Matricarde Falleiro, R. M. 1, Meirelles, A. J. A. 2, Krähenbühl, M. A. 1, *
Laboratório de EXtração,
TeRmodinâmica Aplicada e Equilíbrio
1 Laboratory
of Extraction, Applied Thermodynamics and Equilibrium, ExTrAE, School of Food Engineering, University of Campinas, Brazil
* Corresponding
author e-mail: mak@feq.unicamp.br
OBJECTIVES
RESULTS AND DISCUSSION
Vapor-liquid equilibrium (VLE) data of mixtures involving fatty
acids and ethyl esters are practically nonexistent in literature.
The great importance of such data for the biodiesel industry is
related to the quality of the biofuel. Generally, its quality can be
influenced by several factors, such as presence of Free Fatty
Acids (FFA). These, depending on concentration, inhibit phase
separation of ethyl esters and glycerol, which compromises the
quality standards of biodiesel. Therefore, the present work aims
to determine VLE data for fatty mixtures of acids and esters by
Differential Scanning Calorimetry (DSC).
For every analyzed sample (Figure 2, 3 and 4), the boiling
temperature was determined by the onset temperature measured
as shown in Figure 1-d.
208
x1 ≈ 0.1
x1 ≈ 0.3
This work
NRTL
UNIQUAC
204
x1 ≈ 0.5
Temperature / ºC
2 Laboratory
of Thermodynamic Properties, LPT, School of Chemical Engineering, University of Campinas, Brazil
x1 ≈ 0.7
x1 ≈ 0.9
(a)
200
196
192
(b)
188
0.0
0.2
0.4
0.6
0.8
1.0
x1, y1
Figure 2 – (a) Boiling endotherms to system: ethyl myristate (1) + myristic acid (2)
obtained by DSC; (b) liquid-vapor equilibria: ethyl myristate (1) + myristic acid (2) at
2.67 kPa.
METHODOLOGY
208
Lauric acid, myristic acid, ethyl myristate and ethyl palmitate
with purity greater than 99% were acquired from Sigma, and oleic
acid, with 99% purity, was acquired from Fluka.
This work
NRTL
UNIQUAC
Temperature / ºC
204
The experimental apparatus (Figure 1-a) consists of a DSC (Model
2920) with a vacuum system fitted to it. For the DSC analysis,
binary mixtures of ethyl myristate + myristic acid; lauric acid +
ethyl palmitate and ethyl palmitate + oleic acid was evaluated at
2.67 kPa, covering the entire phase diagram (x1 = 0, 0.1, 0.2 ... to
1.0). Samples of 4 to 8 mg were used in the study, with a heating
rate of 25 °C.min−1 and a small ball placed over the pinhole
(Figure 1-c), in order to avoid the pre-vaporization of the sample,
since it behaves as an exhaust valve, releasing the vapor phase
in a controlled manner.
(a)
200
196
192
(b)
188
0.0
0.2
0.4
0.6
0.8
1.0
x1, y1
Figure 3 – (a) Boiling endotherms to system: lauric acid (1) + ethyl palmitate (2)
obtained by DSC; (b) liquid-vapor equilibria: lauric acid (1) + ethyl palmitate (2) at
2.67 kPa.
240
This work
NRTL
UNIQUAC
236
Temperature / ºC
232
228
224
220
216
(b)
212
(a)
0.0
0.2
0.4
0.6
0.8
1.0
x1, y1
(a)
Figure 4 – (a) Boiling endotherms to system: ethyl palmitate (1) + oleic acid (2)
obtained by DSC; (b) liquid-vapor equilibria: ethyl palmitate (1) + oleic acid (2) at
2.67 kPa.
The data were modelated using the NRTL and UNIQUAC
Equations (Table 1). Both models were able to describe the
experimental data with low deviations.
(b)
Table 1 – Binary interaction parameters of NRTL and UNIQUAC models.
Sistemas
Ethyl Myristate +
Myristic Acid
(c)
Lauric Acid +
Ethyl Palmitate
Ethyl Palmitate +
Oleic Acid
*
gE
NRTL
UNIQUAC
NRTL
UNIQUAC
NRTL
UNIQUAC
N
Mean deviation / º C   Texperiment al  Tg E model
i 1
(e)
(d)
onset temperature
baseline
boiling
endotherm
Figure 1 – (a) General view of experimental apparatus room; (b) Expanded perspective of the
DSC furnace; (c) Tungsten ball being placed on the pinhole; (d) Perspective in more details of
some accessories under of the bench; (e) Differential thermal curve obtained by DSC
Model
A12 /
cal.mol-1
1529.5500
637.3379
-865.8260
-341.7366
844.9378
283.2647
1

N
A21 /
a12
-1
cal.mol
-782.9940 0.2987
-447.0392
1339.6447 0.3181
437.2044
-927.5403 0.3054
-275.0970
-
* Mean
Deviation /
ºC
0.14
0.14
0.17
0.18
0.38
0.38
, N  11
CONCLUSION
The results obtained have shown that the applied method is an
accurate experimental technique that uses a shorter operation
time and small amount of sample. This study also improved the
thermodynamic modeling by determining the binary interaction
parameters of models used to describe non-ideality of the liquid
phase needed for the development of equipment for fuel
production (biodiesel)