Ethylation of benzene: Effect of zeolite acidity and structure

Article abstract of DOI:10.1016/j.apcata.2010.06.041

Zeolite catalysts based on ZSM-5, TNU-9, SSZ-33 and mordenite were prepared and characterized. The catalytic performance of these zeolites containing channels of different sizes has been studied in benzene ethylation with ethanol at three different reacti

Full text of DOI:10.1016/j.apcata.2010.06.041

Applied Catalysis A: General 385 (2010) 31–45
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Ethylation of benzene: Effect of zeolite acidity and structure
∗
T. Odedairo, S. Al-Khattaf
Center of Excellence in Petroleum Refining and Petrochemicals, King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia
a r t i c l e i n f o
a b s t r a c t
Article history:
Zeolite catalysts based on ZSM-5, TNU-9, SSZ-33 and mordenite were prepared and characterized. The
catalytic performance of these zeolites containing channels of different sizes has been studied in ben-
zene ethylation with ethanol at three different reaction temperatures (250, 275 and 300 C) for reaction
Received 22 April 2010
Received in revised form 1 June 2010
Accepted 18 June 2010
◦
times of 3, 5, 7, 10, 13, 15 and 20 s. In the ethylation reaction of benzene with ethanol, SSZ-33 catalyst
comprising 12-12-10-ring channels gave the highest benzene conversion, which is the result of high
acidity of this zeolite together with increased mass transport through large pores. The conversion of ben-
zene in the ethylation reaction of benzene with ethanol over the different zeolite catalysts follows the
order: SSZ-33 > TNU-9 > ZSM-5 > mordenite. TNU-9 catalyst behaves like the 10-ring ZSM-5 with respect
to ethylbenzene selectivity, while the behavior of SSZ-33 is close to that of a large pore zeolite with poten-
tial cage effects. EB selectivity follows the order: ZSM-5 > TNU-9 > SSZ-33 > mordenite, which implies that
this order is not directly related to the benzene conversion. In addition, a simplified kinetic model based on
reactant-converted deactivation model was developed for the reaction and compared with the obtained
experimental data.
Available online 25 June 2010
Keywords:
Zeolites
Benzene ethylation
Ethylbenzene
Kinetic modeling
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
catalysts were extensively evaluated for benzene alkylation using
various alkylating agents [10–14]. The direct use of ethanol, instead
of ethene, as alkylating agent with benzene for this reaction
has gained more attention [15–17]. A long stable catalyst life is
observed when alcohol, rather than ethylene, is used as an alky-
lating agent [18]. Zeolite ZSM-5 was reported to be active in the
ethylation of benzene and toluene [19–22], which led to the great
commercial process (Mobil–Badger). In addition, the effect of zeo-
lite acidity in benzene and toluene alkylation has also been reported
by various researchers [23,24]. More recently, mordenite is used in
the adsorptive separation of gas or liquid mixtures and catalysis
such as hydrocracking, hydroisomerization, alkylation, reforming,
dewaxing and dimethylamines synthesis due to its high thermal
and acid stabilities [25–27]. In addition, MOR zeolite has recently
been considered for applications as hosts of semiconductor mate-
rials, chemical sensors and nonlinear optical materials [28].
A very important investigation of alkylation of benzene with
ethanol by Levesque and Dao [29] over steamed–treated ZSM-5
zeolite and chryso-zeolite ZSM-5 catalysts showed that ethylben-
zene selectivity increased at high benzene to ethanol molar ratio.
Gao et al. [30] studied the alkylation of benzene with ethanol over
ZSM-5-based catalyst. They investigated the effect of zinc salt on
the synthesis of ZSM-5 in the alkylation of benzene with ethanol.
They concluded that the increase in EB selectivity was due to high
Lewis/Bronsted acid ratio and the smaller crystal size. Chandawar
et al. [31] also studied the alkylation of benzene over modified ZSM-
5-based catalyst. The authors reported a significant improvement
in ethylbenzene selectivity.
Ethylbenzene (EB) is important in the petrochemical industry
as an intermediate in the production of styrene, which in turn is
used for making polystyrene, a commonly used plastic material
[
1]. The 90% of EB world’s production is employed in the man-
ufacture of styrene, the other 10% is used in other applications
like paint solvents and pharmaceuticals [2]. The world’s more than
9
0% of ethylbenzene demand is met through benzene ethylation.
Ethylbenzene is produced in bulk quantities by alkylation of ben-
zene with ethene in the presence of aluminium chloride or zeolite
catalyst [3]. Conventionally, ethylbenzene is produced by benzene
alkylation with ethylene using homogeneous mineral acids such
as aluminium chloride or phosphoric acid as catalysts that cause
a number of problems concerning handling, safety, corrosion and
waste disposal [4,5]. The use of zeolite catalysts offers an envi-
ronmentally friendly route to ethylbenzene and the possibility of
achieving superior product selectivity through pore size control
[
6,7]. Lately, Zilkova et al. [8] reported the role of the zeolite channel
architecture and acidity on the activity and selectivity in aromatic
transformation.
Recently, aluminium chloride catalyzed alkylation of benzene
with ethene has been replaced by zeolite-catalyzed alkylation pro-
cesses [9]. Starting from the mid 1960s different zeolite-based
∗ Corresponding author. Tel.: +966 3 860 1429; fax: +966 3 860 4234.
E-mail address: skhattaf@kfupm.edu.sa (S. Al-Khattaf).
0
926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.06.041

3
2
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
Odedairo and Al-Khattaf [38] also studied the ethylation of ben-
Nomenclature
zene over Y-zeolite-based catalysts. Toluene was reported to be
one of the major products, which was formed as a result of EB
cracking.
Zeolite SSZ-33 (CON topology) possesses a channel system com-
plied of intersecting 12-ring and 10-ring pores; it is the first
synthetic zeolite having 4-4 = 1 SBU (secondary building unit) in
the structure [39]. Recently, Corma et al. [40] studied the alky-
lation of benzene with ethanol over SSZ-33 in comparison with
other zeolite-based catalysts. They found that the activity of SSZ-
Ci
concentration of specie i in the riser simulator
3
(
mol/m )
CL
Ei
confidence limit
apparent activation energy of ith reaction (kJ/mol)
3
k1
k2
k3
k4
k
rate constant of reaction 1 (m /kg of catalyst .s)
rate constant of reaction 2 (m /kg of catalyst .s)
rate constant of reaction 3 (m /kg of catalyst .s)
rate constant of reaction 4 (m /kg of catalyst .s)
3
3
3
3
3 for the conversion of ethanol and benzene is the highest among
the four zeolites studied, while the activity of the others (ZSM-5,
Beta/Ge and ITQ-22) is very similar.
3
apparent kinetic rate constant (m /kg of catalyst .s)
k0
pre-exponential factor in Arrhenius equation
3
defined at an average temperature [m /kg of
TNU-9 represents a new three-dimensional zeolite with 10-
ring channel systems, being rather similar to the industrially most
frequently employed ZSM-5. The size of the channels of TNU-9
is 0.52 nm × 0.60 nm and 0.51 nm × 0.55 nm, thus a slightly larger
zeolite compared with ZSM-5 [41]. A recent diffraction analysis
coming as a seminal breakthrough describes the structure solu-
tion of a material, TNU-9, that has the greatest number of unique
symmetry tetrahedral atoms in the structure (22) of any material
solved to date [42].
catalyst .s], units based on first-order reaction
molecular weight of specie i
universal gas constant (kJ/kmol K)
reaction time (s)
MWi
R
t
T
reaction temperature (K)
◦
T0
average temperature of the experiment (275 C)
3
V
volume of the riser (45 cm )
Wc
Whc
mass of the catalysts (0.81 g cat)
total mass of hydrocarbons injected in the riser
In this work, the catalytic activity of a novel zeolite TNU-9 pos-
sessing 10-10-10-ring system is compared with SSZ-33, ZSM-5 and
mordenite in benzene ethylation. Although there are many studies
on ethylation of benzene with ethanol over ZSM-5 and mordenite-
based catalyst, while a few have been reported over SSZ-33 and
none over TNU-9-based catalyst. Most of them are related to cat-
alyst development and elucidation of reaction mechanism. To our
knowledge, the kinetic study over TNU-9, SSZ-33 and mordenite-
based catalyst in the ethylation of benzene with ethanol has not
been reported in the open literature and has not been carried out
in a riser simulator. A simplified kinetic model which describes the
reaction under the present conditions will be developed. The pro-
posed model will be tested with the obtained experimental data
and the model parameters estimated using nonlinear regression
analysis.
(
0.162 g)
Yi
mass fraction of ith component (wt%)
Greek letters
ϕ
ꢀ
Á
apparent deactivation function
catalyst deactivation constant (RC Model)
effectiveness factor, dimensionless
Subscripts
0
1
2
3
4
cat
i
at time t = 0
for reaction (1)
for reaction (2)
for reaction (3)
for reaction (4)
catalyst
for ith component
2. Experimental
Abbreviations
B
DEB
E
EB
T
P/O
benzene
diethylbenzene
ethanol
ethylbenzene
toluene
para to ortho
2
.1. Preparation of catalysts
The uncalcined proton form of mordenite (H-mordenite) zeo-
lite (HSZ-690HOA) used in this work was obtained from Tosoh
Chemicals, Japan. The mordenite has silica to alumina ratio of
1
3
80. An alumina binder (Cataloid AP-3) contains 75.4 wt% alumina,
.4% acetic acid and water as a balance. The alumina binder was
obtained from CCIC Japan. The alumina binder was dispersed in
water and stirred for 30 min to produce thick slurry. The zeolite
powder was then mixed with alumina slurry to produce a thick
paste. The composition of the zeolite-based catalyst in weight
ratio is as follows: Mordenite: AP-3 (2:1). The uncalcined proton
form of ZSM-5 (CT-405) used in this study was obtained from
CATAL, UK. The ZSM-5 has silica to alumina ratio of 30. Simi-
lar procedure used for the preparation of the mordenite-based
catalyst used in this study was also followed for the prepara-
tion of the ZSM-5-based catalyst. The composition of the ZSM-5
zeolite-based catalyst in weight ratio is as follows: ZSM-5: AP-3
In a recent publication, Odedairo and Al-Khattaf [32] investi-
gated the ethylation of benzene with ethanol over ZSM-5-based
catalyst. It was found that, the reaction proceeded via two alky-
lation steps producing EB and diethylbenzene (DEB), respectively.
The apparent activation energy for benzene ethylation was found
to be 32.28 kJ/mol.
Several researchers have reported the studies on alkylation reac-
tions over ZSM type zeolites, [33–36] but only a few researchers
have performed alkylation reactions over mordenite-zeolite. Young
(
(
2:1). Zeolite SSZ-33 and TNU-9 and their characterization data
XRD—not provided in the text, SEM, FTIR) used in this study
[
21] reported alkylation of benzene over H-ZSM-12 and H-
mordenite dealuminated by HCl. The author obtained a benzene
conversion of 98% over H-mordenite and a conversion of 54% over
H-ZSM-12. Recently, Wang et al. [37] reported the alkylation of ben-
zene over a commercial mordenite, which contained 30% Al O as
were obtained from J. Heyrovsky Institute of Physical Chemistry,
Academy of Science of the Czech Republic, Prague, Czech Republic.
Alumina was added to these zeolites (SSZ-33 and TNU-9) adopting
the same procedure used for mordenite and ZSM-5-based cata-
lyst.
2
3
the binder, was studied in a fixed-bed reactor at over 1.0 MPa and
a conversion of 95% was achieved.

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
33
Fig. 1. Schematic diagram of the riser simulator.
2.2. Characterization of catalysts
Surface areas were obtained from N2 physical adsorption
isotherms by applying BET method with Quantachrome AUTO-
SORB-1 (model ASI-CT-8). The samples were preheated at 373 K
for 3 h in flowing N . Type and concentration of acid sites in all zeo-
2
lites were determined by adsorption of pyridine as probe molecules
followed by FTIR spectroscopy (Nicolet 6700 FTIR) using the self-
supported wafers technique. Prior to adsorption, self-supporting
zeolite wafers were activated in situ by evacuation at temperature
◦
◦
4
2
50 C over night. Adsorption of pyridine proceeded at 150 C for
0 min at partial pressure 5 Torr.
The concentrations of Bronsted and Lewis acid sites were calcu-
lated from the integral intensities of individual bands characteristic
−1
of pyridine on Bronsted acid sites at 1550 cm and band of pyridine
−
1
on Lewis acid sites at 1455 cm and molar absorption coefficient
−
1
−1
[
43] of ε(B) = 1.67 ± 0.1 cm ␮mol and ε(L) = 2.22 ± 0.1 cm ␮mol
,
respectively. The infrared spectra of absorbed pyridine on SSZ-33
were recently discussed in detail by Gil et al. [44].
Fig. 2. SEM images of SSZ-33 (A) and TNU-9 (B).
2.3. Reaction procedure
started and set to the desired conditions. Once the reactor and the
gas chromatograph have reached the desired operating conditions,
Benzene ethylation was carried out in the riser simulator (see
Fig. 1). This reactor is bench scale equipment with internal recy-
cle unit invented by de Lasa [45]. A detailed description of various
riser simulator components, sequence of injection and sampling
can be found in Kraemer [46]. Catalytic experiments were car-
ried out in the riser simulator with a feed mixture of benzene
and ethanol for residence times of 3, 5, 7, 10, 13, 15 and 20 s at
0
.162 g of the feedstock was injected directly into the reactor via a
loaded syringe. After the reaction, the four-port valve opens imme-
diately, ensuring that the reaction was terminated and the entire
product stream was sent on-line to analytical equipment via a pre-
heated vacuum box chamber. The products were analyzed in an
Agilent 6890N gas chromatograph with a flame ionization detec-
tor and a capillary column INNOWAX, 60-m cross-linked methyl
silicone with an internal diameter of 0.32 mm. During the course
of the investigation, a number of runs were repeated to check for
reproducibility in the experimental results, which were found to
be excellent. Typical errors were in the range of ± 2%.
◦
temperatures of 250, 275 and 300 C. Analytical grade (99% purity)
pure benzene and ethanol were obtained Sigma–Aldrich. All chem-
icals were used as received, and no attempt was made to further
purify the samples. The feed molar ratio of benzene to ethanol is
1
:1. About 800 mg of the fluidizable catalyst particles (60-␮m aver-
age size) were weighed and loaded into the riser simulator basket.
The system was then sealed and tested for any pressure leaks by
monitoring the pressure changes in the system. Furthermore, the
reactor was heated to the desired reaction temperature. The vac-
3. Results and discussion
◦
uum box was also heated to ∼250 C and evacuated to a pressure
3.1. Characterization of zeolites and catalysts used
of ∼0.5 psi to prevent any condensation of hydrocarbons inside
the box. The heating of the riser simulator was conducted under
continuous flow of inert gas (Ar), and it usually takes a few hours
until thermal equilibrium is finally attained. Meanwhile, before the
X-ray powder patterns of all zeolites under study exhibited good
crystallinity and characteristic diffraction lines (not shown here).
The shape and size of all zeolites were determined by scanning
electron microscopy (Jeol, JSM-5500LV). Fig. 2 shows typical SEM
images of zeolites SSZ-33 and TNU-9. The SEM images evidence
the absence of impurities including an amorphous one. The crys-
tal sizes of all zeolites used in this present study were between
initial experimental run, the catalyst was activated for 15 min at
◦
6
20 C in a stream of Ar. The temperature controller was set to the
desired reaction temperature, and in the same manner, the timer
was adjusted to the desired reaction time. At this point, the GC is

3
4
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
Table 1
Characteristics of samples used in this work.
2
Sample code
Si/Al
Lewis sites (mmol/g)
Bronsted sites (mmol/g)
Lewis sites (%)
Bronsted sites (%)
BET (m /g)
Parent-ZSM-5
F.C.-ZSM-5
Parent-TNU-9
F.C.-TNU-9
Parent-MOR
F.C.-MOR
Parent-SSZ-33
F.C.-SSZ-33
AP-3
30.0
20.3
21.9
19.3
180
135.9
18
13.6
0.13
0.28
0.15
0.25
<0.002
0.12
0.43
0.48
0.64
0.24
0.21
0.43
0.33
0.03
0.03
0.30
0.18
0
35
57
26
43
4
80
59
72
100
65
43
74
57
96
20
41
28
0
304
330
441
448
280
F.C.—Final Catalyst.
◦
0
.3 and 1 ␮m. Table 1 provides the quantitative evaluation of FTIR
A maximum EB yield of ∼10.2% was obtained at 300 C at a ben-
zene conversion of 20.2% for a reaction time of 20 s. Diethylbenzene
was also observed in significant amount in the ethylation reaction
over mordenite-based catalyst. The highest triethylbenzene (TEB)
spectra of parent zeolites and final catalysts. For application of
zeolites in acid catalyzed reactions, the concentration of acid sites
(
Bronsted and Lewis types) is of utmost importance. The concen-
◦
tration of Bronsted acid sites and all acid sites of all catalysts under
study is presented in Table 1. In this case the following sequences
were obtained: mordenite < SSZ-33 < ZSM-5 < TNU-9 for Bronsted
sites and mordenite < ZSM-5 < TNU-9 < SSZ-33 for all sites. It was
observed that addition of alumina binder led to the decrease in
Si/Al ratio for all catalysts and to a simultaneous increase in the
concentration of Lewis acid sites.
yield of ∼3.0% was achieved at 300 C at 20.2% benzene conver-
sion. The product distribution is partially reproduced in Table 2.
Formation of ethylbenzene represents the primary ethylation step,
while the ethylation reaction of EB with ethanol represents the sec-
ondary ethylation reaction. A tertiary ethylation reaction was also
observed in this reaction, which led to the formation of triethyl-
benzene (TEB). The formation of TEB is as a result of the channel
opening of mordenite that is large enough for the alkylation reac-
tion. The channel opening provides larger reaction volume enabling
consecutive reactions of DEB to proceed.
3
3
.2. Ethylation of benzene over mordenite (M-catalyst)
.2.1. Benzene conversion
The formation of ethylbenzene (EB), diethylbenzene (DEB) and
3.2.2. Ethylbenzene and DEB Yield
triethylbenzene (TEB) (Table 2) via ethylation of benzene with
The yield of ethylbenzene was found to increase with both tem-
◦
ethanol was studied over mordenite-based catalyst at 250, 275, and
perature and time, up to a maximum of ∼10.2% at 300 C for a
◦
3
00 C for residence times of 3, 5, 7, 10, 13, 15 and 20 s. Benzene
reaction time of 20 s. This corresponds to an ethylbenzene selec-
tivity of ∼50.7%. Cracking of ethylbenzene to produce toluene was
not observed in this reaction compared to the pronounced EB crack-
ing reported in our recent publication [38] over USY-zeolite-based
catalyst.
conversions of approximately 20.2, 14.5, and 9.7% were achieved
◦
at 300, 275 and 250 C, respectively, for a reaction time of 20 s, as
shown in Fig. 3. At all reaction times studied, the maximum ben-
◦
zene conversion was obtained at 300 C. The experimental results
showed that the conversion occurred via ethylation. The major
product of the ethylation reaction over M-catalyst is ethylbenzene.
Similar to ethylbenzene yield, diethylbenzene yield was noticed
to increase with both time and temperature. A maximum of ∼6.3%
◦
yield of DEB was obtained at 300 C at a benzene conversion of
∼
20.2% for a reaction time of 20 s. The experimental results showed
that m-DEB was found to be in significant amount more than
other DEB isomers detected in this ethylation reaction. The ther-
modynamic equilibrium mixture of diethylbenzenes (para: meta:
ortho = 30:54:16) reported by Kaeding [47] and Halgeri [48] was
compared with the DEB isomers obtained under this present study.
It can be observed from Table 2 that, for all the reaction conditions
studied; p/o ratio was found to be between 1.5 and 3.2, which com-
pared favorably well with the thermodynamic equilibrium values
given by previous researchers.
3
3
.3. Ethylation of benzene over ZSM-5 (Z-catalyst)
.3.1. Benzene conversion
The main products obtained from the ethylation reaction of
benzene with ethanol over ZSM-5-based catalyst were ethylben-
zene and diethylbenzene (Table 3). Small amount of toluene and
gaseous hydrocarbon was also noticed from the reaction products.
◦
A maximum benzene conversion of ∼21.7% was obtained at 300 C
for a reaction of 20 s. The conversion of benzene was observed to
increase with reaction temperature and time, as shown in Fig. 4.
This is in conformity with observation made by Odedairo and Al-
Khattaf [32] during the ethylation of benzene over ZSM-5-based
catalyst, in which kaolin and alumina were used as the filler and
silica sol was used as the binder. About 2.2–8.6% of toluene was
also noticed in this ethylation reaction over ZSM-5-based catalyst,
Fig. 3. Variation of benzene conversion with time-on-stream achieved over
mordenite-based catalyst at 250 (ꢀ), 275 (ꢀ) and 300 C (ꢁ) for catalyst/feed = 5.
◦

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
35
Table 2
Product distribution (wt%) at various reaction conditions for the ethylation of benzene over mordenite-based catalyst.
◦
Tempearture ( C)
Time (s)
Benzene conv (%)
EB
Toluene
Gases
m-DEB
p-DEB
o-DEB
Total DEB
TEB
250
275
300
5
0.81
4.33
7.92
9.72
0.49
–
–
–
–
–
0.08
0.49
1.08
1.44
0.03
0.27
0.44
0.55
0.02
0.13
0.22
0.28
0.13
0.89
1.74
2.27
0.19
0.98
1.90
2.36
10
15
20
2.37
4.05
4.83
0.09
0.23
0.26
5
0
5
0
4.30
9.92
13.51
14.53
2.51
5.24
6.67
7.26
–
–
–
–
–
0.59
1.59
2.28
2.55
0.22
0.65
0.98
1.12
0.12
0.26
0.36
0.37
0.93
2.50
3.62
4.04
0.86
1.98
2.86
2.87
1
1
2
0.20
0.36
0.36
5
0
5
0
8.07
13.55
18.16
20.16
4.61
7.07
9.30
–
–
0.18
0.25
0.12
0.25
0.45
0.46
1.40
2.52
3.47
3.93
0.63
1.13
1.57
1.78
0.21
0.36
0.49
0.55
2.24
4.01
5.53
6.26
1.10
2.22
2.70
2.96
1
1
2
10.23
◦
T = 250–300 C and catalyst/feed = 5.
Table 3
Product distribution (wt%) at various reaction conditions for the ethylation of benzene over ZSM-5-based catalyst.
o
Tempearature ( C)
Time (s)
Benzene Conv (%)
EB
Toluene
Gases
m-DEB
p-DEB
o-DEB
Total DEB
2
2
3
50
75
00
5
4.13
10.76
12.74
16.10
3.01
7.55
8.90
0.09
0.34
0.48
0.70
0.08
0.25
0.30
0.34
0.50
1.27
1.49
1.94
0.43
1.25
1.46
1.72
0.03
0.10
0.11
0.17
0.95
2.62
3.06
3.83
10
15
20
11.23
5
0
5
0
8.24
14.75
17.54
19.29
5.96
10.63
12.21
13.29
0.32
0.84
0.87
1.12
0.20
0.34
0.37
0.52
0.93
1.54
2.13
2.33
0.75
1.25
1.78
1.81
0.08
0.15
0.18
0.22
1.76
2.94
4.09
4.36
1
1
2
5
0
5
0
7.78
14.28
19.02
21.70
5.62
10.12
13.20
14.94
0.54
1.24
1.42
1.88
0.26
0.39
0.68
0.88
0.73
1.36
2.06
2.29
0.55
1.01
1.42
1.44
0.08
0.15
0.23
0.27
1.36
2.53
3.71
4.00
1
1
2
◦
T = 250–300 C and catalyst/feed = 5.
as compared with a negligible amount reported in our recent pub-
lication over another catalyst which was also based on ZSM-5 [32].
The small amount of toluene observed in this ethylation reaction is
due to the higher acidity of the ZSM-5 used in this present study as
compared with the one previously reported, which was of a lower
acidity (0.23 mmol/g). The noticeable amount of toluene observed
at the higher temperatures, is probably formed from the cracking
of ethylbenzene.
A possible mechanism to represent the ethylation of benzene
with ethanol over ZSM-5 is illustrated in Scheme 1. Surface pro-
ton attacks ethanol to form water and surface ethyl cation (ethoxy
cation). An ethyl cation attacks benzene molecule to form proto-
nated ethylbenzene on the surface. The protonated ethylbenzene
returns a proton to the surface and forms ethylbenzene. Thereafter,
surface ethyl cation (ethoxy cation) attacks ethylbenzene ring car-
bon atom at ortho, meta or para position atom to form a surface
protonated diethylbenzene. The surface protonated diethylben-
zene returns a proton to the surface and forms diethylbenzene.
3.3.2. Ethylbenzene and DEB Yield
The yield of ethylbenzene increased with reaction time and tem-
◦
perature, up to ∼14.9% at 300 C, for a reaction time of 20 s. EB yield
◦
◦
of 13.3 and 11.2% was also obtained at 275 C and 250 C, respec-
◦
tively. An optimum yield of ∼4.4% of DEB was noticed at 275 C
for a reaction time of 20 s. As temperature was further increased
◦
Fig. 4. Variation of benzene conversion with time-on-stream achieved over ZSM-
to 300 C, a slight decrease in DEB yield was observed to a value of
4.0%. In the ethylation reaction of benzene with ethanol over ZSM-
◦
5
-based catalyst at 250 (ꢂ), 275 (ꢀ) and 300 C (ꢁ) for catalyst/feed = 5.
∼

3
6
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
Scheme 1. Proposed overall reaction scheme during benzene ethylation.
◦
5
-based catalyst, the formation of para-isomers can be expected to
as temperature was increased from 250 to 300 C; more toluene
was formed due to ethylbenzene cracking, leading to a decrease in
the selectivity of ethylbenzene at higher temperature. Negligible
proceed almost entirely in the zeolite channel system, while the
formation of meta- and ortho-isomers proceeds on the external
surface. The p/o ratio obtained from the ethylation of benzene with
ethanol over ZSM-5-based catalyst was found to range between
◦
amount of DEB was noticed at 250 C, while no DEB was formed at
the higher temperatures.
∼
5.3 and 14.3. This value was noticed to be far above the ther-
modynamic equilibrium value given by Kaeding [47] and Halgeri
48].
It is of paramount importance to note that triethylbenzene
According to the product distribution over SSZ-33-based cat-
alyst, a possible reaction mechanism representing the reaction
of benzene with ethanol over SSZ-33 is given by Scheme 2.
Surface proton attacks ethanol to form water and surface ethyl
cation (ethoxy cation). An ethyl cation attacks benzene molecule
to form protonated ethylbenzene on the surface. The protonated
ethylbenzene returns a proton to the surface and forms ethyl-
benzene. Surface proton reacts with ethylbenzene to form a
carbonium ion I. These carbonium ion I form surface methyl cation
(methyoxy cation) and toluene. The surface methyl cation formed
reacts with benzene molecule to form protonated toluene. The
protonated toluene returns a proton to the surface and forms
toluene.
[
was not formed in the ethylation reaction over ZSM-based cata-
lyst. Even if the alkylation product (TEB) is formed in the large
channel intersections, its chances to leave the channel would be
very small. This also goes to ascertain the fact that zeolite with
1
0 rings usually exhibits restricted transition-state and product
shape selectivity being particularly important in para-selective
reactions such as toluene disproportionation, xylene isomeriza-
tion, and toluene alkylation with methanol, ethylene/ethanol and
propylene/propanol [49–56].
3.5. Ethylation of benzene over TNU-9
3.4. Ethylation of benzene over SSZ-33
Table 5 shows the product distribution of TNU-9 catalyst in the
Ethylbenzene and toluene were observed to be the main prod-
ucts obtained from the ethylation of benzene with ethanol over
SSZ-33. Benzene conversion was found to increase with reaction
ethylation of benzene with ethanol at variable temperatures of 250,
◦
2
75 and 300 C. EB, toluene, DEB and gaseous hydrocarbons were
found in the ethylation reaction. Benzene conversion increased
time for all temperatures studied, reaching maximum of ∼27% at
◦
◦
with reaction time, up to a maximum of ∼25.0% at 300 C.
2
75 C. Toluene selectivity shows a high dependence on reaction
Ethylation was noticed to be the major reaction, while at
elevated temperatures the cracking reaction is also important.
Cracking reaction of EB and DEB was observed at elevated temper-
atures, leading to increased cracking product (toluene) at elevated
temperatures. The ethylation and cracking reaction noticed over
TNU-9-based catalyst in the ethylation reaction of benzene with
ethanol is similar to the reactions observed during benzene ethy-
lation with ethanol over USY catalyst reported in our recent
publication [38]. Both toluene and EB selectivity shows a high
dependence on temperature over TNU-9 catalyst. The selectiv-
ity of toluene increased with increase in temperature, while the
selectivity of ethylbenzene and DEB decreased with increase in
temperature.
temperature over SSZ-33 catalyst, showing a toluene selectivity
rise from ∼25 to 43% and then to ∼54% for temperatures of 250,
◦
2
75, and 300 C, for 20 s reaction time, respectively. The selectivity
of toluene increased with increase in temperature. The significant
increase in the selectivity of toluene indicates that the ethylben-
zene formed undergoes cracking at higher temperature. Gao et al.
[
30] gave similar explanation when traces of toluene were found
in the alkylation of benzene with ethanol over ZSM-5-based cat-
alyst. They attributed the formation of toluene to the cracking of
ethylbenzene over Bronsted acid sites.
Ethylbenzene selectivity decreased steadily with increase in
temperature over SSZ-33-based catalyst to minimum of ∼34% at
◦
3
00 C for a reaction time of 20 s. It is evident from Table 4 that

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
37
Table 4
Product distribution (wt%) at various reaction conditions for the ethylation of benzene over SSZ-33 catalyst.
o
Temperature ( C)
Time (s)
Benzene conv (%)
EB
Toluene
Gases
m-DEB
p-DEB
o-DEB
Total DEB
250
275
300
5
12.08
18.86
24.17
25.28
7.81
2.01
2.86
4.95
6.26
1.33
1.65
2.11
2.32
0.56
0.88
1.04
1.04
0.29
0.41
0.53
0.58
0.08
0.12
0.13
0.14
0.93
1.42
1.70
1.76
10
15
20
12.93
15.41
14.94
5
0
5
0
12.85
18.57
25.43
27.24
8.03
10.01
11.85
12.35
3.25
6.39
10.76
11.75
1.57
2.17
2.82
3.14
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
1
1
2
5
0
5
0
16.26
23.85
24.97
26.38
6.76
7.97
8.25
8.87
7.39
13.34
13.89
14.18
2.11
2.54
2.83
3.33
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
1
1
2
◦
T = 250–300 C and catalyst/feed = 5.
3.6. Comparison of catalysts in the ethylation of benzene with
Based on the comparison of the role of structure of the zeo-
ethanol
lites on benzene conversion in ethylation reaction of benzene with
ethanol, it can be inferred that more open structure of SSZ-33-
based catalyst (12-12-10-ring), which has been known to combine
both large and medium pore channels [8], allows a higher reaction
rate and faster diffusion of both reactants and products in com-
parison with the medium pore (10-ring) ZSM-5 and TNU-9 zeolite.
Similar explanation was given during toluene disproportionation
reaction over zeolite-based catalysts when an increase in toluene
conversion was observed [57]. They attributed the increase in con-
version to the increase in the size of channels from medium to
large pore zeolites. The above trend in which the lowest benzene
Fig. 5 provides time-on-stream dependence of benzene conver-
sion at 300 C over zeolite-based catalysts prepared from ZSM-5,
TNU-9, SSZ-33 and mordenite materials. It is evident from this
figure that, over all the catalysts studied, benzene conversion
increases as expected with increase in reaction time (5–20 s). The
conversion of benzene in the ethylation reaction of benzene with
◦
◦
ethanol carried out at 300 C for a contact time of 20 s, follows
the order: SSZ-33 (26%) > TNU-9 (25%) > ZSM-5 (22%) > mordenite
(
20%).
Scheme 2. Proposed reaction scheme for toluene formation during benzene ethylation reaction with ethanol.

3
8
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
Fig. 6. Product selectivity of benzene ethylation over the different catalysts at 18%
◦
benzene conversion (catalyst/feed = 5, reaction temperature: 250, 275 and 300 C).
Fig. 5. Time-on-stream dependence of benzene conversion in benzene ethylation
◦
over ZSM-5 (ꢁ), TNU-9 (ꢀ), SSZ-33 (ꢂ) and mordenite (ꢃ) at 300 C for cata-
lyst/feed = 5.
and Lewis acid sites. The order of the concentrations of Bronsted
acid sites, calculated for the different catalysts, is mordenite < SSZ-
3
3 < ZSM-5 < TNU-9. If the total amount of acid sites (Bronsted
conversion was noticed over mordenite-based catalyst is expected
since mordenite, with a set of 12-ring channels interconnected by
another set of 8-ring cross channels, in which the cross channels
are sufficiently small that they do not provide a means for trans-
port of molecules between adjacent passageways [58]. It was also
pointed out that with respect to hydrocarbons, the structure is
effectively one-dimensional and may be regarded as an array of
parallel, noninterconnecting channels. Similarly, it has also been
reported by Chen et al. [59] that in mordenite, the eight-membered
ring channels with aperture size of 2.6 Å × 5.7 Å which crosses
the 12-membered ring channels are actually windows rather than
channels and they found it not readily accessible to chromium ions
during metal exchange. Abdullah et al. [60] reported that such nar-
row apertures make it difficult for benzene molecule to diffuse. In
addition, Zilkova et al. [8] found the 8-ring channels of mordenite
inaccessible for toluene as well.
plus Lewis) is taken into account, the order is slightly different:
mordenite < ZSM-5 < TNU-9 < SSZ-33, mainly because of the high
concentration of Lewis sites in zeolite SSZ-33. In both cases, mor-
denite gave the least amount of acid sites for both the Bronsted acid
sites as well as in all acid sites (Bronsted plus Lewis). Mordenite,
possessing the lowest amount of acid sites, gave the lowest ben-
zene conversion as compared with other zeolite-based catalysts.
Catalyst based on ZSM-5 showed comparable benzene conversions
with the one noticed over mordenite-based catalyst. The lowest
benzene conversion over mordenite- and ZSM-5-based catalyst is
due to larger pores of mordenite with low amount of acid sites pro-
viding similar results as ZSM-5 with smaller channels but higher
concentration of acid sites. Therefore, in the case of mordenite,
its unique channel architecture as well as low amount acid sites
accounts for the low benzene conversion noticed over this zeolite-
based catalyst.
The calculated concentrations of Bronsted and Lewis sites are
presented in Table 1. Individual zeolite-based catalysts under study
vary in Si/Al ratio as well as in the concentrations of Bronsted
The product selectivity during the ethylation of benzene with
ethanol over ZSM-5, TNU-9, SSZ-33 and mordenite-based catalyst
is compared in Fig. 6 at constant conversion level of 18% at 250,
Table 5
Product distribution (wt%) at various reaction conditions for the ethylation of benzene over TNU-9 catalyst.
◦
Temperature ( C)
Time (s)
Benzene Conv (%)
EB
Toluene
Gases
m-DEB
p-DEB
o-DEB
Total DEB
250
275
300
5
10.92
17.65
23.78
25.01
6.90
1.43
2.49
3.37
3.95
0.81
1.39
1.63
1.73
1.09
1.79
2.44
2.57
0.58
0.97
1.27
1.34
0.11
0.15
0.23
0.23
0.95
2.88
3.94
4.14
10
15
20
10.89
14.84
15.19
5
0
5
0
12.19
20.94
24.20
24.58
7.46
12.20
13.55
13.60
1.87
4.02
5.07
5.29
0.85
2.10
2.23
2.25
1.23
1.60
2.08
2.12
0.65
0.84
1.07
1.10
0.13
0.18
0.20
0.22
2.01
2.62
3.35
3.44
1
1
2
5
0
5
0
12.41
19.53
24.31
25.06
6.09
10.09
11.54
11.78
2.91
5.11
7.64
7.83
1.38
1.88
2.16
2.26
1.19
1.49
1.72
1.97
0.45
0.80
1.07
1.01
0.10
0.15
0.18
0.21
1.38
2.45
2.97
3.19
1
1
2
◦
T = 250–300 C and catalyst/feed = 5.

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
39
Fig. 7. Effect of benzene conversion on ethylbenzene selectivity (catalyst/feed = 5,
◦
reaction temperature 275 C).
◦
2
75 and 300 C reaction temperatures. The results show that ethyl-
benzene (EB) is obtained as the most predominant product over all
the catalysts. The formation of EB with high selectivity over all the
catalysts might be due to its free diffusion without steric hindrance
through the pores of the catalysts. Catalyst based on ZSM-5 gave
the highest EB selectivity of ∼70%, and mordenite-based catalyst
gave the lowest with an EB selectivity of ∼51%. SSZ-33 gave the
highest toluene selectivity of ∼34% among all the catalysts studied,
at a constant benzene conversion of 18%. It is worth mentioning
that DEB was not observed over SSZ-33 catalyst at all conversion
◦
levels, except at 250 C, which gave negligible amount of DEB. In
contrast, a significant amount of DEB was obtained over morden-
ite as compared to SSZ-33 catalyst. Mordenite, however, gave the
lowest selectivity towards toluene, particularly at lower reaction
temperatures.
With respect to the ethylation products, it can be seen from Fig. 6
that TNU-9 behaves like the 10-ring ZSM-5 for EB and DEB selec-
tivity, while SSZ-33 shows no selectivity towards DEB. The effect of
◦
benzene conversion on ethylbenzene selectivity at 275 C over all
zeolite-based catalysts under study is given in Fig. 7. EB selectiv-
ity over SSZ-33, mordenite and TNU-9-based catalyst shows a high
dependence on benzene conversion and was noticed to decrease as
benzene conversion increases. Ethylbenzene selectivity over ZSM-
Fig. 8. (a) Effect of benzene conversion on diethylbenzene selectivity at 275 ◦C on
catalysts (᭹) mordenite, (ꢃ) ZSM-5, (ꢂ) SSZ-33 and (ꢁ) TNU-9. (b) Benzene ethyla-
tion: effect of temperature on diethylbenzene selectivity at reaction time of 20 s on
(
᭹) mordenite, (ꢃ) ZSM-5, (ꢂ) SSZ-33 and (ꢁ) TNU-9.
5
-based catalyst on the other hand, shows a moderate dependence
on benzene conversion. Fig. 8a shows the effect of conversion of
benzene on diethylbenzene selectivity at 275 C. It can be noted
◦
◦
toluene/EB ratios at 250 C were found in the order: MOR < ZSM-
5 < TNU-9 < SSZ-33 (Fig. 9a). The above trend is completely reverse
to the trend which was noticed in the case of DEB selectivity with
benzene conversion in the sense that, in all cases, SSZ-33 gave
the lowest selectivity. This trend confirms that little or no crack-
ing of ethylbenzene occurred over mordenite-based catalyst, which
accounts for the high DEB selectivity noticed over mordenite cata-
lyst, since the production of DEB is at the expense of ethylbenzene
conversion to DEB (second ethylation step). The ratio of toluene/EB
that the diethylbenzene selectivity depends on both type of catalyst
and temperature (Fig. 8a and b). On the other hand, in all cases mor-
denite has shown higher selectivity, followed by ZSM-5 and TNU-9,
while the SSZ-33 has shown no selectivity of diethylbenzene at
this particular temperature. Both ZSM-5 and TNU-9 catalysts have
◦
shown the maximum selectivity at 250 C, after which an obvious
drop can be seen. In the case of mordenite, diethylbenzene selectiv-
ity increased with conversion and reaction temperature in which
◦
◦
the maximum selectivity was obtained at 300 C. This trend is clear
is between 0 and 0.42 at 250 C for the benzene ethylation over all
in Fig. 8b, where the comparison is performed for 20 s reaction time.
The ratio of toluene to ethylbenzene has been plotted versus
benzene conversion for its ethylation reaction over mordenite-,
ZSM-5-, SSZ-33- and TNU-9-based catalysts as shown in Fig. 9a. The
the zeolite-based catalysts under study. However, as temperature
increases, the ratio increased from 0.42 to about 1.68 (i.e., four times
more) as shown in Fig. 9b. Cracking of the ethylation product (EB)
is responsible for the higher toluene-to-EB ratio at higher tempera-

4
0
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
◦
Fig. 10. Effect of total acid sites on toluene formation at 300 C at 20% benzene
conversion.
slightly higher selectivity towards the formation of EB than TNU-9
and a much higher EB selectivity than SSZ-33 and mordenite-based
catalyst.
4. Kinetic modeling
In this section, a comprehensive universal kinetic model for ben-
zene ethylation over all the zeolite-based catalysts under study was
developed. To develop a suitable kinetic model representing the
overall ethylation of benzene with ethanol, we propose the reaction
network shown in Scheme 3.
4.1. Model development for ethylation reaction over
mordenite-based catalyst
Based on the product distribution observed in the ethylation
reaction over mordenite-based catalyst (Table 2), a suitable kinetic
model representing the ethylation reaction can be developed. Since
toluene was not observed as one of the products in the ethylation
reaction over mordenite-based catalyst, the cracking of ethylben-
zene as well as the cracking of diethylbenzene can therefore be
neglected.
◦
◦
Fig. 9. Toluene/EB ratio with benzene conversion at (a) 250 C and (b) 300 C on
catalysts (᭹) mordenite, (ꢃ) ZSM-5, (ꢂ) SSZ-33 and (ꢁ) TNU-9.
It should be noted that the following assumptions were made
in deriving the reaction network:
tures. The effect of total acid sites on toluene formation at constant
◦
conversion level of 20% at 300 C over all the zeolite-based catalysts
1. An irreversible reaction path is assumed for the ethylation reac-
tions. Similar assumption was made by Sridevi et al. [61].
2. Catalysts deactivation is assumed to be a function of reactant
conversion (RC). A single deactivation function is defined for all
the reactions taking place.
3. Isothermal operating conditions can also be assumed given the
design of the riser simulator unit and the relatively small amount
of reacting species [45].This is justified by the negligible temper-
ature change observed during the reactions.
4. A pseudo-first-order reaction kinetic for all species involved in
the reactions.
5. Negligible thermal conversion.
6. A single effectiveness factor was considered for benzene, ethyl-
benzene and triethylbenzene.
7. The effectiveness factor Á was taken to be unity. This assumption
was made based on the fact that p-xylene formed by catalytic
under study is given in Fig. 10. The yield of toluene increased with
increase in total acid sites. It was observed that a maximum toluene
yield of about 10.3% was obtained at 300 C over SSZ-33-based cat-
alyst at 20% benzene conversion. Therefore, to reduce cracking of
ethylbenzene in the ethylation reaction over all the zeolite-based
catalysts, the total acid sites need to be balanced.
It is a well-established fact that the formation of dialkylbenzenes
provides vital knowledge on the structure and acidity of zeolite
catalysts [8]. From product diffusion shape selectivity effects, it is
expected that the p/o ratio of the DEB isomers would be higher
in medium pore ZSM-5-based catalyst than in the large pore zeo-
lite. Within the distribution of isomers in the DEB, ZSM-5 showed
the highest p/o ratio of ∼9.9 at constant conversion level of 18%,
while the lowest was noticed over SSZ-33. In conclusion, accord-
ing to the results presented above, ZSM-5-based catalyst shows a
◦

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
41
Scheme 3.
reaction can easily escape via the pores of ZSM-5 [62] comparing
mordenite with a larger pore opening; no diffusion restriction is
expected in this reaction.
concentration [66]. For the reactant conversion model the deacti-
vation function is
ϕ = exp(−ꢀ(1 − yB)),
(6)
Therefore, kx2 and kx4 ≈ 0, (where x = m in the case of mordenite).
The following set of species balances and catalytic reactions can be
written:
where ꢀ is catalyst deactivation constant. Substituting Eqs. (5) and
(6) into Eqs. (1)–(4), we have the following first-order differential
equations which are in terms of weight fractions of the species:
Rate of disappearance of benzene:
dyB
dt
Wc
=
−ÁF1km1yByE
exp(−ꢀ(1 − yB))
Wc
(7)
(8)
V dCB
V
−
= Ákm1CBCEϕ
(1)
(2)
(3)
(4)
Wc dt
Rate of formation of ethylbenzene:
V dCEB
dyEB
dt
=
[ÁF2km1yByE − ÁF1km3yEByE]
exp(−ꢀ(1 − yB))
V
dyDEB
dt
Wc
=
[ÁF3km3yEByE − ÁF km5yDEByE]
exp(−ꢀ(1 − yB))
(9)
1
=
(Ákm1CBCE − Ákm3CEBCE) ϕ
V
Wc dt
dyTEB
dt
Wc
Rate of diethylbenzene formation:
= ÁF4km5yDEByE V exp(−ꢀ(1 − y ))
(10)
B
V dCDEB
Wc dt
F , F , F and F are lumped constants given below:
=
(Ákm3CEBCE − Ákm5CDEBCE) ϕ
1
2
3
4
Whc
F1
=
VMWE
Rate of triethylbenzene formation:
V dCTEB
Wc dt
=
Ákm5CDEBCEϕ,
MWEBWhc
F2
=
VMWBMWE
where CB is the benzene concentration, CEB is the concentration of
ethylbenzene, CDEB is the concentration of diethylbenzene, CTEB is
MWDEBWhc
the concentration of triethylbenzene in the riser simulator, V is the
F3 =
VMWEBMWE
3
volume of the riser (45 cm ), W is the mass of the catalyst (0.81 g
C
of catalyst), t is the time (s), ϕ is the apparent deactivation function,
Á = an effectiveness factor and k is the rate constant (cm /(g of cat-
alyst.s)). By definition the molar concentration, Cx, of every species
in the system can be related to its mass fraction, yx (measurable
from GC), by the following relation:
MWTEBWhc
3
F4
=
.
VMWDEBMWE
Eqs. (7)–(10) contain seven parameters, km1, km2, km3, Em1, Em2
Em3 and ꢀ, which are to be determined by fitting into experimental
,
data.
yxWhc
cx =
,
(5)
The temperature dependence of the rate constants was rep-
resented with the centered temperature form of the Arrhenius
equation, i.e.
V MWx
where Whc is the weight of feedstock injected into the reactor, MWx
is the molecular weight of specie x in the system, V is the volume
of riser simulator.
Regarding catalyst deactivation, as deactivation functions can
be expressed in terms of the catalyst time-on-stream [ϕ = exp (-
ꢀ
ꢁ
ꢂꢃ
−
Ei
1
T
1
To
ki = koi exp
−
.
(11)
R
◦
Since the experimental runs were done at 250, 275 and 300 C,
◦
˛
t)], deactivation can also be related to the progress of the reaction
T0 was calculated to be 275 C, where To is an average temperature
[
63]. The deactivation function based on time-on-stream was ini-
introduced to reduce parameter interaction [68], koi is the rate con-
stant for reaction i at To, Wc is the weight of catalyst and Ei is the
activation energy for reaction i.
tially suggested by Voorhies [64], since then this model has been
accepted and used extensively in the FCC literature. A different
approach for deactivation functions, based on coke concentration,
was proposed by Froment and Bischoff [65]. As a result; a sto-
ichiometric relationship can be established, as demonstrated by
Al-Khattaf and de Lasa [66] and Al-Khattaf et al., [63] between
the amount of reactant and amount of coke produced. This allows
for the use of a so-called “reactant conversion” model. This type
of model has been reported to incorporate a sound mechanis-
tic description of catalyst deactivation [67], allows for changes of
chemical species without extra requirement of measuring the coke
4.2. Model development for ethylation reaction over
ZSM-5-based catalyst (Z-catalyst)
A kinetic model development for benzene ethylation over ZSM-
5-based catalyst using an activity decay model based on reactant
conversion, which is based on Scheme 3, was developed for the
ethylation reaction over ZSM-5. The slight increase in toluene for-
mation noticed at elevated temperature, which is probably due

4
2
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
Rate of formation of toluene:
to cracking of ethylbenzene, was not accounted for in the kinetic
model development.
Therefore, kx2, kx4 and kx5 ≈ 0, (where x = z in the case of ZSM-5-
based catalyst). The following set of species balances and catalytic
reactions can be written as follows:
V dCT
Wc dt
=
ÁkS2CEBϕ
(20)
Substituting Eqs. (5) and (6) into Eqs. (18)–(20) results in the
following first-order differential equations which are in terms of
weight fractions of the species:
Rate of disappearance of benzene:
V dCB
−
= Ákz1CBCEϕ
(12)
(13)
(14)
dyB
dt
Wc
Wc dt
Rate of formation of ethylbenzene:
V dCEB
= −ÁG1kS1yByE
exp(−ꢀ(1 − yB))
Wc
(21)
(22)
(23)
V
dyEB
dt
=
[ÁG2kS1yByE − ÁkS2yEB]
Wc
exp(−ꢀ(1 − yB))
V
=
(Ákz1CBCE − Ákz3CEBCE) ϕ
Wc dt
dyT
dt
=
ÁG3kS2yEB
exp(−ꢀ(1 − yB)).
V
Rate of formation of diethyl benzene:
G , G and G are lumped constants given below.
1
2
3
V dCDEB
Wc dt
=
Ákz3CEBCEϕ.
Whc
G1
=
VMWE
Substituting Eqs. (5) and (6) into Eqs. (12)–(14) results in the
following first-order differential equations which are in terms of
weight fractions of the species:
MWEBWhc
VMWBMWE
G2
=
dyB
dt
Wc
=
−ÁH1kz1yByE
exp(−ꢀ(1 − yB))
Wc
(15)
(16)
(17)
MWT
V
G3
=
.
MWEB
dyEB
dt
=
[ÁH2kz1yByE − ÁH1kz3yEByE]
Wc
exp(−ꢀ(1 − yB))
V
The above proposed model equations were based on the
simplified assumptions stated for both mordenite and ZSM-5-
based catalysts, discussed earlier in Section 4.1. Contained in Eqs.
dyDEB
dt
=
ÁH3kz3yEByE
exp(−ꢀ(1 − yB))
V
(
21)–(23) are five parameters to be determined by fitting into
H , H and H are lumped constants given below:
experimental data.
1
2
3
Whc
H1
H2
H3
=
4.4. Model development for ethylation reaction over
TNU-9-based catalyst
VMWE
MWEBWhc
Based on the product distribution presented in Table 5, in
which both ethylation and cracking reaction occur in the ethyla-
tion of benzene with ethanol over TNU-9-based catalyst, a kinetic
model based on Scheme 3 was developed. Therefore, kx5 ≈ 0 (due to
absence of TEB), where x = TNU in the case of TNU-9-based catalyst.
The following set of species balances and catalytic reactions can be
written as follows:
=
=
VMWBMWE
MWDEBWhc
VMWEBMWE
.
Similar assumptions used in the kinetic model development for
benzene ethylation over mordenite-based catalyst are also appli-
cable in the ethylation of benzene with ethanol over ZSM-5-based
catalyst. Eqs. (15)–(17) contain five parameters; kz1, kz2, Ez1, Ez2 and
Rate of disappearance of benzene:
V dCB
−
= ÁkTNU-1C C ϕ
B
E
(24)
Wc dt
Rate of formation of ethylbenzene:
ꢀ
, which are to be determined by fitting into experimental data.
4
.3. Model development for ethylation reaction over
V dCEB
Wc dt
SSZ-33-based catalyst
= (ÁkTNU-1CBCE − (ÁkTNU-2CEB + ÁkTNU-3CEBCE)) ϕ
(25)
(26)
Rate of formation of toluene:
In accordance with the products obtained in the ethylation reac-
tion of benzene with ethanol over SSZ-33 (Table 4), a suitable
kinetic model based on Scheme 3 was developed. Diethylbenzene
was not accounted for in the model development due to the incon-
sistent and insignificant amount of DEB noticed in the ethylation
V dCT
Wc dt
=
ÁkTNU-2CEBϕ
Rate of diethylbenzene formation:
◦
reaction over SSZ-33, except at 250 C, during which traces of DEB
were observed.
V dCDEB
Wc dt
=
(ÁkTNU-3CEBCE − ÁkTNU-4CDEB) ϕ.
(27)
Therefore, kx3, kx4 and kx5 ≈ 0 (due to absence of TEB), where x = s
in the case of SSZ-33-based catalyst. The following set of species
balances and catalytic reactions can be written as follows:
Rate of disappearance of benzene:
Substituting Eqs. (5) and (6) into Eqs. (24)–(27) gives the follow-
ing first-order differential equations which are in terms of weight
fractions of the species:
dyB
dt
Wc
V dCB
= −ÁS1kTNU-1yByE
exp(−ꢀ(1 − yB))
(28)
−
= ÁkS1CBCEϕ
(18)
V
Wc dt
Rate of formation of ethylbenzene:
V dCEB
dyEB
dt
=
[ÁS2kTNU-1yByE − (ÁkTNU-2yEB + ÁS1kTNU-3yEByE)]
Wc
=
(ÁkS1CBCE − ÁkS2CEB) ϕ
(19)
Wc dt
×
exp(−ꢀ(1 − yB))
V
(29)

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
43
Table 6
Table 8
Estimated kinetic parameters for mordenite catalyst based on reactant conversion
Estimated kinetic parameters for ZSM-5 catalyst based on reactant conversion (RC
model).
(
RC model).
Parameters
Values
Parameters
Values
km1
km2
km3
ꢀ
kz1
kz2
ꢀ
Ei (kJ/mol)
59.87
3.61
0.0065
0.0030
16.08
6.54
0.0574
0.0259
16.63
9.61
0.0712
0.0345
3.03
1.12
Ei (kJ/mol)
17.12
2.64
0.0132
0.0069
8.90
5.80
0.0310
0.0164
2.28
0.98
9
5% CL
95% CL
a
3
3
a
3
3
k0i × 10 [(m /kg of catalyst .s)]
k0i × 10 [m /(kg of catalyst.s)]
3
3
9
5% CL × 10
95% CL × 10
a
6
a
6
Pre-exponential factor as obtained from Eq. (11); unit for second order (m /kg
Pre-exponential factor as obtained from Eq. (11); unit for second order (m /kg
of catalyst.s).
of catalyst.s).
Table 9
Correlation matrix for benzene ethylation over ZSM-5-based catalyst.
dyT
dt
Wc
=
ÁS3kTNU-2yEB
exp(−ꢀ(1 − yB))
(30)
kz1
Ez1
ꢀ
kz2
Ez2
ꢀ
V
kz1,2
Ez1,2
ꢀ
1.0000
0.2295
0.7937
0.2295
1.0000
0.2332
0.7937
0.2332
1.0000
1.0000
−0.1668
1.0000
0.4619
0.8922
0.4619
1.0000
dyDEB
dt
Wc
=
(ÁS4kTNU-3yEByE−ÁkTNU-4yDEB)
exp(−ꢀ(1 − yB)). (31)
−0.1668
V
0.8922
S , S , S and S are lumped constants given below:
1
2
3
4
Whc
Table 10
S1
S2
S3
S4
=
VMWE
Estimated kinetic parameters for SSZ-33 catalyst based on reactant conversion (RC
model).
MWEBWhc
Parameters
Values
=
=
=
VMWBMWE
ks1
ks2
ꢀ
Ei (kJ/mol)
12.29
4.72
0.0569
67.97
10.57
3.0000
2.94
0.75
MWT
MWEB
95% CL
k0i × 10 [m3/
a
3
(
kg of catalyst.s)]
3
MWDEBWhc
VMWEBMWE
95% CL × 10
0.0193
0.6512
.
a
6
Pre-exponential factor as obtained from Eq. (11); unit for second order (m /kg
of catalyst.s).
Similar assumptions used in the kinetic model development for
benzene ethylation over mordenite-, SSZ-33- and ZSM-5-based cat-
alyst are also applicable to the model development for ethylation
of benzene with ethanol over TNU-9-based catalyst. Eqs. (28)–(31)
contain nine parameters; kTNU-1–kTNU-4, ETNU-1–ETNU-4 and ꢀ, which
are to be determined by fitting into experimental data.
inferred that the high energy of activation noticed over mordenite-
based catalyst as compared with all other catalysts used in this
study is anticipated, since in mordenite-based catalyst, the cross
channels are sufficiently small and with respect to hydrocarbons,
the structure is effectively one-dimensional and may be regarded
as an array of parallel, noninterconnecting channels [58].
4.5. Discussion of kinetic modeling results
Comparing this to the three-dimensional pore structure of ZSM-
5 [69], containing interconnected channels of 10-ring parallel to
[1 0 0] (5.1 Å × 5.5 Å) and to [0 1 0] (5.3 Å × 5.6 Å), which permits
complete access from one pore to all others in the structure, thereby
permitting the entering of feed molecules as well as moving out of
the product molecules, a much lower apparent energy of activa-
tion is expected over ZSM-5 as compared with mordenite-based
catalyst.
The kinetic parameters k , E , and ꢀ for the ethylation reaction
were obtained using nonlinear regression (MATLAB package). The
values of the model parameters along with their corresponding
5% confidence limits (CLs) are shown in Tables 6, 8, 10 and 11
RC model), while the resulting cross-correlation matrices are also
given in Table 7 for mordenite-based catalyst and Table 9 for ZSM-
0i
i
9
(
5
-based catalyst.
Based on the correlation matrices of the regression analysis
Considering SSZ-33-based catalyst [52], which contains 12- and
10-ring interconnected channels of 6.4 Å × 7.0 Å, 5.9 Å × 7.0 Å (12-
ring parallel to [0 0 1] and [1 0 0], respectively, and 4.5 Å × 5.1 Å (10
MR parallel to [0 1 0]), a lower apparent activation energy for ben-
zene ethylation is expected, as compared with the medium pore
(10-ring) ZSM-5 zeolite. It can be observed that over mordenite
and ZSM-5-based catalysts, benzene ethylation was found to be
higher than the apparent activation energy needed to form DEB, as
a result of ethylbenzene ethylation. The difference in apparent acti-
vation energy for benzene ethylation (59.87 and 17.12 kJ/mol) and
ethylbenzene ethylation (16.08 and 8.90 kJ/mol) noticed over both
presented in Table 7 for ethylation of benzene over mordenite, it
shows the very low correlations between km1–km3 and Em1–Em3
and Em1–Em3 and ꢀ and the moderate correlation between km1–km3
and ꢀ. Similarly, Table 9 reports the very low correlations between
kz1, kz2 and Ez1, Ez2 and Ez1, Ez2, and ꢀ. It can be observed that in
the cross-correlation matrices presented in this study, most of the
coefficients remain in the low level with only a few exceptions.
Comparing the energies of activation for benzene ethyla-
tion over mordenite (59.87 kJ/mol), ZSM-5 (17.12 kJ/mol), SSZ-33
(
12.29 kJ/mol) and TNU-9-based catalyst (10.44 kJ/mol), it can be
Table 7
Correlation matrix for benzene ethylation over mordenite-based catalyst.
km1
Em1
ꢀ
km2
Em2
ꢀ
km3
Em3
ꢀ
km1,2,3
Em1,2,3
ꢀ
1.0000
0.5846
0.8903
0.5846
1.0000
−0.5951
0.8903
1.0000
0.3145
0.7968
0.3145
1.0000
−0.3560
0.7968
1.0000
−0.1222
1.0000
0.1822
0.6900
0.1822
1.0000
−0.5951
−0.3560
−0.1222
1.0000
1.0000
0.6900

4
4
T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
Table 11
energies for benzene ethylation were found to decrease as fol-
Estimated kinetic parameters for TNU-9 catalyst based on reactant conversion (RC
model).
lows: Em1 > EZ1 > ES1 > ETNU-1
.
Parameters
Values
Acknowledgements
kTNU-1
kTNU-2
kTNU-3
kTNU-4
ꢀ
Ei (kJ/mol)
10.44
2.71
0.116
0.067
15.20
1.36
2.700
0.060
24.51
5.79
0.362
0.022
12.82
3.52
16.67
8.13
2.44
0.93
The authors acknowledge the financial support provided by King
AbdulAziz City for Science and Technology for this research under
the Refining & Petrochemicals Program of the National Science and
Technology Plan (NSTP) for the year 2009 (09-PETE86-4). We are
grateful for the support from Ministry of Higher Education, Saudi
Arabia for the establishment of the Center of Research Excellence
in Petroleum Refining and Petrochemicals at King Fahd University
of Petroleum and Minerals (KFUPM). The authors thank J. Cejka
and M. Kubu (Heyrovsky Institute of Physical Chemistry, Prague,
Czech Republic) for providing zeolites TNU-9 and SSZ-33 and their
characterization.
9
5% CL
a
4
3
k0i × 10 [m /(kg of catalyst.s)]
4
9
5% CL × 10
a
6
Pre-exponential factor as obtained from Eq. (11); unit for second order (m /kg
of catalyst s).
catalysts is not unusual, since alkylation reactions of long substi-
tuted benzene generally occur with greater ease compared to their
shorter substituted counterparts [47].
It is evident from Tables 4 and 5 that the increase in temperature
leads to a larger fraction of cracking products, which clearly vali-
date the higher apparent activation energies noticed for EB cracking
References
(
67.97 and 15.20 kJ/mol) over SSZ-33 and TNU-9-based catalysts,
respectively, as compared with the apparent energies of activa-
tion obtained for benzene ethylation (12.29 and 10.44 kJ/mol) over
both catalysts. This important difference in energies of activation
is in perfect agreement with previous study [38]. Sridevi et al. [61]
and Barman et al. [18] reported an apparent energy of activation of
[
[
1] J. Cejka, B. Wichterlova, Catal. Rev. 44 (2002) 375–421.
2] T. Degnan, Appl. Catal. A 221 (2001) 283–294.
[3] N.M. Tukur, S. Al-Khattaf, Catal. Lett. 131 (2009) 225–233.
[
[
[
4] J. Weikamp, L. Puppe, Catalysis and Zeolites, Springer, Berlin, 1999, pp. 81.
5] C. Perego, P. Ingallina, Catal. Today 73 (2002) 3–22.
6] N.Y. Chen, W.E. Garwood, Catal. Rev. Sci. Eng. 28 (1986) 185–264.
6
0.03 and 56 kJ/mol for ethylation of benzene over AlCl3 impreg-
[7] A. Corma, M.E. Domine, S. Valencia, J. Catal. 215 (2003) 294–304.
[
8] N. Zilkova, M. Bejblova, B. Gil, S.I. Zones, A.W. Burton, C.Y. Chen, Z. Musilova-
Pavlackova, G. Kosova, J. Cejka, J. Catal. 266 (2009) 79–91.
9] S. Kato, K. Nakagawa, N. Ikenaga, T. Suzuki, Catal. Lett. 73 (2001) 2–4.
nated 13× zeolite and cerium exchange NaX zeolite, respectively.
These values are in agreement, or have the same order with the
value obtained in this present study for benzene ethylation over
mordenite-based catalyst.
Graphical comparisons between experimental and model pre-
dictions for the reactant conversion model (RC) based on the
optimized parameters for mordenite and ZSM-5 are shown
[
[10] P.B. Venuto, L.A. Hamilton, P.S. Landis, J.J. Wise, J. Catal. 5 (1966) 81–98.
[11] C.G. Wight, US Patent 4, 169 (1979) 111.
12] G. Bellusi, G. Pazzuconi, C. Perego, G. Girotti, G. Terzoni, J. Catal. 157 (1995)
27–234.
13] L. Forni, G. Cremona, F. Missineo, G. Bellusi, C. Perego, G. Pazzuconi, Appl. Catal.
121 (1995) 261–272.
14] J. Cejka, B. Wichterlova, S. Bednarova, Appl. Catal. 79 (1991) 215–226.
15] M. Raimondo, G. Perez, A. De Stefanis, A.A.G. Tomilinson, O. Ursini, Appl. Catal.
A 164 (1997) 119–126.
[
2
[
[
[
in Figs.
3 and 4, respectively. It can be seen that the
model predictions compared very well with the experimental
data.
[16] V.R. Vijayaraghavan, K.J.A. Raj, J. Mol. Catal. A 207 (2004) 41–50.
17] N.N. Binitha, S. Sugunan, Micropor. Mesopor. Mater. 93 (2006) 82–89.
18] S. Barman, N.C. Pradhan, J.K. Basu, Indian Chem. Eng. A 48 (2006) 10–17.
[
[
5
. Conclusions
[19] R.T. Sebulsky, A.M. Henke, Ind. Eng. Chem. Process Des. Dev. 10 (1971) 272–279.
[
[
[
20] T.J.L. Berna, D.A. Moreno, European Patent Appl. EP 0 353 (1991) 813.
21] L.B. Young, US Patent 4, 301 (1981) 317.
22] J. Cejka, A. Vondrova, B. Wichterlova, G. Vorbeck, R. Fricke, Zeolites 14 (1994)
147–153.
The following conclusions can be drawn from the ethylation
reaction of benzene with ethanol over ZSM-5-, TNU-9-, SSZ-33-
and mordenite-based catalyst.
[
[
23] J. Cejka, N. Zilkova, B. Wichterlova, Collect. Czech. Chem. Comm. 61 (1996)
1115–1130.
24] B. Wichterlova, J. Cejka, N. Zilkova, Microporous Mat. 6 (1996) 405–414.
1
2
. The zeolite structure in combination with the reaction temper-
ature determines the types of reactions that occur.
. Ethylbenzene selectivity was found to be highest in the ethy-
lation reaction over ZSM-5-based catalyst as compared to its
ethylation over all other zeolite catalysts.
. SSZ-33-based catalyst produced the highest benzene conver-
sion as compared to its conversion over ZSM-5-, TNU-9- and
mordenite-based catalyst.
. Considerable amount of toluene was obtained in the ethylation
over SSZ-33-based catalyst, while negligible amount was noticed
over mordenite-based catalyst. It can be concluded that acidity
plays a major role in toluene formation over these zeolites.
. An increase in the concentration of Lewis acid sites was observed
after modification of the parent zeolites with alumina binder.
. In the ethylation reaction over mordenite- and ZSM-5-based cat-
alyst, the apparent activation energy for EB ethylation (Em2 and
Ez2) was found to be lower than the apparent energy of activa-
tion obtained for benzene ethylation (Em1 and Ez1), which shows
that reactivity of the alkylbenzenes increases as the number of
ethyl group per benzene ring increases.
[25] P.K. Bajpai, Zeolites 6 (1986) 2–8.
[
[
26] I.E. Maxwell, W.H.J. Stork, in: H. von Bekkum, E.M. Flanigen, J.C. Jonsen (Eds.),
Introduction to Zeolite Science and Practice, Elsevier, Amsterdam, 1991, p. 71.
27] L.D. Fernandes, J.L.F. Moneteiro, E.F. Sousa-Aguiar, A. Martinez, A. Corma, J.
Catal. 177 (1998) 363–377.
[28] J.E. Gilbert, A. Mosset, Mater. Res. Bull. 33 (1998) 997–1003.
[29] Pierre Levesque, Le H. Dao, Appl. Catal., 53 (1989) 157–167.
[30] J. Gao, L. Zhang, J.X. Hu, W. Li, J.G. Wang, Catal. Comm. 10 (2009) 1615–1619.
3
4
[31] K.H. Chandawar, S.B. Kulkarni, P. Ratnaswamy, Appl. Catal. A 4 (1982) 287–295.
[
[
32] T. Odedairo, S. Al-Khattaf, Chem. Eng. J. 157 (2010) 204–215.
33] J.R. Anderson, K. Foger, T. Mole, R.A. Rajyadhyaksha, J.V. Sanders, J. Catal. 58
(
1979) 114–130.
[34] J.R. Anderson, T. Mole, V. Christov, J. Catal. 61 (1980) 477–484.
[
[
35] Y.S. Bhat, A.B. Halgeri, T.S.R. Prasada Rao, Ind. Eng. Chem. Res. 28 (1998)
94–899.
36] B. Lee, I. Wang, Ind. Eng. Chem. Prod. Res. Dev. 24 (1985) 201–208.
8
5
6
[37] I. Wang, S.J. Li, J.Y. Liu, In: Int. Symp. Zeolites and Microporous Crystals, Sendai,
Japan, 2 LC 01, 6-9 August 2000.
[
[
38] T. Odedairo, S. Al-Khattaf, Ind. Eng. Chem. Res. 49 (2010) 1642–1651.
39] P. Wagner, Y. Nakagawa, G.S. Lee, M.E. Davis, S. Elomari, R.C. Medurd, S.I. Zones,
J. Am. Chem. Soc. (2000) 122–263.
[40] A. Corma, F.J. L1opis, C. Martinez, G. Sastre, S. Valencia, J. Catal. 268 (2009) 9–17.
[
41] J. Cejka, N. Zilkova, S.I. Zones, M. Bejblova, 18th Saudi Japan Symposium,
November, 2008.
[
42] F. Gramm, C. Baerlocher, L.B. McCusker, S.J. Warrender, P.A. Wright, B. Han, S.B.
Hong, Z. Liu, T. Ohsuna, O. Terasaki, Nature 444 (2006) 79–81.
43] C.A. Emeis, J. Catal. 141 (1993) 347–354.
[
[
7
. Kinetic parameters for benzene ethylation over all the stud-
ied zeolites were calculated using the catalyst activity decay
function based on reactant conversion. The apparent activation
44] B. Gil, S.I. Zones, S.J. Hwang, M. Bejblova, J. Cejka, J. Phys. Chem. 112 (2008)
2997–3007.
[45] de Lasa, H.I., US Patent 5, 102 (1991) 628.

T. Odedairo, S. Al-Khattaf / Applied Catalysis A: General 385 (2010) 31–45
45
[
46] D.W. Kraemer, Ph.D. Dissertation, University of Western Ontario, London,
[59] N.Y. Chen, F. Thomas Degnan, C. Morris Smith, Molecular Transport and Reac-
tion in Zeolites, VCH, New York, 1994.
Canada, 1991.
[
[
[
[
[
[
[
[
[
[
47] W.W. Kaeding, J. Catal. 95 (1985) 512–519.
[60] A.Z. Abdullah, M.Z. Abu Bakar, S. Bhatia, The Proceedings of RSCE and 16th
SOMCHE, Malaysia, 28–30 October, 2002.
[61] U. Sridevi, B.K. Bhaskar Rao, N.C. Pradhan, Chem. Eng. J 83 (2001) 185–189.
[62] J. Wei, J. Catal. 76 (1982) 433–439.
[63] S. Al-Khattaf, J.A. Atias, K. Jarosch, H. de Lasa Chem, Eng. Sci. 57 (2002)
4909–4920.
[64] A. Voorhies, Ind. Eng. Chem. Res. 37 (1945) 318–322.
[65] G.F. Froment, K.B. Bischoff, Chemical Reactor Analysis and Design, second ed.,
John Wiley & Sons, New York, 1979, pp. 288.
[66] S. Al-Khattaf, H.I. de Lasa, Ind. Eng. Chem. Res. 40 (2001) 5398–5404.
[67] J.A. Atias, G. Tonetto, H. de Lasa, Ind. Eng. Chem. Res. 42 (2003) 4162–4173.
[68] A.K. Agarwal, M.L. Brisk, Ind. Eng. Chem. Process Des. Dev. 24 (1985) 203–207.
[69] H.V. Koningsveld, H.V. Bekkum, J.C. Jansen, Acta Crystallogr. B 43 (1987)
127–128.
48] A.B. Halgeri, Bull. Catal. Soc. India 2 (2003) 184–193.
49] B. Wichterlova, J. Cejka, Catal. Lett. 16 (1992) 421–429.
50] W.W. Kaeding, G.C. Barlie, M.M. Wu, Catal. Rev.—Sci. Eng. 26 (1984) 597–612.
51] T. Tsai, S. Liu, I. Wang, Appl. Catal. A 181 (1999) 355–398.
52] F.J. L1opis, G. Sastre, A. Corma, J. Catal. 227 (2004) 227–241.
53] S. Al-Khattaf, M.A. Ali, A. Al-Amer, Energy Fuels 22 (2008) 243–249.
54] S. Rabiu, S. Al-Khattaf, Ind. Eng. Chem. Res. 47 (2008) 39–47.
55] S.B. Hong, Catal. Surv. Asia 12 (2008) 131–144.
56] S.H. Lee, D.K. Lee, C.H. Shin, Y.K. Park, P.A. Wright, W.M. Lee, S.B. Hong, J. Catal.
2
15 (2003) 151–170.
57] S. Al-Khattaf, Z. Musilova-Pavlackova, M.A. Ali, J. Cejka, Top. Catal. 52 (2009)
40–147.
[
[
1
58] C.N. Satterfield, Heterogeneous Catalysis in Industrial Practice, 2nd ed, Krieger
Publishing Company, Malabar, Florida, 1996.