Kinetics of liquid phase synthesis of ethyl tert-butyl ether from tert-butyl alcohol and ethanol catalyzed by β-zeolite supported on monolith

Article abstract of DOI:10.1002/kin.10057

This paper compared the performance of β-zeolite and Amberlyst-15 catalysts on a liquid phase synthesis of ethyl tert-butyl ether (ETBE) from ethanol (EtOH) and tert-butyl alcohol (TBA) β-Zeolite was synthesized and deposited on monolith support. Its structure was confirmed by an XRD measurement and its composition was analyzed by an XRF measurement. It was found that even though the catalytic activity of β-zeolite was lower than that of Amberlyst-15, the selectivity of ETBE was much higher than that of Amberlyst-15, resulting in almost the same level of ETBE yield. The dehydration of TBA to isobutene (IB) was the major side reaction. The kinetic study of the reaction catalyzed by β-zeolite supported on monolith was carried out by using a semibatch reactor. The effect of external mass transfer was investigated by varying stirring speeds. The activity-based rate expressions were developed taking into account of water inhibition. Three temperature levels of 323, 333, and 343 K were performed in the study to obtain the parameters in the Arrhenius's equation and the Van't Hoff's equation.

Full text of DOI:10.1002/kin.10057

Kinetics of Liquid Phase
Synthesis of Ethyl tert-Butyl
Ether from tert-Butyl Alcohol
and Ethanol Catalyzed
by ␤-Zeolite Supported
on Monolith
S. ASSABUMRUNGRAT,1 W. KIATKITTIPONG,1 N. SEVITOON,1 P. PRASERTHDAM,1 S. GOTO2
1Research Center on Catalysis and Catalytic Reaction Engineering, Department of Chemical Engineering,
Faculty of Engineering, Chulalongkorn University, Bangkok 10330, Thailand
2Department of Chemical Engineering, Nagoya University, Chikusa, Nagoya 464-8603, Japan
Received 12 June 2001; accepted 29 January 2002
DOI 10.1002/kin.10057
ABSTRACT: This paper compared the performance of β-zeolite and Amberlyst-15 catalysts on
a liquid phase synthesis of ethyl tert-butyl ether (ETBE) from ethanol (EtOH) and tert-butyl
alcohol (TBA) β-Zeolite was synthesized and deposited on monolith support. Its structure was
confirmed by an XRD measurement and its composition was analyzed by an XRF measurement.
It was found that even though the catalytic activity of β-zeolite was lower than that of Amberlyst-
15, the selectivity of ETBE was much higher than that of Amberlyst-15, resulting in almost the
same level of ETBE yield. The dehydration of TBA to isobutene (IB) was the major side reaction.
The kinetic study of the reaction catalyzed by β-zeolite supported on monolith was carried out
by using a semibatch reactor. The effect of external mass transfer was investigated by varying
stirring speeds. The activity-based rate expressions were developed taking into account of water
inhibition. Three temperature levels of 323, 333, and 343 K were performed in the study to
C
obtain the parameters in the Arrhenius’s equation and the Van’t Hoff’s equation. 2002 Wiley
Periodicals, Inc. Int J Chem Kinet 34: 292–299, 2002
predominates in the world industries, a recent
investigation revealed that it has tendency to pollute un-
derground water because of its high solubility in water.
As a result, there is a pending legislation in a number
of states in the US [1]. A review on the market expan-
sion of oxygenate fuels and the catalytic aspect of their
synthesis was given by Ancilloti and Fattoro [2].
Similar to MTBE, ETBE showed a considerable
reduction in CO, a small reduction in unburned
INTRODUCTION
Although MTBE, commercially produced from the
reaction of isobutene (IB) and methanol, currently
Correspondence to: S. Assabumrungrat; e-mail: fchsas@eng.
chula.ac.th.
Contract grant sponsor: Thailand Research Fund.
Contract grant sponsor: TJTTP-OECF.
2002 Wiley Periodicals, Inc.
c

SYNTHESIS OF ETBE FROM TBA AND EtOH CATALYZED BY ␤-ZEOLITE
293
hydrocarbon, and a nonsignificant effect on NOx [3,4].
It may be a good alternative compared to lighter alco-
hols because of its lower blending Reid vapor pressure
(bRvp), low vaporization latent heat, very low solu-
bility in water and so on which account for its full
fungibility in the petroleum refining and distribution
system [5], and compared to other tert ethers such as
TAME and TAEE because of their high cost. ETBE is
also attractive from the viewpoint of environment, as
it is derived from EtOH which can be obtained from
renewable resources such as biomass [6,7]. It was ex-
pected that by 2005, 5% of fuel used for transportation
in France should be produced from renewable sources
[8]. ETBE outranks MTBE as an octane enhancer and
is more attractive than MTBE for low bRvp blends as
required (less than 55 kPa) in some hot places during
summer or in tropical countries because ETBE has low
bRvp (28 kPa)thanMTBE (55–69kPa) [5]. Inaddition,
because the water solubility of MTBE (43 kg/m3) is
about 4 times higher than that of ETBE (12 kg/m3), the
use of ETBE reduces the risk of water contamination.
Generally, ETBE can be produced by an exothermic
reversible reaction between EtOH and IB. However, the
supply of IB is mainly limited from refinery catalytic
cracking and steam cracking fractions. Hence, alterna-
tive routes for the synthesis of ETBE were currently
explored [9]. TBA, a major byproduct of propylene
oxide production from IB and propylene in the ARCO
process, can be employed instead of IB as a reactant
[10]. TBA was first investigated for the ETBE synthe-
sis about 60 years ago [11]. There are two methods to
produce ETBE from TBA, namely the indirect and the
direct methods. In the indirect method, TBA is first de-
hydrated to IB in a reactor and then the produced IB is
reacted with EtOH to produce ETBE in a second reac-
tor. In the direct method which is of our interest, ETBE
is produced directly from TBA and EtOH in only one
reactor. It is favorable because it shortens the process
itself [12]. However, selection of proper catalysts with
high activity and selectivity is a major concern for the
success of this process.
direct synthesis of ETBE from TBA and EtOH in reac-
tive distillation columns. NaHSO4 failed to synthesize
ETBE. The homogeneous catalyst KHSO4 was found
to be superior to H2SO4 and Amberlyst-15 which pro-
duced IB as a main product; however, a subsequent
catalyst separation unit was required. Recent research
compared three cation-exchange resins of S-54, D-72,
and Amberlyst-15 [14]. It was observed (at T = 338 K)
that S-54 showed improvements (compared to those of
Amberlyst-15) of activity and selectivity of 6 and 5%,
respectively, while D-72 showed improvements of 10
and 1%, respectively.
The purpose of this investigation was to compare
the performance of ␤-zeolite catalyst with the com-
mercial Amberlyst-15, which is usually used for tert-
ethers synthesis, for the production of ETBE from TBA
and EtOH by the direct method. The zeolitic catalyst
was chosen as it showed promising properties, high
thermal stability and no acid fume emission against
conventional resin-based catalysts [15]. In this study,
the kinetic parameters based on an activity model of
␤-zeolite catalyst were determined. The obtained pa-
rameters were used for modeling the ETBE production
inareactivedistillationwithorwithoutacombinedper-
vaporation unit (as proposed in our earlier work [10])
and a pervaporative membrane reactor.
EXPERIMENTAL
Catalyst Preparation
Synthesis of ␤-zeolite powder was carried out in an
autoclave at 408 K and 300 kPa. Colloid of SiO2 and
NaAlO2, which were the main reagents employed in
the zeolite synthesis, were used to obtain a Si/Al ratio
of 42. A compound of tetraethylammonium hydroxide
was used as a template for crystals. Sodium and potas-
sium ions, contained in NaCl, NaOH, and KCl were
used as seeds for the crystals, as well as to balance
the ionic charges in the crystals. All these components
when mixed together formed a gel. To avoid quick
solidification of the gel, HCl was added to keep pH at
low level. The gel was stirred thoroughly at room tem-
perature before transferring it to the autoclave. Then,
the mixture was stirred and heated for 40 h. In order
to remove the template from the catalyst precursor, the
catalyst was calcined at 813 K for 3.5 h. After this, the
catalyst in the Na+ form was ion exchanged twice with
the solution of 2 M NH4NO3 at 353 K for 1 h. Then, the
ion-exchanged crystals were dried in an oven at 383 K
A number of researches are investigating the ETBE
synthesis by the direct method using various catalysts
such as Amberlyst-15, heteropoly acid (HPA), KHSO4,
NaHSO4, and H2SO4. Yin and coworkers [12] com-
pared the performance of Amberlyst-15 and HPA by
performing the reactions in a semibatch reactor. The
conversion and selectivity of Amberlyst-15 (at 338 K
and 8 h reaction time) were 62% and 43%, respec-
tively. Heteropoly acid was found to yield superior
selectivity (79%); however, it was less attractive be-
cause it was significantly inhibited by the presence of
water. Matouq and coworkers [13] employed KHSO4,
NaHSO4, H2SO4, and Amberlyst-15 as catalysts for the
+
for at least 3 h. The resulting crystals were in the NH4
form. Finally, calcination of the catalyst at 773 K for 2+h
was necessary to dissociate the ammonium ion, NH4
into NH3 and protonated form of hydrogen, H+. NH3

294
ASSABUMRUNGRAT ET AL.
escaped in the atmosphere while H+ stayed onto the
catalyst to balance the ionic charge. The catalyst was
characterizedbyanX-raydiffraction(XRD-SIEMENS
D5000) and its composition was measured using an
X-ray fluorescence spectrometer (XRF-model Fision).
as shown in Fig. 1a. A frame of four catalyst baskets
was equipped with a rotating shaft which was driven
by a motor via an inverter controller to determine the
accurate start-up time in a semibatch reactor [16]. The
cylindrical baskets were made of stainless steel tubes
with a wall made of stainless steel mesh. The frame was
held above the liquid level by upper hooks as shown
in Fig. 1b. When the temperature was maintained at
a desired value, the reaction was started by inverting
the direction of agitation so that the frame of baskets
dropped into the liquid mixture. The lower hooks were
securely connected with slots on the disk turbine and
the frame was rotated without slip. Three temperature
levels, T = 323, 333, and 343 K, were performed under
atmospheric pressure. It is noted that the liquid samples
taken from the reactor were small compared to the total
liquid volume.
Preparation of Supported ␤-Zeolite
Supported ␤-zeolite was prepared by coating the ob-
tained powder catalysts on a cordierite monolith sup-
port (400 cell/in2.) which was cut into small cubes of
0.5 0.5 0.5 cm3. The supports were weighted and
soaked in 2.5 wt% acetic acid solution for 2 min. Af-
ter this they were washed with distilled water several
times to remove residual acid solution, and then dried
in an oven at 383 K until the weight became constant.
␤-zeolite powder was added into 2.5 wt% acetic acid
solution to give 30–50% w/v washcoat. The obtained
slurry was stirred for 10 min. Then, the monolith sup-
ports were dipped into the prepared washcoat for 15
min, followed by drying at 383 K overnight in the oven.
The supports were repeatedly dipped in the washcoat
2–3 times and calcined at 773 K for 3.5 h in air atmo-
sphere. The amount of catalyst was calculated as the
increased weight of the monolith. The uniformity of the
catalyst distribution was examined by measuring dis-
tribution of Al using a scanning electron microscope–
energy dispersive X-ray spectrometer (SEM–EDX). It
was shown that the catalyst was well dispersed on the
monolith surface.
RESULTS AND DISCUSSION
Catalyst Characterization
Synthesized ␤-zeolite was analyzed by an XRD mea-
surement to identify crystal structure. Figure 2 showed
an XRD pattern which agreed well with those reported
in Ref. [17]. This indicates that the synthesized cat-
alysts had the same structure as ␤-zeolite. The particle
size of the obtained ␤-zeolite powder was approxi-
mately 0.5 ␮m. The Si, Al, and Na contents of ␤-zeolite
measured by XRF spectrometer (XRF-model Fision)
were 96.97, 2.30, and 0.21 wt%, respectively, thus
yielding the Si/Al ratio of 42.
Kinetic Study
The powder ␤-zeolite was used in a preliminary study
to compare the catalyst performance between ␤-zeolite
and Amberlyst-15 (obtained from Fluka). The latter is
a sulfonated styrene divinyl benzene copolymer with
a macroreticular structure. The mesh size was in the
range of 20–50 and the ion exchange capacity was 4.96
mol-H+/kg-dry resin. A certain amount of catalyst was
weighted and pretreated by leaving it in an oven at
363 K overnight to remove moisture from the cata-
lyst. A desired amount of EtOH and TBA was mixed
and placed into a slurry reactor consisting of 250 cm3
three-necked flask fitted with a condenser in the cen-
tral opening. The mixture was stirred and heated up to
338 K by circulating hot water through the jacket. The
reaction was started by adding the catalyst into the re-
action mixture. Liquid samples of 1 cm3 were taken to
measure concentrations of H2O, EtOH, TBA, IB, and
ETBE at different reaction times. They were analyzed
by a TCD gas chromatography with a column packed
with Gaskuropack 54.
Comparison Between Catalyst
Performances
Two types of catalysts, i.e. commercial Amberlyst-15
and powder ␤-zeolite were tested in a slurry reactor
to compare the performance on the synthesis of ETBE
from EtOH and TBA. The experiments were carried
out under the following conditions: catalyst weight =
4 g, T = 338 K, stirring speed = 660 rpm, and the
initial amount of ethanol and TBA = 0.5 and 0.5 mol,
respectively. The reactions taking place in the reactor
can be summarized as follows:
TBA + EtOH ⇔ ETBE + H2O
TBA ⇔ IB + H2O
IB + EtOH ⇔ ETBE
(1)
(2)
(3)
Figure 3 shows the number of moles of TBA, EtOH,
ETBE, and H2O in liquid mixtures at different reaction
time of both catalysts. Filled and empty symbols
In the kinetic study, ␤-zeolite supported on monolith
was packed in a specially designed basket-type reactor

SYNTHESIS OF ETBE FROM TBA AND EtOH CATALYZED BY ␤-ZEOLITE
295
Figure 1 Schematic diagrams of (a) experimental setup and (b) catalyst basket.
represent results of Amberlyst-15 and ␤-zeolite, res-
pectively. It should be noted that the side product,
IB was mainly present in the gas phase. Consid-
ering the disappearance of TBA, it was found that
␤-zeolite was less active than Amberlyst-15. However,
when considering the formation of ETBE, it is obvi-
ous that the ETBE yields of both catalysts were almost
the same. By defining selectivity as the ratio of the
produced amount of ETBE to the produced amount of
water, it was found that the selectivities of ␤-zeolite and
Amberlyst-15 were 70% and 34% respectively. It can
be concluded that ␤-zeolite was much more attractive
than Amberlyst-15 and, consequently, the following
studies would consider only ␤-zeolite catalyst.
Kinetic Study
Effect of External Mass Transfer. The kinetic study
was investigated in the basket-type reactor using ␤-
zeolite supported on monolith. The effect of external

296
ASSABUMRUNGRAT ET AL.
Figure 2 XRD pattern of β-zeolite (Si/Al = 41).
mass transfer of catalyst was studied by varying stir-
ring speeds. Figure 4 shows the relationship between
the conversion of TBA (at 7 h) and the stirring speed. It
was found that the conversion increased with increas-
ing speed and, finally, it leveled off at speed of 1210
rpm. It can be concluded that the effect of external mass
transfer can be neglected when the speed is higher than
1210 rpm. In the subsequent studies, a speed of 1210
rpm was used to ensure negligible external mass trans-
fer resistance.
consequently, only small amount of IB can be dissolved
in the liquid. It was also confirmed by our experimen-
tal results that the concentration of IB was very small
in liquid mixture. As a result, the rate laws of reac-
tions (1) and (2) can be expressed in terms of activities
as
aETBEaH2
O
aTBAaEtOH
K1
r1 = k1
r2 = k2
(4)
(5)
1 + KWaH O
2
aTBA
1 + KWaH O
Development of Mathematical Models. Mathemati-
cal models were developed by following our previous
study [6]. The reverse reaction in Eq. (2) and the re-
action in Eq. (3) were neglected since the operating
pressure in this study was at atmospheric pressure and,
2
where kj is the reaction rate constant of reaction j ( j =
1, 2), ai is an activity of species i, and KW is the water
inhibition parameter.
Figure 3 Comparison between different catalysts. Filled
symbol: Amberlyst-15; empty symbols: β-zeolite (catalyst
weight = 4.0 g, mTBA,0 = 0.5 mol, mEtOH,0 = 0.5 mol, T =
338 K, and reactor type = slurry reactor).
Figure 4 The effect of speed on the conversion (catalyst =
␤-zeolite, catalyst weight = 15 g, mTBA,0 = 2 mol, mEtOH,0
=
2 mol, T = 333 K, time = 7 h, and reactor type = basket
reactor).

SYNTHESIS OF ETBE FROM TBA AND EtOH CATALYZED BY ␤-ZEOLITE
297
The expression of K1 is given as follow [18].
K1 = exp(1140.0 14580/T + 232.9 ln T + 1.087 T
7
1.114 10 3 T 2 + 5.538 10 T 3)
(6)
It should be noted that the resistance of pore diffu-
sion plays an important role in the reaction rate because
of the very fine pore size of ␤-zeolite. However, it was
assumed that this effect was included into the kinetic
parameters obtained from the data fitting.
By performing the material balance for a semibatch
reactor, the following expressions are obtained.
dmTBA
dt
dm
H2O
Figure 5 Mole changes with time (catalyst = ␤-zeolite,
=
=
= W (r1 + r2)
= Wr1
(7)
(8)
dt
catalyst weight = 15.0 g, mTBA,0 = 1.91 mol, mEtOH,0
=
1.98 mol, mETBE,0 = 0.01 mol, mH O,0 = 0.10 mol, T =
dmEtOH
dt
dmETBE
dt
2
343 K, and reactor type = basket reactor).
where mi and W represent the number of mole of
species i and the catalyst weight, respectively. It is
noted that the number of moles in the liquid phase at
any time is constant because IB can only slightly dis-
solved in the liquid phase. In addition, every 1 mol of
TBA consumption produces 1 mol of water, and every
1 mol of EtOH consumption produces 1 mol of ETBE.
The activity can be calculated from the following
relation.
of the residual (at T = 343, 333 and 323 K) of 0.061,
0.099 and 0.002, respectively.
Figure 6 shows the Arrhenius’s plot of the reaction
rateconstantsandthevan’tHoff’splotoftheadsorption
parameters using the results at three temperatures. The
following equations were determined from the plots.
k1 = exp(3.55 2286/T )
k2 = exp(36.57 13653/T )
kW = exp( 16.16 + 6636/T )
(10)
(11)
(12)
ai = i xi
(9)
Table I compares the values of the activation energy
of the reactions (1) and (2) and the heat of adsorp-
tion of water with those values of different catalysts.
It was observed that the heat of adsorption of water of
␤-zeolite catalyst was quite close to the values of other
catalysts except that of HPA which showed significant
water-inhibition effect. The activation energy of the
dehydration reaction of ␤-zeolite was within the range
where xi is mole fraction of species i in the liquid
mixture and is the activity coefficient. The activ-
i
ity coefficients can be calculated using the UNIFAC
method [19].
Kinetic Parameter Determination. A set of experi-
ments was carried out at three temperature levels of
323, 333, and 343 K to investigate the kinetic parame-
ters. Figure 5 shows typical results of mole changes
with time at T = 343 K. The initial moles of each
species are given in the figure legend. It should be noted
that the experimental results showed good agreement
with the simulation results. In addition, the experimen-
tal results showed that the production of ETBE became
higher with the increase of temperature as expected in
the Arrhenius’s equation.
A curve fitting method was employed to find the
kinetic parameters k1, k2, and Kw at each temperature.
Initial guess values of the parameters k1 and k2 were
obtained by using an initial rate method [20]. The lines
in the figure represent the simulation results. It was
found that within the ranges of the study, the model
fit the experimental results well with the sum square
Figure 6 Arrhenius and van’t Hoff plots.

298
ASSABUMRUNGRAT ET AL.
Table I Comparison of Activation Energy and Heat of Adsorption of Water Among Different Catalysts
Activation Energy of
Reaction (1) (kJ/mol)
Activation Energy of
Reaction (2) (kJ/mol)
Heat of Adsorption
of Water (kJ/mol)
Catalysts
Ref.
␤-zeolite
Amberlyst-15
HPA
S-54
D-72
19
57
95
48
43
–
114
85
108
74
30
142
55
63
20
67
63
55
This work
[12]
[12]
[14]
[14]
[21]
Amberlyst-15
reported in the other catalysts. This indicates that the
production of the undesired product, IB was greatly
enhanced with increasing temperature. However, the
activation energy of the etherification reaction (1) of
␤-zeolite was much lower than those of the other cat-
alysts. This may indicate the pore-diffusion limitation
arising from the transport of large molecule ETBE from
the pore of ␤-zeolite.
The obtained rate expressions will be used in our
subsequent studies on a reactive distillation column
with and without a pervaporation unit and a pervapo-
rative membrane reactor; however, it should be noted
that it is desirable to operate the reactor at low operat-
ing temperature. This is because the activation energy
of the dehydration reaction (2) is higher than the main
reaction (1), and consequently the selectivity to ETBE
decreases with increasing temperature.
EtOH Ethanol
H2O
k1
Water
Reaction rate constant of reaction (1) in the
activity-based model (mol kg 1 s
)
1
k2
Reaction rate constant of reaction (2) in the
activity-based model (mol kg 1 s
)
1
K1
KW
Equilibrium constant of reaction (1) in the
activity-based model
Water inhibition parameter in the activity-
based model
mi
rj
t
Number of mole of species i (mol)
Reaction rate of reaction j (mol kg 1 s
Reaction time (s)
)
1
T
Temperature
TBA tert-Butyl alcohol, 2-methyl 2-propanol
(IUPAC)
W
xi
Catalyst weight (kg)
Mole fraction of species i in liquid mixture
Activity coefficient of species i
Initial value a t = 0
i
CONCLUSION
( )0
It was found that ␤-zeolite showed superior perfor-
mance over the commercial Amberlyst-15 for the di-
rect production of ETBE from TBA and EtOH. Even
though the activity was moderate, the selectivity was
much higher. The kinetic study of the reaction cat-
alyzed by ␤-zeolite supported on monolith was carried
out by using a semibatch reactor. The effect of external
mass transfer was negligible at the stirring speed higher
than 1210 rpm. The activity-based model taking into
account of the effect of water inhibition was devel-
oped. Three temperature levels of 323, 333, and 343 K
were used in the study to obtain the parameters in the
Arrhenius’s equation of the reaction rate constants
and the van’t Hoff’s equation of the water inhibition
coefficient.
The authors thank Dr. Suphot Phatanasri for providing the
␤-zeolite preparation technique.
BIBLIOGRAPHY
1. Parkinson, G. Chem Eng 1999, June, 49.
2. Ancilloti, F.; Fattore, V. Fuel Process Technol 1998, 57,
163.
3. Fit, C.; Iborro, M.; Tejero, J.; Izquierdo, J. F.; Cunill, F.
Ind Eng Chem Res 1994, 33, 851.
4. Davis, B. C. Hydrocarbon Process 1992, 71(5), 91.
5. Cunill, F.; Vila, M.; Izquierdo, J. F.; Iborra, M.; Tejero,
J. Ind Eng Chem Res 1993, 32, 564.
6. Quitain, A.; Itoh, H.; Goto, S. J Chem Eng Jpn 1999, 32,
280.
7. Quitain, A.; Itoh, H.; Goto, S. J Chem Eng Jpn 1999, 32,
539.
NOMENCLATURE
ai
Activity of species i
8. Poitrate, E. Renewable Energy 1999, 16, 1084.
9. Rihko, L. K.; Quti, A.; Krause, L. Ind Eng Chem Res
1996, 35, 2500.
ETBE Ethyl tert-butyl ether, 2-ethoxy 2-methyl
propane (IUPAC)

SYNTHESIS OF ETBE FROM TBA AND EtOH CATALYZED BY ␤-ZEOLITE
299
10. Yang, B.; Goto, S. Sep Sci Technol 1997, 32, 971.
11. Norris, J. F.; Rigby, G. W. J Am Chem Soc 1932, 54,
2088.
16. Ishigaki, S.; Goto, S. J Chem Eng Jpn 1994, 27, 309.
17. Borade; Ramesh, B.; Clearfield, A.; Abraham J Phy
Chem 1992, 96, 6729.
12. Yin, X.; Yang, B.; Goto, S. Int J Chem Kinet 1995, 27,
1065.
18. Jensen, K. L.; Datta, R. Ind Eng Chem Res 1995, 34,
392.
13. Matouq, M.; Quitain, A.; Takahashi, K.; Goto, S. Ind
Eng Chem Res 1996, 35, 982.
19. Gmehling, J.; Rasmussen, P.; Fredenslund, A. Ind Eng
Chem Proc 1982, 21,118.
14. Yang, B.; Yang, S.; Yao, R. Reactive Functional Polym
2000, 44, 167.
15. Audshoorn, O. L.; Janissen, M.; van Kooten, W. E. J.;
Jansen, J. C.; van Bekkum, H.; van den Bleek, C. M.;
Calis, H. P. A. Chem Eng Sci 1999, 54, 1413.
20. Fogler, H. S. Elements of Chemical Reaction Engi-
neering, 2nd ed.; Prentice-Hall International, U.S.A.,
1992.
21. Abella, L. C.; Gaspillo, P.-A. D.; Maeda, M.; Goto, S.
Int J Chem Kinet 1999, 31, 854.