Kinetic Study of

Transesterification of
Methyl Acetate with
n-Butanol Catalyzed
by NKC-9
BAOYUN XU, WEIJIANG ZHANG, XUEMEI ZHANG, CUIFANG ZHOU
School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, People’s Republic of China
Received 18 December 2007; revised 10 June 2008, 15 July 2008, 20 July 2008; accepted 20 July 2008
DOI 10.1002/kin.20378
Published online in Wiley InterScience (www.interscience.wiley.com).

ABSTRACT: The transesterification of methyl acetate and n-butanol catalyzed by cation-

exchange resin, NKC-9, was studied in this work to obtain the reaction kinetics. The exper-
iments were carried out in a stirred batch reactor at different temperatures (328.15, 333.15,
338.15, 343.15, 345.15 K) under atmospheric pressure. The effects of temperature, molar ratio
of reactants, and catalyst loading on the reaction rate were researched under the condition of
eliminating the effect of diffusion. The experimental data were correlated with a kinetic model
based on the pseudo-homogeneous catalysis. The kinetic equation describing the reaction cat-
alyzed by NKC-9 was developed.  C 2008 Wiley Periodicals, Inc. Int J Chem Kinet 41: 101–106,

2009

INTRODUCTION The esterification of acids and alcohols is one of

the most important applications of resin catalysis. Like
Methyl acetate is a by-product during the production of esterification, transesterification can be catalyzed by
PVA (poly(vinyl) alcohol). From 1 ton of PVA, about ion-exchange resin. The advantage of a heterogeneous
1.68 ton of methyl acetate is produced. The industrial catalyst is well known, and the use of solid ion-
application of methyl acetate is low. But as the prod- exchange resin as the catalyst has the following inher-
ucts of transestrification of methyl acetate with n-butyl, ent advantages: (i) The catalyst can be easily separated
methanol is a raw material for PVA production; n-butyl from the reaction products by decantation or filtra-
acetate is an important solvent in plastics, resins, gums, tion; (ii) continuous operation in column is possible;
and coatings, etc.; what is more, it can be used as an (iii) the side reaction can be eliminated and the product
extracting agent or intermediate in organic synthesis. purity is high, etc. [1]. The strong acid cation-exchange
Thus the most attractive way will be to convert methyl resin has been used commercially as solid acid cata-
acetate into methanol and n-butyl acetate. lyst in many areas. Harmer and Sun [2] summarized
the recent development on the applications in alkyla-
tion, transalkylation, isomerization, esterification, etc.
Correspondence to: Xuemei Zhang; e-mail: zhang-xby@163.
com. Several papers show that the cation-exchange resin has
c 2008 Wiley Periodicals, Inc. an excellent catalysis on the reaction.
102 XU ET AL.

Several papers have reported on this transesterifi- Apparatus

cation, catalyzed by a heterogeneous catalyst. Bott re-
The experiments were performed in a 500-mL
ported the method that is used to produce acetic acid
round-bottom glass reactor dipped in the constant-
esters by alkali-transesterification of acetic acid es-
temperature water bath. The reactor was equipped
ter with an alcohol. The reaction was carried out in
with the temperature indicator (Pt-100) and a speed-
the middle section of a distillation column. The alco-
monitoring facility. The reflux condenser was used to
hol or the mixture of alcohol and ester is drawn from
avoid any possible loss of volatile components.
the top of the column. The higher boiling ester was
taken from the lower zone of the column [3]. Jiménez
et al. reported the production of n-butyl acetate and Procedure
methanol via reactive and extractive distillation. In this
The methyl acetate, n-butanol, and catalyst were
study, Amberlyst 15 was used as a catalyst. The kinet-
charged into the separate vessels dipped in the wa-
ics, mass-transfer, and process modeling were studied
ter bath. Once the desired temperature was attained,
[4]. Steinigeweg and Gmehling reported the transes-
methyl acetate and catalyst were charged into the
terification process of methyl acetate and n-butanol
n-butanol, it was considered as the zero reaction time.
by combination of reactive distillation and pervapora-
One-milliliter samples, which were considered negli-
tion. The result shows that it is favorable since conver-
gible to the total volume of reactants, were withdrawn
sions close to 100% can be obtained with a reasonable
at specified time intervals.
size of the reactive section [5]. Bozek-Winkler and
Samples were analyzed by SP-2100 gas chromatog-
Gmehling studied the kinetic behavior of the reaction
raphy with a thermal conductivity detector (H2 as a
of methyl acetate and butanol, leading to butyl acetate
carrier gas, a stainless-steel column packed with PEG
and methanol catalyzed by Amberlyst 15. Two dif-
20 M 3 m × 2 mm, column temperature 140◦ C, detector
ferent kinetic models have been built to describe the
temperature 160◦ C, injector temperature 200◦ C; reten-
reaction kinetics that can be applied to the designing
tion time: methanol 0.98 min, methyl acetate 1.4 min,
of the reactive distillation process or the membrane
n-butanol 3.6 min, and n-butyl acetate 5.4 min).
reactor [6].
Chromatography data were collected at a workstation
In this paper, the chemical equilibrium and the re-
N2000 (Zhejiang University Zhida Information En-
action kinetics of transestrification of methyl acetate
gineering Co. Ltd.) and processed with the modified
with n-butanol producing n-butyl acetate and methanol
area normalization method to determine the product
were studied. The reaction can be presented as
composition.

CH3 COOCH3 + CH3 (CH2 )3 OH

 CH3 COO(CH2 )3 CH3 + CH3 OH (1) RESULTS AND DISCUSSION

Calculation of Activities
A strong acid cationic exchange resin, NKC-9, was
used as a heterogeneous catalyst. The forward and Nonideality of the liquid phase was corrected by re-
reverse reactions were investigated, and the pseudo- placing components’ concentration with activities. The
homogeneous model was studied in this work. components’ activity coefficients were calculated by
the UNIFAC group contribution methods. The split-
ting of the groups is shown in Table I. The volume
and area parameters of the groups and their interaction
EXPERIMENTAL
parameters are quoted from Peisheng Ma [7].
Materials
Mass-Transfer Resistance
n-Butanol (99.9% w/w) and methyl acetate (99.9%
w/w) were obtained from Ke Wei Co. Ltd., Tianjin, To evaluate the effect of external mass-transfer resis-
China. The catalyst, i.e., the cationic exchange resin, tance on the conversion of methyl acetate, the reaction
NKC-9 in the form of H+ , was purchased from Chem- was carried out at three different stirring speeds, mean-
ical Plant of Nankai University, Tianjin, China. Its ap- while the rest of the reaction conditions were similar.
pearance is like a camel bead, the particle size was As we can see from Fig. 1, the speed of agitation had
0.4–0.7 mm, and the volume exchange capacity was no effect on the reaction rate at third and fourth gears,
4.7 mmol/g. The catalyst was washed with methanol which ensured the absence of external mass-transfer
to remove impurities and dried at 343.15 K. resistance. All further experiments were performed at

International Journal of Chemical Kinetics DOI 10.1002/kin

 TRANSESTERIFICATION OF METHYL ACETATE WITH n-BUTANOL CATALYZED BY NKC-9 103

Table I UNIFAC Group Identification of the Components

Group Identification
Volume Parameter, Area Parameter,
(i)
Molecular Group Name Main Secondary vj Rj Qj

Methyl acetate CH3 COO 11 22 1 1.9031 1.728

CH3 1 1 1 0.9011 0.848
CH3 1 1 1 0.9011 0.848
n-Butanol CH2 1 2 3 0.6744 0.54
OH 5 15 1 1.000 1.200
CH3 COO 11 22 1 1.9031 1.728
n-Butyl acetate CH2 1 2 3 0.6744 0.54
CH3 1 1 1 0.9011 0.848
Methanol CH3 OH 6 16 1 1.4311 1.432

the third gear to ensure that the reaction rate was not
restricted by external diffusion.
To evaluate the effect of internal diffusion on the
conversion of methyl acetate, the commercial NKC-9
resin was screened into four different sizes. Identical
operations were carried out with each of the fractions
obtained. The experimental results in Fig. 2 show that

Figure 3 Effect of temperature on the conversion of

MeOAc (BuOH/MeOAc = 1, catalyst mass concentra-
tion = 87.06 g/L).

the internal mass transfer was to be considered negligi-

ble in the particle diameter range used in this work. Be-
cause the resin was composed of small microspheres
with large internal macropores, mass transfer can be
Figure 1 Effect of the stirring rate on the conversion excluded for this reaction. Unsifted resin was used for
of MeOAc (BuOH/MeOAc = 1, reaction time = 200 min, all further experiments.
343.15 K, catalyst mass concentration = 87.06 g/L).

Effect of the Reaction Temperature

To investigate the effect of the reaction temperature,
operations were carried out at 328.15, 333.15, 338.15,
343.15, and 345.15 K. The initial molar ratio of
n-butanol to methyl acetate was 1, and the catalyst
mass fraction was 20% (by mass of methyl acetate).
The experimental results in Fig. 3 show that the
conversion of methyl acetate increased rapidly with
the increase of the temperature.

Effect of Catalyst Loading

Figure 2 Effect of the internal diffusion on the conversion The effect of catalyst loading on the conversion of
of MeOAc (BuOH/MeOAc = 1, reaction time = 200 min, MeOAc is depicted in Fig. 4. With the catalyst loading
343.15 K, catalyst mass concentration = 87.06 g/L). increasing, the reaction rate increased and thus the time

International Journal of Chemical Kinetics DOI 10.1002/kin

104 XU ET AL.

Figure 4 Effect of catalyst loading on the conversion of Figure 6 Effect of the molar ratio of reactants on the
MeOAc (BuOH/MeOAc = 1,343.15 K). conversion of MeOAc (343.15 K, catalyst mass concentra-
tion = 87.06 g/L).

to reach the reaction equilibrium reduced. The increase

of the reaction rate was due to the increase in the total
number of acid sites available for the reaction with the experiments. Figure 6 shows that the equilibrium
increase of catalyst loading. Furthermore, the catalyst conversion of methyl acetate increased with the in-
loading had no effect on the equilibrium constant. crease in the molar ratio of n-butanol to methyl
The effect of catalyst loading on the reaction rate acetate.
constant is shown in Fig. 5. The forward reaction rate
constant increased with the increase of the catalyst
loading. The relationship between catalyst loading and Kinetic Model
rate constant could be expressed as
Kinetic Equation. Most resin-catalyzed reactions
k0 = k1 × mcat could be expressed as either quasi-homogeneous
(2)
= 1.95 × 10−4 × mcat or two adsorption-based models, i.e., Langmuir–
Hinshelwood (L–H) and Elay–Rideal (E–R) model
[6,8–10]. The idealized homogeneous state required
Effect of the Molar Ratio of n-Butanol to complete swelling of the resin and total dissociation
Methyl Acetate of the polymer-bound-SO3 H group. However, the het-
The initial molar ratio of n-butanol to methyl ac- erogeneous state was characterized by a direct inter-
etate was varied between 1:1 and 1.5:1 during the action of the substance with the polymer-bound-SO3 H
group. A pseudo-homogeneous model could be ap-
plied for the system where the mass-transfer resistance
was absent. One of the reactants or solvents was high
polar; the rate of the reaction could be expressed as
a simple pseudo-homogeneous model. The pseudo-
homogeneous model was based on the Helfferich ap-
proach, which treated catalysis confined within the in-
ternal catalyst mass, wherein the reactants, products,
and solvents were in distribution equilibrium with the
bulk solution. The swelling of the resin particle in the
presence of polar solvents led to an easy accessibility
of the acid groups for the reaction and free mobility of
all components [11].
The transesterification reaction was known to be a
reversible second-order reactions. Therefore, the reac-
Figure 5 Effect of catalyst loading on the forward reaction tion rate could be expressed as a pseudo-homogeneous
rate constant (BuOH/MeOAc = 1,343.15 K). model:

International Journal of Chemical Kinetics DOI 10.1002/kin

 TRANSESTERIFICATION OF METHYL ACETATE WITH n-BUTANOL CATALYZED BY NKC-9 105

1 1 1 dni 1 dCBuOAc
r= =
mcat νi V dt mcat dt
= k1 CBuOH CMeOAc − k2 CMeOH CBuOAc (3)
= k1 (aMeOAc aBuOH − aMeOH aBuOAc /K)

To obtain a general model to produce the reaction ki-

netics with a set of parameters, the data obtained from
the experiment in this work were correlated. Parame-
ters for the models were estimated by minimizing the
sum of the residual squares (SRS) between the exper-
imental and calculated mole concentration of n-butyl
acetate through the simplex Nelder method.
Figure 8 Arrhenius plot for reactions (BuOH/MeOAc = 1,
 catalyst mass concentration = 87.06 g/L).
SRS = (rexp − rcalc ) 2
(4)
all..samples

Equilibrium Constant. The equilibrium constant of indicated by Eq. (6)

the reaction in this work could be obtained from the
equilibrium composition of the mixture 276
ln K = 0.365 − (6)
T
aBuOAc,e aMeOH,e
K= (5)
aMeOAc,e aBuOH,e The reaction enthalpy could be calculated from the plot
of lnK vs. 1/T as indicated by the Van’t Hoff equation.
Most of the kinetic experiments lasted long enough The standard enthalpy of this reaction obtained from
until the chemical equilibrium was reached. The the experiment was  Hro = 2.29 kJ/mol, which agrees
equilibrium constant K could be calculated from the with the anticipation for an endothermic reaction.
equilibrium concentration, which was measured in
each operation. The reaction equilibrium was reached
Activation Energy. The Runge–Kutta method was
in 7–23 h based on the reaction temperature and
used to integrate the rate equation with the kinetic pa-
catalyst loading. Figure 7 shows that the equilibrium
rameter determined by experiments. As could be seen
constant slightly increased with an increase in the
from Figs. 3–6, a good agreement between calculated
reaction temperature.
curve and the experimental points was observed.
The dependence of the equilibrium constant on the
The value of k1 and k2 is a function of the tempera-
temperature was obtained by the traditional method as
ture and could be expressed as the Arrhenius equation:

k1 = A1 e(−Ea,1 /RT ) (7)

k2 = A2 e(−Ea,2 /RT ) (8)

The kinetic parameters could be obtained from the lin-

ear regression of the Arrhenius plot (Fig. 8). The values
of A1 , Ea,1 , A2 , Ea,2 are given in Table II.

Table II Kinetic Parameter for the Pseudo-

homogeneous Model
Ai (L · mol−1 · Ea,i
Reaction i min−1 · g-cat−1 ) (J · mol−1 )
Figure 7 Effect of the reaction temperature on the equilib- Forward reaction 1 3.45 × 104 54,315
rium constant (BuOH/MeOAc = 1, catalyst mass concentra-
Reverse reaction −1 2.39 × 104 52,021
tion 87.06 g/L).

International Journal of Chemical Kinetics DOI 10.1002/kin

106 XU ET AL.

CONCLUSION R general gas constant, 8.314 J · mol−1 · K−1

r reaction rate, mol · L−1 · min−1 · g-cat−1
The kinetic behavior for the synthesis of n-butyl acetate rcalc calculated value
catalyzed by the cationic-exchange resin, NKC-9, from rexp experimental value
methyl acetate and n-butanol was carried out in the SRS residual squares
experiments described in this paper. The reaction was T absolute temperature, K
performed in the stirred batch reactor. The effects of t time, min
temperature, reactants’ molar ratio and catalyst load- V volume of reaction mixture, L
ing were studied. The kinetic data were correlated with νi stoichiometric coefficient of component i
the pseudo-homogeneous kinetic model, and as a re-
sult the experiment agreed with the model very well. Subscripts
Then the kinetic parameters were obtained from the
linear regression of the Arrhenius plot and the kinetic BuOAc n-butyl acetate
equation describing the reaction catalyzed by cationic BuOH n-butanol
exchange resin was developed. MeOAc methyl acetate
MeOH methanol

NOMENCLATURE
BIBLIOGRAPHY
A1 frequency factor for forward reaction,
L · mol−1 · min−1 · g-cat−1 1. Roy, R.; Bhatia, S. J. Chem Technol Biotechnol 1987,
A2 frequency factor for reverse reaction, 37, 1–10.
2. Harmer, M. A.; Sun, Q. Appl Catal A 2000, 221, 45–62.
L · mol−1 · min−1 · g-cat−1
3. Bott, K.; Kaibel, G.; Hoffmann, H.; Irnich, R.; Schaefer,
ai activities of component i, mol · L−1 E. US Patent 4370491, 1983.
ai,e equilibrium activities of component i, 4. Jiménez, L.; Garvin, A.; Costa-López, J. Ind Eng Chem
mol · L−1 Res 2002, 41, 6663–6669.
Ci concentration of component i, mol · L−1 5. Steinigeweg, S.; Gmehling, J. Chem Eng Process 2004,
Ea ,1 activation energy for forward reaction, 43, 447–456.
J · mol−1 6. Boz̀ek-Winkler, E.; Gmehling, J. Ind Eng Chem Res
Ea ,2 activation energy for reverse reaction, 2006, 45, 6648–6654.
J · mol−1 7. Peisheng, M. Chemical Engineering Thermodynamics;
K equilibrium constant Chemical Industry Press: Beijing, People’s Republic of
k0 used in Eq. (2); L · mol−1 · min−1 China, 2005.
8. Liu, W. T.; Tan, C. S. Ind Eng Chem Res 2001, 40,
k1 reaction rate constant of forward reaction,
3281–3286.
L · mol−1 · min−1 · g-cat−1 9. Sanz, M. T.; Murga, R.; Beltran, S.; Cabezas, J. L. Ind
k2 reaction rate constant of reverse reaction, Eng Chem Res 2004, 43, 2049–2053.
L · mol−1 · min−1 · g-cat−1 10. Silva, Viviana V. T. M.; Rodrigues, A. E. Chem Eng Sci
mcat catalyst mass in reaction mixture, g-cat 2006, 61, 316–331.
ni moles of component i in reaction mixture, 11. Helfferich, F. Ion Exchange; McGraw-Hill: New York,
mol 1962; pp 519–550.

International Journal of Chemical Kinetics DOI 10.1002/kin