A THREE PARAMETER CUBIC

EQUATION OF STATE FOR

PREDICTION OF
THERMODYNAMIC PROPERTIES
OF FLUIDS
A.H. Farrokh-Niae, H. Moddarress, M. Mohsen-Nia

Presented by Group 10- Shrestho, Divyasupreta, Ishan, Mahati and

Puneet
Introduction to EOS
What is an EOS? Why is it required?
An EOS or Equation of State is a mathematical model that describes
the relationship between P, V, T or other thermodynamic properties
of a substance in a particular phase like liquid or solid.

They allow us to predict how a substance behaves under various

conditions, making them a cornerstone of thermodynamics.

EOSs can range from simple models, such as the Ideal Gas Law, to
more complex forms, like the cubic equations of state (e.g., Van der
Waals, Soave-Redlich-Kwong). Complex models can capture the non-
ideal behaviour of fluids by introducing parameters to account for
intermolecular forces and molecular sizes.
Types of EOS
EOSs can be divided into three main types:
Cubic EOS Cubic equations of state, like the Van der Waals, Soave-Redlich-
Kwong (SRK), and Peng-Robinson (PR) models, use simple
mathematical forms with two or three parameters to predict
fluid behavior, balancing ease of use with reasonable accuracy
across various conditions.
Virial- The Virial EOS expands pressure as an infinite series in terms
expansion of molar volume and includes virial coefficients to capture
EOS intermolecular interactions, making it accurate for low to
moderate pressures but complex at high pressures.
Complex These models involve numerous parameters to model complex
EOS molecular interactions, achieving high accuracy but often at the
cost of increased computational complexity.
Why another EOS
Limitations of traditional cubic EOSs
Challenges in prediction Especially near the critical region (0.90 < Tr < 1.1)
Resulting in inaccurate predictions of liquid density and
phase behavior due to heightened sensitivity in molecular
interactions.

Issues with Two- Rely on a fixed critical compressibility factor, which limits
Parameter Models their flexibility and results in poor predictions around the
critical point where fluid behavior becomes highly non-ideal.

Need for improvement Adding a third parameter allows for greater adaptability in the
EOS, enabling more accurate modeling of complex phase
behavior and density variations near the critical region by
better accounting for temperature-dependent molecular
interactions.
Why another EOS
Purpose and goals of the new EOS
Enhanced accuracy The new EOS aims to improve predictions of liquid and
vapor densities, particularly near the critical region, where
most pre-existing models at the time fell short.
Improved VLE By accurately modeling vapor-liquid equilibrium, the new
Calculations: EOS provides more reliable phase distribution data

Greater Model Incorporating a third parameter increases the EOS’s

Flexibility adaptability, making it suitable for a wider variety of
substances and thermodynamic conditions.
Comprehensive The main results focus on accurately calculating saturation
Property Calculations properties, volumetric properties in the supercritical region,
sub-cooled liquid density in the sub-critical region, as well as
molar enthalpy and heat capacity. These improvements
enhance the model’s applicability across diverse
thermodynamic conditions.
Cubic Equation of State
GENERAL STRUCTURE :
Consists of two terms: attraction (Patt) and repulsion (Prep)

REPULSION TERM (Prep):

1. Van der Waals:

2. Carnahan and Starling:

3. Mohsen-Nia et al.:

4. Eu and Ohr: A complex but accurate formulation

Cubic Equation of State
ATTRACTIVE TERM: (Patt)
The generic form of Patt was represented by Schmidt and
Wenzel, where u and w are either expressed explicitly or
implicitly

1. SRK: u=1, w=0 5. Patel-Teja: u=b+c,w=-cb

2. P-R: u=2, w=-1 6. Adachi et al.: u=b+c,w=-cb

3. Fuller: u=c, w=0 7. Yu-Lu: u=c/b,w=0

4. Harmens and Knapp: u=c, w=-(c-1)

Cubic Equation of State
ATTRACTIVE TERM: (Patt)
u and w can be expressed implicitly by applying volume
translation techniques
1. Yu-Lu: u+w=1 2. Schmidt and Wenzel: u=w+1

In order to improve the performance of the EOS, empirical relations

are suggested for w versus acentric factor.

The acentric factor provides a measure of how non-ideal a substance

behaves. By linking w to acentric factor, the EOS can adjust for different
molecular complexities, improving the accuracy of predictions for real
substances.
Advantages of three-
parameter EOS
Most two-parameter EOS assume a fixed critical
compressibility factor, which is not always accurate when
compared to actual experimental data.

1. RK : 2. P-R :

3. M4 EOS :

A three-parameter EOS can have variable critical compressibility factor

values over different temperatures, which helps the EOS adjust more
accurately to real-world conditions and improves its ability to predict
saturated vapor pressure, liquid density, and vapor density at different
temperatures.
The three-parameter
EOS
VLE data of a pure substance in the sub-critical region along
P–T curve (from Ttriple to Tcritical = 1) are used for
determining the temperature functionality of the EOS
parameters
The three parameters should satisfy the following thermodynamic
equilibrium conditions and the following equalities:

Using Abbott’s method to derive a,b and c we get:

where Ac is the temperature dependent correction function.

The three-parameter
EOS
A common form of it as suggested by Soave:

By applying the critical condition of the EOS at the critical point,

three necessary equations for the primary variables A, B, C, and
the critical composition factor​are derived.
Using equilibrium conditions, one of the variables is fixed, while the other
three are adjusted to match the experimental volumetric properties
(such as saturated liquid/vapor densities and vapor pressure) over the
specified temperature range.

Only parameter a is temperature-dependent, while b and c remain constant for

each substance. This can lead to inaccuracies in liquid density predictions,
especially near the critical point, causing deviations from experimental values.
What has been proposed
A new three parameter EOS has been proposed in the following form:

P =( )(
RT
V
v+ b
v-b ) -
a
v(v + c)

The P_rep term is the same as suggested by Mohsen-Nia's

simplified hard core model with = 1.3191.

For the P_att term, we used a simple volume translation term [v(v+c)
in the denominator], which also appeared in the Adachi’s EOS.
What has been proposed
Now, utilizing the method of Abbott, and taking B as the input parameter, the
dimension less parameters A, , and C can be derived as
What has been proposed

The input parameters in the EOSs of Patel and Teja, Adachi et al. and
Yu and Lu are , B and B respectively, which was correlated in
terms of , therefore considered as constant for each compound
and independent of temperature.

But, as we have seen in the equations for a and b, both of them are
function of temperature, at least in the sub critical region. Graphical
representation of the experimental data of the same is shown in the
next few slides.
 Calculations for volumetric properties

For sub-critical region

It is observed that the parameter 𝑏
undergoes a sharp decrease near the
critical point.
This explains why conventional cubic
equations of state (EOSs) struggle to
accurately predict (l) and (v) close to
the critical point, as they only
incorporate temperature dependence in
the parameter a, leaving b unaffected
by temperature changes.
Calculations for volumetric properties

So, we propose the following

equation for input parameter B in
sub-critical region, with bi as
constants of equation.
Calculations for volumetric properties
We propose a similar relation for
Ac in the sub-critical region (Tr<1.0),
using the Soave functionality for Ac
with variation power n
Calculations for volumetric properties
Calculations for volumetric properties
For super-critical region
For the supercritical region, since there is only one phase, the calculation
procedure is straight forward.

Ac = 1 (as usual)

For input parameter B, by using volumetric calculations for 11 compounds

and trial and error procedure, we propose the generalized temperature
independent equation in terms of (acentric factor).
Calculations for volumetric properties

For other compounds not listed in the table on page 11, the constants (b0,
b1, b2, F and n) are represented by U in the following equation and are
expressed in terms of of each compound as the equation, where i
are the coefficients of the equation and presented in the following table.
Calculations for other thermodynamic properties
We get the following equation for enthalpy calculations from other papers
used by the author:

Now, we can derive an explicit equation from the known equation by

applying our proposed EOS as

For Cp calculations, the below equation is used with the finite-divided-

difference method as a method of solution
Calculations for Vapour-Liquid Equilibrium

The proposed EOS was extended to VLE calculations of binary

mixtures by using the well-known conventional mixing rules
with the binary interaction parameter kij, as an adjustable
parameter

Over a limited range of temperature, kij is usually considered as

independent of density, composition and temperature.
Calculations for Vapour-Liquid Equilibrium
For VLE calculations, the procedure suggested by Anderson and Prausnitz
and Smith et al. was used, the K-value for component i as defined by the
following equation was calculated

where v and l are the fugacity coefficients of component i in the vapour

and liquid phases of the mixture.

The fugacity equation for pure component and for component i in mixtures
found using our proposed EOS are given as follows.
 Results

1) VOLUMERTIC PROPERTIES FOR PURE COMPOUNDS

ADD% has been calculated for the 5
established EOS’s and the values
have been compared with those
obtained with the new EOS.

The overall AAD% obtained by the

proposed EOS for (l) of simple
fluids is 1.57%, which is lower than
the other 5 EOS’s.

The overall AAD% obtained by the

proposed EOS for (l) of associating
fluids is 1.62%, which is lower than
the other 5 EOS’s.
Results

The proposed EOS indicates considerable improvement in vapor

density (v) calculations of associating fluids over other five
EOS’s.

None of the EOS’s can predict the saturated vapor densities of

acetic acid and ethylene glycol.
Results

For Ethane and Butane, the proposed EOS has a lower deviation than
the other two EOSs, especially in the critical region (0.90 < Tr < 1.0),
where it is closer to the zero line.
 Results
For Supercritical Region :-

The AADs% obtained by the proposed EOS is close to those obtained

by the M4 EOS, but are much lower than those obtained by the EOS of
P–R, P–T, Adachi et al. and Yu–Lu
 Results
P v/s v (CO2) P v/s v (H2O) - subcooled P v/s v (CO2) - supercritical

The new EOS is capable of predicting subcooled liquid volumes and superheated
vapor volumes, across pressures and temperatures.

Predicted deviations can be observed from the experimental values near critical
region.
 Results
Molar Enthalpy and Heat Capacity Predictions

P–T EOS is not able to predict h along the subcooled liquid isotherms, and cp along both
the subcooled and supercritical isotherms.

For sub-cooled water and ammonia, the proposed EOS has a high capability to predict h
up to 50.0 MPa
 Results
Molar Enthalpy and Heat Capacity Predictions

For the supercritical regions, the h prediction of the M4 EOS is better than the others
equations.

The results of cp calculations by the proposed EOS are more accurate than the EOS by
P–R, Yu–Lu, and M4 (P–T EOS cannot determine cp)
Results
P v/s h (water) P v/s h (ammonia)
Results
cp v/s P (water) cp v/s P (ammonia)

The proposed EOS for the critical isotherm predicts the maximum cp near the critical
pressure, which corresponds to the experimental data.
The calculations of h and cp confirm thermodynamic consistency of parameters in the
proposed EOS.
 Results

1) Binary Mixtures :-

As the other 4 EOS had lower

accuracy, our results are
compared with AAD% of three
parameter Yu-Lu EOS for VLE
calculations.

In most cases, the proposed

EOS provides a better VLE
prediction than the Yu–Lu EOS.
Results

But, the higher deviations in VLE results for those with polar
components such as CO2 and H2S can be related to the
shortcomings of the mixing rules as emphasized by other
researchers.
Conclusions
proposed EOS

Improved liquid density predictions in sub-critical and critical regions. This is

confirmed through volumetric calculations of subcooled liquids (specifically
carbon dioxide and water) and supercritical fluids for 11 different substances.

The EOS is utilized for both PVT and VLE calculations, making it versatile for
evaluating the behavior of simple pure fluids, associating fluids , and their binary
mixtures.

Results from the proposed EOS are compared with those obtained from five
well-established cubic EOSs. The average absolute deviation percentage
(AAD%) is used as a metric for comparison.
Conclusions
new EOS

The proposed EOS achieved an AAD% of 1.57% for simple fluids and 1.62%
for associating fluids, indicating high accuracy in liquid density predictions.

The EOS also allows for the calculation of two additional derivative
properties: molar enthalpy (h) and heat capacity (Cp). The results support
the thermodynamic consistency of the EOS parameters, with a meaningful
temperature dependency observed in the subcritical region, enhancing
the model's reliability.

The proposed EOS was tested for VLE calculations involving 41 binary
mixtures. In most cases, the new EOS provided better accuracy than the
Yu–Lu EOS, which is recognized for its performance in VLE applications.
Thank You :)