Application of a sodalite membrane reactor in esterification - Coupling reaction and separation

Article abstract of DOI:10.1016/j.cattod.2010.02.042

A tubular hydroxy sodalite (SOD) membrane was successfully applied in the esterification of acetic acid with ethanol and acetic acid with 1-butanol, to selectively remove water by pervaporation and to overcome the thermodynamic equilibrium limitation of the esterification. The reactions were carried out using equimolar solutions of acetic acid with the appropriate organic alcohol at 363 K using Amberlyst 15 as acid catalyst. The hybrid process drove the esterification reactions almost to completion. The special feature of the SOD-membrane is that it showed absolute selectivity towards water and retained its stability under the reaction conditions. The membrane exhibited a stable water pervaporation performance at pH values above 2.9 for acetic acid-water mixtures.

Full text of DOI:10.1016/j.cattod.2010.02.042

Catalysis Today 156 (2010) 132–139
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Application of a sodalite membrane reactor in esterification—Coupling reaction
and separation
∗
Sheida Khajavi, Jacobus C. Jansen, Freek Kapteijn
Ceramic Membrane Center, “The Pore” – Catalysis Engineering, DelftChemTech, Delft University of Technology, Julianalaan 136, 2628 BL Delft, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 15 March 2010
A tubular hydroxy sodalite (SOD) membrane was successfully applied in the esterification of acetic acid
with ethanol and acetic acid with 1-butanol, to selectively remove water by pervaporation and to over-
come the thermodynamic equilibrium limitation of the esterification. The reactions were carried out
using equimolar solutions of acetic acid with the appropriate organic alcohol at 363 K using Amberlyst
Keywords:
Membrane reactors
Pervaporation
Esterification
1
5 as acid catalyst. The hybrid process drove the esterification reactions almost to completion. The special
feature of the SOD-membrane is that it showed absolute selectivity towards water and retained its sta-
bility under the reaction conditions. The membrane exhibited a stable water pervaporation performance
at pH values above 2.9 for acetic acid–water mixtures.
Hydroxy sodalite
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
as an integrated unit where both reaction and separation take place
simultaneously [35]. Both configurations have been widely studied
in the literature [34]. Next to esterification, other types of equi-
librium limited reactions such as etherification and condensation
reactions have been investigated [36].
Hybrid processes combining reaction and separation are becom-
ing increasingly popular in chemical industry [1–5]. In these, the
use of membranes and membrane technology for the selective
separation of products from equilibrium limited media to shift
equilibria towards higher reaction yields is being progressively
recognized [6–16]. Esterification is a classical case of an equilib-
rium limited reaction [17]. Traditionally the equilibrium limitation
is overcome by either using an excess amount of the alcohol or
by separating the by-produced water from the reaction through
reactive distillation or reactive stripping [18,19]. While the use of
excess alcohol may increase the operation costs on the downstream
reagent recovery and result in unwanted ether formation [20], reac-
tive distillation is only useful when the products and reactants are
not close boiling liquids [19]. In this respect, the use of membrane
reactors is attractive as the process efficiency is not limited by the
thermodynamic equilibrium conversion while the process costs
can be reduced due to the smaller amounts of reactants required
and the higher conversions obtained [21–25].
In integrating reaction and membrane separation, typically per-
vaporation is coupled with the chemical reaction [7,26–33]. In
contrast to distillation which is based on the difference in volatili-
ties of the substances, pervaporation solely relies on the solubility
and transport rate of each compound. Conventionally membrane
reactors are configured as either a batch reactor, where the reaction
takes place, followed with an external pervaporation unit built in
the recycle to continuously remove water from the reactor [34], or
Here we report on the esterification in a membrane reactor
of acetic acid with ethanol and with 1-butanol to produce ethyl
acetate and butyl acetate, respectively, using Amberlyst 15 as cat-
alyst. A tubular hydroxy sodalite membrane has been used as a
highly water selective membrane for water permeation. Hydroxy
sodalite (H-SOD) is a zeolite-like material, consisting of sodalite
cages only [37]. The material does not contain distinct pores or
channels, but consists of an array of cages connected to each other
through 4- and 6-rings. The 4-rings are too small to allow per-
meation of any molecule, but the diameter of the 6-rings is large
enough (0.27 nm) to allow only very small molecules such as water
(kinetic diameter: 0.265 nm) to pass through [37–39]. Owing to
the unique window diameter of the material, absolute separation
of water from various organic alcohol streams has been achieved
[38]. In using membranes under esterification conditions, it is vital
that the membrane preserves its structural integrity. Therefore, a
hydroxy sodalite membrane was applied in dehydration of con-
centrated acetic acid solutions to study its stability under acidic
conditions [40].
2. Experimental
2.1. Membrane preparation
A hydroxy sodalite tubular membrane was made according to
a previously reported procedure [41] by means of hydrothermal
synthesis on the inner surface of an ␣-alumina tubular support
∗ Corresponding author. Tel.: +31 15 2783516; fax: +31 15 2785006.
E-mail address: f.kapteijn@tudelft.nl (F. Kapteijn).
0
920-5861/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2010.02.042

S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
133
Table 1
The tubular membrane was initially sealed in a stainless steel
membrane module, with the membrane on the inner side of the
tube facing the feed (A = 15.0 cm effective membrane surface area
␣
-Alumina support properties.
2
Provider
Inocermic
and a selective layer thickness of about 1 ␮m [38]); the module was
placed in the oven of the pervaporation set-up and was controlled
to maintain the same temperature as the reaction medium.
The feed pressure was regulated using a backpressure controller
and kept at 0.5 MPa to ensure a liquid phase feed at the exercised
temperatures. A two-stage vacuum pump was used to evacuate
the permeate side. The pressure at the permeate side was kept at a
constant value of 300 Pa and was monitored using a digital vacuum
gauge installed between the vacuum pump and the cold traps. The
temperature of the feed inside the membrane cell and the permeate
temperature were measured using two thermocouples. At steady
state conditions, the flux through the membrane was determined
by the rate-of-rise method described elsewhere [38].
Number of layers
3
Pore diameter top layer
Surface treatments
Outer diameter
Inner diameter
Length permeable section
200 nm
None
7 mm
5 mm
5 cm
purchased from Inocermic. The support properties are given in
Table 1. The synthesis gel was prepared by mixing silicate and
aluminate solutions with the final gel molar composition being
5
SiO :Al O :50NaO :1005H O. After the hydrothermal treatment
2 2 3 2 2
at 413 K for 3.5 h, the tube was thoroughly washed with water and
dried under dry air flow overnight. Following, single gas perme-
ation measurements using He and N2 were conducted to verify
formation of a closed membrane prior to use in the esterification
reactions. As water acts as a template during synthesis, the sodalite
cages are fully occupied with water. Therefore neither He nor N2
should be able to permeate through the cages, unless through mem-
brane defects. At room temperature and under a cross membrane
pressure of 0.5 MPa, themembranes were impermeable tonitrogen,
Samples were withdrawn from both the reactor and the perme-
ate side at regular intervals and immediately quenched for analysis
using a GC (HP 6890, column: HP-FFAP) equipped with a thermal
conductivity detector (TCD).
3. Reactor model
−
10
−1
−2
−1
and the helium permeance was as low as ∼10
mol s
m
Pa .
The model reactions which have been used to describe the
esterification reactions investigated in this study can be generally
expressed as:
This was used as a test of membrane quality before coupling it with
the reaction section.
2.2. Pervaporation membrane reactor
A + B ↔ E + W
The experimental set-up used for the reaction-pervaporation
where A is acetic acid, B is the alcohol, being either ethanol or
butanol, E is the ester and W is the water.
For sole esterification, when an excess of the alcohol is used and
if the selectivity of the desired production is 100%, the conversion
of the limiting reactant acetic acid, X, can be expressed in the ratio
of the acid to ester:
experiments is schematically depicted in Fig. 1. The esterification
was carried in a 250 ml round bottom flask fitted with a reflux con-
denser. The flask was inserted in a thermostatic water bath and its
temperature was regulated using a thermocouple. 1 mol ethanol
or 1-butanol were initially added to the reactor with Amberlyst 15
as catalyst (loading: 100 g/l solution) and heated to the reaction
temperature. An equimolar amount of acetic acid was heated sep-
arately, and after reaching the reaction temperature it was added
to the reactor. Simultaneously, the reaction mixture was started
to be continuously pumped through the pervaporation unit at a
rate of 50 ml/min. This time was noted as the starting time of the
experiment. The conventional esterification reactions were carried
out until equilibrium was reached. In membrane facilitated exper-
iments, the time at which the reaction was started parallel to the
start up of the membrane system was noted at the starting time of
the experiment.
acid
ester
1 − X
=
R1
=
(1)
(2)
X
1
X =
R1 + 1
Hence,
Table 2 gives the reaction mixture composition in the feed at
time t . When pervaporation is coupled with esterification, water
1
is removed from the reaction system, and the composition of the
reaction changes as given in Table 2. In the pervaporation-aided
Fig. 1. Schematic drawing of the pervaporation set-up (V, valve; PI, pressure indicator; TR, temperature recorder; BPC, back pressure controller; CT, cold trap).

1
34
S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
Table 2
Feed composition and reaction composition in the coupled membrane reactor.
Using calculated X and Y values from the experimentally deter-
mined concentrations, the amount of water removed (Nw) can be
determined:
Time (h)
Amount (mol)
Acetic acid
Alcohol
Ester
Water
Nw = N (X − Y)(mol)
(5)
0
Feed
Batch reactor
Coupled reactor
t = 0
t = t1
t = t1
N0
N0
0
N0X
N0X
0
N0X
N0Y
N0(1 − X)
N0(1 − X)
with N being the initial amount of acetic acid.
0
N0(1 − X)
N0(1 − X)
The liquid reaction volume changes from the feed volume V to
0
V . If the volume of the removed water is V , and assuming additive
1
2
properties, then:
esterification the ratio of ester to water is equivalent to:
V0 = V + V2
(6a)
(6b)
1
ester
water
X
=
R2 = Y
(3)
(4)
V2 = Molar volume liq water × N0(X − Y)
X
Y =
For simple first order behaviour in the reactants and products,
the rate of the reversible esterification can be written as:
R2
ꢀ
ꢁꢀ
ꢁ
ꢀ
ꢁꢀ
ꢁ
N (1 − X)
N (R − X)
k1
N X
N0Y
Hence,
0
0
1
0
rest = k1
−
(7)
V
1
V
1
Keq
V1
V1
and substitution in the molar balance for a batch reactor leads to:
ꢂ
ꢃ
dX
dt
N k
0
X · Y
Keq
1
=
×
(1 − X) · (R − X) −
(8)
1
V1
The equilibrium constant, Keq, is determined from the composi-
tion of the reaction mixture in the batch experiment without water
removal (so R = 1, Y = X, and V = V ) at equilibrium:
2
1
0
ꢄ
ꢅ
[
[
E][W]
A][B]
Keq =
(9)
eq
The forward rate constant k is determined from the early stages
1
of the reaction (t ∼ 0). Both parameters were estimated simultane-
ously by nonlinear Least Squares Analysis, applying Athena Visual
Studio (www.athenavisual.com).
In the coupled reaction system, the water removal rate at any
time t can be expressed as:
dNw
d(X − Y)
RH
=
= N0
(mol/ min)
(10)
2O
dt
dt
If all water is removed through the membrane, the flux can be
determined as:
R
H2
O
J =
(11)
A
where J is the water flux through the membrane, and A is the
effective membrane surface area. Now the calculated flux can be
compared with the measured flux through the membrane.
As hydroxy sodalite is a highly water selective membrane with
water selectivity values of above 1,000,000, the permeation of the
organic components need not to be considered.
4. Results and discussion
4.1. Heterogeneous catalytic reactions
The catalytic activity of Amberlyst 15 and the thermodynamic
equilibrium conversion were determined from the batch reactor
operation. As shown in Fig. 2, in case of ethanol esterification with
acetic acid, after 2 h of reaction at 363 K equilibrium was reached,
while 1-butanol esterification with acetic acid required 7 h at 363 K.
The equilibrium constant, Keq, and forward rate constant, k , of each
1
reaction were determined at 343 K and 363 K (Table 2).
Table 3.
4.2. Membrane activity
Fig. 2. Concentrations of acetic acid (ꢀ) and ethyl acetate (a) and butyl acetate
The esterification reactions were carried out in the absence
of the catalyst with the addition of freshly synthesized hydroxy
sodalite crystals (10 wt.%). In case of ethanol esterification after 8 h
(
b) (ꢀ) in reaction mixture as a function of reaction time without water removal
(T = 363 K, equimolar feed). The solid lines were calculated according to the kinetic
model.

S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
135
Table 3
Equilibrium constant and forward rate constant in the esterification of acetic acid
with ethanol and 1-butanol with Amberlyst 15 as catalyst.
Esterification
T = 343 K
Keq
T = 363 K
Keq
k1 (l/mol min)
k1 (l/mol min)
Acetic acid with ethanol
Acetic acid with butanol
3.16
2.73
0.003
0.037
4.85
3.51
0.045
0.103
of experiment only a 17% conversion was reached, the same as in a
blank run without any catalyst (Fig. 3). Similarly, for butanol ester-
ification, after 8 h of reaction, a 14% conversion was reached which
is the same as without catalyst. Thus, hydroxy sodalite behaves
as an inert by showing no catalytic activity in either esterification
reaction.
4.3. Pervaporation-aided esterification
In Fig. 4a and b the concentration profiles of acetic acid, water
and ethyl acetate and butyl acetate in the pervaporation-assisted
Fig. 4. Concentration of acetic acid (ꢀ), ethyl acetate (a) butyl acetate (b) (ꢀ), and
water (ꢁ) in the reaction mixture with pervaporation through a hydroxy sodalite
membrane as a function of reaction time (T = 363 K, equimolar feed). The solid lines
were calculated according to the kinetic model.
reactor configuration are presented. The water concentration pro-
file in the membrane reactor initially showed an increase until it
reached a maximum, after which it started to decline continuously
until it almost reached zero. This indicates that at the beginning of
the reaction, the rate of water formation is faster than its removal
from the reaction mixture.
In the pervaporation-aided esterification of both ethanol and
butanol with acetic acid, the concentration profiles show a steady
continuous decline in the amount of acetic acid followed simul-
taneously by the continuous increase of the ester content. The
conversion of acetic acid to ethyl acetate increased from 67% to
9
8% (Fig. 5a) and in case of butyl acetate, conversion was enhanced
from 62% to 92% (Fig. 5b). In both reactions, the water removal over-
came the equilibrium limitation and drove the reaction towards
completion. The conversion profiles in Fig. 5 indicate that the ester-
ification of ethanol initially does not proceed much different from
the batch experiment, while that of butanol shows a faster con-
version in time. Fig. 6 shows the development of water activity as
Fig. 3. Conversion of acetic acid with ethanol (a) and with 1-butanol (b) in a blank
run without catalyst (ꢀ), and in an experiment in the presence of hydroxy sodalite
crystals (᭹), T = 363 K.

1
36
S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
Fig. 5. Conversion of ethanol (a) or 1-butanol (b) as function of time with (ꢁ) and
without pervaporation (ꢀ). The solid lines were calculated according to the kinetic
model.
Fig. 6. Evolution of water activity as function of time in the esterification of (a)
ethanol with acetic acid and (b) butanol with acetic acid.
membrane type [38,41], showing a proportionality with the water
concentration. This explains the long tailing of the process towards
completion.
function of time in the mixture. Similar to the concentration pro-
file of water formation (Fig. 4), evolution of water activity follows
the same pattern whereby it increases with increasing water con-
centration and declines when membrane reaches steady state in
removing water as it is being produced.
Fig. 7 shows the water flux through the membrane measured
experimentally in comparison with the calculated flux values
obtained from conversion data assuming that all water permeated
through the membrane. For approximately the first 2 h of the exper-
iment, the flux through the membrane is significantly lower than
the flux values calculated based on conversion, while at longer time
there is still a steady water flux through the membrane. The high
Hydroxy sodalite is exemplary in demonstrating the potential
of membranes for in situ separation of products in reactions. The
unique property of hydroxy sodalite in comparison with the other
widely studied membranes in hybrid processes [26–33] is its vir-
tually 100% selectivity in removing water solely on the basis of
molecular sieving [38–40]. No traces of the other components were
detected in the permeate (water selectivity∼ 1,000,000). This in
terms of industrial applicability gives sodalite the advantage of pro-
ducing pure water at the permeate side, which eliminates the need
of further downstream recovery or cleaning with direct reflections
in the operation costs.
‘
fluxes’ in the beginning are attributed to the increased volume of
the reaction system in which water can be present in the vapor
phase, constituting an apparent water disappearance in the initial
phase of the process. The steady but low water removal through
the membrane finally drives the reaction slowly to completion. The
experimental flux values are in-line with earlier reports with this
Compared to other hydrophilic inorganic membranes [42],
sodalite also demonstrates a good stability under the ruling reac-
tion conditions. The acidic conditions lead to the loss of membrane
integrity in most hydrophilic membranes due to the leaching of
the framework aluminum by the attack of hydroxonium ions [43].

S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
137
Fig. 8. Membrane stability in pervaporative dehydration of water/acetic acid mix-
tures with feed water content of 93 mol% (ꢃ), 87 mol% (ꢁ), 83 mol% (ꢂ), 80 mol%
Fig. 7. Water ‘flux’ through the membrane measured experimentally (ꢁ), and water
removal rate calculated from the conversion data (ꢂ) for the esterification of ethanol
with acetic acid (Fig. 4a).
(
᭹); Tfeed = 473 K, Pfeed = 2.2 MPa, Pperm = 300 Pa. One hydroxy sodalite membrane was
used in the whole series.
Under mild acidic conditions, pH ≥ 2.9, sodalite does not deteri-
orate in quality. Fig. 8 illustrates a hydroxy sodalite membrane
performance in permeating water in different acidic solutions
over prolonged operation times. The flux through the membrane
remained stable through the process; no traces of acetic acid were
detected at the permeate side. Although hydroxy sodalite is a
hydrophilic material (Si/Al = 1), its window openings are too nar-
row for aluminum to be leached out. In addition, the cages are at
full saturation with water which almost fully occupies the space,
−
and thus does not ease the attack of the dissociated OH groups.
However, if the pH is decreased severely (Fig. 9), the framework
exterior is weakened by the initial breakage of Si–O–Al bridges
[
28]. Under esterification conditions, the amount of acid present
is counterbalanced by the other components throughout the reac-
tion. Nevertheless, at the initial stages of the reaction when 50 mol%
of the solution consists of acetic acid, long-term exposures under
these conditions can initiate the disintegration of the membrane.
Therefore it is crucial for the catalyst to be active enough in boosting
the rate.
Selectivity, membrane stability and flux are three dominating
factors which are crucial to membrane applications. A combina-
tion of the three is seldom offered by one specific membrane. Li
et al. have recently reported on the synthesis of long-term sta-
ble membranes under reactive conditions. While the membranes
show good flux values, they preserve their intactness under the
reaction conditions [44]. Other reports indicate different results,
while in most studies the importance of high flux values is empha-
Fig. 9. Membrane stability in pervaporative dehydration of water/acetic acid mix-
tures with feed water content of 69 mol% (×), 49 mol% (+), 15 mol% (ꢀ); Tfeed = 473 K,
Pfeed = 2.2 MPa, Pperm = 300 Pa. For each condition a new membrane was used.
Table 4
Collection of membranes published in the literature tested for H2O permeation under reactive conditions.
−
2
−1
h )
Membrane
NaA
Reaction type
Flux (kg m
0.104
Selectivity
124,000
Reference
[45]
Knoevenagel condensation between
benzaldehyde and ethyl cyanoacetate,
ethyl acetoacetate and diethyl malonate
Esterification of acetic acid and ethanol
Esterification of acetic acid and ethanol
Esterification of acetic acid and ethanol
Esterification of lactic acid and ethanol
Acetylation of benzyl alcohol
Zeolite T
MOR
Zeolite A
NaA
0.81
830
164
92
1000
640
[46]
[42]
[42]
[47]
[31]
0.075
0.150
0.12
PVA
0.54

1
38
S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
sized, other focus on membrane stability [30–33]. It is often the
process requirements which indicate a membrane’s specific prop-
erties which can be utilized. In processes, where product loss does
not govern the process costs, high flux values may be more desired
compared to a case in which downstream separation of permeate
or product loss can be detrimental to the process. Table 4 gives a
summary of some of the most studied membranes under reactive
conditions with their flux and selectivity values. Compared to these
membranes, sodalite has the remarkable capability of being purely
water selective.
[13] Y.S. Zhu, H.F. Chen, Pervaporation separation and pervaporation–esterification
coupling using crosslinked PVA composite catalytic membranes on porous
ceramic plate, J. Membr. Sci. 138 (1998) 123.
[
[
[
14] M.A. Larrayoz, L. Puigjaner, Study of n-butanol extraction through PV in aceto-
butylic fermentation, Biotechnol. Bioeng. 30 (1987) 692.
15] R.A. Kemp, A.R. Macrae, Esterification process, European Patent (EP) 506159
(
1992).
16] F. Lipnizki, R.W. Field, P.K. Ten, Pervaporationbased hybrid process: a review
of process design, application and economics, J. Membr. Sci. 153 (1999) 183.
[17] W. Riemenschneider, Esters, Ullmann’s Encyclopedia of Industrial Chemistry,
vol. A9, VCH Verlagsgesellschaft, Weinheim, 1993.
[
18] T.J. Schildhauer, F. Kapteijn, J.A. Moulijn, Reactive stripping in pilot scale mono-
lith reactors – application to esterification, Chem. Eng. Process. 44 (2005)
695.
[
19] E.E. Reid, Esterification, in: P.H. Groggins (Ed.), Unit Processes in Organic Syn-
thesis, 4th ed., McGraw-Hill, New York, 1952, pp. 596–642.
5
. Conclusions
[
20] T.J. Schildhauer, I. Hoek, F. Kapteijn, J.A. Moulijn, Zeolite BEA catalysed esterifi-
cation of hexanoic acid with 1-octanol: kinetics, side reactions and the role of
water, Appl. Catal. A: Gen. 358 (2009) 141.
Pervaporation-aided esterification of acetic acid with ethanol
and with 1-butanol was successfully carried out using equimo-
lar solutions of acid and base at 343 K and 363 K using Amberlyst
[21] M.O. David, R. Gref, T.Q. Nguyen, J. Neel, Pervaporation–esterification coupling.
Part I. Basic kinetic model, Trans. IChemE A, Chem. Eng. Res. Des. 69 (1991) 335.
[
22] L. Bagnell, K. Cavell, A.M. Hodges, A.W. Mau, A.J. Seen, The use of catalytically
active pervaporation membranes in esterification reactions to simultaneously
increase product yield, membrane permselectivity and flux, J. Membr. Sci. 85
1
5 as catalyst. A hydroxy sodalite membrane was used to
selectively remove water during reaction to enhance product
formation, overcoming the equilibrium limitation. In the reac-
tion system the membrane was capable of removing water at
a rate compatible with its formation. In both reactions, conver-
sion was increased by 30% almost reaching completeness. The
water selectivity in these experiments was above 1,000,000. The
reactor model to predict conversion and concentration profiles
of the reactions was in good agreement with the experimental
results.
Despite the hydrophilic nature of hydroxy sodalite, the mem-
brane showed a good stability under the reaction conditions. Stand
alone experiments of water pervaporation through a sodalite mem-
brane in acetic acid/water solutions with pH ≥ 2.9 proved the
membrane to be stable for long operation duration. Below this pH
the membrane disintegrated after a certain time.
(
1993) 291.
[
23] M.P. Bernal, J. Coronas, M. Menendez, J. Santamaria, Coupling of reaction and
separation at the microscopic level: esterification processes in a H-ZSM-5
membrane reactor, Chem. Eng. Sci. 57 (2002) 1557.
[24] Q.T. Nguyen, C.O. M’Bareck, M.O. David, M. Metayer, S. Alexandre, Zeolite-
coated ceramic pervaporation membranes; pervaporation–esterification
coupling and reactor evaluation, Mater. Res. Innovat. 7 (2003) 212.
[
[
[
25] T.A. Peters, J. Fontalvo, M.A.G. Vorstman, J.T.F. Keurentjes, Design directions for
composite catalytic hollow fibre membranes for condensation reactions, Trans.
IChemE A, Chem. Eng. Res. Des. 82 (2004) 220.
26] M.-O. David, R. Gref, T.Q. Nguyen, J. Neel, Pervaporation–esterification cou-
pling. II. Modeling of the influence of different operating parameters, Trans.
Inst. Chem. Eng. 69 (1991) 341.
27] H.H. Nijhuis, A. Kemperman, J.T.P. Derksen, F.P. Cuperus, Pervaporation con-
trolled bio-catalytic esterification reactions, in: R. Bakish (Ed.), Proceedings
of Sixth International Conference on Pervaporation Processes in the Chemical
Industry, Bakish Materials Corporation, Englewood, NJ, 1992, p. 368.
28] S.J. Kwon, K.M. Song, W.H. Hong, J.S. Rhee, Removal of water product from
lipase-catalyzed esterification in organic solvent by pervaporation, Biotechnol.
Bioeng. 46 (1995) 393.
[
[
29] X. Ni, Z. Xu, Y. Shi, Y. Hu, Modified aromatic polyimide membrane preparation
and pervaporation results for esterification system, Water Treat. 10 (1995) 115.
30] R. Ronnback, T. Salmi, A. Vuori, H. Haario, J. Lehtonen, A. Sundqvist, E. Tirro-
nen, Development of a kinetic model for the esterification of acetic acid with
methanol in the presence of a homogeneous acid catalyst, Chem. Eng. Sci. 52
Acknowledgement
[
The TU Delft Research Centre for Sustainable Industrial Pro-
cesses (DRC-SIP) is gratefully acknowledged for the financial
support of this work.
(
1997) 3369.
[
31] L. Domingues, F. Recasens, M. Larrayoz, Studies of a pervaporation reactor:
kinetics and equilibrium shift in benzyl alcohol acetylation, Chem. Eng. Sci. 54
(
1999) 1461.
References
[32] R. Krupiczka, Z. Koszorz, Activity-based model of the hybrid process of an ester-
ification reaction coupled with pervaporation, Sep. Purif. Technol. 16 (1999)
55.
[
1] X. Feng, K.Y.M. Huang, Studies of a membrane reactor: esterification facilitated
by pervaporation, Chem. Eng. Sci. 51 (1996) 4673.
[
33] Q.L. Liu, H.F. Chen, Modeling of esterification of acetic acid with n-butanol in
the presence of Zr(SO4)2·4H2O coupled pervaporation, J. Membr. Sci. 196 (2002)
[
[
2] M. Moli, F. Reeasens, M.A. Larrayoz, Ingenierta Qutmica 332 (1997) 175.
3] M.R. Waldburger, F. Widmer, Membrane reactors in chemical production pro-
cesses and the application to the pervaporation-assisted esterification, Chem.
Eng. Technol. 19 (1996) 117–126.
4] B.L. Yang, S. Goto, Pervaporation with reactive distillation for the production of
ethyl tert-butyl ether, Sep. Sci. Technol. 32 (1997) 971.
5] C. Staudt-Bickel, R. Üdiger, N. Lichtenthaler, Integration of pervaporation for the
removal of water in the production process of methylisobutylketone, J. Membr.
Sci. 111 (1996) 135.
6] H. Kita, K. Tanaka, K. Okamoto, M. Yamamoto, The esterification of oleic
acid with ethanol accompanied by membrane separation, Chem. Lett. (1987)
171.
[
[
34] S.Y. Lim, B. Park, F. Hung, M. Sahimi, T.T. Tsotsis, Design issues of pervaporation
membrane reactors for esterification, Chem. Eng. Sci. 57 (2002) 4933.
35] Y. Zhu, R.G. Minet, T.T. Tsotsis, A continuous pervaporation membrane reactor
for the study of esterification reactions using a composite polymeric/ceramic
membrane, Chem. Eng. Sci. 51 (1996) 4103.
36] C. Herion, L. Spiske, W. Hefner, Dehydration in the synthesis of dimethyl urea by
pervaporation, in: R. Bakish (Ed.), Proceedings of the Fifth International Confer-
ence on Pervaporation Process in the Chemical Industry, Heidelberg, Germany,
Bakish Materials Corporation, Englewood, NJ, 1992, p. 349.
[
[
[
[
[
[
[
37] D.W. Breck, Zeolite Molecular Sieves, John Wiley, New York, 1974.
38] S. Khajavi, F. Kapteijn, J.C. Jansen, Application of hydroxy sodlaite films as novel
water selective membranes, J. Membr. Sci. 326 (2008) 153.
2
053.
7] M.O. David, Q.T. Nguyen, J. Neel, Pervaporation membranes endowed with
catalytic properties based on polymer blends, J. Membr. Sci. 73 (1992)
[
[
[
[
[
39] M.P. Rohde, G. Schaub, S. Khajavi, J.C. Jansen, F. Kapteijn, Fischer–Tropsch
synthesis with in-situ H2O removal – directions of membrane development,
Micropor. Mesopor. Mater. 115 (2008) 123.
40] S. Khajavi, F. Kapteijn, J.C. Jansen, Performance of hydroxy sodalite membranes
as absolute water selective materials under acidic and basic conditions, J.
Membr. Sci., accepted.
1
29.
[
[
8] K. Okamoto, M. Yamamoto, Y. Otoshi, T. Semoto, Pervaporation-aided esterifi-
cation of oleic acid, J. Chem. Eng. Jpn. 26 (5) (1993) 475.
9] H. Kita, S. Sasaki, K. Tanaka, K.I. Okamoto, M. Yamoto, Esterification of car-
boxylic acid with ethanol accompanied by pervaporation, Chem. Lett. (1988)
2
025.
41] S. Khajavi, F. Kapteijn, J.C. Jansen, Synthesis of thin defect-free hydroxy sodalite
membranes: new candidate for activated water permeation, J. Membr. Sci. 299
[
10] J.F.F. Keurentjes, G.H.R. Janssen, J.J. Gorissen, The esterification of tartaric acid
with ethanol: kinetics and shifting the equilibrium by means of pervaporation,
Chem. Eng. Sci. 49 (1994) 4681.
11] X. Feng, R.Y.M. Huang, Studies of a membrane reactor: esterification facilitated
by pervaporation, Chem. Eng. Sci. 51 (1996) 4673.
12] L. Bagnell, K. Cavell, A.M. Hodges, A.W.H. Mau, A.J. Seen, The use of catalytically
active pervaporation membranes in esterification reactions to simultaneously
increase product yield, membrane permselectivity and flux, J. Membr. Sci. 85
(
2007) 63.
42] O. de la Iglesia, R. Mallada, M. Menendez, J. Coronas, Continuous zeolite mem-
brane reactor for esterification of ethanol and acetic acid, Chem. Eng. J. 131
[
[
(
2007) 35.
43] Y. Morigami, M. Kondo, J. Abe, H. Kita, K. Okamoto, The first large-scale per-
vaporation plant using tubular-type module with Zeolite NaA membrane, Sep.
Purif. Technol. 25 (2001) 251.
(
1993) 291.

S. Khajavi et al. / Catalysis Today 156 (2010) 132–139
139
[
[
44] X. Li, H. Kita, H. Zhu, Z. Zhang, K. Tanaka, Synthesis of long-term acid-stable
zeolite membranes and their potential application to esterification reactions,
J. Membr. Sci. 339 (2009) 224.
45] X. Zhang, E.S. Man Lai, R. Martin-Aranda, K.L. Yeung, An investigation of Kno-
evenagel condensation reaction in microreactors using a new zeolite catalyst,
Appl. Catal. A: Gen. 261 (2004) 109.
[46] K. Tanaka, R. Yoshikawa, C. Ying, H. Kita, K. Okamoto, Application of
zeolite membranes to esterification reactions, Catal. Today 67 (2001)
12.
[47] J.J. Jafar, P.M. Budda, R. Hughes, Enhancement of esterification reaction yield
using zeolite A vapour permeation membrane, J. Membr. Sci. 199 (2002)
117.