Silicalite-1 as efficient catalyst for production of propene from 1-butene

Article abstract of DOI:10.1021/cs5009255

Reaction of 1-butene was studied over silicalite-1 and H-ZSM-5 zeolites with different Al contents (SiO₂/Al₂O₃ molar ratio (Si/Al₂) = 23, 80, and 280) to explore an efficient catalyst for the formation of propene as well as to elucidate the reaction scheme and the relevant acid sites involved in the reaction. The formation of alkenes, including propene, increased and those of alkanes and aromatics decreased with decreasing Al content. The percentage of alkenes other than n-butene isomers was 60 C-wt % over silicalite-1 at 550 °C with 34.1 C-wt % propene. Over H-ZSM-5 with Si/Al₂ = 23, the formation of alkenes was negligible, and the main products were alkanes and aromatics, the sum of alkanes and aromatics being 65.4 C-wt % at 550 °C. These product distributions are consistently interpreted by the successive reactions of oligomerization, cracking, and hydrogen transfer. For oligomerization and cracking, in addition to strong acid sites on H-ZSM-5 zeolites, weak acid sites present on silicalite-1 act as active sites. For the hydrogen transfer reaction of alkenes to form alkanes and aromatics, strong acid sites are required. The scheme can also be applicable to the reactions of 1-pentene and 1-hexene. The weak acid sites on silicalite-1 are assumed to be the silanol groups that act as Bronsted acid above 300 °C. The presence of strong acid sites on H-ZSM-5 catalysts, which are the OH groups bridging to Si and Al, results in the consumption of alkenes by hydrogen transfer. Removal of a part of Al contained in silicalite-1 as an impurity and enrichment of surface silanol groups on silicalite-1 resulted in the improvement of the propene yield. It is concluded that silicalite-1 is an efficient catalyst for the formation of propene by the reactions of light alkenes because of the absence of strong acid sites and the presence of weak acid sites.

Full text of DOI:10.1021/cs5009255

Research Article
pubs.acs.org/acscatalysis
Silicalite‑1 As Efficient Catalyst for Production of Propene from
1‑Butene
Palani Arudra, Tazul Islam Bhuiyan, Muhammad Naseem Akhtar, Abdullah M. Aitani,
Sulaiman S. Al-Khattaf,* and Hideshi Hattori
Center of Research Excellence in Petroleum Refining and Petrochemicals, King Fahd University of Petroleum & Minerals,
Dhahran 31261, Saudi Arabia
S
* Supporting Information
ABSTRACT: Reaction of 1-butene was studied over silicalite-1 and
H-ZSM-5 zeolites with different Al contents (SiO2/Al2O3 molar
ratio (Si/Al2) = 23, 80, and 280) to explore an efficient catalyst for
the formation of propene as well as to elucidate the reaction scheme
and the relevant acid sites involved in the reaction. The formation of
alkenes, including propene, increased and those of alkanes and
aromatics decreased with decreasing Al content. The percentage of
alkenes other than n-butene isomers was 60 C-wt % over silicalite-1
at 550 °C with 34.1 C-wt % propene. Over H-ZSM-5 with Si/Al2 =
23, the formation of alkenes was negligible, and the main products
were alkanes and aromatics, the sum of alkanes and aromatics being
65.4 C-wt % at 550 °C. These product distributions are consistently interpreted by the successive reactions of oligomerization,
cracking, and hydrogen transfer. For oligomerization and cracking, in addition to strong acid sites on H-ZSM-5 zeolites, weak
acid sites present on silicalite-1 act as active sites. For the hydrogen transfer reaction of alkenes to form alkanes and aromatics,
strong acid sites are required. The scheme can also be applicable to the reactions of 1-pentene and 1-hexene. The weak acid sites
on silicalite-1 are assumed to be the silanol groups that act as Brønsted acid above 300 °C. The presence of strong acid sites on
H-ZSM-5 catalysts, which are the OH groups bridging to Si and Al, results in the consumption of alkenes by hydrogen transfer.
Removal of a part of Al contained in silicalite-1 as an impurity and enrichment of surface silanol groups on silicalite-1 resulted in
the improvement of the propene yield. It is concluded that silicalite-1 is an efficient catalyst for the formation of propene by the
reactions of light alkenes because of the absence of strong acid sites and the presence of weak acid sites.
KEYWORDS: silicalite-1, H-ZSM-5, 1-butene, propene, alkene cracking, hydrogen transfer, weak acid
Zhu et al. studied the effects of zeolite pore structure and
Si/Al2 ratio on the cracking of C4 alkenes to propene.11 A high
selectivity to propene was observed for the medium-pore
10-membered ring zeolites and the small-pore SAPO zeolite.
The smaller the pore size of the zeolites, the greater the extent
of suppression of the hydrogen transfer reaction of alkenes and
the higher the propene selectivity. The H-ZSM-5 with an Si/Al2
ratio of 366 exhibited the best performance. Lin et al. reported
that the H-ZSM-5 modified with phosphorus to reduce the
number of strong acid sites showed the best performance in the
reaction of 1-butene to propene.14 They suggested that
adjusting the acid site distribution is important. Zeng et al.
also reported that the phosphorus-modified PITQ-13 yielded a
high propene yield of 41.6 C-wt %.15 They reported that the
modification with phosphorus weakened the acid sites to result
in the high yield of propene. Recently, Epelde et al. reported
that K-modified H-ZSM-5 showed a high yield of propene in
the reaction of 1-butene. They explained the role of K to hinder
ighly selective propene production has received much
Hattention in recent years because of its use in important
derivatives of polymers, intermediates, and chemicals. The
demand for propene is growing very rapidly, primarily driven
by a high growth rate of polypropylene usage.1
Several propene production technologies have been inves-
tigated including methanol-to-olefin process, steam cracking,
propane dehydrogenation, catalytic cracking of C4-alkenes, and
metathesis.2 Among these methods, the process of butene
cracking has attracted much attention because of the availability
of large and stable supplies of butene isomers from FCC and
stream cracking processes.3−5 Cracking of C4 to C6 alkenes
appears to be a promising technology for production of
propene and ethylene.6,7
The catalysts studied so far for the purpose of propene pro-
duction by cracking include H-ZSM-5, H-ZSM-48, PITQ-13,
and SAPO-34.8−15 H-ZSM-5 showed a high propene yield of
35% after modification with phosphorus and lanthanum.8
H-ZSM-48 showed a higher propene selectivity in C4-alkene
cracking compared with H-ZSM-5, although the reasons for the
high selectivity were not certain.9
Received: February 27, 2014
Revised: October 8, 2014
© XXXX American Chemical Society
4205
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
the second reactions of aromatization and hydrogen transfer
by reducing the density and strength of acid sites.16
Because silicalite-1 was prepared with NH4F, a trace amount
of F might be included in the resulting silicalite-1. To examine
the effect of F, we prepared silicalite-1 without using a com-
pound containing F according to the literature.19 In a typical
synthesis, 6 g of fumed silica was mixed with tetrapropylam-
monium hydroxide (TPAOH) and water. The resulting mixture
was crystallized for 4 days at 90 °C. The composition of the gel
was 1.0 SiO2/0.085 TPAOH/3.72 H2O. After crystallization,
the product was mixed with water and centrifuged. The ob-
tained solid was dried at 90 °C for 12 h and calcined at 750 °C
for 6 h. The silicalite-1 prepared by this method was denoted by
silicalite-1(non-F).
1.3. Catalyst Characterization. The elemental analysis for
Si and Al was performed by ICP-OES with an optical emis-
sion spectrometer, Ultima 2, Horiba Scientific. In a typical
procedure, 50 mg of catalyst sample was fused with 300 mg of
lithium metaborate in a muffle furnace at 950 °C for 15 min.
The fused product was dissolved in 20 mL of 4% HNO3 and
diluted with deionized water to make a total volume of 50 mL.
The obtained solution was analyzed by ICP-OES.
The F content in silicalite-1 was measured by ion-exchange
chromatograpy. Silicalite-1 (50 mg) was fused with lithiumme-
taborate, dissolved into 50 mL of KOH aqueous solution and
analyzed by ion-exchange chromatography. The content of F in
silicalite-1 was 0.000 01 wt % (0.1 ppm).
The zeolite samples were characterized by X-ray powder
diffraction (XRD) with a Rigaku Mini-flex II system using
nickel filtered Cu Kα radiation (λ = 1.5406 Å, 30 kV, and
15 mA). The XRD patterns were recorded in static scanning
mode from 1.2 to 50° (2θ) at a detector angular speed of
2° min−1 with a step size of 0.02°. The crystal size was
estimated by Scherrer’s method for the peak at 2θ ∼ 7.9° for
diffraction of the (011) face.
Silicalite-1 has the same crystalline structure as ZSM-5, but
contains no Al. The acid sites on silicalite-1 are so weak that the
acid sites cannot be detected by ammonia TPD under normal
conditions. Because alkenes are easily protonated to form
carbocations, the weak acid sites on silicalite-1 would be able to
form carbocations to initiate the acid-catalyzed reactions.
Because the strength of acid sites increases with temperature,
the formation of carbocations from alkene would be plausible
over silicalite-1 in particular at a high temperature. To the best
of our knowledge, there have been no reports in the literature
in which silicalite-1 was used as a catalyst in catalytic cracking
of light alkenes except for one patent.17 In the present study,
silicalite-1 was synthesized, and its catalytic activity was
examined in the reactions of C4−C6 alkenes. It was found
that silicalite-1 exhibited a higher yield of propene as compared
with H-ZSM-5 zeolites with a SiO2/Al2O3 molar ratio (Si/Al2)
of 23−280.
1. EXPERIMENTAL METHODS
1.1. Reagents. Tetrapropylammonium bromide (98%),
ammonium fluoride (98%), fumed silica (Cab-O-sil M-5)
(99.8%), 1-pentene (98.5%), and 1-hexene (97%) were
purchased from Sigma-Aldrich and used without further
purification. 1-Butene (99.9%) was procured from Saudi
Industrial Gas Company.
1.2. Catalyst Synthesis. H-ZSM-5 zeolites with Si/Al2
molar ratios of 23, 80, and 280 were obtained by calcination of
the ammonium forms of the ZSM-5 zeolites purchased from
Zeolyst International at 500 °C for 2 h in air and are denoted
by H-ZSM-5(23), H-ZSM-5(80), and H-ZSM-5(280), respec-
tively.
The surface areas of the catalyst sample were measured
by nitrogen adsorption at −195 °C with Autosorb-1 (Quanta
Chrome), the BET equation being applied to the isotherm. The
external surface area was estimated from a t-plot.
SEM images were measured with a JEOL JSM-5800 scanning
microscope. Magnification was 7000. Before taking SEM
photographs, the samples were loaded onto a sample holder,
held with conductive aluminum tape, and coated with a film of
gold in vacuo using a cressington sputter ion-coater for 20 s
with 15 mA current.
The IR spectra of pyridine adsorbed on the sample were
measured for Brønsted acid sites and Lewis acid sites. The thin
wafer sample was pretreated at 450 °C in a vacuum. After
cooling to room temperature, the sample was exposed to pyr-
idine at room temperature for 5 min and evacuated at 150 °C
for 30 min. IR spectra were measured at room temperature.
IR spectra of the silicalite-1, silicalite-1(HCl), and silicalite-
1(NH3) in the OH stretching region were measured for the
wafer samples disks prepared with KBr because self-supporting
wafer transparent to IR light could not be made without KBr.
The sample in the form of a self-supporting wafer was prepared
using 15 mg of sample mixed with 85 mg of KBr. The diameter
of wafer was ∼20 mm. The wafer was then placed in an in situ
cell (Makuhari Rikagaku Garasu Inc., JAPAN) and pretreated
under a vacuum (∼2 × 10−1 Pa) at 300 °C for 1 h. All spectra
were measured at room temperature with Nicolet FTIR spec-
trometer (Magna 500 model).
Silicalite-1 was synthesized according to the procedures
reported in the literature.18 In a typical synthesis, 4.26 g of
tetrapropylammonium bromide (TPABr) and 0.74 g of
ammonium fluoride (NH4F) were dissolved into 72 mL of
water. Then, 12 g of fumed silica was added, and the mixture
was stirred until a homogeneous gel was formed. The gel was
subjected to a hydrothermal crystallization process at 200 °C
for 2 days. The molar composition of the gel was 1 SiO2/0.08
TPABr/0.10 NH4F/20 H2O. The gel was washed with water
and dried at 80 °C overnight. The template was removed by
calcination at 750 °C for 6 h in air.
Because silicalite-1 normally contains a trace amount of Al
that originates from impurity in a Si source, we prepared the
HCl-treated silicalite-1, which is supposed to have a smaller
amount of Al. For the HCl-treated silicalite-1, silicalite-1 (5 g)
was mixed with 100 g of 1 M HCl aqueous solution. The
mixture was stirred for 24 h under reflux conditions. The
sample was filtered, washed with deionized water, and dried at
100 °C for 4 h, followed by calcination at 550 °C for 5 h in air.
The HCl-treated silicalite-1 was denoted by silicalite-1(HCl).
We also prepared NH3-treated silicalite-1, which is supposed
to have a larger number of the surface OH groups. For the
NH3-treated silicalite-1, silicalite-1 (4 g) was mixed with 20 g
of a mixture of aqueous solutions of ammonia (10 g, 25 wt %)
and ammonium nitrate (10 g, 7.5 wt %) in a glass beaker. This
mixture was transferred into a polypropylene bottle and stirred
at 90 °C under autogenous pressure for 1 h. The catalyst was
then washed with deionized water, filtered, and dried at 110 °C
for 4 h, followed by calcination at 550 °C for 5 h in air. The
NH3 treated sample was denoted by silicalite-1(NH3).
Temperature-programmed desorption (TPD) of ammonia
was performed with a BEL-CAT-A-200 chemisorption apparatus.
The measurement conditions were the same as those proposed
4206
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
by Katada et al.20 The sample (100 mg) was pretreated at 500 °C
under a helium flow for 60 min, followed by exposure to
ammonia at 100 °C. The ammonia excessively adsorbed was
flushed by a helium stream for 30 min at 100 °C. TPD was run
from 100 to 600 °C at a ramp rate of 10 °C min−1. Those TPD
measurement conditions are suitable to measure the strong acid
sites on zeolites. To measure the weak acid sites, three experi-
ments in which the flushing temperature was set at 25 °C were
carried out for silicalite-1, silicalite-1(HCl), and silicalite-1(NH3).
1.4. Reaction Procedures. The catalytic cracking of
C4−C6 alkenes was performed in a fixed-bed tubular reactor
packed with 2 mL of catalyst with a particle size of 0.5−1.0 mm
diameter, which was sandwiched between 6 and 2 mL of SiC
particles. The catalyst sample was pretreated in a nitrogen
stream at 550 °C for 1 h, and then a mixture of the feed and
nitrogen (5 mL/min and 25 mL/min, respectively) (GHSV =
900 h−1) was passed through the catalyst bed at 550 °C. For the
reaction of 1-butene, the reaction temperature was varied in a
range of 50−550 °C. The products were analyzed by online GC
equipped with a GS-Gaspro column and a flame ionization
detector (FID).
2θ = 7−10° and 22−25°.21−24 The relative intensity of diffrac-
tion peaks was different, depending on the sample. This is
caused by a difference in crystal growth for the different
samples. The first peak observed at about 2θ = 7.9° is a
superposition of the diffractions from (−101), (011), and (101)
faces, diffraction from the (011) face being the main peak. The
second peak observed at about 2θ = 8.9° is a superposition of
the diffraction from the (020), (200), (−111), and (111) faces,
the diffractions from the (020) and (200) faces being
dominant. From the similarity of the relative peak intensity,
the crystal growth for silicalite(non-F) is the same as those of
H-ZSM-5’s. Crystal sizes estimated by Scherrer’s method for
the peak at about 2θ = 7.9° (011) face are presented in Table 1.
SEM images of all samples are shown in Figure 2. Particles
were larger as the Si/Al2 ratio increased for H-ZSM-5’s. The
image for H-ZSM-5 showed relatively large quadrangular
prismlike crystallites. Silicalite-1, silicalite-1(HCl), and silica-
lite-1(NH3) showed a mixture of large crystallites and small
particles. Silicalite-1(non-F) was composed of only small
particles. Because Scherrer’s method to obtain crystal size
cannot applied to crystals larger than 0.1−0.2 μm, the values of
crystal size listed in Table 1 do reflect the large crystallites
observed for silicalite-1, silicalite-1(HCl), and silicalite-1(NH3).
The BET and external surface areas of the samples are
presented in Table 1. The surface areas were not much different
for all samples, although the surface areas of H-ZSM-5′s and
silicalite-1(non-F) were ≥400 m2 g−1, and those of silicalite-1,
silicalite-1(HCl) and silicalite-1(NH3) were <400 m2 g−1. A
large difference was observed in the external surface area.
Silicalite-1(non-F) showed a high external surface area of
122 m2 g−1, whereas H-ZSM-5’s showed decreasing surface
areas of 82−33 m2 g−1 with an increase in the Si/Al2 ratio. Low
external surface areas were observed for silicalite-1, silicalite-
1(HCl), and silicalite-1(NH3). This is consistent with the SEM
images; large crystallites were observed for the samples with
low external surface areas.
2. RESULTS AND DISCUSSION
2.1. Catalyst Characterization. XRD patterns of silicalite-1
and H-ZSM-5 zeolites are presented in Figure 1. All samples
exhibited characteristic peaks of MFI structure in the ranges
Figure 3 shows the FT-IR spectra of the catalysts recorded
after the adsorption of pyridine and subsequent evacuation at
150 °C. The absorption bands at 1545 and 1455 cm−1 were
assigned to the pyridinium ion formed on the Brønsted acid site
and pyridine coordinated to Lewis acid sites, respectively.25 The
numbers of Lewis and Brønsted acid sites were highest for
H-ZSM-5(23) as compared with other zeolites (Table 1). With
silicalite-1, no absorption peaks were appreciable. There
remained a negligible amount of pyridine after evacuation of
preadsorbed pyridine at 150 °C.
Figure 1. XRD patterns of H-ZSM-5(23), H-ZSM-5(80), H-ZSM-
5(280), silicalite-1, silicalite-1(HCl), silicalite-1(NH3), and silicalite-
1(non-F).
Table 1. Physicochemical Characterization of H-ZSM-5 with Different Si/Al2 Ratios and Silicalite-1
pyridine FT-IR
e
(mmol g−1
)
b
d
d (011)
crystal size
(Å)
Si/Al2 molar
surface area BET/ext. SA by t-plot
NH3-TPD
a
c
sample
(Å)
ratio
(m2 g−1
)
(mmol g−1
)
TA
B
L
H-ZSM-5(23)
H-ZSM-5(80)
H-ZSM-5(280)
silicalite-1
12.09
11.83
11.89
11.94
12.00
12.00
11.78
426
313
423
457
437
458
341
2.4 × 10
8.6 × 10
425/82
425/68
400/33
0.25
0.16
0.07
nd
0.79
0.14
0.02
nd
0.69
0.08
0.01
nd
0.10
0.06
0.01
nd
2.95 × 102
2.17 × 103
2.50 × 103
2.04 × 103
1.95 × 103
348/5.0
374/6.8
295/4.7
446/122
silicalite-1(HCl)
silicalite-1(NH3)
silicalite-1(non-F)
na
na
na
na
na
na
na
na
na
na
na
na
a
b
c
Evaluated by XRD peak at about 2θ = 7.9. Calculated by Scherrer’s method using peak at about 2θ = 7.9. Si/Al2 molar ratio by ICP analysis.
d
e
Acidity calculated using high-temperature NH3 desorption peak. Absorptivity ratio ε1455/ε1545 = 1.33 was used to calculate the acidity.26 TA = total
acidity, B = Brønsted acidity, L = Lewis acidity, nd = not detectable, na = not available.
4207
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
Figure 2. SEM images of (a) H-ZSM-5(23), (b) H-ZSM-5(80), (c) H-ZSM-5(280), (d) silicalite-1, (e) silicalite-1(HCl), (f) silicalite-1(NH3), and
(g) silicalite-1(non-F).
Figure 3. FT-IR spectra of pyridine adsorbed on H-ZSM-5(23),
H-ZSM-5(80), H-ZSM-5(280), and silicalite-1 and subsequent outgassing
at 150 °C.
Figure 4. FT-IR spectra of silicalite-1 (a) and silicalite-1(NH3) (b) in
the OH stretching region after pretreatment at 300 °C in a vacuum.
Figure 4 shows the FT-IR spectra in the OH stretching
region for silicalite-1, silicalite-1(HCl), and silicalite-1(NH3). A
band at 3725 cm−1 is assigned to the terminal (isolated) silanol
groups, and broad bands in the range 3700−3300 cm−1, to the
vicinal and nest silanol groups.26 It is evident that treatment
with NH3 increased the number of the OH groups on the
surface of silicalite-1, which is in accordance with the results
reported.27 Treatment with HCl, however, did not increase the
number of the OH groups to such an extent as the treatment
with NH3.
TPD profiles of desorbed ammonia are presented in Figure 5.
Two peaks appeared for H-ZSM-5 zeolites. The peaks
appearing at a higher temperature correspond to the ammonia
desorbed from the acid sites of zeolites. The peaks appearing at
a lower temperature were assigned to ammonia molecules
adsorbed either on NH4+ species formed on Brønsted acid sites
4208
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
aromatics, and hydrocarbons heavier than C8. Therefore, the
contribution of the SiC used for supporting the catalyst in the
catalyst bed to the composition of the product was neglected.
In the catalytic reaction, more than 20 hydrocarbons were
produced. To simplify the presentation, the products were
classified into seven groups according to the expected pathway
to form different products. These groups were 1-butene
(reactant), cis- and trans-2-butene (double-bond isomerization
products), propene (target product), alkenes other than
propene and n-butene isomers (skeletal isomerization and
cracking products), alkanes (hydrogen transfer products),
aromatics (benzene, toluene, xylenes, and ethylbenzene;
hydrogen transfer products), and hydrocarbons with a carbon
number >8 (C8+, aromatics other than benzene, toluene,
xylenes, and ethylbenzene, alkanes, and alkenes; oligomeriza-
tion and hydrogen transfer products).
The changes in the compositions of propene and the
aromatics as a function of the time on-stream are shown in
Figure 6 for H-ZSM-5(23) and silicalite-1. The changes in the
Figure 5. NH3-TPD profiles of H-ZSM-5(23), H-ZSM-5(80),
H-ZSM-5(280), and silicalite-1 after flushing excess NH3 at 100 °C.
Inset: TPD profiles of silicalite-1, silicalite-1(HCl), and silicalite-
1(NH3) after flushing excess NH3 at 25 °C.
or on Na+ cations.28 From the areas of the peaks at a higher
temperature, the numbers of acid sites were estimated and are
given in Table 1. The number of acid sites increased with a
decrease in the Si/Al2 ratio of H-ZSM-5 zeolites, which was
expected. No desorption peaks were observed in the high
temperature range for silicalite-1. Because the TPD measure-
ment conditions were suitable for measurement of strong acid
sites present on zeolites, no peak for silicalite-1 indicated that
no acid sites existed on silicalite-1 that were as strong as those
on zeolites.
We measured the ammonia TPD under the conditions that
can detect the ammonia adsorbed on weakly acidic sites. The
temperature at which preadsorbed ammonia molecules were
flushed with a helium flow was lowered to 25 °C, and TPD was
measured for silicalite-1, silicalite-1(HCl), and silicalite-1(NH3).
The TPD profiles are shown in the inset of Figure 5. TPD peaks
appeared at ∼100 °C, which indicated the existence of the sites
that weakly adsorbed ammonia and desorbed at ∼100 °C on the
surfaces of these samples. The numbers of the sites weakly
adsorbing ammonia were 0.028, 0.042, and 0.174 mmol g−1 for
silicalite-1, silicalite-1(HCl), and silicalite-1(NH3), respectively.
ICP-AES analysis showed that the silicalite-1 contained
400 ppm of Al as an impurity. This corresponds to the Si/Al2
molar ratio of 2.17 × 103. There might be a small number of
strong acid sites, but these strong acid sites on silicalite-1 were
below the detection limit by the pyridine-IR and the ammonia-
TPD under normal conditions. The Si/Al2 molar ratios of
silicalite-1(HCl) and silicalite-1(NH3) were 2.50 × 103 and
2.04 × 103, respectively. By treatment with HCl, the Si/Al2
ratio increased only 15%. By treatment with NH3, the Si/Al2
ratio decreased, indicating that desilication occurred to some
extent. The Si/Al2 ratio of silicalite-1(non-F) was 1.95 × 103.
2.2. Catalytic Reactions. 2.2.1. Reaction of 1-Butene.
Because we used SiC for supporting the catalyst in the catalyst
bed, we carried out a blank test in which 8 mL of SiC was
placed in the reactor without catalyst (Supporting Information
Table S1). Only double bond isomerization took place to a
small extent. At a reaction temperature of 250 °C, no reaction
occurred with SiC, whereas the conversion of 1-butene was
85% with silicalite-1. At 550 °C, 1-butene converted only to
trans- and cis-2-butene by 30%, but with silicalite-1, 92% of
1-butene converted to 2-butene isomers, other alkenes, alkanes,
Figure 6. Variations in the yields of propene and aromatics as a
function of time on-stream at 550 °C, GHSV = 900 h−1. Propene over
▲
○
silicalite-1 (Δ) and H-ZSM-5(23) ( ), aromatics over silicalite-1 ( )
●
and H-ZSM-5(23) ( ).
composition with time on-stream were small for the time on-
stream of 1−5 h. The activity decay was small, except for the
production of aromatics over silicalite-1 on which the
production of aromatics decreased during the initial 3 h.
The compositions of the grouped products are plotted
against the reaction temperature and shown in Figures 7−13
for H-ZSM-5(23), H-ZSM-5(80), H-ZSM-5(280), silicalite-1,
silicalite-1(HCl) and silicatlite-1(NH3), and silicalite-1(non-F),
respectively. The changes in the composition with reaction
temperature varied with the type of catalysts. Numerical data
are presented in the Supporting Information (Table S2−S8).
Over H-ZSM-5(23), double bond isomerization mainly
occurred in the temperature range 50−200 °C. At 250 °C,
the main products became alkenes, and C8+. As the reaction
temperature increased to 350 °C, the formation of alkenes
became very small, and the formation of alkanes and aromatics
became dominant. The dominant formation of alkanes and
aromatics continued up to 550 °C, and the sum of alkanes
and aromatics reached 65.4%. The formation of propene was
4209
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
Figure 7. Variation in the composition of the products as a function
of temperature for the reaction of 1-butene over H-ZSM-5(23).
Aromatics, (benzene + toluene + xylene + ethylbenzene); C8+
(hydrocarbons containing more than 8 carbons, including aromatics,
alkenes, and alkanes) GHSV = 900 h−1; time on-stream = 1 h; total
pressure, atmospheric pressure.
Figure 9. Variation in the composition of the products as a function of
temperature for the reaction of 1-butene over H-ZSM-5(280).
Expressions of compounds and reaction conditions are the same as
those in Figure 7.
Figure 10. Variation in the composition of the products as a function
of temperature for the reaction of 1-butene over silicalite-1.
Expressions of compounds and reaction conditions are the same as
those in Figure 7.
Figure 8. Variation in the composition of the products as a function of
temperature for the reaction of 1-butene over H-ZSM-5(80).
Expressions of compounds and reaction conditions are the same as
those in Figure 7.
differences were apparent. In the temperature range 50−200 °C,
double-bond isomerization occurred faster over silicalite-
1(HCl) than over silicalite-1. Above 250 °C, the changes in
composition as a function of temperature were similar to those
for silicalite-1. Over silicalite-1(HCl), the yield of propene
reached 35.6 C-wt % at 550 °C. The formation of C8+,
including alkenes, alkanes, and aromatics was smaller over
silicalite-1(HCl) than over silicalite-1.
Over silicalite-1(NH3), the changes in composition with
temperature were similar to those over silicalite-1 and silicalite-
1(HCl). In the temperature range 50−200 °C, double-bond
isomerization occurred faster over silicalite-1(NH3) than over
silicalite-1, but at a rate similar to that over silicalite-1(HCl).
Above 250 °C, the changes in composition as a function of
temperature were similar for silicalite-1(NH3), silicalite-
1(HCl), and silicalite-1. Over silicalite-1(NH3), the yield of
observed in the temperature range 250−350 °C, but the
compositions were <5%.
Over silicalite-1, no reaction was appreciable at 50 °C. In the
range 100−200 °C, only double-bond isomerization occurred.
At 250 °C, the main products were still 2-butene isomers. As the
temperature increased to 300 °C, the formation of alkenes
became predominant, and the formation of C8+ occurred to some
extent. The formations of alkanes and aromatics were limited to
small percentages. As the reaction temperature increased further,
the compositions of alkenes including propene, C8+, total alkanes
and aromatics did not change much; however, the percentage of
propene in alkenes steadily increased and reached a maximum
percentage of 34.1% of the total products at 550 °C.
Over silicalite-1(HCl), the changes in composition with tem-
perature were similar to those over silicalite-1, although small
4210
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
Figure 11. Variation in the composition of the products as a function
of temperature for the reaction of 1-butene over silicalite-1(HCl).
Expression of compounds and reaction conditions are the same as
those in Figure 7.
Figure 13. Variation in the composition of the products as a function
of temperature for the reaction of 1-butene over silicalite-1(non-F).
Expressions of compounds and reaction conditions are the same as
those in Figure 7.
of H-ZSM-5(23) and silicalite-1. The change for H-ZSM-5(80)
was close to that observed for H-ZSM-5(23), and the change
for H-ZSM-5(280) was close to that observed for silicalite-1.
The formations of alkanes and aromatics were dominant for
H-ZSM-5(80) in the temperature range 300−550 °C, whereas
the formations of alkenes, including propene, were dominant
for H-ZSM-5(280). The percentage of propene in alkenes over
H-ZSM-5(280) increased with the reaction temperature up to
500 °C. The maximum propene formation with H-ZSM-5(280)
was 22.8% at 500 °C, which was close to those of alkanes and
aromatics.
The changes in product distribution as a function of reaction
temperature shown in Figures 7−13 suggest that the reactions
of 1-butene proceed according to the pathway shown in
Scheme 1. 1-Butene is protonated by Brønsted acid site to form
Scheme 1. Simplified Reaction Pathway of Reaction of
1-Butene over H-ZSM-5 and Silicalite-1
Figure 12. Variation in the composition of the products as a function
of temperature for the reaction of 1-butene over silicalite-1(NH3).
Expressions of compounds and reaction conditions are the same as
those in Figure 7.
propene reached 39.1 C-wt % at 550 °C, the highest values
among those for four silicalite-1 catalysts. The formation of
C8+, including alkenes, alkanes, and aromatics was smaller over
silicalite-1(NH3) than over silicalite-1.
Over silicalite-1(non-F), the general trend of the changes in
composition with temperature was similar to those for the
other silicalite-1’s. The formation of C8+ was not extended, as
compared with that observed for silicalite-1 in the temperature
range 350−550 °C, and the formations of aromatics and
alkanes were not extended, as compared with those for
silicalite-1(HCl) and silicalite-1(NH3) in the temperature
range 400−550 °C. The maximum yield of propene over
silicalite-1(non-F) reached 31.6 C-wt % at 550 °C, which was a
little smaller than those obtained for the other silicalite-1’s.
Over H-ZSM-5(80) and H-ZSM-5(280), the changes in the
composition with the reaction temperature were between those
2-butyl cation. The 2-butyl cation converts to trans- and cis-2-
butene by deprotonation, octane and dodecene isomers by
oligomerization and isobutene by skeletal isomerization. Octene
and dodecene isomers undergo cracking to form alkenes. Alkenes
undergo hydrogen transfer to form alkanes and aromatics.
Among these reactions, the double-bond isomerization is the
easiest reaction. The reaction proceeds at a low temperature.
The equilibrium conversions of 1-butene to 2-butenes are 93.1,
87.7, and 83.2% at 100, 200, and 250 °C, respectively.29 The
1-butene conversions over H-ZSM-5(23) at 100 °C, over
silicalite-1(NH3) at 200 °C, and over silicalite-1 at 250 °C were
4211
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
89, 83, and 83%, respectively. The temperatures at which the
conversions closely reached the equilibrium conversions were
100, 200, and 250 °C for H-ZSM-5(23), silicalite-1(NH3), and
silicalite-1, respectively.
from the raw materials perhaps in fumed silica. The amount of
Al in silicalite-1 was 2 orders of magnitude smaller than that in
H-ZSM-5(23). The activities of silicalite-1 and H-ZSM-5(23)
were also largely different; the conversions of 1-butene at 50 °C
were 0% for silicalite-1 and 45% for H-ZSM-5(23). The
correlation between the Al content and the activity could not
be obtained properly. Although the possibility that the strong
acid sites originating from the trace Al act as active sites on
silicalite-1 could not be excluded, it is worth considering the
other possibility, that weakly acidic OH groups on silicalite-1
act as active sites.
It is known that four types of OH groups exist on the surface
of silicalite-1. These are terminal silanol, geminal silanol, vicinal
silanol, and nest silanol.30,31 The strength of these OH groups
was reported to be in the order, terminal < germinal< vicinal <
nest.30,31 It was reported that the pKa of the terminal silanol on
silica was 4.9 at room temperature.32,33 The OH groups on
silicalite-1 are assumed to increase their acid strength as the
temperature increases and act as Brønsted acid at a high
temperature. Actually, the nest silanol groups are recognized to
be active sites for an acid-catalyzed reaction of Beckmann re-
arrangement of cyclohexanone oxime to ε-caprolactam carried
out in the temperature range 350−400 °C.26,29−31
Because alkanes and aromatics formed to a large extent at a
higher temperature over H-ZSM-5(23) possessing strong acid
sites, it can be assumed that strong acid sites are required
for the hydrogen transfer reaction. The hydrogen transfer
reaction forming alkanes and aromatics consists of several
steps: a protonation step, an oligomerization step, three hydride
transfer steps, three proton transfer steps, one cyclization step,
and one ring expansion step in the case of the reaction of
1-butene to butane and xylene (Supporting Information
Scheme S1). The proton transfer step may be either to alkenes
to form carbocation or to the surface to restore the original
Brønsted acid sites. The active sites for the hydrogen transfer
have not been studied extensively; however, it is plausible to
assume that the existence of strongly acidic sites close to each
other promotes the reaction efficiently because the proton
transfer steps of the former type and hydride transfer steps
involved in the hydrogen transfer are the bimolecular reactions
between the molecules formed on acid sites and the
carbocations adsorbed on the other acid sites. Accordingly, it
is assumed that a high concentration of strong acid sites favors
the hydrogen transfer reaction. Although it is not certain which
step is the slow step among these steps, the slow step should
need strong acid sites.
It is noted for silicalite-1 that the percentage of propene
increased with the reaction temperature, even though the per-
centage of total alkenes did not change much in the tem-
perature range 300−550 °C. A similar tendency was observed
for H-ZSM-5(280), silicalite-1(HCl), silicalite-1(NH3), and
silicalite-1(non-F). Although the reason for this tendency is not
certain, one of the possible reasons is as follows: The alkenes
formed by skeletal isomerization, oligomerization, and cracking
underwent further skeletal isomerization, oligomerization, and
cracking repeatedly. Propene is less reactive than the other
alkenes with C4 or higher for cracking because the secondary
propyl cation has no β-C−C bond to be cleaved by β-scission.
It is plausible that propene tended to remain unreacted and
increased its concentration with repeating skeletal isomer-
ization, oligomerization, and cracking of alkenes. The other
possibility is a high diffusivity of propene as compared with
higher alkenes. Propene formed by cracking in the cavities of
silicalite-1 or H-ZSM-5 could be diffused out of the cavities
more rapidly as compared with higher alkenes before
undergoing further oligomerization and cracking.
To examine which is more plausible as active sites for
propene formation, strong acid sites originating from a trace
amount of Al in silicalite-1 or the surface silanol groups, we
treated silicalite-1 with HCl to dealuminate and with NH3 to
enrich the surface silanol groups. It is anticipated that if the
strong acid sites act as the active sites, the propene yield would
be decreased by HCl treatment, and if the surface silanol
groups act as active sites, the propene yield would be increased
by NH3 treatment.
By HCl treatment, 15% of Al in silicalite-1 was removed.
The propene yield at 550 °C over the dealuminated silicalite-1
(silicalite-1(HCl)) was increased to a small extent (34.1 to 35.6
C-wt %). By NH3 treatment, the surface OH groups were
enriched, as evidenced by IR. The propene yield at 500 °C over
the NH3 treated silicalite-1 (silicalite-1(NH3)) increased to a
considerable extent (34.1 to 39.1 C-wt %). These results are in
favor of the possibility that silanol groups are relevant to the
formation of propene from 1-butene, although it is not certain
which types of silanol groups are relevant to the reaction.
One might argue that because of the presence of fluorine
in silicalite-1, acidic properties of silanol groups might be
strengthened to show the catalytic activity. The catalytic activity
of silicalite-1(non-F) was similar to those of silicalite-1,
silicalite-1(HCl), and silicalite-1(NH3) in the sense that the
products composed mainly of alkenes and the formations of
alkanes and aromatics were small in the temperature range
350−550 °C. The presence of fluorine in silicalite-1 does not
much affect the catalytic activity. It is suggested that the silanol
groups on silicalite-1 act as an active site without fluorine.
Silicalite-1(non-F) has a morphology different from that of
silicalite-1, silicalite-1(HCl), and silicalite-1(NH3), as shown in
SEM images (Figure 2). No large crystallites were observed for
silicalite-1(non-F). In contrast, large crystallites were observed
for slicalite-1, silicalite-1(HCl), and silicalite-1(NH3); how-
ever, the crystal size estimated from Scherrer’s method was not
greatly different for these four silicalite-1’s. A large difference
between silicalite-1(non-F) and the other silicalite-1’s was
observed in external surface area. The external surface area was
about 20 times larger for silicalite-1(non-F) than for the other
silicalite-1’s. Regardless of the large differences in morphology,
2.2.2. Active Sites on H-ZSM-5 and Silicalite-1. It is well
established that the Brønsted acid sites of H-ZSM-5 zeolites are
the OH groups bridging to Si and Al atoms of the zeolite
framework. Ammonia molecules adsorbed on the bridged OH
groups are desorbed around 300−500 °C in TPD. These acid
sites are strong and catalyze many acid-catalyzed reactions. The
active sites of H-ZSM-5(23) for the reaction of 1-butene should
be the bridged OH groups. These acid sites catalyzed double-
bond isomerization in the temperature range 50−150 °C and
above; oligomerization, skeletal isomerization; and cracking at
200 °C and above; and hydrogen transfer reaction at 250 °C
and above.
On the surface of silicalite-1, no strongly acidic OH groups
were observed. In the TPD of ammonia under the conditions
normally employed, no desorption peaks appeared. A trace
amount of Al was contained in silicalite-1, which originated
4212
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
Table 2. Compositions of the Products in C-wt % for the Reactions of 1-Butene, 1-Pentene, and 1-Hexene over H-ZSM-5(23),
a
H-ZSM-5(80), H-ZSM-5(280), and Silicalite-1 at 550 °C
H-ZSM-5(23)
H-ZSM-5(80)
H-ZSM-5(280)
silicalite-1
=
=
=
=
=
=
=
=
=
=
=
=
catalyst
1-C4
1-C5
1-C6
0
1-C4
0
1-C5
1-C6
1-C4
2
1-C5
1-C6
1-C4
1-C5
1-C6
=
=
1-C4
2-C4
0
0
0
0
1.1
2.4
1.4
3.0
2.8
6.2
2.4
5.4
6.2
14.3
13.5
9.3
3.5
7.7
6.9
7.4
19.3
48.6
1.9
3.7
1.0
0
3.3
7.2
0.5
0
0.4
0.6
4.9
4.4
0
0.9
1.5
5.4
2.3
=
i-C4
2.3
2.8
5.9
5.0
5.5
=
C2
1.7
0.5
0
3.7
3.3
0
12.5
11.2
0
8.9
10.1
14.1
1.7
16.9
21.1
5.1
12.1
22.6
6.1
11.0
20.0
5.7
4.9
=
C3
11.7
1.4
34.1
7
39.1
10.6
26.6
1.6
=
C5
=
C6
0
0
0
0
0
0
0
0.9
0.8
0.7
alkanes
aromatics
C8+
14.8
50.5
32.6
2.2
0.2
39.3
32.7
20.5
7.0
0.6
45.6
33.9
10.4
9.3
0.6
25.4
40.7
7.9
36.2
31.5
4.6
33.1
28.9
4.9
14.7
22.5
10.2
38.0
0.8
23.5
16.6
3.4
17.1
16.9
15.9
31.0
1.2
5.1
5.3
0.5
4.5
0.9
=
=
C2 + C3
23.7
0.6
20.6
0.9
24.2
0.9
34.7
1.3
43.4
2.4
26.7
1.8
44.0
6.5
=
=
C3 /C2 molar ratio
a
GHSV of 900 h−1 for butenes; LHSV of 6 h−1 for 1-penetene and 1-hexene; time on-stream, 1 h; total pressure, atmospheric pressure.
Figure 14. Variations in the yield of propene as a function of
Figure 15. Variations in the yield of aromatics as a function of the
Al2/Si ratio of catalyst at 550 °C. Aromatics from reactions of 1-butene
( ), 1-pentene ( ), and 1-hexene ( ). The reaction conditions are
the same as those for Figure 14.
Al2/Si ratio of catalyst at 550 °C. Propene from reactions of 1-butene
−1
□
○
( ), 1-pentene (Δ), and 1-hexene ( ) at a GHSV of 900 h for
1-butene and LHSV of 6 h−1 for 1-pentene and 1-hexene; time on-
stream = 1 h; total pressure, atmospheric pressure.
■
●
▲
and silicalite-1 were examined for the reactions of 1-pentene and
1-hexene to see whether the catalytic performances observed for
1-butene were also observed for other alkenes. The reactions
were carried out at 550 °C, and the results are summarized in
Table 2, in which the results of 1-butene reaction are included for
comparison.
The product distributions for the reactions of 1-pentene and
1-hexene were essentially the same as those observed for the
reaction of 1-butene. Over silicalite-1, the yield of propene was
the highest for the reactions of both 1-pentene and 1-hexene
except for the reaction of 1-pentene, in which H-ZSM-5(280)
showed a slightly higher yield (22.6 C-wt %) than silicalite-1
(19.3 C-wt %). In particular, 39.1 C-wt % of propene yield was
obtained for the 1-hexene reaction. The formations of alkanes
and aromatics were very small: <6 wt %. Over H-ZSM-5(23),
on the other hand, the yields of propene were less than 5 wt %
from both 1-pentene and 1-hexene. Large amounts of alkanes
and aromatics were formed from these reactants; the yields of
them exceeded 30 C-wt %.
crystal size, and external surface area among silicalite-1(non-F)
and other silicalite-1’s, the activity and selectivity in the reaction
of 1-butene were similar for all silicalite-1 catalysts. The activity
and selectivity are governed neither by morphology, diffusion in
the cavities, or F-content, nor by concentration of the terminal
silanol groups, which are estimated to be high for silicalite-
1(non-F) having a large external surface area. It is suggested
that the silanol groups other than the terminal silanol groups
act as catalytically active sites.
A hydrogen transfer reaction occurred above 250 °C for
H-ZSM-5(23), but a hydrogen transfer reaction scarcely occurred,
even at 550 °C, for silicalite-1. The high yield of propene over
silicalite-1 results from the fact that silicalite-1 possesses only weak
acid sites that can promote isomerization, oligomerization, and
cracking, but cannot promote a hydrogen transfer reaction. The acid
sites of H-ZSM-5 zeolites are strong enough to promote a hydrogen
transfer reaction in the temperature range 300−550 °C, by which
the formed alkenes are consumed to form alkanes and aromatics.
2.2.3. Reactions of 1- Pentene and 1-Hexene. The catalytic
performances of H-ZSM-5(23), H-ZSM-5(80), H-ZSM-5(280),
On the last line in Table 2, the propene/ethylene molar
ratios are included. It is noted that the ratio increased with an
4213
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214

ACS Catalysis
Research Article
(5) Lu, J.; Zhao, Z.; Xu, C.; Duan, A.; Zhang, P. Catal. Lett. 2006,
109, 65−70.
increase in the Si/Al2 molar ratio of the catalyst. Silicalite-1 gave
the highest ratio of propene/ethylene. In particular, a high ratio
of 6.5 was observed for the 1-hexene reaction over silicalite-1.
The formation of ethylene involves primary carbocation as an
intermediate, which needs strong acid sites. The formation of
ethylene is assumed to be suppressed over silicalite-1 because of
the absence of strong acid sites.
The features of the catalysts became clearly seen when we
plotted the yields of propene and aromatics in the reactions of
1-butene, 1-pentene, and 1-hexene against a reciprocal of Si/Al2
ratio (Al2/S) as shown in Figures 14 and 15. The propene
yields increased with a decrease in the Al2/Si ratio, except for
the reaction of 1-pentene over H-ZSM-5(280) and silicalite-1
(Figure 14). In contrast to the yields of propene, the yields of
aromatics increased with an increase in the Al2/Si ratio for the
reactions of all alkenes (Figure 15). As the number of strong
acid sites increased, the aromatic formation increased, and the
propene formation decreased. The absence of strong acid sites
maximized the propene formation.
(6) Zhu, X.; Liu, S.; Song, Y.; Xu, L. Catal. Lett. 2005, 103, 201−210.
(7) Zhu, X.; Liu, S.; Song, Y.; Xie, S.; Xu, L. Appl. Catal., A 2005, 290,
191−199.
(8) Xue, N.; Nie, L.; Fang, D.; Guo, X.; Shen, J.; Ding, W.; Chen, Y.
Appl. Catal., A 2009, 352, 87−94.
(9) Zhao, G.; Teng, J.; Zhang, Y.; Yue, Y.; Chen, Q.; Tang, Y. Appl.
Catal., A 2006, 299, 167−174.
(10) Sazama, P.; Dedecek, J.; Gabova, V.; Wichterlova, B.; Spoto, G.;
Bordiga, S. J. Catal. 2008, 254, 180−189.
(11) Zhu, X.; Liu, S.; Song, Y.; Xu, L. Appl. Catal., A 2005, 288, 134−
142.
(12) Vora, B.; Marker, T. L.; Barger Enhanced Light Olefin
Production, U.S. Patent 6,049,017, April 11, 2000.
(13) Tang, X.; Zhou; Qian, W.; Wang, D.; Jin, Y.; Wei, F. Catal. Lett.
2008, 125, 380−385.
(14) Lin, L.; Qiu, C.; Zhuo, Z.; Zhang, D.; Zhao, S.; Wu, H.; Liu, Y.;
He, M. J. Catal. 2014, 309, 136−145.
(15) Zeng, P.; Liang, Y.; Ji, S.; Shen, B.; Liu, H.; Wang, B.; Zhao, H.;
Li, M. J. Energy Chem. 2014, 23, 193−200.
(16) Epelde, E.; Gayubo, A. G.; Olazar, M.; Bilbao, J.; Agyayo, A. T.
Ind. Eng. Chem. Res. 2014, 53, 4614−4622.
CONCLUSIONS
■
(17) Colombo, F., Alberti, G., Padovan, M., Paparatto, G., Contessa,
S. Process for the Conversion of Linear Butenes to Propylene. Eur.
Pat. EP 0109060, March 11, 1987.
Silicalite-1 acts as an efficient catalyst for the formation of
propene by the reaction of light alkenes because of the presence
of weak acid sites and the absence of strong acid sites. It was
proposed that the weakly acidic silanol groups on silicalite-1 act
as active sites for the production of propene, starting with
protonation of alkenes to form carbenium ions, followed by
oligomerization and cracking. Strongly acidic bridged OH
groups on H-ZSM-5 catalysts promote hydrogen transfer of
alkenes to form aromatics and alkanes, which lowers the yield
of propene. The absence of strongly acidic sites on silicalite-1
prevents the alkenes from further converting to alkanes and
aromatics by hydrogen transfer. The silicalite-1 treated with
ammonia to enrich the surface silanol groups exhibited the
maximum propene yield of 39.1 C-wt % from 1-butene at 550 °C.
(18) Aubert, E.; Porcher, F.; Souhassou, M.; Petricek, V.; Lecomte, C.
J. Phys. Chem. B 2002, 106, 1110−1117.
(19) Fyfe, C. A.; Darton, R. J.; Mowatt, H.; Lin, Z. S. Microporous
Mesoporous Mater. 2011, 144, 57−66.
(20) Katada, N.; Igi, H.; Kim, J.-H.; Niwa, M. J. Phys. Chem. B 1997,
101, 5969−5977.
(21) Treacy, M. M. J.; Higgins, J. B.; von Ballmoos, R. Collection of
Simulated XRD Powder Diffraction Patterns for Zeolites, 4th revised ed.;
Elsevier: London, 2001; p 237.
(22) Baerlocher, C. H.; Meier, W. M.; Olson, D. H. Atlas of Zeolite
Framework, 6th ed.; Elsevier Science: Amsterdam, 2007; p 212.
(23) Hardenberg, T. A. J.; Mertens, L.; Mesman, P.; Muller, H. C.;
Nicolaides, C. P. Zeolite 1992, 12, 685−689.
(24) Feng, F.; Balkus, K. J., Jr. Microporous Mesoporous Mater. 2004,
69, 85−96.
ASSOCIATED CONTENT
* Supporting Information
■
(25) Parry, E. P. J. Catal. 1963, 2, 371−379.
S
(26) Emeis, C. A. J. Catal. 1993, 141, 347−354.
(27) Hoelderich, W. F.; Roeseler, J.; Heitmann, G.; Liebens, A. T.
Catal. Today 1997, 37, 353−366.
Blank test, numerical data for Figures 7−13, and Scheme for
hydrogen transfer. This material is available free of charge via
the Internet at http://pubs.acs.org.
(28) Kadata, N.; Niwa, M. Catal. Surv. Asia 2004, 8, 161−170.
(29) Ono, Y.; Hattori, H. Solid Base Catalysis; Springer Series in
Chemical Physics 101; Springer: Heidelberg, 2011; pp 42−47.
(30) Heitmann, G. P.; Dahlhoff, G.; Hoelderich, W. F. J. Catal. 1999,
186, 12−19.
AUTHOR INFORMATION
Corresponding Author
*Phone: +966 13 8602029. E-mail: skhattaf@kfupm.edu.sa.
■
(31) Ichihashi, H.; Ishida, M.; Shiga, A.; Kitamura, M.; Suzuki, T.;
Suenobu, K.; Sugita, K. Catal. Surv. Asia 2003, 7, 261−270.
(32) Ishida, M.; Suzuki, T.; Ichihashi, H.; Shiga, A. Catal. Today
2003, 87, 187−194.
(33) Ong, S. W.; Zhao, X. L.; Eisenthal, K. B. Chem. Phys. Lett. 1992,
192, 327−335.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge the support from the Ministry of
Higher Education, Saudi Arabia, in establishment of the Center
of Research Excellence in Petroleum Refining & Petrochemicals
at King Fahd University of Petroleum & Minerals (KFUPM).
The support of KFUPM is also highly appreciated.
REFERENCES
■
(1) Liu, H. J.; Zhang, L.; Li, X. J.; Huang, S. J.; Liu, S. L.; Xin, W. J.;
Xie, S. J.; Xu, L. Y. J. Nat. Gas Chem. 2009, 18, 331−336.
(2) Zhao, Q. F.; Chen, S. L.; Gao, J. S.; Xu, C. M. Transition Met.
Chem. (N.Y.) 2009, 34, 621−627.
(3) Zhu, X.; Li, X.; Xie, S.; Liu, S.; Xu, G.; Xin, W.; Huang, S.; Xu, L.
Catal. Surv. Asia 2009, 13, 1−8.
(4) Plotkin, J. S. Catal. Today 2005, 106, 10−14.
4214
dx.doi.org/10.1021/cs5009255 | ACS Catal. 2014, 4, 4205−4214