E can be utilised in devising processes to reduce ofdes that are above by
the metalwhose oxide is belcnltr.
Example14.4: Thusif 2<Mn> + 02 = 2<MnO> (14.261

lGfts.zol = -184000 + 34'63 T

and <Si> + Q = <SiOz> (4.2n
aG?r+.zzl = - 208000 + 43"3 T
Hence the free energy for reac{ion u
2<MnO> + <SiU> = <2Mn> +1$ioz> (14.28\
is obtained by subtracting one from the other as
AGtre.zel = -24o00 + 8.67 T
or AG?eoo"t<= -10128 caUrnole
Since this is negative this reaction is {easible or manganese oxide shall be reduced by
silicon.
Similarly commercially uranium oxide is reduced by calcium to produce uranium.
There are many such commercial processqs. The thermit process, wherein iron oxide is
reduced by aluminium to produce molten iron, which is used as a filler metalfor welding large
objects. Ferro-alloys which are nothing but iron and some other metal combination like,
feno- manganese, ferro-chrorne, ferro-silicon, ferro-aluminium, ferro-titanium, leno-tungsten'
feno-rnolybdenum, ferro- nickle and host of such alloys which are produced by using reducers
like Si, Al, Ca, etc., individually or in combination, to reduce their oxides. The reducers so
used have fairly negative AGo values of oxide formation as compared with the values of the
oxides to be reduced. ln other words elements at the lcnrer side of the diagram are used as
reducers to reduce the oxides which are relatively above.
+

Example 14.5 : Example given in Section 2.10.2 may be recalled here. lt was for calculation
of enthalpy. ln place of AH if AG values are inserted it could show the method of calculation
of metallothermic reduction of oxides.

ELLINGHAM DIAGRAMS FOR NON.STANDARD CONDITIONS

'4.6
The pattern of lines on the standard Ellingham diagram is well established since the
reactants and the products are assumed to be in their standard states i.e. al unit activities'
What will happen if the activitiqs are less than one r.e. the reactants and the products,
individually or both are under non-standard conditions?
The fiee energy change for the reaction in Fq. (1a'1{ is given as

AG = AGo + RT ln -,1?-
aM x pO,
lf all the reactants and the products aFe in their standard states then the RTln term becornes
zero and
AG = AGo
But if M is not in standard state then

174
 AG = ld + RT ]n aMl-x pOz

The values of the product depends only on temperature'because it irrii

*+*
nothing but equilibrium constant. and whose value is fixed for a fixed temperature' ln
orcE'e'n

to maintain the value of K i.e. the above quotient the sanp even if activity of M decreasesli'

the po, will harre to increase. ln other words if the activity of reac'tant decreases
then tlit'llr
equlibrium partial pressure of oxygen shall proportionately increase. By the sane logir:r
i1

po, value
the reactant ac{ivity is unity but that of product oxide decreases then the
woutl|il

decrease to maintain the value of thequotient the same for a fixed temperature. ln othe'r
words the equilibrium partial pressure of ox.ygen shall decrease if the prgduct actirrity
decreases.
The situation is shown in Fig.14.11 by plotting the standard AGo vs'T line ior
r:

on this line by point

generalised reaction as in Eq.(14.14). The tenperature T is marked
"'Arl"

This point is connected to point O, which when extended upto the p6, axis shows lllht:
corresponding Poz value as depicled by point'A. Because of decrease in the
activity cr'l'
reactant (metaq the equilibrium pe. increases to say 'B'value. The'straight line ioining
1[l

with o has a free energy value equivalent to .point B at the same temperature
T. lf lhisr
exercise is repeated for several temperatures the AG lor this non-Sandard oxidatiicrnt
reaction as it varies with temperature shall be available. The enthalpy
change even irl

!o
:
d-l PRODUCT ,tr",ys
(,d
I-
o.'

oK
t*".iXotrRE

Fig.14.tI - The rotation of lines on Ellingftam diagram

when the activities of prod'ucts andlor the reactants change'
175
non-standard conditions rernains the same since it is independant of activities. The non-
standard line is thus shownoriginatingatthesarnepoint i.e. atAGo = AHo and is shourn
€ts one passing through the point B. By the same logic when the activity of product oxide
decreases keeping the nretal activity at unity the line originating at AGo = Al-.fo and passing
through point C shall be obtained.
ln fact these findings can be qualitatively stated as : 'lf the activity of the reactant
decreases the standard free-energy-temperatureplotshallrotate anti-clockwise and
if the product activity decreases it shall rotate ln clockwise direction lrom the plvot,
AGo = AHo on the free energy axis.'
lf both,theactivitiesoftheproductas well as thereactantdecreasethentheresultant
be the final state of affairs. Coincidentally it may so happen that their
of their effect shall
variation cancel each other and the standard line may rernain the same even in the
non-standard conditions.

14.6.1 Specific Non-Standard Reactions

The rotation of the standard line on the Ellingham diagram because ol non-stand-
ard conditions of the reactant and the product may lead to drastic changes in the
relative positioning of the various lines. lf the relative positions of the lines undergo reversals,
the processes otherwise not possible, may becorne possible as was shown in Sec 14.3.21or
intersecting lines. Severalsuch processes have been developed and carried out in practice
in non-s{andard conditions, which otherwise are not possible under the standard condi-
tions. Some such specific cases are dealt with belovv.
isafactthatsolid MgO isfarmorestablethanSiOz and hence solid MgO cannot
It
be reduced by silicon at any temperature. But if the activity of SiOz is decreased by
combining it with CaO the standard line of Si-SiOz rotates downwards considerably. At
thesametimeifthe activity of magnesium isdecreasedbyusingvaccumthentheMg-MgO
line rotates upwards. The net result is that the two lines intersect each other at a much
lcruver temperature and the process:
CaO. MgO + Fe.Si = _FeO + Mg (vacuum)
Ca0-SiOz
Dolomite Feno Silicon Slag Silicab Vapour
s readily possible and is carried out under vaccum at around 12000 C commercially to
produce magnesium. This is known as Pidgeon's Process. The relavent thermodynamic
data are
2{Mg} + {Oz} = 2(MgO) (14.2e1
AGo = - 358700 + 102.67 cailg mole
< Si> + {Oz} = SiOz (1e.30)
AGo = -208300 + 43"3 T cal/gmole
Therefore the reaction
2< MgO> + < Si> = < SiOe> + 2{Mg} (1 arnos) (14.31)
is obtained from combination of ( a.291 and (1 4.30) . lf the reaction is to be carried out under
vaccum of 10€ atmosphericpressure then (14.31) will reduce to '
2<MgO> + <Si> = <SiOz> + 2{Mg}ro-3atmos (14.32l.

176
The aG?r+.ga = lG?t+.so) - aGfra.eg
and lqr+.sA = Gfr+.sg - Gfrq.ag + RT ln pMs (14'33)
putting the actual vralues it becomes
=+ 130400 - 59.3 T +
RT ln 10€
=+ 130400 - 59.3 T + 4.575 x T x (-3)

=+ 130400 - 59.3 T - 13.72 T

lf CaO is added to the reactant the resultant reaction would be
2<MgO> + 2<CaO> + <Si> = <2CaO.SiOz> + 2{Mghc'.tno" (14.34)
and z<CaO> + <SiO2> = <2CaO.SiO2> (14.35)

therefore lGfra.sq = -30200 - 1.27

and hence lGtra.s4) = AG(r+.sz) + AGlt+.ss1
= + 100200 - 71.82 T
Atequlibrium AGlr+.3+; 0 =
therefore the temperature at which equilibrium holds true is
+1100200-171.82T=0
or T=13950K=11220C
No wonder in commercial praaice the extraction of Mg is carried out around 12000 C under
a vaccum of 10€ atmos. This is shown in Fig 14-1.2
The other typical example is reduction of silica by high manganese feno-mana-
ganese. The silicon line is well belorr managanese line in a standhrd Ellingham diagram

- !50

i:...--/-"
-2o0
c
d
{Pl
to ;e.t
d
filfi"Jil?l-l#
d
lJ
e{F?}.{ott*<?
o, *.ti k-3
,6$
!-o
o,g ',;{J;' ,/+S9-
\T{l

-3oo
500 1000 ,5o0 209.0
TEMFSRATU;,g c11 ---.e

Fig. 14.12 - Free etergl consi.derations in the Pi.dgeon process for

the extruction of magnesium by reduction of magnesia by silicon.
177
and therefore reduction of solid silica by manganese under standard conditions does not
arise. However if silicon is dissolved in iron the activity of silicon is reduced drastically
and hence the silicon-silica line shall rotate upwards. Concurrently as the MnO activity is
reduced because of its conbination with silica as manganese silicate the Mn-MnO line shall
rotate downwards" The net result is that the two intersect at resonably lol temperature like
1000o K and hence it is possible for the reaction

[Mn]re-un + [SiOz]sand mould = [Si]re-un + (Mn0-Si0a)sus

to occur in real practice. For the reaction
2[Mn]+{Oz}=(MnO) (14.36)

AGlr+.sd) = -162200 + 25'25 T

and [Si] + {Oz} = lSiOz) (14.34

lG?t+.sa = -215400 + 47.73 T

hence for !{Mnl + (Sioz) = [Si] +Z(MnO) (14.38)

AGo 1u.ae; = AGo (14.36) - AGo 6a.sz;

= + 53200 - 22.8 T
The equlibrium constant of reaction (1a.38) is

(r+'ss; 3si)x a(Mnoi

= a(MnCl x a(Sioz)
- [%Sil :fsifxNrr,tnoz' yr,,rr6,
'fsioz
[%Mn]:fufr Nsioe
Assuming approximate values as
(i) %Si = 2%, fs = 0'005
(ii) oh Mn = 75"/o fun = 0.01
(iii) Nuno = 0'1 Y Mno = 0.5
(iv) Ns1s, = 0.9 YSio2 = Q.g
Putting these values in equation (14.38) gives
K1ra.sel = approximatelY 10-s

AG6e.se1 = AGftc.sol + RT ln K
AG(r+.ss) (16000r; = 16800 + 4'575 x 1600 x log 10-s
approximately = - 20OO02al
Slnce free energy change is fairly - vetne reaction of reduction of silica by Mn is possible
and which is shown in Fig 14.13.

14.6.2 Reactlons Under Vacuum

This belongs to the category of reactions under non- standard conditions but being
of its own type has been described separately. Pressure is also one of the standard state
and 1 atmos. is taken as the standard state for plotting the Ellingham diagram. lf the reactions
are carried out at less than 1 atmos l.e. under vaccum or at higher than 1 atrnos pressure
the standard [ines, since all of these reactions involve a gaseous component i.e' at least

178
 d
fg
.9
x
og-

Fig. 14.13 - The free energl-temperature relationships ind.icating

the possibility of reduction of silica by manganse.

oxygen, undergochangeswith pressure. Forageneralreaction as shown in Eq. (14.14),

if pressure is reduced, the oxide would tend to get reduced and produce metal and vice
versa. But when the metal and its oxide are not at unit activities and in that case if the orygen
partial pressure is decreased then the equilibrium shall move in such a way as to maintain
the equilibrium constant the sarne and vice versa.
Pressure is important when gaseous product is formed in these reactions and which
is possible only when carbon-monoxide or carbon-dioxide are fornred. Carbon dioxide is
not a high temperature stable phase and hence only monoxide is considered here. The
decrease in the partial pressure means decreased pco and hence for the same activity of
carbon the equilibrium oxygen partial pressure has to decrease orthe C-GO line shall rotate
in clock-wise direction. lt means that this line shall cut the oxide lines at still lorer
ternperatures and thereby make the carbothermic reduction possible at lorrrer tempera-
tures. lf the prssure is increased aborre one atmasphere then just the reterse shall happen.
It also means that at lower pressures r.e. under vacuum carbon will be readily oxidised by
oxygen and that the product of activities of oxygen andTarbon shall be decreased. This
isfullyexploitedincarryingouteitherdecarburisation ofsteelmeltundervacuumby supplying
oxygen or deoxidation of steel batrr by supplying carbon readily. This is what is known .as
vacuum treatrnent of rnohen steel. This pressure dependency of carbon oxidation reaction
can pose serious threat because of the possibility of disintegration of refiactories (oxides)
by carbothermic reduction under vacuum.

179
Example 14.5 : Find out the levels of carbon that can be achiewd when molten steel is
decarburised by supply of oxygen at 1 atmosphere pressure and under vacuum of 10€
atnnsphere.
Girrcn : The oxidation of carbon in steel rnelt takes place as
tcl +[ol ={co}
and logK= #- 2.191
Solution: Thevalueolequilibrium constant at the steelmaking temperatureof 16000C
or 1g73o K is given as
log Ktezsox = 2'69 or KtezsoK = 2.02 x 10€ atrnos.
K can also be expressed for the above reaction as
Pco
'. = atclr ..(o)
The level of oxygen in both the cases shall be proportionalto the oxygen pressure
supplied and it is the same in either case and therefore the carbon level achievable shall
be proportionalto the pressure of carbon monoxide in the surrounding atmasphere. Therefore
assuming that Henry's law is obeyed then

K(rarnos) =*gt'7q =2.o2x110

therefore %C in steel melt at under one atmos will be about 0.3-0.4%
and %C in steelmelt at 103 atmos will be about 0.1
This shows why vacuum treatrnent is given to steelto reduce carbon to a very lcnru level, which
otherwise is not achievable in open atnnsphere.
14.7 CONCLUDINGREMARKS
The free energy temperature relationships in the form of Ellingham diagrans is a
fundamental knowledge in the understanding of oxidation-reduction reactions for predicting
the equilibria and which in turn is essentialto understand the smelting and refining processes.

I
I

180
 CHAPTER t5

KINETICS OF PROCESSES
rates of reactions

15.1 INTRODUCTION
So farthe reader is quite conversant with state and path properties. Therrnodynarn
ias mainly deals with state properties and the conclusions therefrom are independent of path
of the process. Taking the example of overhead water tank, one can say that the,energy
exchange associated with the transfer of water from tank to the ground level is fixed no
matter whether water florls down, via 15 m,m dia pipe or 100 mm dia pipe or 'through a
nanow capillary like tube (disregarding the frictional component in each). Whereas in each
of these cass the tinre required forflorr of a certain annunt of water depenCs on the path
r.e the dianreter of the tube carrying the water. The rate of flcnv shall further get complicated
if the section of the tube was non-uniform and having various restrictions in the form of bends
and the like. The rate of flow of water therefore is a path function. The study of rate of any
such thermodynamic change constitutes the subject matter of the field of 'kinetics'.
The study of rates of chemical or physicochemical processes ought to be very
rigorous and in detail in order to appreciate the processes and evaluate the time taken for
a certain change to occur. Conversely during a certain time interval it would be possible
to evaluate the extent of change that might occur, from such kinetic studies. The industrial
or commercial reactors wherein the processes are carried out ln actual practrce, are often
fairly complicated and quite often these are studied in simplified models in labooratory and
the findings are compared with the cornmercial results to assess the clqsseness of suclr
simulations.
The possibilities of a process occuring tn practice is indicated by its - ve free energ,*
change or decrease in the free energy. ln reglity a thermodynamically feasrble process may
occur so slowly that apparently it might give lan rmpression of not occuring of all, The fact,
in such cases, is that the processs is taking place at so lovv a rate that within the reasonable
time no perceptible change is observed. This is best illustrated from the example of a mixture
of Hz + Oz contained in a vessel. Factually Hz and Oz are reacting but the rate of such
reaction is egligibly small.
Kinetic, studies may be performed under 'steady' or lunsteady' status oi the
system. At this stage the discussion is confined to steady state condition only. Processe*;
if occur within a single homogeneous phase, the study rs referred to as study of kinetics ol
 homogeneous reactions but if it is between two or more phases then it is a case of ktnefics
of hetrogeneous reactlon"
The honngeneous kinetics is quite uncommon whereas heterogeneous kinetics
is normally possible. The examples of systens like solid-solid, solid-gas, liquid-liquid, etc.
are all examples of heterogeneous systems. Sonretimes a new phase is forned during the
process and hence the role of nucleation and growth may dominate the kinetic studies in
such cases. The formation of CO bubbles in steelmaking, precipatation reactions, etc. are
typically the best illustrations of this category.
Since kinetics is path dependant, the process may be sub-divided into various
sub-steps to make up the oterall process. These are called elementary or kinetic steps.
This is referred to as study of mechanisin of reaction and rate of the change in each of
these sub-step may not be the sarne depending upon the nature of these sub-steps.
ln such a case the overall rate of the process will be dictated by the rate of the slorest
kinetic step in the overall path. This is known as the rate controlling step and which is
revealed through the study of rnechanism of the process. ln a given overall mechanism the
rate of the rate-controlling step is virtually the maximum rate the process shall exibit
approximately. The aim of the kinetic studies ls therefore to know the overall rate of
the process under a given set of conditions (i.e path), to evaluate its mechanism and
therefrom the rate controlling step and, to predict the condition under which the rate
controlling step would be enhanced to achive hlgher overall rate of the process.
The rate or velocity of a reaction is defined as the rate of change of concentration
(decrease generally).of reactants and (generally increase) of products of the reaction. lf 'C'
is the concentretion of the reactants then the rate is given by -dC/dt. The reaction rates
of some reactions like the thermit reaction may be extremely fast or for others may be quite
slol. lt is a common observation that the rate is a function of concentration and temprature.
15.2 KINETICS OF HOMOGENEOUS REACTIONS
The reactions which take place within a single phase are known as honrogeneous
reactions. Solid state honngeneous reactions are just not possible. Liquid-state homoge-
neous reactions may occur but rather uncommon. However gaseous homogeneotrs reac-
tions are not only possible but numerous. e.g 2Hz+02=ZHzO
at sorne temperature abore 1000 C or Hz + Cl2 = zHC/. and so on.
These are often catalytic reactions meaning thereby that catalyst help the reactions to occur
readily i.e accelerate them immensely.
15.2.1 Order of Reaction
The sum of the powers to which the concentration of the reacting atoms or molecules
that have to be raised to determine, the rate of'the reaction is known as the 'order of
reaction'. The order of a reaction is quite independent from molecularity of the reaction. lf
the process is expressed as
aA+ bB+ € product. (15.1)
then the rate of the reaction is expressed as

182
 dc Cl,
kx CB..... (1s.2)
dt
where k is the 'velocity constant or specific reactlon rate', or 'rate constant'and Cl ,

C8,..., areconcentrationsattimet. The valueof(a + b +....)lsthentheorderof

the reaction.
For a first order reaction

- * = *' Ce x ce ...... (15.3)

and since the concentrations of A or B or .... are going to vary in proportion to one another

-# = k x cn (1s.4)

= kJidt
-,f # {15.s)

OnintegrationwithinlimitsC=Co€ttt=to and Q=Q at t=t i

r, = 4Q
t
rog;L
-Uo- (
(15.6)

where r is that amount of C reacted in tinre t. Similarly lor a reaction if

-9= -ft=kxcsxpe (15.7)

dt dt
shall be a second order reaction. ln this way the higher order reaction can be evaluated.
The order can be zero also if the rate does not depend on concentration of the reactants. Also
it can be a fraction depending upon the reaction'
Far a zero order therefore dr/dt = k or k = rlt and a straight line relationship
exists betweep r and t. For a half order reaction drldt = k (Co - r)V2 and
k = Zlt pf - 121
- r)121 and hence t vs. [CY'? - Co -4 should be a straight line
(Co .

Likewise for first order reaction drldt = k(Co-$ or R = 1ltflog (ColOo-r)] and
t vs. ilog (Co /Co - r)] should be a straight line relationship. For ond order reaction
drldt = k (Co-r) 2 or ft = 1A [r/Co (Co-r)] and tvs.[r/Co (Co-r)]
has to be a straightline relationship.
The order can thus be verified from experimental data by plotting the respective
variations as a function of time. For example if log flog (Co /Co - 0l vs. time is a straightline
it is first order process and so on. n

Example15.1 : Therate constants for the lea'ching of alumina inaqueotrsNaOHsolution

at different temperatures are :
logk -12.9 -13.1 \13'3 -13'5 -13'7 -13.9
1/T x 103 2.25 2.30 '2.35 2.40 2.45 2'50
where k is the rate constant and T is in 0o K. Calculate the activation energy of the process,
Solution : According to Arrhenius equation on log scale
logk=losA-q€'?'+
where the factor 0.4342 converts the natural log to log to the base ten. The slope is thereforer
equalto
-13.1 + 12.9 x103=-4 x 103
2.fi - 2.25

183
 This is atso equatto g*,? x e from
the equarion and

hence # e = -4 x 103 or e = -1go0ca/gnnle.

The - ve sign indicates only the
- \€ nature of the slope in this exercise. The data are such
that the same value is true even if other pairs are
taken into account because the
relationahip is exact straight line.
Example 15'2 : For kinetic studies of a certain process
follorruifig data were obtained at
certain temperature.
Time min. 248121s202s35
%wt lms 1527385263779299,
(cumulative)
(the %wt loss shown above is of reactant). Find out
the order of this reaction and ado find
out the time required to achiere 50% of the reaction.
solution : The data are anarysed further by evaruating the
foilowing varues
Time min. 2 4 B 12 15 20 25
:

3s
%wt loss 15 27 g8 52 63 TZ gZ 99
(cunu.rlative)
{reroo-r) Bs zs 62 4s sz zs I 1
(unreacted)
pY - (co-+)ral o.7s 1.6 2.1s 3.08 3.s2 4.81 7.18 e
log Co/ (Co - r) 0.070 ,0.136 O.2OT 0.gl8 0.2131 0.638
LDzg 2
fr tc.-rt
F,
12'75 ts.71 23.56 24.s6 23.31 17.71
7.36 1.e8
i wr - (co - r11t21 ols o.ss o.ss 0.51 o.ss 0.4s 0.s7 0.51
theaveragevalueof rr'P! = 0.562 Thesedataaredeveropedtoprotgraphs
lo?-(co-
oftimetvs'eachone ofthevaluesgibeninhorizontat columnsasshown
Refening back the conditions for straight line relationships for
inFig. 15.1.
certain order of a reaction noir
gives that since in the graph prf _(Co_r)1pl vs.timet
s almqst a straight line relationship whereas the other relationships
are far aruay from this
nature therefore the process must be an 'half order'reaction.
Since for ,""ltion in" l-","
constant / "urh
k - | rc'* - (cq *,r;121 and since these varues vary to sonre Lxent
the
a\€rage value can be taken as 0.562.
The tinre required for loc* of the reaction to occur can be calculated by putting the values
in the above equation wherein altyalues except that of
time 't' are known as k is 0.562 and
'r' is 50"6 and Co is 100% and hence
time t for 50% reduction
= 10.42 min"
15.3 THEORIES OF REACTION RATES

The earlier collision theory of reaction rates is based on the assumption that reaction
occurs because of collision of a reactant atoms or molecules with a certain minimum

184
 T
4^
t
g
cl"
AI

of order of a reaction'
Fl1.- 15.b Auantitative method for evaluatinn

intensityofimpact,Theminimumimpactenergyisnothingbutactivationenergyofthe
reacting atoms or molecules and
it depends on the concentration (mass) and tempera-
place the
successful in making the reaction take
ture of the reactants. For collision to be Hence the rate
orientation of the colliding atoms or molecules should be appropriate'
-"-onr (15'8)
Rate = Or tN". atoms/moles/cc/sec
WhereNcisthecollisionpersecandQistheactivationenergyandopisthe
form as that of Anhenius
same
orientationprobabilityor stericfactor'Thiseqlrationhasthe
equationwhereQpxNcisconstant,indicatingtheprobabilityofareactingparticlebeing
activation energy Q'
correctly oriented and possesbing the
further modified into theory of Absolute
reaction Rates or Transition
This has been
This is based on Arrhenius concept of energy
state Theory or Activated complexTheory.
For a reaction -
barrier and principles of quhntum mechanics'
A+ 8..".-C+ D'"^'" sinto
asperthetheory,firstthereactantsformaactivatedcomplexwhichthendisintegrate other chemical
has a very short life but it exists as any
the products. The activated complex disintegration of the
reactants' The specific rate of
species and is in equilibrium with the

185
 acti\rated complex is a universal rate independant of the volume of reactants and the
complex fornred therefrom. The reaction path is often shown as in Fig.l5.2.
It shows the reactants at energy level R and products
at energy level P indicating that
as reaction occurs R -*
P energy will be liberated. But before this can happen the reactants
must acquire additionalenergy Q as activation enerqy and attain the activated leveli.e. the

-- -
T-=-
I
"l
6 Fig. 15.2 - The schematic path
a
llt
zur
of the reaction showing the I

energl barcier

PROOUCT EN€REY LEVEL

REACTION CO.ORDINATE

activated complex of energy Q + R is formed and then the reaction occurs to produce
product at P in the form
reactants + activated complex + product I
The net energy release is stillthe sarne. The reactants acquire this additional energy by
way of mutual collusion and interaction. The reactant at tines may acquire additional energy
more than even Q. But if such a particle is not correctly oriented meaning thereby that the path
is not due for conect collision the reaction may not occur. Hence for the reaction to take place
the minimum additional energy equivatent to activation energy and correct orlenta-
tion must precede the actual collision. The net useful equation finally has the form
Rate = T tG (15.e)
"-aFl'fir
This equation differs from the Arrhenius equation by an additional linear temperature
term. But over a nanow temperature range the term 'T x l(s' can be treated as another
constant and has thus the sarne form as that of{he Anhenius equation - which is therefore
considered adequate for mqst purposes. The activation energy is equalto AH the enthalpy
change in the formation of the acitvated complex.
A reaction having a large negative free energy change can still occur very slowly if
the activation energy is too large. The. reaction is still strongly temperature dependent
because of the temperature dependent exponential term.
15.3.1 Effect of Concentration on Kinetics
For a reaction A + 8...... - C + D +.."".
Froward rate is K. Cc. Co......
and backward rate is Ku . Cc . Co. , . , "
where Ku and Kl are respectively the specific reaction rate constants of backward and forard
reactions.

186
At equilibrium by definition ilself these two rates are same and hence
Kr . Cn . Ce....'. Ku . Cc'Co..'.' (15.10)
dividing one by the other =
= K {equili. constant) (15.11)

For irreversible reactions the bachrvard rate is negligibly small and hence
Rate = Kr Cn Ce. . . ... (15.12)

15.3.2 Effect of TemPerature

From his experimental data, Arrhenius empirically showed that the rate of reaction is
givenas Rate = AexpQRr (15'13)
where A & e are constants. Now the meaning of the terms have been understood as A being
the frequency factor or the probability of a reactant being correctly oriented on the reaction
path wiih energy Q. The constant Q is now \nown as activation energy, the minimum of which
must be possassed by a particle before a reaction takes place. Taking logarithm
log Rate = log A-CyRT which showsthat log Ratevs. lnverseoftemperatureshall
be a straight line relationshiP.
Reaction rates are measured at various temperatures and plotted as rates vs. inverse
of temperature. The -ve slope of this line is the activation energy of the process.
15.3.3 Diffusion Laws
The Ficks First Law states that the flux of the diffusing specie is proportional to
the concentration gradient and area across which diffusion takes place. or
flux J=-D!q
dx
(15.14)

is the amount of material that diffuses per unit time in a perpendicular direction with
,J'
where
reference plane having a unit cross sectional area, 'C' is the concentration "nF x is the
coordinateinthedirectionofdiffusionandDisthecoefficientofdiffusion'
Ficks Second Law states that
ac_na2c
At -" a'x
(15.15)

wherein dO/dt means the rate of change of concentration"

The diffusion does not occur just because of concentration gradient. ln fact
to
thermodynamically diffusion shall occur only from where the cheniical potential is high
gradient is strictly the driving force for
where this is lcnar. ln other words cherflical activity
gradient is
diffusion to occur. This is rather complicated and hence simple concentration
considered sufiicient to work out diffusion for general purposes.

15.4 KINETICS OF HETEROGENEOUS,REACTION$

The reaction which occurs between two or rnore phases are like reduction of solid iron
of solid
oxide by a reducing gas, between two immiscible phases like slag and metal, leaching
phases
ore by a liquid acid, distillition, and so on. ln this either reactants are from different
phases and as a consequence the
or the reactants and products are frorn different
reaction takes place only upon their contact i.e. at the interface of the two phases alone'
the
ln either case the reactants have to be transported to the interface and then intereact for

187
reaction to occur and the products to be transported away from the interface. The proces.s
of transportation of these takes place in various ways depending upon the nature of the phase'
ln solid state this nrass transferis known as diff*aon which is a result of rnorrement
of atoms/molecules down the concentration gradient. ln a liquid phase the mass transfer can
take place by either diffusion as well as by fluid florrrr in either lamellar or terbulant form. ln
lamellar way the fluid moves in the direction of flovv and in terbulant by forming localeddies
resulting in flow at right angles to the direction of flory'
15.4.1 lntedace of Heterogenous Systems
A solid-solid interface is merely a contact without in any way altering the structure of
the two solid phases in the contact region. The story however is quite different if a fluid
phase is involved. /
The velocity of the fluid in contact with a solid is practically zero, although the Velocity
in the bulk may be a finite value. This is because the layer in contact with the solid cannot
slip over the solid surface. lf it is so the flow would be a frictionless flcrur which is not the case'
Therefore there exist a velocity gradient from zero at the interface to whatever apparent value
is there in the bulk. The fluid flow near the solid interface is thus lamellar and in the bulk
terbulant with a buffer zone in between. Any reaction between solid and fluid phase
therefore has to take into account this rrelocity gradient in the region of velocity-boundary-
layer while considering the mass transport.
This relative stagnancy of a fluid layer at the interface, affecting mass transpott
also results in concentration boundary-layer atthe interface on the fluid side. lf both phases
are fluid then this layer exists on either side of the interface. ln case of two fluid phases in
contact with each other,the interface is not a plane; it is rather waly. There is likely to be a
emulsion of both the phases i.e. dispersed phase at theinterface.Althoughthetendency
of the interface is to remain plane (as minimum energy is associated with plane surface)
the eddies do not allow it to remain so.
The idealised structure of the fluid interface is shown in Fi9.15.3. Although the
concentration varies continuously from the interface to the bulk, the actual contour is not
readily available for mathematical treatment. Therqfore a tangent drawn at the point of interface
conccentration in the form of Ct -N is assunre'd to be the concentration gradient and the
Ca { is supposed to be the stagnant boundary layer for the purpose oi concentration
assegsment
The concentration gradient can now be mathematically expressed as
t-l - \rB
existing at the interface.

The diffusion ian be exPressed as

nux J=DF=a)
The idealised structure of is shown
the intLrface foi a two immiscible liquid phases system
in Fig.15.4. lt is pressumed that the bulk concentration anywhere in the bulk is the same
because of inherent terbulance.

188
 CONCENTRATION
A.r INTERFACE
I
ACTUAL CONCFNTFA'tloN
z FADIEN'T
o
Fig. 15.3 - Concentration gradbnt
kG in a hetero- geneous sYstem. The
F ASSUMEO CONCENTRArION
zld assumed concentration gradient
(,
z is also shown by the tengent.
o
(J

lf a reaction is to occur between these two phases then the reaction shall take place
in the folloting stePs :
1. Tbansport of reactants from the bulk scross the boundary layer to the
interface.
2. Adsorption of reactants into the interfacial layen
3. Chemical resction amongst the adsorped species in the inter{ecial layer
(bond sxsftqngie)
4. Desorption of products from the interfacial layen
5. Tlanslnrt of products across theboundary layer into thebulk.
,1
Although physical aOso*ption ls possible, chemisorption is more dominant in
the

metallurgical world chemisorption is a irreversible process and it may tend to cover

the

whole intedace and once covered fully further adsorption occurs with difiiculty. The transporl
of reactants or products can be worked dut from Fick's lzurs. The fraction
of the interface
with increase in concentration of the
covered by the chemisorbed reactants increases
surface active reactants in the bulk phase'
electric aK
Example 15.2 : Find out the rate of desulphurisatisn of molten iron by slag in an
fumace und the follorruing conditiohs
[%S] = 0.067o
bulk metal
lnterface %S = 0.01%
Difiusioncoeff.of S = 3x10aqmz/sec
Stagnant boundary layer thickness = 0'05 cm
Average bath depth in the furnace = 20 cm
dS n" x ACs/X g/cm2 /sec
Solution : - - "o
dt, =
189
 ca0I)
\
BULK PI.IASE II

ASSUMED EOUNDARY Fig. 15.4 - Idealised structure

nrrnrace\!r{II) s?AqMNT LAyFR (E
of an interface of contacting
fluid phases I and II.
ASSUMED STAANANT
oOUXDARY LAYER (I'

(I) Cgcl
SUt'( FHASE

where Ds is the diffusion coefficient of sulphur in metal bath. x is the boundary layer thickness
and ACs is the difference in concentrations of bulk and equilibrium or interfacial sulphur
concentration. Conwrting this into more convenient form in %S
gives -d%s/dt=DsxAcs/ r(#)
where p is the density of molten iron and can be taken as 7.5 glcc. Putting the other values
in the above eluatiols";;;*
= z x 1a4 o/orsec. or = 0.7 yorhr.
Example 15.3 : ln the previous example find out the time required to attain 80% of the
equilibrium.
: !9 x / x g/"-]/"""
Solution - OT
= P. ACs
converting it into more conwnient form using lo as
_ = Ds x A %s/x (o/1oo; glcnf lsec
S
Since change in 7oS in the bath changes the concentration gradient responsible for
diffusion of sulphur into A%S, therefore

dt%sl =9i#=-d[^%s]
px d
or ds = - d[A%sJ:1 Px d
100
putting this into the above gives

d,=lffiS'# or =#x drnt^%sl

Therefore for a bath depth of 20 cm, boundary layer thickness of 0.05 cm and diffusion
coefficient of sulphur of 3 x 104 and between the limits of [A %S] at t = 0 to be one to
[A %S] at time t to be 0.20 (left orrer to attain equilibrium); it gives
t = 158 mins.

190
15.5 RATE CONTROLLING STEP . -'
in series for a process to occur the
out of the above mentioned 5 consgcutive steps this
or decided upon by the slc,vrrest step in
overall rate o{ the process shall be controiled considered
4 are fast enough so that they are not
series. ln general step No. 2 and step No. are mass transport or
out of the remaining' two
as limiting tne overatirate of reaction.
chemical reaction step'
diffusion steps and the third one as
activation energies and that step with
These various sub-steps harrc their individual
the one limiting the orrcrall rate of the reaction' lf
the
the highest activation energy shall be the overall
slour, and the rest of the steps are
fast enough
chemical reaction
"t "u"n"i"
process rate will be controlled by the rate
of chemical reaction and the process is terrned
if
as under chemical control or it is
chemically controlled process' on the contrary
difusionormasstransportrateisslowandotherstepsarefastenoughthenthe
transport or is said be a diffusion or
mass
process is controlled by diffusion or mass
transport controlled Process'
Thiscanbeexplainedintermsoftheflowofliquidinupsidedownpositionofa
bottlewhereintheflowrateiscontrolledbytheconstrictioninthebottleandthatconstriction
thb flotrr'
or the bottleneck in the overall flow
of liquidfrom the bottle is controlling
reaction' both are enhanced by increasing
The rate of diffusion or rate of chemical
stirring and it has nc
rate of diffusion is alone enhanced by
temperature or vice versa. But the
influence on the rate of chemical reaction'
process anc
It is very pertiment to indentify the
slow step in the orerall kinetics of the
bywhichthis bottle-neck step would be removetl
therefromwork out the ways and means
under diffusion controll will be accelerated
bV
or diminished in its influence' A process not only hastet
diffusion' lncrease in temperature would
I

adopting stirring which would hasten control' thtr

diffusion but, if it is not controlling
the rate and if the process is under chemical
These tricks have been full/
rate as we, would;;;;;;;;"i"bta"temperaturerise.
process techniques for improvin::t
exploited in oewloping a variety of -.steelmaking
kineticsofsteelmakingreactionsandtherebytheproduclivityofsteelmakingshops.
as an indication of the type of contrtl
Acwation energy values are often taken
itis not possible tosimplil'r
possibty existing in theprocess.lf theprocessiscomplexandif
it for laboratory studies ior estimation
of activation energy values, the apparent rates vl'
inverseoftempraturcarcplottedandfromthemon,theapparentactivationenergy|: in th ri
indication of the possible control existing
estimated. Even this gives approxmate controllt
often taken as an indication of diffusion
tl:

process. Low values of activation are

proces-ses' The intermediate valutrri
processs and high values for chemically controlled
be interpreted carefully'
can however be misleading and should
15.5.1 Mechanism of.solid-Fluid Reactions
reaction is shown in Fig' 15'5' ln su';il
The mecahnism of soild-fluid interfacial
g",r assimirated in the burk fruid and :r
reactions if the product of the r"".tion'i!'irriJi
'

the reaction may be controlled by such transport

t:

transport through ut"g"nt fluid layer cQ

as such' Gasification of-carbon by
-:c;
reactants or products or by chemicalreaction
;'tCorl = 2{co}
(15'1'il
as
191
 is a typical example of this type and the reaction has peen
found to be controlled by chemical
reaction step.
The other type is where product is a solid condensed phase and in that case the
situation after a certain amount of reaction has occured is shown in Fig. 15.6.
The reaction resufts in accumulation of product layer which
separates the main
interacting phases' Further reaction can occur only by diffusion
of reactants across this
product layer, in addition to the usual diffusion of reactants
across the stagnant fluid layer.
ln this case the accumulation may result in contraction of the area of original interface
depending upon how the accumulation occurs. The shape of the
interfacial area hower,er

lrl l--t
lc
*\ REQUI REI
DtFFust0r,
--PAl.Hg
q I trr
3r COI{C ul l>
l{,
IN BUU
('l l-,
r F A IF
l((.l -.r
lo BULK
o. z t:
lo
FLUID
coNc.
E
o
f locl I

/ LJ
J o. I

I q
conc At ci Q
15
lr al
crl 'o
INTERFACE I

E
o d Y o AI 3i
Jl o-
z l'lI trl
tt
c)
(, J FI r+J
(,
<l Jl FI
zl :
-.1 ltl tl o 3l <tr
o c,l
ul EI
f,l
I
3
;tl
(4
c6
I

(n FI I (, I
-,
(, I
=l It,

Fig. 15.5 - The idealised corrcentra- Fig. 15.6 - The idealised concentration
tion gradient in a soli"d-fluid system gradi.ent in a solid-fluid system when
when the prdrct of the reacti.on does
the product of the reaction does
nat accumulate at the interface.
accumulate at the interfaee.
remains the same,as like the original. Such processes where the shape of the original
reaction interface remains the same but actual areab altered are known
as topochemical
processes. The ray of such reaction will decrease continuously with
the progress of the
reaction because of increasing accumulation of the product at the interface.
The reduction of
Fez og/F€3 o+ by H2 or Co gas is a typical case of this type.
tt can be shown that if
[1-(1-R1/31 vs' time is a straight line then it is a pro.""" .ontrolled by chemical reaction
ana if tr{ F -(1-R2/1 vs. time is a straight line then it is under diffusion control; where F is
the fractional reaction occurred at time t. This is true when the reaction
interface is well definsl
and rernains so during. the progress of the precess. Hower,er if the solid partigle is porous

192
then the kinetics will be afiected by the tortuoslty factor (interlinking of pores) and the pore
volume.
15.52 Nucleatton and Growth !

lf reactions or processes resuh in the formation of a new phase then the nuleation
and growth of that phase has to be taken into account while evaluating their kinetics' The
nucleii are too small and reiuire volume free energy and interfacial free energies to be
supplied from the pror,ss for its formation. lt is not easy to obtain so much energy and hence
it is not easy to nucleate a new phase homogeneously. But if part of this energy requirememt
The
of nucleation is met by providing readymade interface the nucleation becornes easier.
and can prevent homogeneous reaction
necessity of hornogeneous nucleation is a barrier
from taking place at all. The typical case is ol carbon-oxygen reaction in steelmaking.
lf an
interface is made available for this then the reaction rates hare been observed
to be
reasonaly large. The rate of reaction of oxidation of carbon during steelmaking as
(15.16)
lQ+[o] ={co}
is very slor when iron ore is added for its oxidation but when gaseous o{rgen is supplied
gaseous
for its oxidation it is much fasfier. This has been attributed to the fact that when
for the reaction to occur but
oxygen is supplied the gas-metal interface is readily available
this is not the case whLn iron oxide is added and hence the sluggishness ol this reaction
in the latter case.
15.5.3 Reactlon Rates and Heat Supply
A process may be endothermic such that unless adequate heat supply is available
the reaction may not take place for want ol heat. The case of carbon oxidation as in Eq,
(15.16) in open hearth process of steelmaking where oxygen is made available from
Fe2 Os , nnking the or,erall process occur as

assured then only

which is a net endotherrnic and hence simmultaneous heat supply has to be
it may be under diffusin or chemical control. But if heat flow itsell is lacking for the
available diffusion rate or chemical reaction rate, it is the heat rate that willthen be controlling
the orcrall rate of this reaction. The rate of heat transfer depends on
ternperature difference
geonretry and thermal
between the source of heat and the reaction interface, physical
of carbon can be
eonstants of such a reaction. But the rate of such a reaction like oxidation
reaction into a net exothermic reaction if
enhanced by converting this net endothermic
.13 when the net reaction that takes
oxygen gas supplied in place of iron oxide for oxidation
pucl is as giwn in Eq.(15.16) which is net exothermic in nature.
15.6 CONCLUDING REMARKS
The study of rates of procasses is a must for understanding the time required for
carrying out such processesin actual practice. This has a direct bearing on
the process
econorny.The understanding of the rate controlling steps can and has actually<resuhed

ir inproving the process economy by reducing the time required for carrying ot'tt these
pr"*="* in practice. lt also indicates the practical limitations of improving the proc€ss
ffiromy by improving the productivi$.

193
 Bibliography

Books recommended for additional readinE.

1. Physical Chemistry of Metals, by L.s. Darken and R.w.curry,

ll cG ran E in 1953 .
2. CfiemicalandMetallurgicalThermodynamics by u. L. Kapoor, VoI
f and II, NaciandandBnos, Roorkee j,979.
3 lntoduction to Metallurgical rhermodynamics by o. R. Garkerrr
ll ccril B ilf Tokyo, 1973.
4 ThermochemistryforSteelmaking by g. F. EIIiot, Irt. Gleiser,
and. Ramkrishna Vol II, Addjrruweslr, London, 1963.
5 Problems in MetallurgicalThermodynarnics and Kinetics by,
c. S. Upadhyaya and R. K. Dube, p€rgm@Or6d, 1977.
6 OassicalThermodynamics by A. B. pippard - 1963.
? Physbal Chemistry of lron and SteelMaking by R. G. word,
t&ronilAnoE, L962.
8 PlrysicalChemistryoflronandSteelManufacture, by G. Bodsworth,
LagEa'r 1953.
9 ll€{dlurgicalThermochemistry by o. Kubaschewski, E. L. Evans
and C. B. Alcock p€(ges - 19G9.
Ls Ptryshal Chemistry of Melts in Metallurgy by F. D. Richardson,
VoI I & fI Acadqbprres London, New york, 1974.
 SYMBOLS
Activation energy o
Activity - Raoultian a
Henrian h
Activity coeffi cient R aoultian v
F{enrian f
Chemical potential F
Concentration c
Diffusion coefficient D
Entropy s
Electrochemicai valency n
Flectro- rnotivq-force E
Equilibrium constant .K
,

Entropy s
Excess proper$ XS

Farraday's constant r
Finite change in extensive property A
Fugacity t
Gas constant R
General extenslve property z
Gibb's free energy G
Gram g
l'{eat capacity c
Heat exchanged q
${elinholtz {ree energy A
lniernalenergy U
I nteraction pararneter eii
Meter m
Mixing function zM
Partial property 2
Specific heat constant volume Cv
cqlstant pressure Cp
Standard state with supe5;cript o
Specific rate constant k
oK
Temperature degrees'absolute
O6
degrees Celcius
Volume V
Work done mechanical w
any other w'
 INDICATORS

Compositlon in.weight percentages wt%

,Gas phase { }
Liquid oxide phase ( )
Molten metallic phase t I
Solid phase

ABBREVIATiONS

atrnm atmospheric pressu.re

cal calories
Kcal kilo-calories
cc cubic centimeters
hr time in hours
ln naturallogarithm
log , logarithm to the base ten
min time in minutes
sec time in seconds

CONVERSION FACTORS

Farraday's constant, F 96,500 coulombs/g equiralent

23061 calories/volVg equivalent
Gas constant R ":iT'tJ,,1li.es/more
l:3ii
82.07 cc atmos/de g/nnle
0.08207 litre atmos/deg/mol
ln 2.303logro
one atmosphere pressure
;::[T "n ,., ,n,.,., = 1 torr]
' 1p33 g/cm2
I

ill
 INDEX

Enerry 4
Absolub n"""tjorrni" Theory- 185 altemate sources - 7
Activated complex - 185
avail"bility - 5,59

Activatbn enerry - 186

forrns - 6
Acti\dty - 69,111,147 prcpertbs - E
coefficient - 108,129 sou:pes - 7
determinatbn - 127,147 zero bvel 70
emf - 147 Ebctochembalprccess - t4S
Adiabatic flame temperable - 36 Ebchbalwork- LA.$
Adiabatb Process - LL,24
Ebchomotive force - 145
Abha funcfron- 727,L}L
-. coeffrcbnt- L49
equation - 186
,,-dhenius
\t/ -?ningham diagem - 166
B Electode pobntial - L44
Boundary hyer- 189 Enthalpy - 5,20
C phase charge - g2

Cabric equatbn -
reaction - g6
18
Carbofirermic rcductbn -
temperatue - 29
L73
Equation of State - g,48
.,,4amotcycle - 40
F4r'ilibrium - L2,59
Chemicalconbol - 190
constant- 8g
f .nhembalpotential- 63
metastable ZB
of cell L49
neukal- 7A
a€lapeyron equatbn - ot pafial - 1g
Qldusius - Clapeyron equatbn - 93 stabb - 7A
,_
Concenfuatbn cell - LM thermodJrnarnb - Lz
emf- 145 unstable - 76
Criticaltemperature - 79 Enthalpy 5,42
chembalchange - 51-
D
naturalplocess - 43
Debye's temperahle - 28 ' I phase hansformatbn -
I

Diffilsbnconbol-
I
190
prlcess -
Dubng and Petitss law - '27 51
randomness - 45
E temperature 48
Eirsbin'a theory of zero level - 7L
specific heats - 27 Excess,fi.ulctbn -
'\
l4z

iv
 Brbrrsive ppPedY - 9,59 Ivfiscibility gap - . 139

Nfixing furctiorrs - 11ll

F
lvlneupnic squaie - €lii
Fanday- L48
Ivlobl quan&ies - '€i3l
Ftcls law- 187
Ivbbrquantities - 63,102,12rEi
Ftstbw- 15
Mufti-compo rre nt sohrtb n - 1ll
Flee erergr - 53
57 N
Gbbs -
Ibbnhoh - 56 Nlafual process - :litri

ofnixing - LL7
*.D6-ms't Ib at Theorem' 6t I t:

F\rgacity - 66 1(,|i;t
Nucbation -
G o
Gibbs fiee energr' 57
orrilerofreactbn- 1l:l;1

- Duhemequatbn' 101, 125 1tl{ri

Oxygen potential -
-.e{*
Gibbs ' Ilehnholtz equation' 66,87
P
H
Partbl mohrProPertY' 63,101;ll

lbat caPacitst - 19
Partblprcpe*Y' ll).1t.

Ibruy's bw- 108

Path function' rL ril

lbss's bw- 26
lPtrase mb - i:ll:ll

I stabrlitY - '11,fi$

Heal sohrtion - 104 harsformatiors - ,lNlil

Inbnsive ProPerty' 9 Polbopb process - 11,:2iiri

Inbrnal energr - 5,L6,21 Process

bobaric PrPcess' 10 fureversible' ,t,,$

bochorb Pmcess - LI,24'- path - 1{)

bothermal Process - 10,23 rcvesible' 1l:

K 0
I&netbs - 151 Quasi-chembalmodel- l"Cl;

I(uchoffs hw- 33R

I6PP rule - 28 Ramsay-YoungRub -
(;#ii

:- L Law' :[ePl
"fraouHs
Iaws of &rermd5rmtttics - 6 R€gubrsolutbn - liil0
(:}'ill

M Richard's Rule -

Maxwelfs rcbtiors - 65s

' lvletalb{rermb - L74 Secondbw- iiil[i;

V
!

i
I
 Sieverf,s hw- 113
Solil elecholte - 150
Solutbnmodeh - 106
Speeific heat-
Corstantvolune - 19
Corrsbntpressure - 20
Specific rcaetbn mte - 183
Stardad state - 109
Siate furctbn - 1"1

Sbady state - 181

Sublimation - 94
System I
state - 9

T
Thermit process'- 31
Thermal dissocbtbn - L7T
Thermochembky - 25
Tlremrodynamic pote ntial - 53
Third law - 73
Thrctling - 25
Tbpochemicalprocess - L92
Tlarrsport eonbnl - 191
Thoutan's Rub - 96

vanderWaal's equatbn- 25
-.}
-,rr{arft }bffequatbn' 87,94

,w
Workfirrction- 57

Zeroth bw- 76,77

vt
I
I

Llt
ltl
ilil
lirll
It
It
Il
I

lt
LIL