DOI: http://dx.doi.org/10.18180/tecciencia.2018.24.

Experimental Data and New Binary Interaction Parameters for Ethanol-

Water VLE at Low Pressures Using NRTL and UNIQUAC
Datos Experimentales y Nuevos Parámetros de Interacción Binaria para el ELV Etanol-Agua a
Bajas Presiones Utilizando NRTL and UNIQUAC

Marco Andrés Guevara Luna1,2,3*, Fredy Alejandro Guevara Luna1, Luis Carlos Belalcázar Cerón1
1
Universidad Nacional de Colombia, Bogotá, Colombia
2
Smart and Simple Engineering SAS, Bogotá, Colombia
3
Universidad Nacional Abierta y a Distancia, Bogotá, Colombia

Received: 3 Aug 2017 Accepted: 28 Jan 2018 Available Online: 14 Feb 2018

Abstract
Ethanol is currently one of the most important and attractive sources of energy with a heating value of 12800 kJ/kg. Ethanol
is commonly obtained from many sources worldwide, mainly renewable sources, and its separation is achieved typically by
atmospheric distillation. Even so, there is new technologies looking for cheaper and more efficient ethanol separation and
purification: pressure-swing distillation, extractive distillation, adsorption with molecular sieves and vacuum membrane
distillation. The design, rating and optimization of these process technologies requires a reliable and universal thermodynamic
modeling approach capable of represents the ethanol-water system properties, and in particular the vapor-liquid equilibrium
(VLE) near the azeotropic point. This study summarize VLE experimental data for ethanol-water system at 3 vacuum
pressures (13.15, 19.71 and 101.32 kPa), and 17 azeotropic point data. Thermodynamic consistency test of the data was
performed using the Redlich-Kister method. The parameters and constants for the detailed thermodynamic modeling of the
ethanol-water system using polar fluid Soave-Redlich-Kwong equation of state (polar-SRK) are summarized as part of this
study, as well as the standard binary interaction parameters for NRTL and UNIQUAC excess Gibbs energy models. Both
models are used simultaneously for the simulations involved in the chemical process engineering activities in science,
academy and industry. The main target of this study consist in the regression of new binary interaction parameters in a
temperature dependent form compatible with the most process simulation software, for NRTL and UNIQUAC, and its
validation using experimental data for azeotropic points at 3 different low pressures with errors of less than 1%, for
temperature and ethanol molar fraction in the vapor phase. The regressed new parameters were tested using isobaric
experimental data for VLE at the 3 under-atmospheric pressures mentioned, obtaining correlation coefficient (R 2) values of
about 1. The calculations were performed using Python 3.4® codes developed and supplied by S&S and Aspen properties®
V8.6 provided by the Universidad Nacional de Colombia.

Keywords: Ethanol-Water, Experimental Data, Vapor-liquid Equilibrium, Azeotropic Point, NRTL, UNIQUAC, Binary
parameters.

Resumen
El etanol es actualmente una de las fuentes de energía más importantes y atractivas, posee un poder calorífico de 12800 kJ /
kg. El etanol se obtiene comúnmente de diversas fuentes en todo el mundo, principalmente fuentes renovables, y su separación
se logra comúnmente mediante destilación atmosférica. Aun así, existen nuevas tecnologías que buscan una separación y
purificación de etanol más económica y eficiente, tales como: destilación de presión y destilación, destilación extractiva,
adsorción con tamices moleculares y destilación por membrana de vacío. El diseño, evaluación y optimización de estas
tecnologías de proceso requiere de un enfoque de modelado termodinámico confiable y universal capaz de representar las
propiedades del sistema etanol-agua, y en particular el equilibrio vapor-líquido (ELV) cerca del punto azeotrópico.

*Corresponding Author. How to cite: Guevara, M. A., Guevara, F. A., Belalcázar, L. C.,
E-mail: [email protected] Experimental Data and New Binary Interaction Parameters for
Ethanol-Water VLE at Low Pressures Using NRTL and UNIQUAC,
TECCIENCIA, Vol. 13 No. 24, 17-26, 2018
DOI: http://dx.doi.org/10.18180/tecciencia.2017.23.11
Este estudio resume los datos experimentales de ELV para el sistema etanol-agua a 3 presiones de vacío (13.15, 19.71 y
101.32 kPa), y 17 datos de puntos azeotrópicos. La prueba de consistencia termodinámica de los datos se realizó utilizando
el método de Redlich-Kister. Los parámetros y constantes para el modelado termodinámico detallado del sistema etanol-agua
usando la ecuación de estado para fluidos polares Soave-Redlich-Kwong (polar-SRK) se resumen como parte de este estudio,
así como los parámetros de interacción binarios estándar para NRTL y UNIQUAC exceso de modelos de energía Gibbs.
Ambos modelos se utilizan simultáneamente para las simulaciones involucradas en las actividades de ingeniería de procesos
químicos en la ciencia, la academia y la industria. El objetivo principal de este estudio consiste en la regresión de nuevos
parámetros de interacción binaria en una forma dependiente de la temperatura que es compatible con la mayoría del software
de simulación de procesos en la industria y el mercado, para NRTL y UNIQUAC, y su validación utilizando datos
experimentales para puntos azeotrópicos a 3 diferentes bajas presiones, esta validación permitió verificar que el modelo
empleado tiene errores de menos del 1%, para la temperatura y la fracción molar de etanol en la fase de vapor. Los nuevos
parámetros regresados se probaron usando datos experimentales isobáricos para VLE a las 3 presiones atmosféricas
mencionadas, obteniendo valores de coeficiente de correlación (R2) de aproximadamente 1. Los cálculos se realizaron usando
códigos Python 3.4® desarrollados y suministrados por S&SE S.A.Sy Aspen properties® V8.6 proporcionado por la
Universidad Nacional de Colombia.

Palabras clave: Etanol-Agua, Datos Experimentales, Equilibrio Vapor-líquido, Punto Azeotrópico, NRTL, UNIQUAC,
Parámetros Binarios.

1. Introduction Liquid Theory) and UNIQUAC (Universal Quasi-chemical

Theory), at low pressures. The binary interaction parameters
The increasing world energy demand has been historically are adjusted to be easily implemented in many
covered by fossil fuels, but in recent years the ecological thermodynamic and process simulators due to the simple
repercussions of fossil fuels as energy sources have make the temperature dependency form proposed.
engineers and scientists to turn their attention to renewable
NRTL and UNIQUAC were chosen for this study, these
18
energy sources [1]. Bio-ethanol, as a clean, condensed and
renewable fuel, is gaining increasing attention, mostly models are useful in process design due to their accuracy in
through its major environmental benefits [2]. Azeotropic (non-ideal) systems such as ethanol-water,
propanol water, n-butanol-water and many others [6].
The worldwide bio-ethanol production is increasing WILSON model is also commonly used due to is simplicity,
constantly [3], and with it the design, rating and optimization but considering the actual computation capabilities and
of old and new process technologies for ethanol purification reliability its implementation doesn’t result suitable, in great
are necessary. To successfully develop new processes, such part because more accurate results can be achieved from
as pressure-swing distillation, extractive distillation, NRT and UNIQUAC models instead.
adsorption with molecular sieves and vacuum membrane
distillation, the implementation of simulations for the The first aim of this study is to summarize the isobaric VLE
conceptual design is one of the most important stages. For experimental data and azeotropic point reported at 3
ethanol production processes, at any scale, an accurate and different under-atmospheric pressures for the ethanol-water
flexible thermodynamic model for the process simulations is system [5] (13.15, 19.71 and 101.32 kPa). This study also
required, a single failure in thermodynamic calculations may includes a thermodynamic consistency test, using the
cause significant error propagations leading to false Redlich-Kister method, for the data used in the parameters
solutions or simulation failures [4]. regression. The second aim is to present new, flexible and
accurate binary interaction parameters, with errors of less
In the case of ethanol separation from water, the binary than 1%, for the NRT and UNIQUAC excess Gibbs energy
system has been widely investigated, mostly at atmospheric models, at low pressures in a temperature dependency form
pressure and high temperatures [4]. Some authors report compatible with most of the process simulators used by
experimental data for low pressures [5], and adjust binary process engineers to design ethanol separation and
interaction parameters for excess Gibbs energy models using purification. The third aim is to provide the parameters and
quadratic form for their temperature dependency. Even so, constants for the detailed thermodynamic modeling of the
the quadratic form proposed is not flexible or compatible ethanol-water system using polar-SRK (polar fluid Soave-
with most of the process simulators available and widely Redlich-Kwong equation of state) for vapor phase and
used worldwide for process synthesis. In this study we NRTL or UNIQUAC for liquid phase with the new
present a new set of binary interaction parameters for two parameters adjusted for low pressures in a suitable way to be
excess Gibbs energy models, NRTL (Nonrandom Two- implemented in design, optimization and rating of processes.
2. Methodology For the data presents in the Table 1: the pressure
measurements have an accuracy of ±0.01 kPa, the
The thermodynamic modeling and calculations were temperature values reported have an accuracy of ±0.01 K
developed using high performance hardware: 128 GB RAM and the molar fractions has an uncertainty of about ± 0.001.
DDR3, Two processors Intel® Xeon sixteen core 2.0 GHz.
The software used was: python 3.4®, the codes implemented 2.2 Thermodynamic modeling
were supplied by S&SE (Smart and Simple Engineering)
company and Aspen properties® V8.6 provided by the Most of the sources for bioethanol production processes are
Universidad Nacional de Colombia. The software was fermentation broths [8]. The purification of the ethanol from
executed using the 100% of computation power available the aqueous non-ideal mixtures requires an accurate
(32 cores x 2.0 GHz). thermodynamic approach for its modeling. Without a
reliable model the simulation will get inaccurate results
2.1 Experimental Data leading to bad designs and finally high economic losses.

The experimental data used for this study is of two types: This work presents an accurate model to simulate the water-
isobaric data at 3 different under-atmospheric pressures ethanol non-ideal system at under-atmospheric pressures.
(13.15, 19.71 and 32.86 kPa). Table 1 and azeotropic points The Equations of state and activity models are summarize,
at 17 different under-atmospheric pressures (Table 9). as well as the constants and parameters involved.

Table 1. Isobaric experimental data for 3 under-atmospheric pressures (13.15, 19.71 and 32.86 kPa) [6]

P=13.15 kPa P=19.71 kPa P=32.86 kPa

19
T (K) x1 y1, T (K) x1 y1 T (K) x1 y1
324.53 0 0 333 0 0 344.57 0 0
313.71 0.117 0.461 323.53 0.076 0.407 335.16 0.063 0.356
310.70 0.241 0.567 320.52 0.141 0.510 332.54 0.104 0.463
309.53 0.383 0.632 318.05 0.260 0.592 330.96 0.150 0.494
308.82 0.492 0.668 317.04 0.390 0.632 329.26 0.248 0.556
308.36 0.590 0.714 316.35 0.499 0.679 326.77 0.454 0.662
308.10 0.656 0.737 315.90 0.599 0.706 326.10 0.581 0.696
307.70 0.771 0.812 315.55 0.688 0.760 325.89 0.657 0.732
307.54 0.840 0.855 315.29 0.799 0.826 325.67 0.760 0.792
307.56 0.864 0.874 316.26 0.805 0.828 325.53 0.839 0.850
307.51 0.897 0.904 315.19 0.841 0.854 325.51 0.854 0.868
307.46 0.908 0.910 315.15 0.883 0.888 325.46 0.913 0.913
307.50 0.911 0.914 315.16 0.913 0.915 325.43 0.931 0.931
307.48 0.934 0.934 315.13 0.931 0.932 325.42 0.932 0.934
307.45 0.947 0.947 315.13 0.945 0.945 325.45 0.965 0.963
307.44 0.971 0.969 315.14 0.960 0.957 325.65 1 1
307.46 1 1 315.14 0.964 0.964
315.23 1 1
2.2.1 Activity models

Gij = exp -cijτij  (3)
The activity models considered for the liquid non-ideal bij
phase of the binary mixture ethanol-water are NRTL τij = aij + (4)
T(K)
(Nonrandom Two-Liquid Theory) and UNIQUAC
(Universal Quasi-chemical Theory). These excess Gibbs
energy models have been widely used to reproduce Here τij are the dimensionless interaction parameters,
accurately the behavior of water-alcohols systems [5], [9]. calculated using the binary interaction parameters adjusted
and presented in this study, aij and bij are the binary
Another model available for its use is the WILSON model interaction parameters adjusted according to the particular
sometimes preferred due to its simplicity and faster conditions, in this case low pressures. cij =cji are called non-
implementation, even so, WILSON model is not accurate randomness parameters, in this case set to 0.3.
enough and with the actual computational capacity, due to
tis reason the use of NRTL or UNIQUAC is currently both The quadratic form presented in 2011 by E.C. Voutsas, et al
fast and reliable. [6] to estimate the interaction parameters is accurate and
consistent but its form is not universal in its temperature
For this highly non-ideal binary system, NRTL and dependency.
UNIQUAC models are suitable for the consideration of the
azeotropic point, which is considerably sensible to changes The form of the equation (4) is compatible with the input
in the system pressure. forms for most of the simulators available for process
analysis and design. Many authors used alternative forms to
2.2.1.1 NRTL estimate the interaction parameters but those forms are not
supported by the simulation platforms making slow the
The NRTL model used for this study corresponds to the process development and inflexible.
expanded form for binary systems to determine the activity
coefficients of ethanol and water in the mixture. The activity 2.2.1.2 UNIQUAC 20
coefficients can be determined with the equations (1) and (2)
[9] The UNIQUAC model was also considered in this study due
to its high accuracy for thermodynamic modeling of non-
 2  ideal systems. This model is a so-called lattice model and
  G  τ G  has been derived from a first order approximation of
ln(γ )= x2  τ  21  + 12 12
1 2 21  x + x G  2


 1 2 21  x +x G 
2 1 12 

(1)
interacting molecule surfaces in statistical thermodynamics,
this study considered the form of the model for binary
systems [9].
 2 
  G  τ G 
2
ln(γ )= x  τ  12  + 21 21 (2) Equations (5) and (6) allow the calculation of the activity
2 1 12  x + x G  2


 2 1 12  
x1+ x G 
2 21   coefficients for the two components in the mixture:

Φ  z θ   r   τ τ 
1 x  2 1 Φ  2  1 r 2  1 
ln(γ )= ln  1  + q ln  1  +Φ  l - 1 l  - q' ln θ' +θ' τ
1 2 21 
+θ' q' 
2 1
21
 θ' +θ' τ
- 12
θ' +θ' τ 

 1  1  2   1 2 21 2 1 12  (5)

Φ  z θ   r   τ τ 
2 x  2 2 Φ  1  2 r 1  2 
ln(γ )= ln  2  + q ln  2  +Φ  l - 2 l  - q' ln θ' +θ' τ
2 1 12 
+θ' q' 
1 2
12
 θ' +θ' τ
- 21
θ' +θ' τ 
 (6)
 2  2  1   2 1 12 1 2 21 

xiqi
Here r, q and q’ are parameters related to each pure θi = (8)
xiqi + x jq j
component involved in the system. li, θi, θi' and are
calculated using the equations (7), (8), (9) and (10). xiq'i
θ'i = (9)
xiq'i + x jq'j
xiri
Φi = (7)
xiri + x jrj li =
z
 
r - q - ri - 1
2 i i
 (10)
2.2.2 Equations of state, constants and parameters This is because the condition is isobaric for the
thermodynamic modeling and the main effect over the
The detailed modeling of all the properties involved in the density of the liquids is due to the equilibrium temperature,
VLE requires of EOS (equations of state) to consider in this EOS take in to account the change of the liquid density
detail the non-ideal behavior of the ethanol water-system. In as a function of the temperature, in terms of reduced
order to quantitative describe the vapor phase the Polar Fluid temperature (Tr) in a simple way.
Soave-Redlich-Kwong Equation of State (polar-SRK) [10]
EOS is used.  (1-T )(2/7) 
l  
v cm3mol-1 = vc  Zc r 

(14)
2.2.2.1 Polar Fluid Soave-Redlich-Kwong Equation of  
State (polar-SRK)
The implementation of the Rackett equation implies the use
The four parameters extension of the SRK EOS to reproduce of the fluid properties at critical conditions, for the system
the properties of polar fluids is presents in equation (11), in under study the critical constants are presents in the Table 2.
an explicit form for pressure. This equation of state requires
two additional empirical parameters to be implemented (m 2.2.2.3 Antoine equation
and n), these parameters for the two substances involved in
the study are presents in the Table 2. In the VLE calculations the prediction of the saturation
pressures (Psat) has great effects on the results, in order to
RT a(T) estimate properly this properties the Antoine equation (15)
P= - was used [11]
(v - b) (v(v + b))
(11)
 n  B
a(T)=1+(1- Tr ) m+  
ln Psat (kPa) = A -T(°C)+C
(15)
 Tr 
 (12)
 RT 
b=0.08664  c 
 P 
The Antoine equation requires for its use at least 3 empirical
constants, for the ethanol water system these constants are
21
 c  (13) presented in the Table 3.
2.2.2.2 Rackett equation 2.3. Regression approach
The implementation of the Rackett equation for the accurate The data regression was developed with the Maximum
estimation of the molar volume of the components in liquid likelihood method. The O.F. (objective function) minimized
as function of the temperature allows to improve the for the adjustment of the parameters was the equation (16):
accuracy of the results obtained.

exp calc 2 exp calc 2 exp calc 2

1 NP  Ti - Ti 1 NP  Pi - Pi 1 NP  y i - yi
O.F. =    +    +   
NP i=1  σT NP i=1  σP NP i=1  σy
     
exp calc 2 azeotrope,exp azeotrope,calc 2
1 NP  Hi - Hi 1 NP  x EtOH - x EtOH
+    +   
NP i=1  σH NP i=1  σx 
   

Table 2 Critical properties and SRK-polar parameters for ethanol and water [10], [11].

Critical Constants Polar-SRK

Tr range
Substance Tc (K) Pc (kPa) Vc (cm3mol-1) Zc ω m n

Ethanol 513.9 6148 167 0.24 0.645 61.40 1.1546 0.5696-0.7058

Water 647.1 22055 55.9 0.229 0.345 0.9499 0.1633 0.4395-0.8493

Table 3 Antoine coefficients for ethanol and water [11]. It should be noted that σ is the standard deviation in the
determination of each property, for bubble point calculations
Substance A B C their value is adjusted to 0.1 as was recommended by E.C.
Ethanol* 16.8958 3795.170 230.918 Voutsas, et al [6].
Water* 16.3872 3885.700 230.17
2.4. VLE definition
Table 4 Thermodynamic consistency test results for the
experimental data.
The VLE approach used for the modeling is the γ-ϕ
formulation [10], where ϕ is the fugacity coefficient for the
Data Type P(kPa) Value Tolerance vapor phase non-ideality estimation. In this case the
13.15 -3.4483 10% formulation was implemented using the detailed modeling
Isobaric data 19.71 -8.3517 10% of the VLE and solved numerically with Python 3.4®
32.86 -4.6524 10% scripts. This approach allows to calculate dew point and
bubble point based in the mathematical models summarized
above in a more accurate way than the typical one, even
Table 1 NRTL binary interaction parameters, standard
when the algorithm is essentially the same described by Smit
parameters [13], [14] and low pressure parameters.
& et al [11].

set Interaction Subindex a b c 3. Results and Discussion

Standard
Ethanol/Water 12 -0.81 246 0.3 3.1 Thermodynamic consistency
Parameters* Water/Ethanol
21 3.46 -586 0.3 The consistency of the experimental data used for the
Low Ethanol/Water 12 -27.39 67 0.3
regression of the new binary interaction parameters was
tested using the Redlich-Kister area-test method. This is the
22
pressure
suitable way to assess the consistency of the thermodynamic
Parameters* Water/Ethanol 21 1.92 -71 0.3
data for systems with interactions that cause non-ideal
*The range of temperature recommended to use these behavior in the VLE [12]. The results of the consistency test
parameters is from 298.14 to 373.15 K. performed using this method are summarized in the Table 4.

Table 2 UNIQUAC binary interaction parameters, standard The experimental data is thermodynamically consistent,
parameters [13], [14] and low pressure parameters. using the area test we realize that the data for the pressure of
19.71 kPa presents the lowest consistency, and the pressure
set Interaction Subindex a b of 13.15 kPa presents the highest consistency. Taking as a
criteria of successful consistency 10% of tolerance the three
Standard Ethanol/Water 12 2.01 -729 datasets passed the test, being the lowest pressure the more
Parameters* Water/Ethanol 21 -2.49 757 consistent dataset. The consistency test allows to establish
Low Ethanol/Water 12 -0.25 22 that the experimental data can be used for interaction
pressure parameters regression.
Parameters* Water/Ethanol 21 -0.04 -54
*The range of temperature recommended to use these 3.2 Binary interaction parameters for low pressures
parameters is from 298.14 to 373.15 K.
Using the regression approach mentioned and the
experimental data presented in Table 1, the binary
Table 3 UNIQUAC pure substances parameters for ethanol interaction parameters was regressed for the two activity
and water [15]. models, NRTL and UNIQUAC. The binary interaction
parameters for the ethanol-water system are adjusted to be
Substance r q q' used in the form of the equation (4), such form of the
Ethanol 2.10547 1.9720 0.9200 temperature dependency is compatible with most of the
Water 0.9200 1.4 1 commercial simulators used by process engineers in the
*The range of temperature recommended to use these design and rating of separation units for ethanol.
parameters is from 298.14 to 373.15 K.
Figure 1 Thermodynamic modeling results (this study) and mathematical-thermodynamic modeling of the ethanol-
experimental data [6] for: a) NRTL and b) UNIQUAC at 3 water system were developed.
low pressures (13.15, 19.71 and 32.86 kPa).
The Figure 1 shows the results for 3 low pressures (13.15,
19.71 and 32.86 kPa) and the points corresponding to the
experimental data at those pressures using the two activity
models with the new binary interaction parameters. Here it
can also be seen that the experimental data is accurately
represented with both models, NRTL and UNIQUAC. This
behavior at low pressures is of importance for the
development of chemical processes involving the VLE for
the ethanol-water system.

Figure 2 shows the dispersion of the results for the VLE

thermodynamic modeling using NRTL and UNIQUAC
proposed parameters. The results in the Figure 2 are the
Bubble temperature and the vapor composition as molar
fraction calculated using the python 3.4® codes for the VLE
modeling with the new binary interaction parameters.

The variables plotted for the 3 under-atmospheric pressures

in the Figure 2 presents low dispersion, this analysis was not
developed for the VLE estimations presented by E.C.
Voutsas, et al [6] in 2011 when they proposed a quadratic
form for the binary interaction parameters estimations. To
validate the binary interaction parameters proposed in this
study a correlation analysis was developed, the Table 8
23
contains the results for the analysis of the variables involved
in the VLE calculations mentioned. The values for the
correlation coefficient (R2) are almost 1, which means there
is a good quantitative representation of the thermodynamic
behavior of the binary mixture VLE. The P-Value values
calculated are very low, that is coherent with the results of
the thermodynamic consistency test developed for the
experimental data.

3.4 Azeotropic Point Prediction

The ethanol-water system is a non-ideal binary mixture that

means the representation of the VLE is dominated mostly by
the azeotropic point. To validate the reliability of the new
binary interaction parameters, and the thermodynamic
modeling proposed, at under-atmospheric pressures, 17
The binary interaction parameters adjusted for the NRTL experimental azeotropic points were considered at 17
model are presented in the Table 5, and the binary different pressure values [6].
interaction parameters for the UNIQUAC model are
For each of the 17 experimental azeotropic points the VLE
presented in the Table 6.
calculations, using the new parameters and modeling
3.3 Phase Equilibria - NRTL and UNIQUAC approach, were developed with the Python 3.4® codes, and
the results compared directly with the experimental data, the
Implementing the two models mentioned (NRTL and results obtained are summarized in ¡Error! No se encuentra
UNIQUAC), the binary interaction parameters regressed el origen de la referencia..
from the experimental data collected and the EOS for liquid
volumes, saturation pressures and vapor phase behavior, the The results allows estimating the azeotropic point with error
of less than 1%.
 24

Figure 2 Thermodynamic modeling dispersion analysis for 13.15 kPa, 19.71 kPa and 32.85 kPa using python 3.4® codes: a)
Temperature calculated using NRTL, b) Temperature calculated using UNIQUAC, c) Ethanol vapor phase molar fraction
calculated using NRTL and d) Ethanol vapor phase molar fraction calculated using UNIQUAC.

Table 8 Correlation parameters (R2 and P-Value) for the thermodynamic modeling of the ethanol-water VLE using NRTL
and UNIQUAC new binary interaction parameters for low pressure at 13.15 kPa, 19.71 kPa and 32.85 kPa.

P=13.15 kPa P=19.71 kPa P=32.86 kPa

Correlation Activity
Dew T y1 Dew T. y1 Dew T. y1
Parameter Model
NRTL 0.9996 0.9996 0.9959 0.9998 0.9986 0.9992
R2
UNIQUAC 0.9995 0.9996 0.9957 0.9998 0.9986 0.9992
P-Value NRTL 9.593e-27 1.554e-26 1.46e-20 4.772e-31 1.954e-21 5.089e-23
 UNIQUAC 5.900e-26 1.246e-26 2.096e-20 1.922e-30 2.103e-21 3.931e-23

Table 9 Azeotropic point prediction error with NRTL and UNIQUAC new low pressure binary interaction parameters.

Experimental
Isobaric Data NRTL UNIQUAC
Azeotrope [6]
Pressure (kPa) x1 T(k) x1 T(k) ΔT(%)* Δx(%)* x1 T(k) ΔT(%)* Δx(%)*
17.52 0.9410 312.910 0.9495 312.596 0.100% 0.903% 0.9394 312.593 0.101% 0.171%
17.84 0.9540 313.150 0.9495 312.945 0.065% 0.472% 0.9596 312.946 0.065% 0.587%
29.53 0.9320 323.150 0.9293 323.044 0.033% 0.290% 0.9293 323.039 0.035% 0.290%
29.58 0.9260 323.150 0.9293 323.079 0.022% 0.356% 0.9293 323.074 0.024% 0.356%
36.78 0.9150 327.960 0.9192 327.686 0.084% 0.458% 0.9192 327.680 0.085% 0.458%
37.62 0.9200 328.150 0.9192 328.172 0.007% 0.088% 0.9263 328.166 0.005% 0.684%
50.66 0.9060 334.750 0.9097 334.742 0.002% 0.405% 0.8987 334.734 0.005% 0.807%
60.34 0.9080 338.730 0.8999 338.749 0.006% 0.891% 0.9091 338.739 0.003% 0.120%
72.18 0.9030 343.020 0.9091 342.972 0.014% 0.675% 0.9091 342.961 0.017% 0.675%
72.54 0.9020 343.150 0.9077 343.091 0.017% 0.633% 0.9067 343.080 0.020% 0.522%
72.62 0.9000 343.150 0.8999 343.117 0.010% 0.010% 0.8991 343.107 0.013% 0.101%
72.67 0.8950 343.150 0.8991 343.134 0.005% 0.457% 0.8990 343.124 0.008% 0.446%
80.30 0.9010 345.580 0.9039 345.544 0.010% 0.322% 0.9091 345.533 0.014% 0.898% 25
87.19 0.8980 347.940 0.8901 347.561 0.109% 0.881% 0.8990 347.549 0.112% 0.110%
88.45 0.8970 347.990 0.8889 347.916 0.021% 0.902% 0.8990 347.904 0.025% 0.222%
92.14 0.8960 349.090 0.8990 348.930 0.046% 0.334% 0.8990 348.917 0.049% 0.334%
101.32 0.8940 351.320 0.8971 351.314 0.002% 0.345% 0.8989 351.300 0.006% 0.546%

exp calc exp calc

T T x x
* T (%)  Azeo exp Azeo * 100; x (%)  Azeoexp Azeo * 100
TAzeo x Azeo

The UNIQUAC model is more accurate than NRTL, the liquid phase composition in molar fraction the error is
highest error for the liquid composition of ethanol in molar slower using the UNIQUAC model.
fraction using NRTL is 0.903%, using UNIQUAC the 4. Conclusions
highest error for the same variable is 0.898% in terms of
temperature the highest error is 0.112% obtained with This study examined the mathematical-thermodynamic
UNIQUAC and the highest error with NRTL is 0.109%. modeling of the ethanol-water system using experimental
data at 3 different under-atmospheric pressures for VLE, and
The mean error for NRTL for the variable temperature is 17 azeotropic point data, to regress new and more accurate
0.033% and for the liquid composition of ethanol in molar binary interaction parameters.
fraction the mean error is 0.495%.
For the UNIQUAC model the mean error for temperature is The parameters were adjusted for two excess Gibbs energy
0.034% meanwhile for composition of ethanol in liquid models (NRTL and UNIQUAC), and are presented in a
phase in molar fraction the mean is 0.431%. simple temperature dependency form compatible with most
of the process simulators available, and commonly used for
Considering this results for the azeotropic point prediction chemical processes design and rating.
the NRTL model is more accurate than the UNIQUAC
model for the bubble point representation. In the case of
The use of these parameters will help the engineers and financing this project and offers the computation equipment,
scientists in the development of new technologies for python 3.4® codes for the thermodynamic modeling and
ethanol-water separation, and in the improvement of technical support required for the development. Also to the
conventional technologies actually implemented. Universidad Nacional de Colombia providing the licensed
software used during this research.
The thermodynamic consistency test of Redlich-Kister
based on areas difference was implemented to test the
experimental data collected and presented in this study, References
aiming to provide reliable data set for future developments
in the field. [1] J. Baeyens, Q. Kang, L. Appels, R. Dewil, Y. Lv, and T. Tan,
“Challenges and opportunities in improving the production of
Detailed mathematical-thermodynamic modeling was bio-ethanol,” Prog. Energy Combust. Sci., vol. 47, pp. 60–88,
2015.
developed using python 3.4® codes and high computation
power. The approach performed was the most rigorous, [2] A. Gupta and J. P. Verma, “Sustainable bio-ethanol production
considering model for all the VLE variables in order to avoid from agro-residues: A review,” Renew. Sustain. Energy Rev.,
ideal assumptions: vapor phase was considered using the vol. 41, pp. 550–567, 2015.
polar-SRK EOS, Antoine and Rackett Equations for [3] M. Petitfrere and D. Vladimir, “Multiphase equilibrium
saturation pressures and liquid densities respectively. The calculations using a reduction method,” Fluid Phase Equilib.,
results were compared with the experimental data at 3 vol. 401, pp. 110–126, 2015.
different under-atmospheric pressures (13.15, 19.71 and
[4] A. F. Cristino, S. Rosa, P. Morgado, A. Galindo, E. J. M. Filipe,
32.86 kPa) for the two activity models (NRTL and A. M. F. Palavra, and C. A. Nieto de Castro, “High-temperature
UNIQUAC), and using the new adjusted binary interaction vapour-liquid equilibrium for the water-alcohol systems and
parameters. The correlation coefficients calculated to modeling with SAFT-VR: 1. Water-ethanol,” Fluid Phase
validate the modeling are close to 1with both models. P- Equilib., vol. 341, pp. 48–53, 2013.
values calculated for the VLE are lower of 1x10-5 in all cases [5] E. C. Voutsas, C. Pamouktsis, D. Argyris, and G. D. Pappa,
supporting the validity of the thermodynamic modeling
results.
“Measurements and thermodynamic modeling of the ethanol-
water system with emphasis to the azeotropic region,” Fluid
26
Phase Equilib., vol. 308, no. 1–2, pp. 135–141, 2011.
To consider the azeotropic point estimation at low pressures [6] A. Farajnezhad, O. A. Afshar, M. A. Khansary, S. Shirazian,
in detail, the experimental azeotropic data for 17 different and M. Ghadiri, “Correlation of interaction parameters in
pressure values were used to evaluate the accuracy of the Wilson, NRTL and UNIQUAC models using theoretical
thermodynamic modeling approach using NRTL and methods,” Fluid Phase Equilib., vol. 417, pp. 181–186, 2016.
UNIQUAC with the new proposed binary interaction [7] L. Axelsson, M. Franzén, M. Ostwald, G. Berndes, G. Lakshmi,
parameters through the python 3.4® codes provided by and N. H. Ravindranath, “Perspective: Jatropha cultivation in
S&SE S.A.S. in all cases the error is lower than 1%, meaning southern India: Assessing farmers’ experiences,” Biofuels,
that the approach proposed and the parameters regressed Bioprod. Biorefining, vol. 6, no. 3, pp. 246–256, 2012.
have a high accuracy. For the bubble temperature the mean [8] Y. Dadmohammadi, S. Gebreyohannes, B. J. Neely, and K. A.
error is smaller than 0.034% and for the liquid phase M. Gasem, “Multicomponent phase behavior predictions using
composition of ethanol in molar fraction the mean error is in QSPR-generalized NRTL and UNIQUAC models,” Fluid Phase
both cases smaller than 0.495%, which validates the Equilib., vol. 409, pp. 318–326, 2016.
capabilities of the models and the new parameters proposed [9] J. A. Sandarusi, A. J. Kidnay, and V. F. Yesavage, “Compilation
to reproduce quantitatively the behavior of the non-ideal of parameters for a polar fluid Soave-Redlich-Kwong equation
binary system of ethanol-water. of state,” Ind. Eng. Chem. Process Des. Dev., vol. 25, no. 4, pp.
957–963, 1986.
The thermodynamic model presented in this study can be [10] J. M. Smith, H. Van Ness, and M. M. Abbott, Introduction to
fully implemented in parallel with CFD (computational fluid Chemical Engineering Thermodynamics, 7th ed. 2005.
dynamics) simulations to consider the detailed
thermodynamic behavior in a coupled way with the transport [11] M. Llano-Restrepo and J. I. Carrero-Mantilla, “Futility or
usefulness of common implementations of the area and slope
phenomena, a very recent field of research due to its consistency tests for partial molar properties in binary
technical possibilities. mixtures,” Fluid Phase Equilib., vol. 398, pp. 72–79, 2015.

[12] B. Daniel, M. Peters, D. Glasser, and D. Hildebrandt,

Acknowledgements Understanding Distillation Using Column Profile Maps, 1st ed.
2013.
The authors wish to express their gratitude to: S&SE S.A.S
(Smart and Simple Engineering S.A.S) from Colombia for [13] Dormund-Data-Bank, “Dortmund Data Bank,” 2016. [Online].
Available: http://www.ddbst.com/ddb.html.
[14] Chemicals Enigeer´s Guide, “Chemicals Enigeer´s Guide,”
2016. [Online]. Available: http://cheguide.com/. [Accessed: 01-
Jun-2016].

27