Modification of Cs-X for styrene production by side-chain alkylation of toluene with methanol

Article abstract of DOI:10.1016/j.cattod.2013.08.004

Effects of modification of cesium ion-exchanged zeolite X (Cs-X) with Cs₂O and zirconium borate (ZrB₂O₅₎ on the catalytic activity in the side-chain alkylation of toluene to styrene and ethylbenzene were studied to prepare

Full text of DOI:10.1016/j.cattod.2013.08.004

Catalysis Today 226 (2014) 117–123
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Modification of Cs-X for styrene production by side-chain alkylation
of toluene with methanol
Wahab O. Alabi, Balkrishna B. Tope, Rabindran B. Jermy, Abdullah M. Aitani,
Hideshi Hattori∗, Sulaiman S. Al-Khattaf
Center of Research 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:
Received 11 May 2013
Received in revised form 29 July 2013
Accepted 3 August 2013
Available online 31 August 2013
Effects of modification of cesium ion-exchanged zeolite X (Cs-X) with Cs2O and zirconium borate (ZrB2O5)
onthe catalyticactivityin the side-chain alkylation oftoluenetostyreneandethylbenzenewerestudied to
prepare an effective catalyst for styrene production. The toluene conversion to styrene and ethylbenzene
increased but the selectivity to styrene decreased on modification of Cs-X with Cs2O. For the toluene
conversion, the optimum loading of Cs2O was in the range ca. 1–5 wt%. Modification of Cs2O/Cs-X with
ZrB2O5 increased both toluene conversion and styrene selectivity. Characterization of acidic and basic
properties by IR spectroscopy of adsorbed pyridine and CO2 indicated that basic properties were markedly
enhanced but acidic properties were not changed by modification of Cs-X with Cs2O. It is suggested that
the enhancement of toluene conversion by modification with Cs2O is due to the generation of strong
basic sites which promote the activation of toluene by an abstraction of a proton from methyl group.
Modification with ZrB2O5 facilitates the formation of formaldehyde to increase the toluene conversion
and suppresses the transfer hydrogenation of styrene with methanol, the main pathway to ethylbenzene,
to increase the styrene selectivity.
Keywords:
Toluene
Methanol
Styrene
Ethylbenzene
Cesium oxide-loaded cesium
ion-exchanged zeolite (Cs2O/Cs-X)
Zirconium borate
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
Sidorenko et al. that methanol is first dehydrogenated to formalde-
hyde which undergoes aldol-type condensation with toluene to
Alkylation of toluene with methanol yields xylenes over acidic
catalysts and styrene and ethylbenzene over basic catalysts. The
former reaction is utilized for production of xylene isomers. The
later reaction has been known since 1967 when Sidorenko et al.
reported that K and Rb ion-exchanged zeolite X (K-X and Rb-X,
respectively) catalyzed the reaction to produced styrene and ethyl-
benzene [1]. Recently, the side-chain alkylation of toluene with
methanol has drawn attention since the reaction might be utilized
for styrene production which is currently undertaken by alkylation
of benzene with ethylene to ethylbenzene followed by dehydro-
genation [2].
produce styrene. This mechanism is widely accepted [4–10]. Ethyl-
benzene produced simultaneously with styrene is suggested to
result from hydrogenation of styrene [3,9,11].
Two functions are required for the catalyst active for the side-
chain alkylation; dehydrogenation of methanol to formaldehyde
and activation of the methyl group of toluene probably by abstrac-
tion of an H+. Both functions are related to the basic sites of the
catalyst. It was suggested that the dehydrogenation of methanol
might proceed on less basic sites and the activation of toluene might
proceed on more strongly basic sites [9].
In order to enhance the activity of Cs-X, different kinds of mod-
ifications have been attempted on Cs-X. Different kinds of metal
components were added to Cs-X. Cu and Ag were reported to be
effective metal components [11]. Besides the metal components,
promoting effects of addition of B and P on the activity were
reported in literature [10,12] and patents [13–15]. In our preceding
paper, we have reported that addition of metal borates, in particu-
lar zirconium borate (ZrB2O5), enhanced the activity of Cs-X as well
as the selectivity to styrene [16].
Soon after the report by Sidorenko et al., Yashima et al. studied
the reaction in detail and reported that Rb-X and Cs-X were effi-
cient catalysts as compared to other alkali ion-exchanged zeolite X
and Y [3]. They also supported the reaction pathway proposed by
∗
Corresponding author at: Catalysis Research Center, Hokkaido University, Kita-
ku, Kita 21, Nishi 10, Sapporo 001-0021, Japan. Tel.: +81 11 706 9121;
fax: +81 11 667 3734.
E-mail addresses: woa792@Mail.Usask.Ca (W.O. Alabi), balatope@Yahoo.Co.In
(B.B. Tope), rabindra@kfupm.edu.sa (R.B. Jermy), maitani@kfupm.edu.sa
(A.M. Aitani), hattorih@khaki.plala.or.jp, hattori@cat.hokudai.ac.jp (H. Hattori),
skhattaf@kfupm.edu.sa (S.S. Al-Khattaf).
Since the side-chain alkylation of toluene involves basic sites,
modification of basic properties is expected to alter the catalytic
activity. It was reported that impregnation of Cs-X with Cs com-
pound to add Cs ions in excess of the ion-exchanged capacity
increased the basic properties of Cs-X [6,17–19]. The resulting
0920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.08.004

118
W.O. Alabi et al. / Catalysis Today 226 (2014) 117–123
catalysts, Cs2O/Cs-X, showed higher activity than Cs-X for many
base-catalyzed reactions such as 2-propanol dehydrogenation [20],
1-butene isomerization [19,21–23], and Knoevenagel condensa-
tion [24–26]. It is interesting to see how the addition of Cs2O to
Cs-X affects the catalytic properties in the side-chain alkylation of
toluene.
were analyzed with online gas chromatograph equipped with a
WCOT fused silica capillary column coated with squalane (Varian,
Cat. # CP7520, length: 100 m, ID: 0.25 mm) and FID detector. The
products were identified by comparison with authentic samples.
Because of FID detector, H2, CO and H2O could not be analyzed by
gas chromatography.
The catalytic activities of Cs2O/Cs-X for side-chain alkylation of
toluene with methanol were reported by Lacroix et al. [11], Engel-
hardt et al. [27] and Hathaway and Davis [6]. All of them reported
that the toluene conversion and selectivity to ethylbenzene were
increased by addition of Cs2O to Cs-X. However, they examined
only a limited number of catalysts; no systematic study has been
reported to elucidate optimum loading of Cs2O.
In the present study, Cs2O/Cs-X catalysts with different load-
ings of Cs2O were prepared to examine their catalytic behaviors
in side-chain alkylation of toluene together with basic properties.
Increased toluene conversion was obtained with the catalyst loaded
with Cs2O loading in a range ca. 1–4 wt%. An increase in the activity
seems to result from generation of strong basic sites by modifi-
cation with Cs2O. Modification of Cs2O/Cs-X with ZrB2O5 further
enhanced the toluene conversion as well as styrene selectivity;
10 wt% ZrB2O5 being the optimum amount of addition for activity
and 15 wt% or more for styrene selectivity.
The activity was expressed by the conversion of toluene defined
by
ꢀ
ꢁ
1 − [toluene outlet]
Conversion of toluene (%) =
× 100%
[toluene inlet]
The styrene selectivity was defined by
ꢀ
ꢁ
[styrene]
[styrene] + [toluene]
Styrene selectivity (%) =
× 100%
For the hydrogenation of styrene with hydrogen, a benzene solu-
tion containing 5 mol% styrene was fed at the rate of 0.12 ml/min
into a hydrogen carrier flowing at 40 ml/min (STP). For the hydro-
genation of styrene with methanol (transfer hydrogenation), a
mixture containing styrene (4.3 mol%), methanol (17 mol%) and
benzene (78.7 mol%) was fed at the rate of 0.12 ml/min into an N2
carrier flowing at 40 ml/min (STP). The amount of catalyst and the
reaction temperature were the same as those for the side-chain
alkylation, 400 mg and 683 K, respectively.
2. Experimental methods
2.3. Characterization
2.1. Catalyst preparation
IR spectra of adsorbed CO2 and pyridine were measured with
in situ IR cell with CaF2 windows. Samples were pressed into a
thin wafer (∼50 mg/20 mm dia) and placed in the cell which was
connected to a vacuum system. The samples were pretreated at
723 K for 2 h in a vacuum prior to exposure to the adsorbate. All
spectra were measured at room temperature.
Powder XRD patters were measured with Rigaku Miniflix II XRD
powder diffraction system, Cu-K␣ being used as a radiation source
at 30 kV and 15 mA.
SEM images were measured with a JEOL JSM-5800 scanning
microscope. Magnification was 5000. Before taking SEM pho-
tographs, the samples were loaded on sample holder, held with
conductive aluminum tape and coated with a film of gold in vacuo
with Cressington sputter ion-coater for 20 s with 15 mA current.
Surface areas were measured by N2 adsorption at 77 K, the BET
equation being adopted to calculate the surface areas. Samples
were pretreated at 573 K in a vacuum prior to the measurement.
The Na-X (Junsei Chemical Co., Si/Al = 1.24) was ion-exchanged
with Cs+ to prepare Cs-X. CsOH was used as a source of Cs+ accord-
ing to Engelhardt et al. [27]. Into an aqueous solution of CsOH
(75 ml, 0.5 M), NaX (15 g) was immersed, stirred for 5 min and stand
for ca. 8 h. The slurry was filtered with Buchner funnel. The fil-
tered cake was again immersed in an aqueous solution of the CsOH
(75 ml, 0.5 M), stirred and stand for ca. 8 h. The immersion and fil-
tration procedures were repeated two more time (total 4 times).
The resulting slurry was filtered. To minimize the ion-exchange
of Cs+ with H+ during washing with water, the resulting cake was
immersed in 100 ml water and filtered. The filtered cake was dried
in an oven at 353 K and calcined at 753 K for 3 h in air.
Cs-X containing Cs in excess of ion exchange capacity (Cs2O/Cs-
X) was prepared by incipient wetness impregnation of Cs-X with
CsOH aq. solution. The loading of Cs2O was adjusted by the con-
centration of CsOH aq. solution. After impregnation, the catalysts
were dried at 353 K and calcined at 773 K for 3 h in air. The catalysts
were designated as Cs2O(D)/Cs-X, where D referred to the weight
% of Cs2O in the catalyst.
3. Results and discussion
The Cs-X’s modified with ZrB2O5 were prepared by grinding
a mixture containing the Cs-X and ZrB2O5 borate with a mor-
tar for 30 min in a dry state [28]. A ground mixture was calcined
with a slow ramp rate of 2 K/min at 773 K for 3 h. ZrB2O5 was
prepared from aqueous solutions of disodiumtetraborate and zir-
conium nitrate according to Knyrim et al. [29]. The precipitate was
washed with water, dried and calcined at 623 K in air. The Cs-
X’s modified with ZrB2O5 were designated as ZrB(E) Cs2O(D)/Cs-X,
where E referred to the weight % of ZrB2O5.
3.1. Morphology and acid-base properties of modified Cs-X
3.1.1. BET
The BET surface areas of the catalysts are listed in Table 1.
Since the weight of the zeolite unit cell increased by ion-
exchange of Na with Cs and addition of other compo-
nents, the normalized surface areas as defined below were
calculated and included in Table 1.Normalized surface area =
(measured surface area)×(unit cell weight of zeolite X containing Cs and borate)
unit cell weight of Na-X
The normalized surface areas decreased with the loading of
Cs2O, in particular, for the sample containing 7.8 wt% Cs2O. How-
ever, crystalline structure was retained for Cs2O loaded Cs-X by
XRD measurements (not shown). It was reported on the basis of
133Cs and 23Na MAS NMR study that all Cs species added to Cs-X
were occluded in the zeolite cavities in the form of Cs2O for the
catalysts containing less than 2.9 Cs atoms per supercage (14.9 wt%
Cs2O), while parts of Cs2O were located on the outer surface of the
zeolite crystals for the catalysts containing more than 5.1 Cs atoms
2.2. Reaction procedures
A fixed bed continuous flow reactor system (Model # 401C-
0286, Autoclave Engineers) was employed for carrying out the
reaction. The catalyst sample (400 mg) was pretreated in a flowing
N2 at 723 K for 2 h, then cooled to a reaction temperature of 683 K. A
mixture containing toluene and methanol in 6/1 mole ratio was fed
at 0.12 ml/min into an N2 flowing at 40 ml/min (STP). The products

W.O. Alabi et al. / Catalysis Today 226 (2014) 117–123
119
Table 1
Surface areas of catalysts.
Catalyst
Cs2O wt%
ZrB2O5 wt%
BET surface area (m2/g)
Normalized BET
surface area (m2/g)
Na-X
Cs-X
0
0
0
0
0
0
0
0
5
10
15
100
527
373
367
332
308
187
317
201
159
527
511
507
465
440
279
453
319
267
–
Cs2O(1.1)/Cs-X
Cs2O(2.1)/Cs-X
Cs2O(4.1)/Cs-X
Cs2O(7.8)/Cs-X
ZrB(5)/Cs2O(4.1)/Cs-X
ZrB(10)/Cs2O(4.1)/Cs-X
ZrB(15)/Cs2O(4.1)/Cs-X
ZrB2O5
1.1
2.1
4.1
7.8
4.1
4.1
4.1
–
8.1
per supercage (23.6 wt% Cs2O) [30]. The Cs2O contents used in the
present study were below 14.9 wt%. Accordingly, it is suggested
that all Cs2O species were occluded inside the cavities. A marked
decrease in surface area for Cs2O(7.8)/Cs-X is due to the filling some
of the cavities of the Cs2O(7.8)/Cs-X with Cs2O species.
From the results described above, it is suggested that modi-
fication of Cs-X with Cs2O generates both weak basic sites and
strong basic sites in addition to mild basic sites present on Cs-
X. Some parts of Cs species exist in the form of Cs2CO3. It was
reported in literature that addition of Cs2O generated strong basic
sites [17,19], which was confirmed by the present study. In addi-
tion to the strong basic sites, generation of weak basic sites was
observed in the present study, which has not been reported so
far.
3.1.2. SEM
Fig.
1 shows SEM images of Cs2O(4.1)/Cs-X (A) and
ZrB(10)/Cs2O(4.1)/Cs-X (B), SEM images of Cs-X (C) and ZrB(10)/Cs-
X (B) which were reported in our preceding paper are also shown
for comparison [16]. SEM images of Cs2O/Cs-X and Cs-X are similar,
and those of ZrB(10)/Cs2O(4.1)/Cs-X and ZrB(10)/Cs-X are similar.
Cs-X and Cs2O(4.1)/Cs-X were composed mostly of crystallites in
the range 1–3 ␮m, while ZrB(10)/Cs-X and ZrB(10)/Cs2O(4.1)/Cs-X
were composed of crystallites in the range 1–3 ␮m and small
particles of 0.1–1 ␮m attached to the larger crystallites. It is likely
that the small particles attached to the larger crystallites are
ZrB2O5 particles.
Fig. 3 shows IR spectra of pyridine adsorbed on Cs-X and
Cs2O(4.1)/Cs-X. Because of a strong band below 1450 cm−1 due to
carbonate ion species, the band at 1450 cm−1 ascribed to pyridine
coordinated to Lewis acid site was not observable for Cs2O(4.1)/Cs-
X. Lack on the band at 1540 cm−1 for both catalysts indicated that
no protonic acid sites were present on both catalysts. The spec-
tral changes with evacuation temperature were the same for both
catalysts, indicating that the acid sites were not changed much by
addition of Cs2O to Cs-X. It should be noted that the acid sites which
can adsorb pyridine after evacuation at 523 K exist on the surfaces
of both catalysts.
3.1.3. Basicity and acidity
Changes in basicity and acidity with modification of Cs-X with
Cs2O were studied by IR of adsorbed CO2 and pyridine, respectively.
3.2. Catalytic activities of modified Cs-X
Fig.
2 shows IR spectra of CO2 adsorbed on Cs-X and
3.2.1. Modified with Cs2O
Cs2O(4.1)/Cs-X. The spectra on Cs-X were taken from our preced-
ing paper [16] and shown together in Fig. 2 for comparison with
those on Cs2O/Cs-X. No carbonates remained on the surface of Cs-
X after pretreatment in a vacuum at 723 K. Two bands appeared
at 1380 and 1650 cm−1 after evacuation of the CO2 adsorbed-Cs-
X at 373 K. These bands are ascribed to the bidentate carbonate
[31]. The bidentate carbonate was eliminated by evacuation at
523 K. The basic sites on Cs-X were in such a strength that they
cannot retain CO2 after evacuation at 523 K. CO2 was adsorbed
much stronger on Cs2O(4.1)/Cs-X than on Cs-X. After pretreatment
at 723 K, a broad band was observable at about 1400 cm−1. This
broad band is composed of different carbonates. The carbonates
consisted mainly of ionic carbonate (CO32−) having a peak at about
1400 cm−1 and unidentate carbonate having peaks at about 1420
and 1380 cm−1. The carbonates formed during preparation partly
remained on the surface after the pretreatment at 723 K. Four bands
appeared at 1380, 1440, 1650 and 1680 cm−1 after evacuation of the
CO2 adsorbed-Cs2O(4.1)/Cs-X at 373 K. The two bands at 1380 and
1650 cm−1 were the same as those observed for Cs-X, and assigned
to bidendate carbonate [31]. The band intensity was much stronger
for Cs2O(4.1)/Cs-X than for Cs-X. Two bands at 1440 and 1680 cm−1
were observed only for Cs2O(4.1)/Cs-X. Since splitting of the bands
was 240 cm−1, these bands could be assigned to a hydrogencar-
bonate different from the carbonate showing bands at 1380 and
1650 cm−1 [31,32]. The bands at 1440 and 1680 cm−1 were elim-
inated on evacuation at 523 K. The bands at 1380 and 1650 cm−1
remained a little after evacuation at 523 K.
Modification of Cs-X with Cs2O enhanced the catalytic activity
of Cs-X and reduced the selectivity to styrene as shown in Fig. 4. The
optimum amount of Cs2O addition in toluene conversion was in the
range ca. 1–5 wt%. The activity (toluene conversion) of the catalyst
containing 2.1 wt% Cs2O was about twice as high as that of non-
added Cs-X catalyst. Addition of 7.8 wt% Cs2O resulted in a decrease
in the activity. Unlike non-added Cs-X, Cs2O/Cs-X produced much
more ethylbenzene than styrene. The styrene selectivity dropped
from 87% to 16% on addition of 1.1 wt% Cs2O to Cs-X, though
further addition of Cs2O did not change much the styrene selec-
tivity.
The effects of Cs2O addition to Cs-X on the activity and selec-
tivity observed in the present study were qualitatively similar to
those reported in literature [11,6,27]. All reported that addition of
Cs2O to Cs-X resulted in an increase in the activity and reduction
of styrene selectivity, though no reports described the changes in
activity and selectivity with Cs2O loading.
Modification of Cs-X with Cs2O also affected the methanol
conversion. The methanol conversion was 28.8% on Cs-X at the
toluene conversion of 1.5%, while it was 60.0% on Cs2O(4.1)/Cs-X at
the toluene conversion of 2.8%. Addition of Cs2O increased much
the methanol conversion. If methanol were consumed only for
the formation of styrene and ethylbenzene, the methanol conver-
sions would have been 9% and 16.8% over Cs-X and Cs2(4.1)/Cs-X,
respectively. The differences should be due to the methanol decom-
position to CO and H2. Accordingly, the modification with Cs2O
accelerated the methanol decomposition.

120
W.O. Alabi et al. / Catalysis Today 226 (2014) 117–123
Fig. 1. SEM images of (A) Cs2O/Cs-X, (B) Zr(10)/Cs2O(4.1)/Cs-X, (C) Cs-X, and (D) ZrB(10)/Cs-X.
3.2.2. Modification with ZrB2O5
decreased the activity but brought about further increase in the
styrene selectivity.
In our preceding paper, we reported the effects of ZrB2O5 addi-
tion on the activity and selectivity of Cs-X [16]. Essentially the
Fig. 5 shows the variations in the activity and styrene selectivity
as a function of ZrB2O5 borate loading under a reaction condi-
tion of toluene/methanol feed ratio of 6. Addition of ZrB2O5 to
Cs2O(4.1)/Cs-X enhanced both the activity and styrene selectiv-
ity up to ZrB2O5 loading of 10 wt%. Further addition of ZrB2O5
1
1
d
e
f
d
e
f
a
a
b
c
b
c
1700
1650
1600
1550
1500
1450
1800
1700
1600
1500
1400
1300
Wavenumber / cm-1
Wavenumber / cm-1
Fig. 2. IR spectra of adsorbed CO2 on Cs-X (a, b and c) and Cs2O(4.1)/Cs-X (d, e and
f). c and f, background (pretreated at 723 K in a vacuum); a and d, CO2 adsorbed at
room temperature and evacuated at 373 K for 30 min; b and e, further evacuated at
523 K for 30 min.
Fig. 3. IR spectra of adsorbed pyridine on Cs-X (a, b and c) and Cs2O(4.1)/Cs-X (d,
e and f). c and f, background (pretreated at 723 K in a vacuum); a and d, pyridine
adsorbed at room temperature and evacuated at 423 K for 30 min; b and e, further
evacuated at 523 K for 30 min.

W.O. Alabi et al. / Catalysis Today 226 (2014) 117–123
121
4
3
2
1
0
100
[toluene] / [methanol]
1/1 2/1
1/2
6/1
1/6
12
10
100
80
60
40
20
0
80
60
8
40
6
4
20
0
2
0
0
2
4
6
8
10
Wt % of Cs2O in Cs2O/Cs-X
0
20
40
60
80
100
Fig. 4. Conversion of toluene (left axis) and styrene selectivity (right axis) as a
function of Cs2O loading.
Mol % toluene in feed
Fig. 6. Conversion and styrene selectivity as a function of toluene mole % in the
reactant mixture for Cs2O(4.1)/Cs-X (circle) and Cs-X (triangle). Upper abscissa is
toluene to methanol molar ratio in the reactant.
same results were obtained with Cs2O(4.1)/Cs-X; both the activ-
ity and selectivity for styrene were enhanced by addition of a
proper amount of ZrB2O5. In both catalysts, the highest activity
was observed when ZrB2O5 was added by 10 wt%. The activity
of ZrB2O5-added Cs-X was not so high as compared to that of
ZrB2O5-added Cs-X(4.1)/Cs2O. In contrast, the styrene selectivity
exceeded 90% for ZrB2O5-added Cs-X under the same reaction con-
ditions, while the styrene selectivity was 36.4% for ZrB2O5-added
Cs2O(4.1)/Cs-X. Concerning the styrene yield (toluene conversion
x styrene selectivity) for ZrB2O5-added Cs-X/Cs2O, the addition of
15 wt% ZrB2O5 gave higher yield of styrene (yield 2.4%) than the
addition of 10 wt% ZrB2O5 (yield 1.8%).
It was reported that addition of B component suppressed the
decomposition of methanol to H2 and CO [8]. It was also reported
that addition of metal component together with B increased the
activity [11]. In the preceding paper, we reported that metal
borates were effective agents to prepare the Cs-X catalysts con-
taining both metal components and B [16]. We proposed the role of
metal borates as facilitation of the formation of formaldehyde from
methanol. Among the metal borates, ZrB2O5 was most effective in
promoting the activity and styrene selectivity for Cs-X based cat-
alysts. Addition of ZrB2O5 had never been reported. In the present
study on Cs2O/Cs-X based catalysts, it was also revealed that ZrB2O5
borate was an efficient additive not only to Cs-X but also to Cs2O/Cs-
X.
obtained at 33.3 toluene mole %, that is toluene/methanol = 1/2, for
both catalysts. The styrene selectivity, on the other hand, increased
monotonically with toluene mole % for Cs-X, but kept constant for
Cs2O(4.1)/Cs-X.
It should be pointed out that the attainable conversion of
toluene does not reach 100% for the reaction with a feed in which
toluene/methanol is higher than unity. Toluene should remain
unreacted to a certain extent even methanol were consumed all
for the reaction with toluene. The attainable conversions for the
feeds of toluene/methanol ratios of 1, 2, and 6 are 100, 33.3, and
16.7%, respectively.
3.3. Ethylbenzene formation
3.3.1. Styrene hydrogenation with H2 and transfer hydrogenation
with methanol
Styrene was allowed to react with hydrogen and with methanol
to elucidate the main pathway to form ethylbenzene in side-chain
alkylation of toluene with methanol over Cs-X, ZrB(10)/Cs-X, and
Cs2O(4.1)/Cs-X. The results are shown in Table 2. For all catalysts,
the transfer hydrogenation proceeded faster than the hydrogena-
tion. It can be concluded that ethylbenzene is formed mainly
through the transfer hydrogenation of styrene with methanol,
though the hydrogenation of styrene with hydrogen contributes
to ethylbenzene formation to a small extent. Although Garces et al.
described that styrene was readily hydrogenated with methanol
but only minor amount of ethylbenzene was produced with hydro-
gen over Cs-X, no experimental data were given [33].
3.2.3. Toluene/methanol molar ratio
The toluene/methanol mole ratio in a feed affected much on
both activity and selectivity. Fig. 6 shows the variations in the activ-
ity and styrene selectivity as a function of mole % of toluene in
the feed for Cs-X and Cs2O(4.1)/Cs-X. The maximum activity was
Addition of ZrB2O5 to Cs-X suppressed the transfer hydro-
genation to a significant extent (15.9–11.9%) but increased the
hydrogenation to a small extent (3.4–3.9%). On the other hand,
addition of Cs2O to Cs-X caused an increase in activity for both
the hydrogenation and the transfer hydrogenation. In particular,
an increase in the transfer hydrogenation was large.
6
5
4
3
2
1
0
60
50
40
30
20
10
0
The styrene selectivity in the side-chain alkylation increased
monotonically with toluene/methanol molar ratio in the reactant
Table 2
Conversion of styrene to ethylbenzene by hydrogenation with H2 and transfer
hydrogenation with methanol.
0
2
4
6
8
10
12
14
16
Catalyst
With H2 (%)
With methanol (%)
Wt % of ZrB2O5 in ZrB2O5/Cs2O(4.1)/Cs-X
Cs-X
ZrB(10)/Cs-X
Cs2O(4.1)/Cs-X
3.4
3.9
10.0
15.9
11.9
91.8
Fig. 5. Variations in toluene conversion and styrene selectivity as a function of
ZrB2O5 loading on Cs2O(4.1)/Cs-X.

122
W.O. Alabi et al. / Catalysis Today 226 (2014) 117–123
Step 1
Step 2
CH3-OH
HCH=O
HCH=O + H2
CO + 2H2
A: acidic site
B: basic site
B
B
Dehydrogenation and decomposition of methanol
δ+
C
Step 3
H
H
A
O
A
Step 4
+ H+
B
CH3
CH2
CH2
B
δ+
H
H
Step 6
Step 5
+ H+
B
+
C2H4-OH
CH=CH2 + H2O
C
O
A or B
A
Aldol-type condensation
Step 7
Step 8
CH2-CH3
CH2-CH3
+
H2
CH3-OH
CH=CH2
+
+
HCH=O
CH=CH2
Hydrogenation and transfer hydrogenation of styrene to ethylbenzene
Fig. 7. Reaction mechanisms involved in side-chain alkylation of toluene with methanol.
as shown in Fig. 6. The results are interpreted by a slow transfer
hydrogenation of styrene with methanol at a low partial pressure
of methanol when toluene/methanol ratio is high.
H2 and CO (step 2). In both steps 1 and 2, basic sites and acid sites
may participate. To proceed aldol-type condensation of formalde-
hyde with toluene, formaldehyde is activated by acid site to be
polarized (step 3) [32], and toluene is activated by basic sites to
form benzyl anion and proton (step 4). The benzyl anion attacks
the polarized formaldehyde at the carbonyl C␦+ and the proton
attacks the carbonyl O to form 2-phenylethanol (step 5) (aldol-
type addition). 2-Phenylethanol undergoes dehydration to form
styrene (step 6) either on acidic sites or basic sites to complete
aldol-type condensation. Styrene undergoes transfer hydrogena-
tion with methanol to ethylbenzene (step 8), and hydrogenation
with H2 formed in steps 1 and 2 (step 7). As shown in 3.3, step 8
contributes much more than step 7 to the formation of ethylben-
zene. Basic sites are supposed to be involved in both the transfer
hydrogenation and the hydrogenation.
Modification of Cs-X with Cs2O results in the formation of strong
basic sites. The strong basic sites would facilitate the activation
of toluene to form benzyl anion to increase the toluene conver-
sion. At the same time, the strong basic sites would promote step
2 to decompose methanol to CO and H2. Also, the strong basic
sites promote the transfer hydrogenation of styrene with methanol
to ethylbenzene (step 8). Thereby, enhancements of the activity
and the ethylbenzene selectivity were observed for Cs2O/Cs-X. The
strong basic sites of Cs2O/Cs-X were reported to be O atoms of Cs2O
particles occluded in the cavities of Cs-X based on the 135Cs NMR
study [30].
3.3.2. Effects of carrier gas, N2 and H2
The toluene conversions and styrene selectivities in a nitrogen
stream and a hydrogen streams over Cs-X and Cs2O(4.1)/Cs-X are
given in Table 3. The toluene conversion did not change much, and
the styrene selectivity decreased in a hydrogen stream as com-
pared to those in a nitrogen stream. The results indicate that the
hydrogenation of styrene with hydrogen takes place. However,
taking account of the results given in Table 2 that the transfer
hydrogenation of styrene is much faster than the hydrogenation,
the contribution of the hydrogenation of styrene is not large as
compared to that of the transfer hydrogenation to ethylbenzene
formation.
3.4. Effects of modification with Cs2O and with ZrB2O5 borate on
the activity and selectivity
The reaction of toluene with methanol to form styrene and
ethylbenzene is composed of several steps in which basic sites
and acidic sites are involved in different ways as shown in Fig. 7.
Although it is speculative in some parts at present, the reaction is
supposed to proceed as follows. Methanol undergoes dehydrogena-
tion to form formaldehyde (step 1) and further dehydrogenation to
Modification of Cs-X with ZrB2O5 caused enhancements of both
the activity and the styrene selectivity. Changes in the strength of
basic sites and acidic sites caused by the modification with ZrB2O5
borate could not be detected by IR of adsorbed CO2 and pyridine. It
was observed by SEM that small particles of ZrB2O5 are dispersed
on the surfaces of Cs-X and Cs2O/Cs-X crystallites. ZrB2O5 is mostly
in the form of the particle, though penetration of ZrB2O5 into the
cavities could not be excluded.
Table 3
Effects of carrier gas on the toluene conversion and styrene selectivity in side-chain
alkylation of toluene with methanol.
Catalyst
Carrier gas
Toluene conv. (%)
Styrene select. (%)
Cs-X
Cs-X
Cs2O(4.1)/Cs-X
Cs2O(4.1)/Cs-X
N2
H2
N2
H2
1.0
1.2
3.3
3.0
80.1
73.8
12.2
8.2
In the preceding paper, we reported that formation of uniden-
tate formate species was observed by IR when methanol was

W.O. Alabi et al. / Catalysis Today 226 (2014) 117–123
123
adsorbed on ZrB/Cs-X. It was suggested that the presence of the
Acknowledgements
unidentate formate indicated that formaldehyde is present in a
high concentration. ZrB2O5 form formaldehyde from methanol.
In the present study, however, the formation of unidentate for-
mate could not be detected when methanol was adsorbed on
ZrB(10)/Cs2O(4.1)/Cs-X at 623 K followed by cooling down to
room temperature, the same condition we could observe the
unidentate formate on ZrB/Cs-X. The formaldehyde formed on
the dispersed ZrB2O5 particles might undergo further dehydro-
genation over strong basic sites of Cs2O/Cs-X, and undetectable
by IR. The enhancement of activity by modification is sug-
gested to be due to the formation of formaldehyde on the
ZrB2O5 dispersed on the external surface of the Cs2O/Cs-X
crystallites.
Enhancement of styrene selectivity caused by the modifica-
tion of Cs2O/Cs-X with ZrB2O5 results from the suppression of the
transfer hydrogenation of styrene to ethylbenzene. The suppres-
sion effect cannot be interpreted by the existence of the dispersed
ZrB2O5 particles on the Cs2O/Cs-X surfaces. Some chemical inter-
action should occur between ZrB2O5 or elements of ZrB2O5 and
the active sites for the transfer hydrogenation. There are, how-
ever, no data to show such interaction at present. The mechanism
of the suppression effect of ZrB2O5 is an issue to be elucidated in
future.
The authors are grateful to King Abdulaziz City for Science &
Technology (KACST) for financial support of this research through
Project #AR-30-253. The authors also appreciate the support from
the Ministry of Higher Education, Saudi Arabia in establishment
of the Center of Research Excellence in Petroleum Refining and
Petrochemicals at KFUPM.
References
[1] Y.N. Sidorenko, P.N. Galich, V.S. Gutrya, V.G. Ill’in, I.E. Neimark, Dokl. Akad. Nauk
SSSR 173 (1967) 132–134.
[2] N. Edwards, V. Vadhri, M. Mukherjee, ACS annual meeting, Session, (MON-PM)
SciMix, Div. I&EC, Poster No. 19143, Philadelphia, August, 2012.
[3] T. Yashima, K. Sato, T. Hayasaka, N. Hara, J. Catal. 26 (1972) 303–312.
[4] M.L. Unland, J. Phys. Chem. 82 (1978) 580–583.
[5] S.T. King, J.M. Garces, J. Catal. 104 (1987) 59–70.
[6] P.E. Hathaway, M.E. Davis, J. Catal. 119 (1989) 497–507.
[7] X. Wang, G. Wang, D. Shen, C. Fu, M. Wei, Zeolites 11 (1991) 254.
[8] W.S. Wieland, R. Davis, J.M. Garces, Catal. Today 28 (1996) 443–450.
[9] A.E. Palomares, G. Rder-Mirth, J.A. Lercher, J. Catal. 168 (1997) 442–449.
[10] W.S. Wieland, R.J. Davis, J.M. Garces, J. Catal. 173 (1998) 490–500.
[11] C. Lacroix, A. Deluzarche, A. Kinnemann, A. Boyer, Zeolite 4 (1984) 109–111.
[12] M.L. Unland, G.E. Baker, in: W.R. Moser (Ed.), Catalysis in Organic reactions,
Marcel Dekker, New York, Basel, 1981, p. 51.
[13] H.-C. Lin, R.J. Spohn, US Pat. 4,483,936, Exxon Res. Eng. (1984).
[14] H.-C. Liu, US Pat. 4,483,937, Exxon Res. Eng. (1984).
[15] D. Barthomeuf, V.Q. deQuirillic, US Pat. 5,068,483, CNRS (1991).
[16] B.B. Tope, W.O. Alabi, A.M. Aitani, H. Hattori, S.S. Al-Khattaf, Appl. Catal. A
443/444 (2012) 214–220.
[17] P.E. Hathaway, M.E. Davis, J. Catal. 116 (1989) 263–276.
[18] S.V. Bordawekar, R. Davis, J. Catal. 189 (2000) 79–90.
[19] H. Tsuji, F. Yagi, H. Hattori, Chem. Lett. 20 (1991) 1881–1884.
[20] E.J. Doskocil, R.J. Davis, J. Catal. 188 (1999) 353–364.
[21] J.C. Kim, H.-X. Li, C.-Y. Chen, M.E. Davis, Microporous Mater. 2 (1994) 413–423.
[22] J. Li, R.J. Davis, Appl. Catal. A 239 (2003) 59–70.
[23] F. Yagi, H. Tsuji, H. Hattori, Microporous Mater. 9 (1997) 237–245.
[24] I. Rodoriguez, H. Cambon, D. Brunel, M. Lasperas, P. Geneste, Stud. Surf. Sci.
Catal. 78 (1993) 623–630.
[25] I. Rodrigues, H. Cambon, D. Brunel, M. Lasperas, J. Mol. Catal. A 130 (1998)
195–202.
[26] U. Rymsa, M. Hunger, H. Knoezinger, J. Weitkamp, Stud. Surf. Sci. Catal. 125
(1999) 197–204.
[27] J. Engelhart, J. Szanyi, J. Valyon, J. Catal. 107 (1987) 296–306.
[28] Y. Kito, T. Uda, A. Kashiwaya, JP 57-68144(A) (1982).
[29] J.S. Knyrim, H. Huppertz, Z. Naturforsch. B: Chem. Sci. 63 (2008) 707–712.
[30] F. Yagi, N. Kanuka, H. Tsuji, S. Nakata, H. Kita, H. Hattori, Microporous Mater. 9
(1997) 229–235.
4. Conclusions
1. Modification of Cs-X with Cs2O enhances the activity of the
side-chain alkylation, which is caused by generation of strong
basic sites to efficiently deprotonate toluene to form benzyl
anion. Addition of ca. 1–5 wt% of Cs2O to Cs-X gives the highest
activity.
2. Modification of Cs-X with Cs2O reduced the styrene selectiv-
ity, because the modification with Cs2O facilitates the transfer
hydrogenation of styrene to ethylbenzene which is the main
pathway to form ethylbenzene.
3. Modification of Cs2O/Cs-X with ZrB2O5 enhances the activity,
optimum amount of ZrB2O5 being 10 wt%. Enhancement of the
activity by addition of ZrB2O5 is caused by facile formation of
formaldehyde from methanol by ZrB2O5.
4. Modification of Cs2O/Cs-X with ZrB2O5 enhances the styrene
selectivity, which is caused by suppression of the transfer hydro-
genation of styrene to ethylbenzene. The styrene selectivity
increases with an increase in ZrB2O5 loading up to 15 wt%.
[31] M.L. Hair, Infrared Spectroscopy in Surface Chemistry, Marcel Dekker, New
York, 1967, pp. 204.
[32] Y. Ono, H. Hattori, Solid Base Catalysis Springer Series in Chemical Physics, 101,
Springer-Verlag/Tokyo Institute of Technology Press, Berlin/Heidelberg, 2011.
[33] J.M. Garces, G.E. Vrieland, S.I. Bates, F.M. Scheidt, Stud. Surf. Sci. Catal. 20 (1985)
67.