See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/266505830

A-UNIFAC Modeling of Binary and Multicomponent Phase Equilibria of Fatty

Esters+Water+Methanol+Glycerol

Article

CITATIONS READS

0 98

6 authors, including:

Olga Ferreira Jean-Charles de Hemptinne

Instituto Politécnico de Bragança IFP Energies nouvelles
32 PUBLICATIONS   271 CITATIONS    87 PUBLICATIONS   1,595 CITATIONS   

SEE PROFILE SEE PROFILE

Susana Bottini
Universidad Nacional del Sur
83 PUBLICATIONS   1,774 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Fuel physical properties - ReFGen View project

E-learning from Nature View project

All content following this page was uploaded by Jean-Charles de Hemptinne on 28 October 2014.

The user has requested enhancement of the downloaded file.

 A-UNIFAC Modeling of Binary and Multicomponent Phase Equilibria of Fatty
Esters+Water+Methanol+Glycerol

N. Garridoa, O. Ferreirab, R. Lugoc, J.-C. de Hemptinnec, M. E. Macedoa, S.B. Bottinid,*

a
Department of Chemical Engineering, FEUP, University of Porto, Porto, Portugal
b
Departamento de Tecnologia Química e Biológica - Instituto Politécnico de Bragança,
Campus de Santa Apolónia, Ap. 1134 5301-857 Bragança, Portugal
c
Institut Français du Pétrole – Direction Chimie et Physico-Chimie Appliquées, 1 & 4 av de
Bois Préau, 92852 Rueil Malmaison, France
d
Planta Piloto de Ingeniería Química – PLAPIQUI, Universidad Nacional del Sur, Camino La
Carrindanga Km 7.5 – CC 717, Bahía Blanca, Argentina
Corresponding author: Tel.: +54 2914861700; fax: +54 2914861600; E-mail address:
sbottini@plapiqui.edu.ar (S.B. Bottini)

Keywords: thermodynamic modeling, A-UNIFAC, biodiesel

Abstract
The production of methyl and ethyl esters of fatty acids is of great industrial interest,
considering the direct application of these esters as biodiesel. Prediction of the phase
behaviour of mixtures containing fatty esters, alcohols, glycerol and water is of utmost
importance for the design and optimization of biodiesel purification and by-products recovery
processes. In this work we show the capability of A-UNIFAC to correlate and predict the
phase behaviour of these mixtures. This GE model is an extension of UNIFAC that explicitly
includes association effects between groups, based on the statistical Wertheim theory [1].
A-UNIFAC with the final set of parameters is able to predict with good agreement
experimental data on binary and ternary liquid-liquid equilibria of glycerol, methanol and
fatty esters as well as infinite dilution activity coefficient for these systems. Additionally,
satisfactory results are obtained for the prediction of liquid-liquid equilibrium of binary
mixtures containing water and fatty acid esters.

1. Introduction
Biodiesel can be prepared by transesterification of triglycerides with an excess of alcohol, to
produce methyl or ethyl esters of fatty acids and glycerol. After reaction, several separation
processes are required for the purification of biodiesel and the recovery of by-products.
Knowledge of phase equilibria of the reactive mixture comprising fatty esters, alcohols,
glycerol and water is therefore essential for the design and optimization of these processes.
Considering the presence of several hydrogen bonding components, the associating A-
UNIFAC activity coefficient model is applied to describe binary and ternary liquid-liquid
equilibria (LLE) of glycerol, methanol and fatty acid esters mixtures, as well as LLE of water
and methyl esters.
The parameterization methodology is described in section 3 and the results obtained are
shown and discussed in section 4. The main conclusions are finally presented in section 5.

2. A-UNIFAC model
Mengarelli et al. [1] have proposed a modified UNIFAC model that takes into account
association effects. This GE model adds a group-contribution association term to the original
UNIFAC combinatorial and residual expressions (Fredenslund et al., 1975). The association
term is based on Wertheim’s theory for fluids with highly directed attractive forces [2]. The
general expression for the association contribution to the activity coefficient is a function of
the fraction of non-bonded sites in the solution ( X Ak ) and in pure-component i ( X iAk )::
NGA
X Ak X iAk − 1 1 − X Ak
ln γ iassoc = ν ki ln + + ri ρ k
k =1 Ak X iAk 2 Ak 2
(1)

In this equation ν ki is the number of groups of type k in a molecule of component i, ρj

represents the density of associating group k in the mixture and ri represents the UNIQUAC
molecular volume of species i (evaluated from UNIFAC group volume parameters Rk in the
NG
usual way; i.e. summation over all NG groups within molecule i: ri = ν ki Rk ). The
k =1
summation in eq. (1) extends to all NGA associating groups and Ak associating sites.
The fraction of non-bonded sites X Ak is a function of the associating group density ρj and the
A B
association strength ∆ k j between site A of group k and site B of group j . This fraction is
given by the following expression [1]:
1
X Ak = NGA
(2)
Ak B j Bj
1+ ρ j∆ X
j =1 B j

A similar equation is applied to caculate the fraction of non-bonded sites in pure component i:
1
X iAk = NGA
(3)
Bj
Ak B j
1+ ( ρ j )i ∆ X i
j =1 B j

In these equations ρ j and (ρ j )i represent the density of associating group j in the

mixture and in pure associating component i, respectively, evaluated through the value of the
UNIQUAC molecular volume of specie i:
NC
ν ij xi
ρj = i =1
NC
(4)
ri xi
i =1

ν ij
(ρ )
j i = (5)
ri
where xi is the mole fraction of component i in the mixture. The densities calculated by the
last equations are dimensionless and characterize the concentration of associating group j in
the solution and in pure component i.
AB AB
The association strength ∆ k j is a function of two association parameters: the energy ( ε k j )
AB
and volume ( κ k j ) of association.
A B A B A B
∆ k j = κ k j [exp( ε k j / kT ) − 1] (6)
3. Model Parameterization
3.1 Association parameters
Association effects in methanol, water and glycerol are represented by the same hydrogen-
bonding hydroxyl group (OH) having two associating sites (one electronegative O and one
electropositive H). There is one associating group in methanol and water and three in
glycerol. Group CCOO in fatty esters can cross-associate with the electropositive site of the
hydroxyl OH group.
The values for the hydroxyl (OH) and ester (CCOO) self- and cross-association parameters
were already available [3] and are presented in Table 1.

Table 1. Energy /k and volume association parameters for the A-UNIFAC model [3]
/k (K) κ
Self-association OH 3125.0 0.0062
Cross-association OH – CCOO 1975.0 0.071

3.2 Residual interaction parameters

The introduction of an association contribution to the activity coefficient, makes it necessary
to determine a new set of parameters for the interaction between associating and non-
associating functional groups.
The residual parameters for the interactions between water and ester functional groups were
previously estimated using low-pressure VLE data and infinite dilution activity coefficienta
γ ∞ in binary mixtures of esters and water [3].
For the quantification of the residual non-idealities, both methanol and glycerol molecules are
represented by the molecular groups CH3OH and C3H8O3. The residual interaction parameters
between CH3OH and C3H8O3 were obtained by fitting isothermal liquid-liquid equilibrium
data on the ternary system dodecanoic acid methyl ester-methanol–glycerol [4].
The glycerol/paraffin (C3H8O3/CH2) and glycerol/ester (C3H8O3/CCOO) interaction
parameters were estimated from experimental data on liquid-liquid equilibria and γ ∞ of the
binary systems dodecanoic acid methyl ester (DAME)-glycerol and hexanoic acid methyl ester
(HAME)-glycerol in the range of 320 K to 438 K [4].
The residual group interaction parameters used in this work are reported in Table 2.

Table 2. Residual group interaction parameters am,n (K) for the A-UNIFAC model
n
m CH2 C3H8O3 CH3OH CCOO H2O
c b a
CH2 0.0 101.5 122.7 232.1 380.5b
C3H8O3 -51.2c 0.0 -193.6e 886.3c n.a.
CH3OH -19.78b 427.6e 0.0 200.2d -167.6b
CCOO 114.8a 20.0c 18.49d 0.0 -72.73b
H2O 136.8b n.a. 251.2b 150.3b 0.0
a: original UNIFAC parameters [5]; b: [3]; c: [6]; d: [7]; e: this work; n.a.: not available.
4. Results and discussion
In this section several correlation and prediction results are presented.
4.1 Binary systems of esters and water
An important biofuel quality parameter is its maximum water content. Thus, the correct
description of the solubility of water in fatty acid esters is of major importance during
production and storage of biodiesel.
The A-UNIFAC model has been previously applied to describe low pressure VLE data and
γ ∞ of binary systems of water and short-chain esters [3]. In this work the same association
and residual parameters are used to predict liquid-liquid equilibrium of water and several fatty
acid esters. Figures 1 and 2 show that A-UNIFAC predictions are in good agreement with the
experimental information available.

420 420

390 390
T/K
360 360
T/K

330 330

300 300
0.E+00 2.E-05 3.E-05 5.E-05 0.80 0.85 0.90 0.95 1.00
x ester x ester

Figure 1. Liquid-liquid equilibrium of octanoic acid methyl ester + water: A-UNIFAC

predictions (solid lines) and experimental data (symbols) [4].

0.07 methyl hexanoate

methyl heptanoate
0.06 methyl octanoate
methyl tetradecanoate

0.05 methyl octadecanoate

methyl dodecanoate
x water

0.04

0.03

0.02

0.01
285 295 305 315 325 335 345
T/K

Figure 2. Water solubility in several fatty methyl esters: A-UNIFAC predictions (solid lines)
and experimental data (symbols) [8].
4.2 Binary systems of esters and glycerol
Figures 3-8 compare experimental data [4] with A-UNIFAC correlation of liquid-liquid
equilibrium and infinite dilution activity coefficients of the system ester / glycerol.
500 500

450 450
T/K

T/K
400 400

350 350

300 300
0 0.01 0.02 0.03 0.04 0.05 0.980 0.985 0.990 0.995 1.000
x glycerol x glycerol

Figure 3. Comparison between A-UNIFAC Figure 4. Comparison between A-UNIFAC

correlated (solid lines) and experimental correlated (solid lines) and experimental
(symbols) solubility of glycerol in HAME (symbols) solubility of HAME in glycerol
phase. phase.
500 500

450 450
T/K

T/K

400 400

350 350

300 300
0 0.01 0.02 0.03 0.04 0.05 0.980 0.985 0.990 0.995 1.000
x glycerol x glycerol

Figure 5. Comparison between A-UNIFAC Figure 6. Comparison between A-UNIFAC

correlated (solid lines) and experimental correlated (solid lines) and experimental
(symbols) solubility of glycerol in DAME (symbols) solubility of DAME in glycerol
phase. phase.
15000 160
γ∞ γ∞
12000
120
9000
80
6000
3000 40
0 0
370 390 410 430 450 370 390 410 430 450
T/K T/K

Figure 7. Comparison between A-UNIFAC Figure 8. Comparison between A-UNIFAC

correlated (solid lines) and experimental correlated (solid lines) and experimental
(symbols) infinite dilution activity (symbols) infinite dilution activity
coefficients of DAME in glycerol. coefficients of glycerol in DAME.

4.3 Ternary systems of esters, methanol and glycerol

 The A-UNIFAC model with the parameters listed in Tables 1 and 2 was used to calculate the
fluid phase equilibrium boundaries of fatty acid methyl ester-methanol-glycerol mixtures at
several temperatures [4,6].
Table 3 reports the mean standard deviations in the calculation of methanol distribution
coefficients by the A-UNIFAC model. The distribution coefficient was defined as
the ratio between the methanol mole fraction in the fatty acid methyl ester and glycerol
phases. Table 4 gives the mean and maximum absolute differences between calculated and
experimental molar fractions of glycerol in the fatty ester phase and molar fractions of fatty
ester in the glycerol phase.

Table 3. Mean standard deviations in the calculation of methanol distribution coefficients by

the A-UNIFAC model
System Standard Deviations % (*)
HAME – glycerol – methanol 12.73
DAME – glycerol – methanol 6.13
Methyl oleate (MO) – glycerol – methanol 16.36
(*)
= 100 × [( − ) ]2 ; NP = number of experimental points

Table 4. Absolute deviations between calculated and experimental mole fractions of diluted
components using A-UNIFAC
Mol fraction of glycerol in Mol fraction of fatty ester
System the fatty ester rich phase in the glycerol phase
Mean Maximum Mean Maximum
difference difference difference difference
HAME + methanol + glycerol 0.0384 0.1064 0.0084 0.0368
DAME + methanol + glycerol 0.0179 0.0388 0.0080 0.0154
MO + methanol + glycerol 0.0050 0.0381 0.0056 0.0111

Figures 9 to 12 compare experimental data with A-UNIFAC calculations.

xXmethanol
Methanol
0.0 1.0
T=333.5K 1
0.2 0.8 0.9
0.8
0.7
0.4 0.6 0.6
K meth
Kmethanol

0.5
0.6 0.4 0.4
0.3
0.2 T=333.5K T=352.8K
0.8 0.2 0.1
0
1.0 0.0 0 0.1 0.2 0.3 0.4 0.5 0.6
0.0 0.2 0.4 0.6 0.8 1.0 x methanol
X methin HAME
HAME phasephase
X Glycerol
x glycerol xXHAME
HAME
 Figure 9. Comparison between A-UNIFAC Figure 10. Comparison between A-UNIFAC
predicted (solid lines) and experimental predicted (solid lines) and experimental
(dashed lines) [4] tie-lines of the ternary (symbols) [4] distribution coefficients of
system HAME-methanol-glycerol at 333.5 methanol in ternary system HAME-
K. methanol-glycerol.

xXmethanol
Methanol
0.0 1.0
T=313.15K
0.2 0.8
0.9
0.8
0.4 T=313K T=333KC T=353K
0.6 0.7

K methanol
0.6
0.6 0.4
0.5

Kmethanol
0.4
0.3
0.8 0.2 0.2
0.1
1.0 0.0
0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
xX Glycerol
glycerol xX MO
MO X methin
x methanol MOMO
phase
phase

Figure 11. Comparison between A-UNIFAC Figure 12. Comparison between A-UNIFAC
predicted (solid lines) and experimental predicted (solid lines) and experimental
(symbols) [6] distribution coefficients of (symbols) [6] distribution coefficients of
methanol in ternary system MO-methanol- methanol in ternary system MO-methanol-
glycerol at 313.15 K. glycerol.

5. Conclusions
The A-UNIFAC model gives a good representation of the liquid-liquid equilibria of mixtures
containing glycerol, methanol and fatty esters as well as infinite dilution activity coefficients
in these systems. This group-contribution model can be further extended to describe the phase
equilibria of multicomponent mixtures of interest in biodiesel production.
References
[1] A.C. Mengarelli, E.A. Brignole, S.B. Bottini, Fluid Phase Equilib. 163 (1999) 195-207.
A. Fredenslund, R.L. Jones, J.M. Prausnitz, AIChE. J. 21 (1975) 1086-1099
[2] M. Wertheim, J. Stat. Phys. 35 (1984) 19.
[3] O. Ferreira, E.A. Macedo, S.B. Bottini, Fluid Phase Equilibria 227 (2005) 165-176.
[4] F.M. Korgitzsch, Study of Phase Equilibria as a Fundament for the Refinement of
Vegetable and Animal Fats and Oils. Ph.D. Dissertation, TU Berlin, 1993.
[5] J. Gmehling, P. Rasmussen, A. Fredenslund, Ind. Eng. Chem. Process Des. Dev. 21
(1982) 118–127.
[6] A.E. Andreatta, L.M. Casás, P. Hegel, S.B. Bottini, E.A. Brignole, Ind. Eng. Chem. Res.
47 (2008) 5157-5164.
[7] L.M. Casás, B. Orge, O. Ferreira, J. Chem. Eng. Data 53 (2008) 89–93.
[8] M.B. Oliveira, F.R. Varanda, I.M. Marrucho, A.J. Queimada, J.A.P. Coutinho, Ind. Eng.
Chem. Res. 47 (2008) 4278–4285.

View publication stats