SOLUTION THERMODYNAMICS

Partial molar properties, Gibbs-Duhem equation, fugacities in gas and liquid mixtures,
activity coefficients, Ideal and Non-ideal solutions, azeotropes, Excess Functions;
Calculation of vapor liquid equilibria using equations of state; Classical and excess free
energy based mixing rules, Wilson, NRTL, and UNIQUAC equations, UNIFAC method.
Relations for Partial Molar Properties
• Gibbs-Duhem Equation

• Substituting for dG
• Simplifying, we obtain the Gibbs-Duhem equation,
• Therefore, we conclude at constant T and P:
Gibbs-Duhem Relation for Activity
Coefficients
• To extend the Gibbs-Duhem equation to excess properties, the excess
Gibbs energy can be manipulated

• resulting in
• Inserting the relation between excess chemical potential and activity
coefficients gives
Relations between Various Excess
Properties
• For the development of accurate process calculations, a thermodynamic
model should accurately represent the temperature and pressure
dependence of deviations from ideal solution behavior. The excess
functions follow the same relations as the total functions, HE = UE + PVE,
GE = HE – TSE, and AE = UE –TSE.
• The derivative relations are also followed,

• The Gibbs-Helmholtz relation applies:

Modified Raoult's law

Ideal gas behavior is the simplest type of mixture behavior

because the particles are non-interacting.
deviations from the ideal gas model,
For the liquid phase, we use an activity coefficient

Typically the Poynting method is used to calculate the pure

component liquid phase fugacities,

Combining these expressions,

At the low pressures

 Modified Raoult’s Law and Excess Gibbs Energy

• Note that the bubble line for Raoult’s law

misses the shape completely. The bubble
pressure formula for Raoult’s law is linear in
x1 because x2 = 1 – x1 and the Psat values
are constants with respect to x1.
• Suppose that we were to apply a “correction”
to VLE K-ratio for each component and
attribute the deviations to the liquid phase,
where the molecules are closely packed and
molecular geometry and intermolecular
potentials are much more important than in
the vapor phase. The resultant method
provides

When a system has positive deviations from Raoult’s law, the bubble
line lies above the Raoult’s law bubble line , therefore γ i > 1.
When a system has negative deviations, γi < 1. This correction factor γi, is
referred to as the activity coefficient.
 • From the values of the activity coefficients, a value for excess Gibbs energy can be calculated:

• The One-Parameter Margules Equation

• The simplest expression for the Gibbs excess function is the one-parameter Margules equation
(also known as the two-suffix Margules equation). For a binary system,

Applying for i = 1, and using the product rule on n1/n = n1(1/n),

 Redlich-Kister and the Two-Parameter Margules
Models

• a shortcoming in the one-parameter model’s representation of the

skewness of experimental excess Gibbs energy. In principle, adjusting
both the magnitude and skewness of GE is possible with a two-
parameter model equation.
• Many models can be rewritten in the Redlich-Kister form

Two-Parameter Margules Model

The two-parameter Margules model is a simplification of the Redlich-Kister,

where we relate the constants to the Redlich-Kister via A21 = B12 + C12, A21 = B12 – C12, and D12 = 0. The constants A21 and A12
are fitted to experiment as we show below. Note that if A21 = A12, the expression reduces to the one-parameter model. The
expression for the activity coefficient of the first component can be derived as
Azeotropes - Activity Coefficients at Special Compositions
• The location of an azeotrope is very important for distillation design because it represents a point at which further
purification in a single distillation column is impossible
• Relative volatility crosses 1 at an azeotrope. Any deviation from ideality will create an azeotrope at a Bancroft point.
• A plot of logPsat versus 1/T with both components may show a point where the two curves cross when the heats of
vaporization are different. This point is called a Bancroft point. Since the vapor pressures are exactly equal at the
Bancroft point, any small non-ideality generates an azeotrope. This might be avoidable if the system pressure is
raised or lowered to circumvent the Bancroft point in the temperature range of a distillation column.

Note that the bubble line for Raoult’s law misses the shape completely. The
bubble pressure formula for Raoult’s law is linear in x1 because x2 = 1 – x1
and the Psat values are constants with respect to x1.
Suppose that we were to apply a “correction” to VLE K-ratio for each
component and attribute the deviations to the liquid phase, where the
molecules are closely packed and molecular geometry and intermolecular
potentials are much more important than in the
vapor phase.
The resultant method provides
Purity and Infinite Dilution
• A component is said to be infinitely dilute when only a trace is
present. Thus, when a binary mixture is nearly pure in component 1,
it is infinitely dilute incomponent 2.
• The activity coefficients take on special values at purity and infinite
dilution.

• consider the infinite dilution composition limits of the following

equation
• Then,
Preliminary Indications of VLLE
• azeotropes occur at x=y, where a maximum or minimum appears in all the
plots. Also that the bubble and dew lines do not cross, but they touch at the
azeotrope composition.

Phase diagram calculations for the 2-propanol + water system at 30°C compared
with data. The parameters where selected as described in the text to illustrate how a
numerical error can result in thermodynamically unstable loops. Note the dew line
has a has a loop and the maximum in the bubble line is not at the azeotropic
Relations between Various Excess Properties

• Thermodynamic model should accurately represent the temperature and

pressure dependence of deviations from ideal solution behavior.
• HE = UE + PVE, GE = HE – TSE, and AE = UE – TSE
Liquid Phase Properties from VLE Data

• Calculate activity coefficient at infinite dilution or

Margule’s constants or limiting activity coefficient
13
Infinite dilution
• It is the particular behavior of a two substances mixture when the
concentration of one component approaches the limit of zero and the
other approaches 1(second component being the solvent).
• Behavior when the mixture is sufficiently dilute that solute-solute
interactions are negligeble.

14
Importance and use of infinite dilution
activity coefficient data
• Most difficult and costly stage of a separation process is
the removal of the last traces of impurity.
• Greatest departure from ideality occurs in the very dilute
regions
• Synthesis, design and optimization of separation
processes
• Predicting the existence of an azeotropes
VLE data for methyl ethyl ketone(1)/toluene(2) at
500C

16
17
Method to predict i

18
19
Example

20
Positive deviation from Raoult’s Law

Plot of experimental P-x-y, activity coefficient and molar excess Gibbs free energy data

21
Negative deviation from Raoult’s Law

Plot of experimental P-x-y, activity coefficient and molar excess Gibbs free energy data

22
Azeotropes
• Importance of boiling point diagram(T-x-y diagram) &
P-x-y diagram.
• An Azeotrope is a mixture of two or more liquids in
such a way that its components cannot be altered by
simple distillation.
• This happens because, when an azeotrope is boiled,
the vapor has the same proportions of constituents as
the unboiled mixture.
• Because their composition is unchanged by
distillation, azeotropes are also called constant
boiling mixtures.
23
Azeotropes
• In some real systems, the temperature / composition curve is far from
ideal.
• A maximum or minimum in the curve is possible; this is an azeotrope.
• At the azeotrope, the liquid and vapor have the same composition.
• Word azeotrope is derived from the Greek words, overall meaning,
"no change on boiling".

24
Ethanol-water system
•A example of a positive azeotrope is
89.44% ethanol and 11.56% water (by weight).
• Ethanol boils at 78.4 °C, water boils at 100 °C, but the
azeotrope boils at 78.2 °C, which is lower than either
of its constituents.
• 78.2 °C is the minimum temperature at which any
ethanol/water solution can boil at atmospheric
pressure.

25
 Temperature-Composition Diagram
for Ethanol-Water, P = 1 atm
100

95
Superheated Vapor Phase
Two Phase
90
T( C)
o

85

80

Subcooled Liquid Phase

75
z EtOH
0.0 0.2 0.4 0.6 0.8 1.0
xEtOH or yEtOH

26
• This limit is 0.8943 mole fraction of ethanol for an ethanol-
water mixture at P = 1 atm.
• To separate a mixture containing 0.9 ethanol liquid mole
fraction at 1 atm
• Azeotropes are dependent upon pressures and one way to
avoid this problem is to vary the pressure.
• For example, under a vacuum (P = 70 mmHg), no azeotrope
exists for the ethanol-water mixture. This is one reason
that vacuum distillation is often done for many
separations.

• Azetropes are often also component dependent.

• For example, the addition of a 3rd component to a binary
system, may change or “break” the azeotrope.
Example
• Construct the P-x-y diagram for the cyclohexane(1)-benzene(2)
system at 313 K given that at 313 K the vapour pressures are
and

• The liquid phase activity coefficients are given by

28
Answer
X1 0 0.2 0.4 0.6 0.8 1.0

1.5809 1.3406 1.1793 1.0760 1.0185 1.0

1.0 1.0185 1.0760 1.1793 1.3406 1.5809

P 24.41 26.49 27.37 27.41 26.61 24.62

Y1 0 0.2492 0.4243 0.5799 0.7540 1.0

29
Example
•A mixture of benzene(1) and ethanol(2) for which the
vapor pressures and activity coefficients given by the
following equations.
ln γ1 = 1.25 x22 ln γ2 = 1.25 x12
• Assuming the validity of Modified Raoult’s Law, show
whether or not the system exhibits an azeotrope at
temperature 70 oC.
• What is the azeotropic pressure at 70oC?
sat 2726.81 3795.17
ln P1 ( kPa)  13.7819  sat
ln P2 (kPa)  16.8958 
T ( C )  217.572
o
T ( o C )  230.918

30
VAPOR-LIQUID EQUILIBRIUM AT
LOW TO MODERATE PRESSURE

31
Activity Models
• Wohl’s equations & Redlich Kister Expression
• Margules model
• Two suffix equation or one constant equation
• Porter equations
• Three suffix equation or two constant equation
• Vanlaar equation
• Wilson equation
• NRTL model
• UNIQUAC (Universal Quasi Chemical) model
• Abrams and Prausnitz
• UNIFAC (UNIQuac Functional Activity Coefficient model) model

32
Activity Models
• “Simple” empirical models
• Symmetric, Margule’s, vanLaar
• No fundamental basis but easy to use
• Parameters apply to a given temperature, and the models usually cannot be
extended beyond binary systems.
• Local composition models
• Wilsons, NRTL, Uniquac
• Some fundamental basis
• Parameters are temperature dependent, and multi-component behaviour can
be predicted from binary data.

33
 VAPOR-LIQUID EQUILIBRIUM AT LOW
PRESSURE
• Various activity coefficient models commonly used for
VLE calculations.
• Divided into two major groups depending upon their
applicability to various types of solutions:
1. Margules/Van Laar models
• Homogeneous Mixture models
2. Wilson/NRTL (Non-Random Two Liquid)/UNIQUAC (Universal
Quasi Chemical) Models
• Local Composition models
• Non-ideal behavior of liquid solutions derives from
two sources:
• difference in molecular size/shape of constituent
• difference between the molecular interaction energies.
• Type 1 models generally apply to systems in which the
inter-species and molecular interaction energies differ
from each other.
• Type 2 models are principally useful for describing
systems where the constituent molecular species
differ on account of both size/shape as well molecular
interaction energies.

35
Models for Excess Gibbs free energy

36
 Activity models
• GE/RT is a function of T, P, and composition
• For liquids at low to moderate pressure, it is very week function of P.
• P dependence of the activity coefficients is usually neglected
• Thus for data at constant T, GE/RT is a function of composition (x1, x2,..,xN)

GE
 g ( x1 , x 2 ,....., x N )
RT
• For binary mixture, the function that are most often represented is G E/x1x2RT
and it can be expressed as a power series in x 1.

• An equivalent power series with certain advantages known as

Redlick/Kister expression

GE
 A  B( x1  x2 )  C ( x1  x2 ) 2  ....
x1x2 RT
37
Derivation of Margules Equations

38
Margules equation
Max Margules introduced in 1895 a simple
thermodynamic model for the excess Gibbs free
energy of a liquid mixture.

He had introduced the concept of the activity

coefficient, the model could be used to derive an
expression for the activity coefficients of a
compound i in a liquid.

39
Case I: Ideal Solution

• A=B=C=0

Case II: One-Parameter Margules Euation

A ≠ 0, B =C = 0

For binary mixture:

 Margules one parameter Equations
• This model provide a good representation for many
simple liquid mixtures, for mixtures of molecules
which are similar in size, shape and chemical nature.
• When lny1 and lny2 are plotted against x2(or x1), the
two curves are mirror images.
• Symmetrical nature of activity coefficients for this
model

41
Case III- Two-Parameters Margules Equation
Margules Two parameter Equations

A12 → single molecule of 1 surrounded by 2

A21 → single molecule of 2 surrounded by 1

43
Margules Two parameter Equations

44
Margules Two parameter Equations

45
Margules Two parameter Equations

A12 → single molecule of 1 surrounded by 2

A21 → single molecule of 2 surrounded by 1

 1  ln  2 2 ln  1
A21   2   
 x1  x1 x2
 1  ln  1 2 ln  2
A12   2   
 x2  x2 x1 46
Vanlaar Equation

47
Van Laar Equations

48
Vanlaar Equation
• Van Laar equation is an activity model, which was
developed by Johannes van Laar in 1910-1913, to
describe phase equilibria of liquid mixtures.

• Where A and B are known as vanlaar constants, if the

constants are known the activity coefficients can be
estimated.

49
Estimation of Vanlaar constants
• First method:
• If the activity coefficients are known at any one composition, then vanlaar
constants A and B can be estimated by rearranging the equation.

50
Estimation of Vanlaar constants
• Second method:
• For a systems forming azeotrope, if the temperature and pressure are known
at azeotropic composition then activity coefficients can be calculated as,
• At azeotrpic composition x1=y1

51
52
 Example - 1
• A mixture of two compounds, 2-propanol(1) and water(2),calculate
excess Gibbs energy at x1=0.6369 and fit one parameter Margules
equation. Data provided
• T=300C & P=66.9 mmHg
• P sat 1= 60.7 mmHg
• P sat 2 = 32.1 mmHg
• y1 = 0.6462
• X1 = 0.6369
• Also fit the two-parameter Margules equation(Determine the
numerical values of A12 G
and
E A21 which describe this mixture)
 A12 x1  A21 x 2
RTx1x 2

53
Answers
• Activity coefficient of component 1=1.118
• Activity coefficient of component 2=2.031
• Excess Gibbs energy=0.328
• A =1.42
• A12 = 1.99
• A21 = 1.09

54
 Example - 2
• A mixture of two compounds, A and B, has an azeotrope at T
= 301.4 K, P = 136.94 kPa and xA = yA = 0.392. The vapor
pressures at this temperature are = 101.57 kPa and =
86.42 kPa. To describe the properties of this mixture, use
the Margules equation in the form:
GE
 A12 x1  A21 x 2
RTx1x 2
• Determine the numerical values of A12 and A21 which
describe this mixture.

55
 Example - 3
• An azeotrope consists of 42 mol per cent acetone(1) and
58 mol per cent methanol(2) at 760 mm Hg and 313 K. At
313 K, vapor pressure of acetone and methanol are 786
mm Hg and 551 mmHg respectively. Calculate the van
Laar constants.

56
Estimation of Vanlaar constants

57
Answer
• A = -4.838
• B = 0.274

58
Example - 4
• Liquids A and B form an azeotrope containing 46.1 mole per cent A
at101.3 kPa and 345 k.At 345 K, the vapor pressure of A is 84.8 kPa
and that of B is 78.2 kPa. Calculate the van Laar constants.

59
Answer
• A=1.2955
• B=0.6530

60
 Example - 5
• The azetrope of the benzene(1) & cyclohexane(2) system
has a composition of 53.2 mol % benzene with a boiling
point of 350.6 K at 101.3 kPa. At this temperature the
vapor pressure of benzene(1) & cyclohexane(2) is 100.59
kPa and 99.27 kPa respectively. Determine the activity
coefficient for the solution containing 10 mol % benzene.

61
Answer
• Activity coefficient of benzene = 1.062
• Activity coefficient of cyclohexane = 1.002

62
 Local composition models
• Concept of LCMs is based on the hypothesis of that the local
concentration around a molecule is different from the bulk
concentration.
• Local composition is incorporated to model to account for the
nonrandom molecular orientations resulting from differences
in molecular size and intermolecular forces.
• This difference is due to a difference between the
interaction energy of the central molecule with the
molecules of its own kind and that with the molecules of
the other kind .
• Energy difference also introduces a non-randomness at the
local molecular level.

63
Local composition model

64
Wilson equation
• Concept of local composition was introduced by
G.M.Wilson in 1964 with the development of model in
which he was proposed that the local composition is
different from the overall mixture composition or bulk
composition.
• Local composition model is postulated to account for
the non-random molecular orientations that result
from differences in molecular size and intermolecular
forces.

65
Wilson equation for binary solution
• Wilson proposed the following expression for the
excess Gibbs energy of a binary solution

• Wilson equation have two adjustable parameters

and known as Wilson parameter. These are related
to pure component molar volume and characteristic
energy difference and are given by,

66
Wilson equation for binary solution

67
Wilson equation for binary solution

• V1 and V2 – molar volumes of pure liquids at

temperature.
• - Energies of interaction between molecule
• a - constant for a given pair of components
independent of temp & prssure.

68
Wilson equation
• Strength of Wilson’s approach resides in its ability to describe multi-
component (3+) mixtures using binary data.
• Experimental data of the mixture of interest (ie. acetone, ethanol, benzene) is
not required
• We only need data (or parameters) for acetone-ethanol, acetone-benzene
and ethanol-benzene mixtures

69
Wilson’s Equations for 3-Component Mixtures

70
Wilson equation-advantage
• Provides a good representation of VLE of a variety of miscible
mixtures.
• Considered the temperature dependent of the parameter
• Applicable to the completely miscible system
• Suitable for solutions of polar or associating components like alcohols
in non-polar solvents for which the Margules and van Laar equations
are generally inadequate.

71
Wilson equation-Disadvantage
• It is not suitable for systems showing maxima or minima on the lnY
versus x curves.
• It is not suitable for systems exhibiting limited miscibility
• recommended only for liquid systems that are completely miscible, or for
partially miscible systems in the region where only one phase exists.

72
Non-Random Two-Liquid(NRTL) equation
• NRTL equation was developed by Renon & Prausnitz
in 1968 & this equation is applicable to miscible or
partially miscible systems.
• NRTL equation for the excess Gibbs free energy is
given by,

73
Non-Random Two-Liquid(NRTL) equation

• Activity coefficients are

74
 Non-Random Two-Liquid(NRTL)
equation

• Here are independent of composition & temperature & it is

related to the non-randomness of the solution.
• If = 0, the solution is completely random & NRTL equation reduces
to the one parameter Margules model.
• The value of varies from 0.20 to 0.57, in absence of the
experimental data = 0.3.

75
NRTL equations
• Applicable to partially miscible as well as totally
miscible systems.
• For moderately non-ideal systems, it offers no
advantage over the van laar and margules equations.
But for strongly non-ideal solutions and especially
partially miscible systems, the NRTL equations provide
a good representation
• Disadvantage:
• For moderately non-ideal systems, it offers no advantage
over the Van Laar & Margules equation

76
Characteristics of Models
• NRTL is especially suited for acid gas adsorption, which includes the removal
of carbon dioxide and hydrogen sulfide from a gas stream. Refineries routinely
use this process when making hydrogen.
• UNIFAC model is a group contribution method that allows the model
parameters to be estimated using the molecular structure of each chemical.
When experimental data is not available, this is the only method that can be
used.

77
Characteristics of Models
• UNIQUAC model uses binary parameters, which must be determined
from experimental data. Once found, however, the same parameters
can be used in multicomponent mixtures of three or more chemicals.
• Both UNIFAC and UNIQUAC can be used when two liquid phases or
azeotropes are present.
• Wilson equation is an option if there is only one liquid phase,and it
does handle azeotroipes.

78
Applicability of Activity Coefficient Models

79
Models for the Excess Gibbs Energy and Activity Coefficients for Binary Systems

80
Models for the Excess Gibbs Energy and Activity Coefficients for Binary Systems

81
Relation between the model parameters and infinite dilute activity
coefficients

82
83