Kinetic Modeling of Ethylene Glycol Monoesterification

Article abstract of DOI:10.1002/kin.20938

The monoesterification of ethylene glycol under isothermal conditions was conducted using benzoic acid in methane-sulfonic acid/Al₂O₃ as a catalyst. Using this reagent, glycol was selectively monoesterified with high yield. The reactions were performed within an automated batch reactor under equimolar conditions, constant rotational frequency of the stirrer, and within the temperature range from 65 to 85°C. The rate constant related to this reaction and to the corresponding reverse reaction, activation energy, and preexponential factor was derived from experimental data. It has been concluded that under these conditions the formation of dibenzoate was successfully prevented.

Full text of DOI:10.1002/kin.20938

Kinetic Modeling of
Ethylene Glycol
Monoesterification
ˇ
ˇ
DARJA PECAR, ANDREJA GORSEK
University of Maribor, Faculty of Chemistry and Chemical Engineering, Smetanova 17, SI-2000, Maribor, Slovenia
Received 8 January 2015; revised 17 May 2015; accepted 2 July 2015
DOI 10.1002/kin.20938
Published online 24 August 2015 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: The monoesterification of ethylene glycol under isothermal conditions was con-
ducted using benzoic acid in methane-sulfonic acid/Al2O3 as a catalyst. Using this reagent,
glycol was selectively monoesterified with high yield. The reactions were performed within
an automated batch reactor under equimolar conditions, constant rotational frequency of the
stirrer, and within the temperature range from 65 to 85°C. The rate constant related to this re-
action and to the corresponding reverse reaction, activation energy, and preexponential factor
was derived from experimental data. It has been concluded that under these conditions the
formation of dibenzoate was successfully prevented. ꢀC 2015 Wiley Periodicals, Inc. Int J Chem
Kinet 47: 658–663, 2015
INTRODUCTION
rity of the products is higher because side reactions can
be completely eliminated or are less significant; it is
possible to work under milder and solvent-free reaction
conditions and also what is of great importance we can
develop safer and nonwaste producing reactions [2–5].
Nevertheless, there are economic demands toward the
uses of solid catalysts instead of the liquid ones. For the
replacement of those corrosive liquid acids that are cur-
rently used, solid acid catalysts such as zeolites, clays,
sulfated metal oxides, heteropolyacids, ion-exchange
resins can be applied [6,7].
Esterification reactions are reversible equilibrium-
limited reactions in which water is produced as a by-
product. Water effects the conversion and activity of
an acidic catalyst. Therefore, in some cases water is
removed from the reaction mixture to improve the effi-
ciency of the reaction [8,9]. The downside of such reac-
tions is also the poor selectivity leading to undesirable
side reactions and low yields. These weaknesses can be
overcome by the application of specific catalysts [5,6].
Esters are industrially very important products. Esters
of different alcohols and acids are used as pharmaceu-
ticals and pharmaceutical intermediates, flavors, lubri-
cants, emulsifiers, and additives applicable within the
food, beverages, cosmetics, and utility chemicals in-
dustries [1]. They can be synthesized by esterification
of carboxylic acids, transesterifications of methyl or
ethyl esters, or alkylation of carboxylic anions [2].
The applications of solid catalysts for the esterifica-
tion and/or transesterification of carboxylic acids or
esters with alcohols compared to homogeneous cata-
lysts (mineral acids, metal hydroxide, and metal alkox-
ide) are preferable. Heterogeneous catalysts are much
easily separated from the reaction mixture, and the pu-
Correspondence to: Darja Pe cˇ ar; e-mail: darja.pecar@um.si.
ꢀ
2015 Wiley Periodicals, Inc.
C

KINETIC MODELING OF ETHYLENE GLYCOL MONOESTERIFICATION
659
Figure 1 Reaction scheme for the monoesterification and diesterification of ethylene glycol with benzoic acid.
The synthesis of esters attracts the attention of many
researchers because of its widespread application. De-
spite of the disadvantages of using homogeneous acid
catalyst, the kinetic studies of homogeneous esterifica-
tions are still very important [10–12]. But there is an
increasing tendency toward development of a process
that meets the requirements of safety and minimum
waste. Cation exchange resins are there for a reason-
able choice of solid catalysts [13–15].
The presented paper reports on a kinetic study of
ethylene glycol monoesterification with benzoic acid
that was carried out in Al2O3/MeSO3H as a catalyst.
This catalyst was chosen because it is very effective,
it has high selectivity toward the monoester, and it is
inexpensive [2]. Various parameters were determined
such as equilibrium reaction rate constants, activation
energies, and preexponential factors. A considerable
amount of published work is available on the esterifi-
cation systems, but none of these studies considered the
kinetics of the mentioned esterification using a speci-
fied catalyst.
chloroform (w = 99.0–99.4%; Merck, Darmstadt,
Germany) are commercially available and were used
as received without further purification.
Equipment
An automated, computer in-line controlled and reg-
ulated reaction calorimeter (Mettler Toledo RC1e,
Greifensee, Switzerland) was used to conduct the ex-
perimental work. The 0.8-L double-walled glass-made
laboratory batch reactor was equipped with a propeller
glass stirrer, a glass AP01 temperature sensor, and the
TM
FTIR-based ReactIR iC10 analysis system coupled
TM
with a flexible Silver Halide (AgX) FiberConduit
and 6.3 mm DiComp
,
TM
probe, which was used for
the real-time monitoring of 2-hydroxyethyl benzoate
concentration profiles.
Experimental Procedure
Monoesterification of ethylene glycol under isother-
mal conditions was conducted using benzoic acid in
Al2O3/MeSO3H as a catalyst. The reaction scheme in
Fig. 1 shows two reversible competitive esterification
reactions, which led first to monoester and further to
diester. The purpose of our work was to maximize
the yield of monoester. The synthesis procedure was
proven to be very effective and highly selective for
monoesterification and was reported by Sharghi and
Sarvari [2]. The selectivity toward the formation of mo-
noester was also increased by using equimolar amounts
of ethylene glycol and benzoic acid.
EXPERIMENTAL
Materials
All the chemicals, ethylene glycol (w ࣙ 99.0%; Fluka,
Belgium), benzoic acid (w ࣙ 99%; Merck, Hohen-
brunn, Germany), methane-sulfonic acid (anhydrous;
Aldrich, France), aluminum oxide (anhydrous; Merck,
Darmstadt, Germany), calcium chloride (w ࣙ 97.0%;
Sigma-Aldrich, China), sodium hydrogen carbonate
All the experiments were carried out within a
RC1e reactor system (Mettler Toledo) at a rotational
(w ࣙ 99.0%; J.T. Baker, Deventer, Holland), and
International Journal of Chemical Kinetics DOI 10.1002/kin.20938

ˇ
ˇ
PECAR AND GORSEK
6
60
frequency of the stirrer fm = 250 min−1, and reaction
media temperatures ϑ = 65, 70, 75, 80, and 85°C.
The individual experiment was started by charging
the RC1e reactor with methane-sulfonic acid (m =
2
50 g) and benzoic acid (m = 20.63 g). The medium
was stirred at the defined rotational frequencies of the
stirrer and heated up to the desired reaction medium
temperature. Thereupon, Al2O3 was slowly charged
to the reaction medium (m = 45.61 g). The monoes-
terification was triggered by the addition of ethylene
glycol (m = 10.49 g), and the progress of the reaction
TM
was monitored using the FTIR-based ReactIR iC10
analysis system. The ReactIR iC software was used
to control the spectrometer and to collect the spectra
every 2 min. All the spectra collected during the exper-
iments were measured against a background spectrum
of air.
Analytical Procedure
Figure 2 FTIR–ATR spectra of ethylene glycol, benzoic
acid, and 2-hydroxyethyl benzoate (y axis represents trans-
mittance 5–100%.
2
-Hydroxyethyl benzoate is not commercially avail-
able, thus for real-time monitoring the end product
of monoesterification was further cleaned to remove
the catalyst and nonreactive reagents as described by
Sharghi and Sarvari [2]. The reaction mixture was
cooled and poured into distilled water. The mixture
was twice extracted with chloroform. Both phases were
separated. The organic phase was washed with a satu-
rated solution of sodium hydrogen carbonate and dried
over calcium chloride. After the filtration, the chlo-
roform was evaporated under reduced pressure, p =
4
2
0 mbar, and temperature, ϑ = 50°C. The residue (pure
-hydroxyethyl benzoate) was used to prepare stan-
dard solutions. FTIR–ATR spectra of ethylene glycol
a), benzoic acid (b), and 2-hydroxyethyl benzoate (c)
(
were collected on a Shimadzu IRAffinity-1 spectropho-
tometer (Fig. 2).
Figure 3 Calibration curve for 2-hydroxyethyl benzoate in
methane-sulfonic acid.
For comparing the spectra of ethylene glycol, diben-
zoate was gathered from the Spectral Database for Or-
ganic Compounds, which was compiled by the Na-
tional Institute of Advanced Industrial Science and
Technology, Japan. The main difference among mo-
noester and diester spectra is in the lack of a broad
peak in the diester spectra that corresponds to O–H
group stretching within the range of wave numbers
baseline of the corresponding peak within the wave
−
1
number range, ν = 1440–1470 cm
.
Kinetic Model
For the bimolecular-type second-order reaction,
−1
3
200–3550 cm
.
For the preparation of the calibration curve (Fig. 3),
-hydroxyethyl benzoate was diluted in methane-
A + B ↔ R + S
2
sulfonic acid to obtain solutions with concentrations,
w = 0, 2, 4, 6, 8, 10, 12, and 14%.
with the restrictions that cA0 = cB0 and cR0 = cS0 = 0
and the equilibrium constant must be
The real-time concentration profiles of 2-
hydroxyethyl benzoate during the reaction were ob-
tained by calculating the areas (S) of two point
k1 cRcS
K =
=
(1)
k2 cAcB
International Journal of Chemical Kinetics DOI 10.1002/kin.20938

KINETIC MODELING OF ETHYLENE GLYCOL MONOESTERIFICATION
661
Figure 4 Waterfall plot of FTIR spectra during the monoesterification of ethylene glycol at a temperature, ϑ = 70ºC.
the integrated rate equation is as follows:
reactor. As an example, the FTIR spectra collected dur-
ing one selected experiment is presented in Fig. 4.
It can be clearly seen that the intensity of the
peak (absorbance, A) within the wave number range,
ν = 1440–1470 cm , increases over the reaction
time. After a certain period of time, depending on
the experimental conditions, equilibrium steady state is
formed.
As already mentioned, we wanted to maximize the
production of 2-hydroxyethyl benzoate and that was
achieved by conducting the reaction under equimolar
conditions and with the selection of a specific catalyst
Al2O3/MeSO3H. For this catalyst, the high 94% yield
toward monoester was already reported by Sharghi and
Sarvari [2]. Thus, all the assumptions regarding the
experimental procedure were confirmed by gathering
the FTIR spectra of our end product.
ꢀ
ꢁ
XAe − (2 XAe − 1) XA
XAe − XA
1
ln
= 2 k1
− 1 cA t
0
XAe
−
1
(
2)
where cA0 is the initial concentration of ethylene gly-
col or benzoic acid, k1 the reaction rate constant for
the forward reaction, XA conversion, XAe equilibrium
conversion, and t time.
The adequacy of this kinetics can be tested using
the plot of Eq. (1) and fitting this equation within the
experimental data.
The average absolute deviation (AAD) between
experimental and calculated values was obtained as
follows:
After analytical integration of the rate expres-
sion for second-order reversible reaction, the perfor-
mance equation was developed. By plotting the term
ln(XAe – (2XAe – 1)XA)/(XAe – XA) versus time, it
was possible to calculate the reaction rate constant,
k1. Another constant, k2, for the reverse reaction was
obtained from the equilibrium constant equation (2).
Experiments at five different temperatures were stud-
ied. Each single experiment was conducted at least
twice.
N ꢀ
ꢁ
ꢂ
1
00
XA cal i − XAexp i
XAexp i
AAD (%) =
(3)
N
i=0
where, XA cal i is calculated conversion, XA exp i is ex-
perimental conversion, and N is the number of experi-
mental points.
RESULTS AND DISCUSSION
From the Arrhenius plot (Fig. 5), we determined the
activation energy and the preexponential factor for the
forward reaction and for the reverse reaction, respec-
tively. The obtained values of main kinetic parameters
are listed in Table I.
The FTIR-based analysis system was used for the real-
time monitoring of 2-hydroxyethyl benzoate concen-
tration profiles within an automated laboratory batch
International Journal of Chemical Kinetics DOI 10.1002/kin.20938

ˇ
ˇ
PECAR AND GORSEK
6
62
in pharmaceutical industry. Using a mixture of
Al2O3/MeSO3H as a catalyst effectively prevents the
concurrent formation of diesters, forcing into a te-
dious separation procedure. In spite of considerable
amount of published work on monoesterifications, ki-
netics of the reaction involved has not been studied.
Since the reaction rate also depends on the type of cat-
alyst, the complete study was performed for monoes-
terification of ethylene glycol with benzoic acid using
Al2O3/MeSO3H as a highly selective ability reagent.
Consequently, the comparison of obtained kinetic pa-
rameters was impossible because for the selected reac-
tion and used catalyst; no literature data were found.
Figure 5 Experimental and Arrhenius model–based values
of the logarithm of reaction rate constants, ln k1 and ln k2, in
relation to 1/T.
CONCLUSION
The literature contains numerous reports on kinetic
models of consecutive chemical reactions. However,
much less attention has been paid to the design for the
equally prevalent class of reversible reactions.
Table I Activation Energies and Preexponential
Factors for Forward and Reverse Reactions
Forward
Reaction
Reverse
Reaction
Variable
Ea (kJ mol 1)
The forward and reverse reaction rate constants,
activation energies, and preexponential factors of the
reversible monoesterification of ethylene glycol with
benzoic acid catalyzed by Al2O3/MeSO3H were esti-
mated from the experimental data. It has been indicated
that the reaction can be interpreted using the second-
order reversible reaction kinetic model. By providing
equimolar conditions and by the use of a specific cata-
lyst, we effectively produced a monoester. As the syn-
thesis of glycol monoesters of diol has received consid-
erable interest in view of their widespread applications,
prevention of concurrent forming of diesters is notable.
The results of this study could be applied to the con-
trol of investigated chemical reactions at the molec-
ular level, which strongly depends on kinetics and
mechanism.
−
100.4 ± 1.8
118.3 ± 1.8
14
−
1
−1
12
k0 (L mol min ) (3.6 ± 3.1) × 10 (1.1 ± 1.0) × 10
The comparison of experimental and calculated val-
ues of the conversion for one selected experiment is
presented in Fig. 6.
As it can be seen from Fig. 6, the comparison be-
tween experimental and calculated values of the con-
version is satisfactory. The average absolute deviation
for the 1844 experimental data was 4.7%.
The selected reaction was chosen for the kinetic
study because the synthesis of glycol monoesters
of diols has a widespread applications particularly
NOMENCLATURE
A
Absorbance
cA0 Initial concentration of ethylene glycol or ben-
zoic acid, mol L−1
Ea
fm
k
Activation energy, kJ mol−1
Rotational frequency of the stirrer, min−1
Reaction rate constant, L mol−1 min−1
Preexponential factor in Arrhenius equation, L
mol−1 min−1
k0
m
N
S
t
Mass, g
Number of experimental points
Surface area
Time, s
Figure 6 Experimental and calculated values of conversion
in relation to time at a temperature, ϑ = 65ºC.
T
Temperature, K
International Journal of Chemical Kinetics DOI 10.1002/kin.20938

KINETIC MODELING OF ETHYLENE GLYCOL MONOESTERIFICATION
663
v
V
w
XA
Wave number, cm−1
Volume of reaction mixture, L
Mass concentration, %
Conversion
5. Vahteristo, K.; Laari, A.; Haario, H.; Solonen, A. Chem
Eng Sci 2008, 63, 587–598.
6
7
8
9
. Harmer, M. A.; Sun, Q. Appl Catal Gen 2001, 221,
5–62.
. Shanmugam, S.; Viswanathan, B.; Varadarajan, T. K. J
Mol Catal Chem 2004, 223, 143–147.
. Beers, A. E. W.; Spruijt, R. A.; Nijhuis, T. A.; Kapteijn,
F.; Moulijn, J. A. Catal Today 2001, 66, 175–181.
. Nijhuis, T. A.; Beers, A. E. W.; Kapteijn, F.; Moulijn, J.
A. Chem Eng Sci 2002, 57, 1627–1632.
4
XAe Equilibrium conversion
GREEK SYMBOL
ϑ Temperature, °C
BIBLIOGRAPHY
1
0. Duque-Bernal, M.; Quintero-Arias, J. D.; Osorio-Viana,
W.; Dobrosz-G o´ mez, I.; Fontalvo, J.; G o´ mez-Garc ´ı a, M.
´
A. Int J Chem Kinet 2013, 45, 10–18.
1
1. Molinero, L.; Ladero, M.; Tamayo, J. J.; Garc ´ı a-Ochoa,
F. Chem Eng J 2014, 247, 174–182.
1
2
3
4
. Chand, S.; Adlercreutz, P.; Mattiasson, B. Enzyme Mi-
crob Technol 1997, 20, 102–106.
. Sharghi, H.; Sarvari, M. H. Tetrahedron 2003, 59, 3627–
633.
. Altıokka, M. R.; C¸ ıtak, A. Appl Catal Gen 2003, 239,
41–148.
. Gorzawski, H.; Hoelderich, W. F. Appl Catal Gen 1999,
79, 131–137.
12. Mekala, M.; Goli, V. R. Chin J Chem Eng 2015, 23,
100–105.
13. Ali, S. H. Int J Chem Kinet 2009, 41, 432–448.
14. Buluklu, A. D.; Sert, E.; Karaku s¸ , S.; Atalay, F. S. Int J
Chem Kinet 2014, 46, 197–205.
3
1
15. Yadav, V. P.; Maity, S. K.; Biswas, P.; Singh, R. K. Chem
Biochem Eng Q 2011, 25, 359–366.
1
International Journal of Chemical Kinetics DOI 10.1002/kin.20938