Insight into the impact of Al distribution on the catalytic performance of 1-octene aromatization over ZSM-5 zeolite

Article abstract of DOI:10.1039/c9cy01672d

To correlate the relationship between the Al distribution and the catalytic performance of long-chain olefin aromatization, several ZSM-5 zeolites with different Al locations and proximities were prepared via adjusting the hydrothermal synthesis condition

Full text of DOI:10.1039/c9cy01672d

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Insight into the impact of Al distribution on the
catalytic performance of 1-octene aromatization
over ZSM-5 zeolite†
Cite this: Catal. Sci. Technol., 2019,
9, 7034
b
Liwei Zhang,ac Huaike Zhang,ab Zhiqiang Chen,ac Qiang Ning, ab Suyao Liu,
*
ab
Jie Ren,
Xiaodong Wen ab and Yong-Wang Liab
*
To correlate the relationship between the Al distribution and the catalytic performance of long-chain olefin
aromatization, several ZSM-5 zeolites with different Al locations and proximities were prepared via adjusting
the hydrothermal synthesis conditions. Characterization by XRD, SEM and N2-adsorption indicated that all
the as-prepared ZSM-5 zeolites exhibited high crystallinities, uniform morphologies and open channels.
However, a great difference in the framework Al (AlF) positions was observed by 27Al MAS NMR and UV-vis-
DRS of CoIJII) ions, thus leading to the distinct acidity of various zeolites as determined by NH3-TPD and
Py-IR. In 1-octene aromatization reaction, ZSM-5 catalysts exhibited significantly different catalytic
performances and reaction pathways, which were closely related to the types and locations of the AlF
atoms. The ZSM-5 zeolite catalyst with the most abundant single Al sites exhibited the lowest aromatics
selectivity, while Al pairs in the framework were more favorable to the formation of aromatics. Moreover,
the product distribution revealed that the Al pairs in the channel intersections of the ZSM-5 zeolite were
responsible for the improved selectivity to heavy aromatics due to the fewer space restrictions; however,
the Al pairs in the sinusoidal and straight channels with severe size restrictions suppressed the generation
of bigger aromatics on the Al sites. This result provided a new potential method for precisely tailoring the
catalytic selectivity in olefin conversion and upgrading the Fischer–Tropsch oil.
Received 20th August 2019,
Accepted 14th November 2019
DOI: 10.1039/c9cy01672d
rsc.li/catalysis
aromatization of short chain alkenes.6,7 Although extensive
research efforts focused on C3 and/or C4 olefin aromatization
1. Introduction
Fischer–Tropsch synthesis which converts syngas (CO and H2)
into liquid fuels provides a practical option to solve the
increasing demand for oil resources and other chemicals.1
Liquid fuels produced from the Fischer–Tropsch synthesis are
generally of good quality to emit less nitrous and sulfur
oxides. However, the high content of long chain olefins (C6–
C10) in light-hydrocarbon distillates always causes the low
knock rating of the oil products.2,3 The conversion of olefins
into aromatics over zeolite catalysts is a promising strategy to
meet the expanding demand for platform chemicals, to
increase the octane number of gasoline4 and to produce
important high-value alkyl-aromatics such as ethyltoluene,
ethylbenzene and xylenes.5
with ZSM-5 zeolite catalysts,8–10 less attention has been paid
to the aromatization of longer chain olefins. It was reported
that optimizing the acidity and the diffusion pathway of
intermediates can enhance the catalytic performance,
suggesting the close correlation between the acid sites and
long chain alkene aromatization.11,12 Moreover, as for the
aluminosilicate zeolites which are assembled with the corner-
oxygen sharing of SiO4 and AlO4 tetrahedra, the cations (H+)
as the acid sites are located near the AlO4 structure within
the zeolite framework to balance the electric charge and form
catalytically active sites.13,14 It is therefore a reasonable
assumption that the position and proximity of the Al atoms
in the framework should be considered as crucial factors for
the aromatization performance,15,16 due to the fact that the
adsorption of reactant molecules, spatial constraints of the
intermediates, and the diffusion behaviour of products are
strongly influenced by the AlF positions.17,18
Due to its unique pore structure and strong acidity, ZSM-5
zeolite has been widely investigated in the catalytic
a State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese
Academy of Sciences, Taiyuan 030001, PR China. E-mail: renjie@sxicc.ac.cn
b National Energy Research Center for Clean Fuels, Synfuels China Co., Ltd,
Beijing, 101400, PR China. E-mail: liusuyao@synfuelschina.com.cn
c University of Chinese Academy of Sciences, Beijing 100049, PR China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9cy01672d
Recently, several studies about the distribution of the Al
atoms in the framework of ZSM-5 zeolite materials have been
performed. Liang and co-workers suggested that the AlF
locations and the Brønsted acid distribution in H-ZSM-5
zeolites strongly affected the reaction pathway and
7034 | Catal. Sci. Technol., 2019, 9, 7034–7044
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
determined the catalytic performance in the methanol-to-
olefin (MTO) reaction.19 Kim et al. prepared hierarchically
structured ZSM-5 zeolites with various Al distributions and
demonstrated that the Al atoms located in the straight
channels promoted the reaction mechanism through an
alkene-based cycle, leading to a high olefin selectivity.20
Yokoi synthesized ZSM-5 zeolites with different locations of
Al atoms by using various organic structure directing agents
with or without Na cations. They found that the distribution
of AlF affected the catalytic lifetime in the methylcyclohexane
cracking reaction, and therefore the zeolite with Al atoms
located at the intersection exhibited a shorter catalytic
lifetime.21 The result of Sazama presented that the
distribution of Al species in the framework of ZSM-5 zeolites
obviously affected the product composition during the olefin
cracking reactions. The enrichment of Al siting in the
channel intersections was beneficial to the catalytic
performance in 3-methylpentane cracking and aromatics
(TEOS, SiO2 = 28 wt%), sodium sulphate (Na2SO4, 99%),
aluminium sulphate (Al2IJSO4)3·18H2O, 99%), aluminium
nitrate (AlIJNO3)3·9H2O, 99%), cobalt nitrate (CoIJNO3)2·6H2O,
99%), sodium nitrate (NaNO3, 99%) and urea (99%) were
obtained from Sinopharm Chemical Reagent Co., Ltd.
Sodium hydroxide (NaOH, 99%) and ammonium chloride
(NH4Cl, 99%) were purchased from Beijing Chemical Co.,
Ltd. Silica sol (SiO2 ≥ 30 wt%) was obtained from Qingdao
Haiyang Chemical Co., Ltd. Benzene (99%) and 1-octene
(98%) were obtained from Aladdin Reagent Co., Ltd.
2.2 Preparation of various ZSM-5 zeolites
ZSM-5 zeolite with a spherical morphology was prepared
using silica sol and Al2IJSO4)3·18H2O as silica and aluminium
sources, respectively. TPAOH was used as an organic
structure-directing agent (OSDA). The initial gel was obtained
by adding a solution of TPAOH, silica gel, and Na2SO4 to a
solution of aluminium sulphate under stirring, successively.
The molar composition for the spherical ZSM-5 zeolites was
100SiO2 : 1.0Al2O3 : 40TPAOH: 15Na2O : 6400H2O. Then, the
resulting gel was transferred into a stainless-steel autoclave
and hydrothermally crystallized at 180 °C for 36 h with
autogenous pressure.
disproportionation.22 Janda reported that
a
high
concentration of AlF distributed in the channel intersections
of ZSM-5 zeolites could be obtained by increasing the Al
content,
which
was
beneficial
for
n-butane
dehydrogenation.23 Thus, it is of great importance to deeply
research the impacts of the position and proximity of AlF in
the ZSM-5 zeolite on the distribution and strength of the acid
sites which strongly determine the reaction pathway and
catalytic performance, providing a useful perspective for the
design and synthesis of desired catalysts.
ZSM-5 zeolite with
a
hexagonal morphology was
synthesized using TPABr and TPAOH as OSDAs. The gel for
this sample synthesis was prepared by adding TPAOH and
silica gel successively to a mixed solution of Al2IJSO4)3, urea,
NaOH, and TPABr under stirring. A typical mixture had a
molar composition of 100SiO2 : 1Al2O3 : 7.5TPABr: 6TPAOH:
2.5Na2O : 16urea : 3000H2O. After being stirred for 1 h at room
temperature, the mixture was placed in a stainless-steel
autoclave and heated at 170 °C for 36 h in a furnace at
autogenous pressure under stirring.
In the present work, several ZSM-5 zeolites were prepared
using
hydrothermal
synthesis
and
characterized
systematically. All ZSM-5 samples exhibited similar SiO2/
Al2O3 ratios, high crystallinity, and uniform morphology.
Moreover, various measurements including UV-vis-DRS of
CoIJII) ions and MAS NMR spectroscopy were used to
differentiate the Al siting in the straight and sinusoidal
channels and the channel intersections within the ZSM-5
framework, which is closely related to the acidity properties
determined by NH3-TPD and Py-IR. Then, the 1-octene
aromatization performance over ZSM-5 zeolites with various
Al distributions was evaluated, and a correlation between the
product selectivity and acid site distribution of the ZSM-5
zeolites was built. Furthermore, the alkylation of benzene
with 1-octene was also used to investigate the formation
pathway of the heavy aromatics. This work provides useful
perspectives and better understanding for the selective
conversion of olefins originating from the Fischer–Tropsch
synthesis.
To obtain nano-ZSM-5 zeolite with
a
cylindrical
morphology, an aqueous solution of TEOS and TPAOH was
stirred at 80 °C for 24 h in a stainless-steel autoclave. Then, a
solution of NaOH and AlIJNO3)3·9H2O was added into this
mixture, successively. The starting molar composition was
100SiO2 : 1Al2O3 : 25TPAOH : 5Na2O : 5000H2O. The synthesis
gels were heated to 175 °C and finally crystallized for 24 h.
After crystallization, various zeolites were filtered and
washed with deionized water to neutral, and then dried at
120 °C overnight. The as-synthesized samples were calcined
at 550 °C (ramping rate: 1 °C min−1) for 8 h under an air
atmosphere to remove OSDAs. The calcined samples were
ion-exchanged with a 0.5 M NH4Cl aqueous solution at 80 °C
for 8 h to form NH4-ZSM-5 samples, followed by washing,
drying and calcination steps at 550 °C for 4 h to obtain
protonic form zeolites. The ion-exchange and calcination
processes were repeated twice. The final products with
spherical, hexagonal and cylindrical morphologies are
denoted as ZSM-5-S, ZSM-5-H, and ZSM-5-C, respectively.
To determine the Al pairs and single Al atoms, Co ion-
exchanged ZSM-5 zeolites were prepared by following the
method introduced by Dědeček.24,25 Firstly, the as-prepared
2. Experimental
2.1 Materials
All chemicals were purchased from commercial suppliers and
used without further purification. Tetrapropylammonium
bromide (TPABr, 99%) was purchased from Tianjin Guangfu
Chemical Co., Ltd. Tetrapropylammonium hydroxide
(TPAOH, 25% solution in water), tetraethyl orthosilicate
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 7034–7044 | 7035

View Article Online
Paper
Catalysis Science & Technology
ZSM-5 zeolites were ion-exchanged with a 1.0 M NaNO3
aqueous solution at 80 °C for 7 h to obtain Na-ZSM-5
samples. Then, the Na-ZSM-5 samples were ion-exchanged
with an aqueous solution of cobalt nitrate (0.05 mol L−1) at
80 °C for 7 h. Finally, the resulting samples were washed with
deionized water several times and dried at 120 °C overnight.
The ion-exchange and drying steps were performed three
times to obtain Co-ZSM-5.
NH3 (balance He, flow rate: 50 mL min−1) for 10 min at 100
°C. Subsequently, the physically adsorbed ammonia was
removed by flowing He (50 mL min−1) for 60 min. Finally, the
NH3 desorption was performed by heating the samples to 600
°C at a ramping rate of 10 °C min−1 under a flow of pure
helium (50 mL min−1). The temperature and NH3
concentration were monitored continuously with a thermal
conductivity detector (TCD) and
a
quadrupole mass
spectrometer (Omnistar Pfeiffer, Germany), simultaneously.
Pyridine-adsorbed Fourier-transform infrared (Py-FTIR)
spectra of the various ZSM-5 zeolites were collected using an
in situ cell on an EQUINOX 70 (Bruker, Germany). A self-
supporting wafer with about 15 mg of the sample was placed
in an in situ cell equipped with a CaF2 window and a vacuum
system. The powder zeolite was activated at 400 °C for 4 h
under vacuum and then pyridine vapor was introduced into
the cell. Afterward, prior to the collection of the IR spectra,
the sample was evacuated at 200 °C and 350 °C to remove
the weakly bound pyridine, respectively. The adsorption
amount of pyridine degassed at 200 °C was ascribed to the
total acid amount, whereas the adsorption amount of
pyridine, degassed at a relatively high temperature of 350 °C,
corresponded with the amount of strong acidity. The
difference in the adsorption amounts between 200 °C and
350 °C was attributed to the weak acid sites. The quantities
of Brønsted and Lewis acid sites were estimated from the
bands at 1545 cm−1 and 1455 cm−1 of the Py-FTIR spectra,
according to the Lambert–Beer law26 and the molar
absorption coefficients of εB = 1.13 cm μmol−1 and εL = 1.28
cm μmol−1,27 respectively.
2.3 Characterization
Powder X-ray diffraction (XRD) patterns of the various ZSM-5
zeolites were recorded on a D8 Advance diffractometer
(Bruker, Germany) with Cu Kα, operated at 40 kV and 40 mA.
N2 adsorption–desorption isotherms were measured with
an ASAP 2020 instrument (Micromeritics, USA) at −196 °C.
Prior to the adsorption–desorption test, all zeolites were
degassed under 5 × 10−3 Torr at 350 °C for 8 h. The
micropore surface area and volume were determined using
the τ-plot method.
The bulk SiO2/Al2O3 ratios of the various samples were
determined by X-ray fluorescence (XRF) on a ZSX Primus II
spectrometer (Rigaku, Japan).
Scanning electron microscopy (SEM) images of the ZSM-5
samples were taken with a QUANTA 400 (FEI, USA) operating
at an electron acceleration voltage of 10 kV.
Elemental analysis was carried out to determine the Co
contents using inductively coupled plasma-optical emission
spectroscopy (ICP-OES) with an Optima 2100DV (Perkin-
Elmer, USA).
Ultraviolet-visible light-diffuse reflectance spectroscopy (UV-
vis-DRS) was performed using a UH4150 spectrophotometer
(Hitachi, Japan) equipped with a photodiode array detector and
a diffuse reflectance attachment. The scan speed was set at 600
nm min−1. Prior to the test, all the samples were dehydrated at
400 °C for 6 h. The concentrations of single Al and Al pairs
were calculated by:
2.4 Catalytic performance
1-octene aromatization was carried out in a continuous
flowing fixed-bed stainless-steel reactor. In
a
typical
experiment, 5.0 mL of sieved zeolite catalyst (20–40 mesh)
was loaded into the centre of the reactor and pre-treated at
450 °C under a N2 flow for 2 h. The activated sample was
cooled down to the target temperature (340 °C, 360 °C, and
380 °C) in the flow of N2. Subsequently, the feedstock of
1-octene was fed to the reactor using a microscale pump at a
certain flow rate, and the reaction was carried out at 1.0 MPa.
The volume ratio of N2/feed was 300, and the liquid hourly
space velocity (LHSV) of each catalyst was set at 2.0 h−1. The
products were separated at the outlet of the reactor, leading
to the liquid products being collected, while the gaseous
products were evacuated. The gas products were analysed
using an on-line gas chromatograph (Agilent 7890A,
equipped with a flame ionization detector and a capillary
column, Al2O3 30 m × 0.32 mm × 0.25 μm). The liquid
products were analysed using an off-line gas chromatograph
(Agilent 7890A, equipped with a flame ionization detector
and a capillary column, DB-PONA 50 m × 0.20 mm × 0.55
μm). The liquid products were identified using an Agilent/HP
6890 gas chromatograph equipped with a 5973 mass selective
detector (GC-MS). In each experiment, the carbon balance
[single Al] = [Altotal] − 2[Comax
[Al pairs] = 2[Comax
]
]
where [Altotal] and [Comax] are the Al and Co contents in the
ion-exchanged Co-ZSM-5 zeolites, respectively, and both of
these were determined by ICP-OES analysis.
Solid state 27Al magic angle spinning nuclear magnetic
resonance (27Al MAS NMR) spectroscopy was performed on an
Advance 600 NMR spectrometer (Bruker, Germany). 27Al MAS
NMR spectra were recorded using a 4 mm ZrO2 rotor and spin
at a spinning frequency of 12 kHz with a pulse angle of π/12
and a recycle delay of 1 s. The chemical shifts were given
relative to a 1 M aqueous solution of aluminium nitrate.
The acidity of the ZSM-5 zeolites was determined by
temperature programmed desorption of ammonia (NH3-TPD)
using an Autochem II 2920 instrument (Micromeritics, USA).
Firstly, the ZSM-5 zeolite samples were pre-treated at 550 °C
in a He flow (50 mL min−1) for 1 h, and then saturated with
7036 | Catal. Sci. Technol., 2019, 9, 7034–7044
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
was maintained above 95%. The calculation methods of
conversion and selectivity are shown as follows:
w0 − w
w0
Conversion of 1‐octene ð%Þ ¼
× 100
wi
w0 − w
Product Selectivity ð%Þ ¼
× 100
where w0 and w1 represent the weights of 1-octene in the feed
and product, respectively, and wi denotes the weight of the
products to which the reactant was converted.
Fig. 1 XRD patterns of various ZSM-5 zeolites.
3. Results and discussion
3.1 Structure and morphology
zeolite presents a uniformly spherical morphology with an
average diameter of 5.0 μm (Fig. 2a). Moreover, a clear and
regular aggregation structure of small flake-like particles is
observed within the particles of the ZSM-5-S sample as shown
in Fig. 2b. The SEM images of the ZSM-5-H sample
(Fig. 2c and d) shows coffin-shaped particles with a uniform
size of about 5.0 μm, which is a typical morphology of ZSM-5
single crystals. For the ZSM-5-C sample, the SEM images
(Fig. 2e and f) provide the visual indication that this material
consists of cylindrical-like particles with an average crystallite
size of 250 nm.
The N2 adsorption–desorption isotherms and the
corresponding textural properties of the various zeolites are
shown in Fig. 3 and Table 1. All the ZSM-5 zeolites exhibit
high uptake at a relatively low pressure (P/P0 = 0.01),
illustrating the established microporous structure in all the
samples. However, the distinct hysteresis loops in the range
The bulk SiO2/Al2O3 molar ratios of the various ZSM-5 zeolites
were analysed using XRF. As summarized in Table 1, the
SiO2/Al2O3 ratios of all the zeolite samples are almost the
same regardless of the initial compositions.
The XRD patterns of the as-synthesized zeolites are shown
in Fig. 1, and the calculated crystallinities are listed in
Table 1. All the samples exhibit characteristic peaks of the
MFI-type structure at 2θ = 7.9°, 8.8°, 22.3°, 23.1°, and 23.8°.28
Apparently, the ZSM-5-C sample exhibits much broader
diffraction peaks in comparison to the other zeolites, because
the optimized pre-aging step before the crystallization
produces a large number of nuclei particles in the initial
synthesis gel and finally leads to the formation of ZSM-5
zeolites with small particles.29 Compared to the ZSM-5-C and
ZSM-5-S samples, the ZSM-5-H zeolite exhibits a higher
intensity of the diffraction peak at 2θ = 8.76° which can be
assigned to the (020) crystal plane that is perpendicular to
the b-axis of the MFI crystal structure.30,31 The relative
crystallinity of the as-synthesized ZSM-5 zeolite is defined as
the percentage of the diffraction peak intensity at 2θ = 22.3°
based on that of sample ZSM-5-H at this angle.32 As shown in
Table 1, all the samples exhibit high relative crystallinity,
suggesting negligible impurity phases. Taking the
crystallinity of the ZSM-5-H sample as 100%, the relative
crystallinities of the ZSM-5-S and ZSM-5-C zeolites are 93%
and 89%, respectively.
of P/P0
= 0.4–1.0 indicate the different types of pore
properties. An obvious hysteresis loop is observed in the
isotherm of the ZSM-5-S sample at relatively intermediate
pressure (P/P0 = 0.4–0.8) due to N2 adsorption on the layer
accumulation of the small particles.33 However, no hysteresis
loop in the isotherm of the ZSM-5-H sample is observed,
suggesting that no intercrystalline mesoporosity is present.34
As for the ZSM-5-C sample, the hysteresis loop (H4 type) at
high pressure demonstrates the generation of meso–
macropores which probably arise from the inter-crystal voids
caused by the stacking of the small zeolite crystals. As shown
in Table 1, the BET surface areas are found to be 364.8,
367.0, and 395.0 m2 g−1 for the ZSM-5-S, ZSM-5-H, and ZSM-
5-C samples, respectively. Similarly, the external surface area
of the various samples increases with the decrease of crystal
size. The micropore surface areas, which are defined as the
The morphology of the various ZSM-5 samples was
evaluated by SEM observations (Fig. 2). Obviously, all the
samples exhibit a uniform size and shape, and amorphous
aluminosilicates cannot be observed, confirming their high
crystallinity consistent with the XRD result. The ZSM-5-S
Table 1 Crystallinity and textural properties of various ZSM-5 zeolites
Textural propertiesc
Crystallinityb
Samples SiO2/Al2O3 (%)
a
BET surface area (m2 g−1
)
Micropore area (m2 g−1
)
External area (m2 g−1
)
Micropore volume (cm3 g−1
)
ZSM-5-S 82.4
ZSM-5-H 84.2
ZSM-5-C 81.8
93
100
89
364.8
367.0
395.0
301.9
309.4
331.8
62.9
57.6
63.2
0.14
0.14
0.15
a
b
SiO2/Al2O3 molar ratios measured by XRF. Calculated from the XRD patterns by comparing the intensity of the diffraction peak at 2θ =
22.3°. c Determined by N2-adsorption.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 7034–7044 | 7037

View Article Online
Paper
Catalysis Science & Technology
Fig. 2 SEM images of the ZSM-5-S (a and b), ZSM-5-H (c and d), and
ZSM-5-C (e and f) samples.
Fig. 4 27Al MAS NMR spectra (a) of different ZSM-5 samples and
simulated spectra of ZSM-5-S (b), ZSM-5-H (c), and ZSM-5-C (d).
species, respectively.35 The relative concentrations of
framework and extra-framework Al species, which are
determined by integrating the 27Al MAS NMR signal, are listed
in Table 2. Apparently, the AlF sites are dominant in all the
samples. The Al concentrations in the extra-framework of the
various samples follow the order ZSM-5-C > ZSM-5-S > ZSM-
5-H, which is consistent with the crystallinity shown in
Table 1. Moreover, the peak assigned to the AlF species can be
deconvoluted into five peaks, centered at 52, 53, 54, 56, and
58 ppm.36 Generally, the peak at 54 ppm is associated with
the Al species located in the intersection of the straight and
sinusoidal channels, while the peak at 56 ppm can be
assigned to the Al sites in the straight and/or sinusoidal
Fig. 3 Nitrogen adsorption/desorption isotherms of various ZSM-5
zeolites.
difference between the SBET and SEXT, are above 300.0 m2 g−1
for all the samples. In addition, all the samples exhibit
similar micropore volumes (~0.14 cm3 g−1).
Table 2 Al distributions of various ZSM-5 zeolites determined from 27Al
MAS NMR spectra
3.2 Framework structure
AlF distribution (%)
The location and content of the aluminum species in the
framework determine the several properties of zeolites
including the acidity and catalytic performance. Thus, it is
necessary to investigate and compare the Al distributions in
the various ZSM-5 zeolites.
a
b
c
c
Samples
AlFW
AlEFW
Al(54)
Al(56)
ZSM-5-S
ZSM-5-H
ZSM-5-C
94.8
97.5
93.3
5.2
2.5
6.7
12.6
37.4
26.0
36.2
26.9
31.5
27Al MAS NMR was used to determine the coordination
environments of Al species of the different samples. As shown
in Fig. 4, all the samples exhibit two peaks at 54.5 and 0 ppm
corresponding to the tetrahedral Al species in the framework
and the octahedral coordination of the extra-framework Al
a
The relative concentration of AlF, which is determined by
integrating the 27Al MAS NMR signal between the chemical shifts at
b
20 and 85 ppm. The relative concentration of extra-framework
aluminium, which is determined by integrating the 27Al MAS NMR
c
signal between −10 and 10 ppm. Determined by deconvolution of
the 27Al MAS NMR spectra.
7038 | Catal. Sci. Technol., 2019, 9, 7034–7044
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
channels.37 The concentration of these sites, calculated from
the area of these individual peaks, is given in Table 2. The
concentrations of Al sites located at the intersection follow
the order ZSM-5-H > ZSM-5-C > ZSM-5-S, while the Al
concentrations in the independent channels decrease in the
order ZSM-5-S > ZSM-5-C > ZSM-5-H, suggesting that the Al
sites at the channel intersections can be suppressed by the
accumulation of these nanocrystals.
Kim and Sazama suggested that the tetrahedral Al sites
can be separated as a single Al atom or Al pairs, according to
the number of (Si–O) groups between two Al sites in the
zeolite framework. When the number of (Si–O) groups
between two tetrahedral Al sites exceeds two, the Al sites are
named single Al atoms owing to the long distance between
them. Moreover, only the Al–O–(Si–O)2–Al groups can be
formed, which are denoted as Al pairs.22 According to the
ICP-OES result of Co ion-exchanged ZSM-5 zeolites, the
fractions of single Al atoms and Al pairs are obtained and
summarized in Table 3. Almost all of the Al atoms in the
ZSM-5-H sample exist in the form of single Al atoms.
However, as for the ZSM-5-S sample, 78.1% of the Al sites exist
as Al pairs which are slightly more prevalent than those in the
ZSM-5-C sample. To further investigate the difference in the
location of AlF sites in the ZSM-5-S and ZSM-5-C samples,
more detailed locations of the Al pairs were determined by
UV-vis-DRS of the Co-exchanged ZSM-5 samples.
Fig. 5 UV-vis-DRS spectra of Co-exchanged ZSM-5-S (a) and ZSM-5-
C (b) zeolites. The fractions of α-, β-, and γ-type Al pairs are also
shown.
locate at both the straight and sinusoidal channels. In
contrast, the ZSM-5-C zeolite holds more Al pairs (β-type) in
the channel intersections (70.3%) than the ZSM-5-S sample
(59.5%) due to the dissimilarity of silica sources used in the
zeolite synthesis procedure.25,38
In summary, by combining the 27Al MAS NMR results,
XRF data, ICP-OES analysis, and UV-vis-DRS spectra, it can be
concluded that the coordination environments of AlF can be
quite different for various zeolites with similar bulk SiO2/
Al2O3 ratios. Furthermore, 92.2% of Al atoms in ZSM-5-H
exist in the form of the single Al. However, the Al atoms
prefer to locate in channel intersections of the ZSM-5-C
zeolite as Al pairs, while the ZSM-5-S zeolite has more Al
pairs in the straight and sinusoidal channels.
As shown in Fig. 5, the spectra of the Co-exchanged ZSM-
5-S and ZSM-5-C samples exhibit a broad band ranging from
13 000 to 24 000 cm−1, corresponding to the d–d transitions
of bare CoIJII) cations in several cationic sites. The amount of
these sites is obtained by deconvolution of the UV-vis-DRS
spectra in Fig. 5 using Gaussian functions. The α-type
cationic sites in the straight channels are identified by the
single peak at about 15 100 cm−1, and the peaks at 16 000,
17 150, 18 600, and 21 200 cm−1 are assigned to the sites (β-
type) at the intersections between the straight and sinusoidal
channels. The γ-type sites in the sinusoidal channels are
3.3 Acidity
The acidic characteristics of the as-prepared ZSM-5 samples
were determined by NH3-TPD and Py-IR. The NH3-TPD curves
of the different zeolites are shown in Fig. 6. All spectra of the
ZSM-5 zeolites exhibit two desorption peaks at about 195 °C
and 370 °C, which can be attributed to NH3 adsorbed on weak
and strong acid sites, respectively.39 The amounts of various
acid sites are fitted for the quantitative analysis using Gaussian
deconvolution (as shown in Table 4). The amounts of weak acid
sites on the different samples follow the order ZSM-5-S > ZSM-
5-H > ZSM-5-C, while the amounts of strong acid sites decrease
in the order ZSM-5-S ≈ ZSM-5-H > ZSM-5-C.
defined by peaks at about 20 100 and 22 000 cm−1 19
The
.
contents of these sites in the ZSM-5-S and ZSM-5-C samples
are listed in Table 3. Due to the low Co content as
determined by ICP-OES analysis, it is difficult to distinguish
the location of the Al pairs in the ZSM-5-H sample. Obviously,
the fraction of the Al pairs as the α-type and γ-type sites in
the ZSM-5-S sample is more than that of the ZSM-5-C sample,
suggesting that the Al pairs of the ZSM-5-S zeolite prefer to
The type and concentration of Brønsted (B) and Lewis (L)
acid sites for the various samples are determined by Py-IR as
Table 3 Framework Al distributions of different ZSM-5 samples
Al contenta (%)
Al pair distributionb (%)
Samples
Al pairs
Single Al
α
β
γ
ZSM-5-S
ZSM-5-H
ZSM-5-C
78.1
7.8
76.2
21.9
92.2
23.8
22.1
—
14.9
59.5
—
70.3
18.4
—
14.8
a
Calculated from the ICP-OES analysis of Co-exchanged ZSM-5
b
samples. Determined from the fitting peaks of the UV-vis-DRS
spectra of Co-exchanged ZSM-5 samples.
Fig. 6 NH3-TPD spectra of various ZSM-5 zeolites.
Catal. Sci. Technol., 2019, 9, 7034–7044 | 7039
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Catalysis Science & Technology
Table 4 Acidity properties of various ZSM-5 zeolites measured by NH3-TPD and Py-IR
Aciditya (μmol g−1
)
Acidity/(μmolPy g−1 b
)
Brønsted
Lewis
L/(B + L)c
(%)
Samples
200 °C
350 °C
200 °C
350 °C
Weak
Strong
ZSM-5-S
ZSM-5-H
ZSM-5-C
105.8
91.1
82.0
164.6
163.6
128.2
124.5
155.4
77.5
112.2
119.4
70.9
8.0
7.1
8.1
8.0
5.5
6.9
6.3
4.4
9.2
a
The quantitative results of weak and strong acid sites determined by NH3-TPD were measured by monitoring the desorbed ammonia at 195
b
c
and 370 °C, respectively. The quantitative results of Brønsted and Lewis acid sites were calculated from the Py-IR spectra. This ratio was
calculated using the number of Lewis acid sites over the number of all acid sites determined from the Py-IR spectra.
shown in Fig. 7 and the quantitative results are summarized
in Table 4. Apparently, all the samples exhibit small amounts
of L acid sites, whereas the ZSM-5-H and ZSM-5-S zeolites
have many more B acid sites than the ZSM-5-C sample. It
should be noted that the L/(B + L) ratios over the different
samples follow the order ZSM-5-H (4.4) < ZSM-5-S (6.3) <
ZSM-5-C (9.2), regardless of the pyridine desorption
temperature. This result is consistent with the content of ex-
framework Al species in the different samples determined by
27Al MAS NMR (as shown in Table 2), suggesting that the L
acid sites might mainly originate from the ex-framework Al
species. It is generally accepted that the B acid sites are
derived from the framework aluminium.40,41 Therefore, the
amounts of B acid sites over the different samples follow the
order ZSM-5-H > ZSM-5-S > ZSM-5-C, showing a correlation
with the contents of ex-framework and framework Al obtained
from the 27Al MAS NMR results, respectively. It is worth
noting that a slight difference between the quantitative results
of NH3-TPD and Py-IR for the ZSM-5-S and ZSM-5-H samples
is observed. This phenomenon is probably due to the re-
adsorption of ammonia42,43 on the ZSM-5-S zeolite with layer
accumulation and higher diffusion resistance compared to
the ZSM-5-H zeolite with a more b-axis structure.
zeolites due to it having the largest amount of single Al in
the zeolite framework.
3.4 Catalytic performance
To test the effect of the reaction temperature on the catalytic
activity and product selectivity, the 1-octene aromatization
performance was investigated at 340 °C, 360 °C and 380 °C,
as shown in Fig. S1 in the ESI.† At different reaction
temperatures, the conversion of the various catalysts is above
99%, suggesting that 1-octene can be totally converted.
Apparently, the selectivity to total aromatics and BTEX
(benzene, toluene, ethylbenzene, and xylene) increases with
the increase of the reaction temperature. This result indicates
that the increase of the reaction temperature in this range is
beneficial for the formation of aromatics, due to the
aromatization activity of olefins being favourable with high
temperature. However, high temperature always leads to the
+
cracking reaction, thus resulting in the decrease of C9
aromatics selectivity. The distribution of aromatics and
cracked products over the different catalysts at a higher
temperature (380 °C) is further classified and analysed to
reveal the impact of Al distributions on the catalytic
performance for 1-octene aromatization.
Table 3 indicates that the most of the AlF atoms exist as
the single Al type in the ZSM-5-H sample. Borade et al.
reported that the percentage of strong acid sites increased
with the increase of SiO2/Al2O3 ratios.44 The larger the
distance between two Al atoms, the stronger the acidity
strength of zeolite.45 Thus, the ZSM-5-H sample shows a
larger amount of strong B acid sites in comparison to other
The catalytic performance and product distribution of the
various ZSM-5 zeolites for 1-octene aromatization obtained
with time-on-stream of 6.0 h at 380 °C are summarized in
Fig. 8, and the detailed product selectivity is listed in Table S1
of the ESI.† The conversions of the 1-octene feed over all the
catalysts are almost 100%. Great differences in the product
behavior are observed between these catalysts. Apparently, the
selectivity to BTEX over the various catalysts follows the order
ZSM-5-S (27.0%) > ZSM-5-C (21.2%) > ZSM-5-H (18.8%), and
the total aromatics selectivity exhibits the same trend under
the same reaction conditions. As for the heavy aromatics
(C10+), the highest selectivity is observed on the ZSM-5-C
catalyst (13.0%), and the selectivities on the ZSM-5-S and
ZSM-5-H catalysts are 10.7% and 8.7%, respectively.
It is widely accepted that the olefin aromatization has
several
reaction
steps
including
olefin
cracking,
isomerization, oligomerization, cyclization and hydrogen
transfer reactions,46 in which almost all steps occur on the B
acid sites. Thus, it can be inferred that the B acid sites
Fig. 7 Pyridine adsorbed FT-IR spectra of various ZSM-5 zeolites at
200 °C (a) and 350 °C (b).
7040 | Catal. Sci. Technol., 2019, 9, 7034–7044
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
selectivity to the heavy aromatics (C10+) among all the
catalysts probably due to its less diffusion resistance and
more external acid sites, leading to the further
+
transformation of the BTEX to form C10 aromatics. The
higher concentrations of the Al pairs in the intersections of
two types of 10-membered channels in the ZSM-5-C catalyst
also provide
limitations for the formation of C10 aromatics mainly
including 1,3,4,5-tetramethylbenzene and 1,2,4,5-
a more suitable space with fewer space
tetramethylbenzene as the biggest products, which can only
be formed in the channel intersections of the ZSM-5
zeolite.47,48 However, the shape selectivity of the straight and
sinusoidal channels of the ZSM-5 zeolite suppresses the
formation and the diffusion of much bigger aromatics, which
can only be formed on the acid sites at the external surface
by isomerization and/or alkylation.49 Thus, the ZSM-5-S
+
catalyst exhibits a lower selectivity to the C11 aromatics
mainly including dimethylindane and naphthalene with its
derivatives than the ZSM-5-C catalyst, due to the larger
amount of B acid sites of the ZSM-5-S sample as determined
by NH3-TPD and Py-IR which may lead to the further cracking
and dealkylation reaction.
Fig. 8 Catalytic performance and the detailed product distribution
over the different ZSM-5 catalysts for 1-octene aromatization (Ci= and
Ci0 mean the alkene and alkane hydrocarbons with i carbon atoms,
respectively; reaction conditions: time on stream = 6.0 h, pressure =
1.0 MPa, temperature = 380 °C, N2/1-octene = 300, LHSV = 2.0 h−1).
In addition, the distribution of cracking products can
reflect the structural features of these catalysts and the initial
step of the 1-octene aromatization.50 As shown in Fig. 8, C3
hydrocarbons (mainly C3 paraffins) are the primary small
cracked molecule products, followed by the C4, C5, and C6
components for the ZSM-5-S and ZSM-5-C catalysts. However,
for the ZSM-5-H catalyst, the most abundant cracked product
is C4 hydrocarbon followed by the C3, C5, and C6 components,
further suggesting the different cracking pathways among the
various zeolites. It is important to mention that the C3= favors
the formation of toluene precursors.51 Therefore, compared
to the ZSM-5-C catalyst, the ZSM-5-S catalyst shows less C3
and higher selectivity to C9 aromatics.
derived from the AlF with different positions and proximities
can significantly influence the aromatization performance. As
listed above, the AlF atoms in the ZSM-5-H catalyst are
enriched in the intersections between the straight and
sinusoidal channels as single Al atoms, while the Al pairs in
the ZSM-5-C catalyst are concentrated in the channel
intersections and those in the ZSM-5-S catalyst prefer to
locate in the straight and sinusoidal channels.
Therefore, for the 1-octene aromatization reaction, the
single Al species with corresponding strong acid sites in the
ZSM-5-H catalyst are responsible for the highest selectivity to
cracking products and the lowest aromatics selectivity,
leading to more C4 paraffins and C5 hydrocarbons,
suggesting that the independent Al atoms in the framework
are probably not favorable to the aromatization. Furthermore,
the ZSM-5-S and ZSM-5-C catalysts with more Al pairs exhibit
higher aromatics selectivity because the Al pairs are more
favourable for the hydrogen transfer reactions, thus leading
to the formation of the aromatics.5 Compared with the ZSM-
5-C catalyst, the more B acid sites and the corresponding
higher hydrogen transfer of the ZSM-5-S catalyst guarantee
the easier aromatization to form BTEX. This result indicates
the increasing occurrence of the oligomerization and
hydrogen transfer reactions with higher concentration of acid
sites, favoring the formation of aromatics. However, the
difference in the distribution of Al pairs in the framework
and the crystal sizes also results in different BTEX and total
The TGA profiles of the spent ZSM-5 catalysts after 24 h
on stream for 1-octene aromatization are shown in Fig. 9.
The different temperatures of the sharp weight loss can
reflect the crystal size of the spent ZSM-5 catalysts suggesting
that the carbonaceous deposits can be burnt more easily in
Fig. 9 TGA profiles for the spent ZSM-5 catalysts after 1-octene
aromatization for 24 h (reaction conditions: time on stream = 24 h,
pressure = 1.0 MPa, temperature = 380 °C, N2/1-octene = 300, LHSV
= 2.0 h−1).
+
aromatics selectivities, especially for the C10 aromatics. The
ZSM-5-C catalyst with a small crystal size exhibits the highest
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 7034–7044 | 7041

View Article Online
Paper
Catalysis Science & Technology
the small crystals. Quantitative analysis by TGA indicates that
the coke content of the spent catalysts increases in the order
ZSM-5-S (10.8%) < ZSM-5-H (12.0%) < ZSM-5-C (13.2%),
which is correlated with the AlF properties of the different
ZSM-5 catalysts. It is clear that the largest amount of coke is
formed during the 1-octene aromatization over the ZSM-5-C
catalyst because AlF sites as Al pairs prefer to locate in the
intersections providing a broad space for the formation of
big aromatics which will eventually lead to coking. However,
the formation of heavy aromatics is strongly suppressed on
the AlF sites located in the 10-membered ring channels of the
ZSM-5-S catalyst, leading to less carbonaceous deposits
compared to the ZSM-5-C catalyst. As for the ZSM-5-H
catalyst, single Al sites are responsible for the serious
cracking reaction (as shown in Fig. 8), thus resulting in the
formation of light products; on the other hand, the large
amount of Al sites in the intersections (Table 2) is favorable
for the formation of heavy products.
selectivity to ethylbenzene over various catalysts is very low
(<2.0%), indicating that the direct cyclization and hydrogen
transfer are probably not the main pathways for the
formation of aromatics. Meanwhile, the low selectivity to iso-
octene also suggests that the route of 1-octene isomerization
(route 2) is difficult to occur. Therefore, the cracking reaction
of 1-octene is dominant via β cleavage to produce light
products, which is a typical result.52,53 As shown in Fig. 8, it
can be inferred that the single Al sites of ZSM-5-H are
=
=
probably more favourable to the formation of C4 and i-C4
through route 4, whereas the Al pairs of the ZSM-5-S and
ZSM-5-C catalysts are more favourable to the generation of
=
=
i-C5 and C3 . The BTEX products are formed via oligo-
polymerization, cyclization and hydrogen transfer over the
ZSM-5 catalysts by coupling different cracked products (route
5). In addition, the coupling reaction of the 1-octene and
cracked products (routes
6 and 7) and the alkylation
reactions (routes 8 and 9) are probably responsible for the
formation of heavy aromatics (C10+).
+
Fig. 8 shows the high fraction of C10 aromatics in the
3.5 Aromatization mechanism of 1-octene
total aromatics, and therefore the formation pathway of these
products should also be well considered because the
formation mechanism of the heavy aromatics is still unclear.
Thus, a typical alkylation reaction by using a mixture of
benzene and 1-octene as reactants was performed over the
ZSM-5-S catalyst under the same conditions as the 1-octene
aromatization. The detailed experimental process and
catalytic performance are shown in the ESI.† The 1-octene
conversion over the ZSM-5-S catalyst is above 99% consistent
with the 1-octene aromatization reaction, while the benzene
conversion is 47.5%. Apparently, compared to the product
distribution of 1-octene aromatization, different product
selectivities over the ZSM-5-S catalyst for the alkylation of
benzene with 1-octene are observed in Table S2 of the ESI,†
whereas more TEX (35.1%) and heavy aromatics (17.1%) are
formed. The alkylation reaction catalysed by acidic zeolites is
commonly considered via the carbonium ion mechanism.54
Under these reaction conditions, 1-octene is firstly cracked to
light olefins (Fig. 10, routes 3 and 4). Then, the formed light
alkenes are protonated by the B acid sites to form the
intermediates.55 The latter can be further converted following
two major routes: they can be coupled with benzene to
The above catalytic results demonstrate that the ZSM-5
zeolites with various properties of Al positions and proximity
exhibit similar 1-octene conversions and apparently different
product selectivities. Moreover, the selectivity and product
distribution are significantly related to the location of acid
sites and the siting of AlF as well as the type of Al species,
because the reaction pathway and process for 1-octene
aromatization are affected by the properties of the B acid
sites. The enrichment of single AlF atoms in the channel
intersections of ZSM-5 probably leads to the cracking
reaction rather than aromatization, while the Al pairs are
more favourable for the formation of BTEX, whereas the Al
+
pairs in the interconnection channels enhance the C10
aromatics selectivity by providing
a
more suitable
environment. Combining the above results and literature
data, a detailed mechanism of 1-octene aromatization over
the ZSM-5 catalysts with different Al distributions is
proposed. As shown in Fig. 10, 1-octene can directly cyclize
and hydrogen transfer (route 1) over the ZSM-5 zeolite,
leading to the formation of ethylbenzene. However, the
+
produce C9 aromatics through alkylation or they can react
with another cracked olefins producing toluene which can
undergo further oligomerization, cracking, isomerization and
alkylation, giving olefins and other alkylbenzenes. This result
+
further provides solid evidence that the C10 aromatics in the
1-octene aromatization can be formed through two main
routes including the alkylation of the formed benzene with
light olefins and the aromatization reaction between the
1-octene and formed light olefins.
4. Conclusions
Fig. 10 Simplified reaction pathway for 1-octene aromatization over
ZSM-5 zeolites with different Al positions (reaction conditions:
pressure = 1.0 MPa, temperature = 380 °C, N2/1-octene = 300, LHSV
= 2.0 h−1).
Converting the long chain olefins in light-hydrocarbon
distillates from the Fischer–Tropsch synthesis over ZSM-5
7042 | Catal. Sci. Technol., 2019, 9, 7034–7044
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
zeolite catalysts can not only increase the octane number of
gasoline but can also produce heavy aromatics to improve
the diesel quality, while the key properties and catalytic
performance are determined by the coordination
environments of Al species in ZSM-5 zeolites. In the present
work, ZSM-5 zeolites with different Al positions and
proximities were obtained by tuning the synthesis conditions.
Their phase, morphology, textural properties, acidity, Al
siting, and catalytic performance in 1-octene aromatization
were well investigated. Although all zeolites exhibit high
crystallinity, uniform morphology, and open textural
properties, an obvious difference in the AlF position and
proximity was found via 27Al MAS NMR and UV-vis-DRS of
CoIJII) ions, thus resulting in the different acidities as
determined by NH3-TPD and Py-IR. As a result, the catalytic
performance of the ZSM-5 catalyst in 1-octene aromatization
is found to be strongly related to the AlF positions and
proximity. The product selectivity demonstrated that the
single Al atoms in the ZSM-5 framework are responsible for
the highest activity in the cracking reaction, leading to the
lowest aromatics selectivity. However, the Al pairs in the
framework were more favorable to the formation of
aromatics. In addition, from the distribution of the aromatics
and cracked products, the Al pairs in the channel
intersections of the ZSM-5 zeolite resulted in the
enhancement of the selectivity to heavy aromatics due to the
fewer space restrictions, while the Al pairs in the sinusoidal
and straight channels with severe shape selectivity
suppressed the generation of bigger aromatics. Nevertheless,
the alkylation of benzene with 1-octene provided solid
evidence that heavy aromatics can be formed through both
the alkylation reaction and the aromatization reaction. This
finding here provides useful perspectives on developing
aromatization catalysts for olefin conversion and quality
upgradation of Fischer–Tropsch synthesis products.
5 S. Jongpatiwut, P. Sackamduang, T. Rirksomboon, S.
Osuwan and D. E. Resasco, J. Catal., 2003, 218, 1–11.
6 A. Corma and V. Orchillés, Microporous Mesoporous Mater.,
2000, 35–36, 21–30.
7 S. M. Opalka and T. Zhu, Microporous Mesoporous Mater.,
2016, 222, 256–270.
8 A. H. Janssen, A. J. Koster and K. P. De Jong, J. Phys. Chem.
B, 2002, 106, 11905–11909.
9 K. P. de Jong, J. Zečević, H. Friedrich, P. E. de Jongh, M.
Bulut, S. Van Donk, R. Kenmogne, A. Finiels, V. Hulea and
F. Fajula, Angew. Chem., 2010, 122, 10272–10276.
10 D. Verboekend and J. Pérez-Ramírez, Chem.
2011, 17, 1137–1147.
– Eur. J.,
11 H. Long, X. Wang, W. Sun, G. Xiong and K. Wang, Fuel,
2008, 87, 3660–3663.
12 H. Long, X. Wang and W. Sun, Microporous Mesoporous
Mater., 2009, 119, 18–22.
13 J. Macht, R. T. Carr and E. Iglesia, J. Am. Chem. Soc.,
2009, 131, 6554–6565.
14 J. R. Sohn, S. J. DeCanio, P. O. Fritz and J. H. Lunsford,
J. Phys. Chem., 1986, 90, 4847–4851.
15 E. M. Gallego, C. Li, C. Paris, N. Martín, J. Martínez-
Triguero, M. Boronat, M. Moliner and A. Corma, Chem. –
Eur. J., 2018, 24, 14631–14635.
16 A. B. Pinar, L. Gómez-Hortigüela, L. B. McCusker and J.
Pérez-Pariente, Chem. Mater., 2013, 25, 3654–3661.
17 R. A. Beyerlein, G. B. McVicker, L. N. Yacullo and J. J.
Ziemiak, J. Phys. Chem., 1988, 92, 1967–1970.
18 J. R. Di Iorio and R. Gounder, Chem. Mater., 2016, 28,
2236–2247.
19 T. Liang, J. Chen, Z. Qin, J. Li, P. Wang, S. Wang, G. Wang,
M. Dong, W. Fan and J. Wang, ACS Catal., 2016, 6,
7311–7325.
20 S. Kim, G. Park, M. H. Woo, G. Kwak and S. K. Kim, ACS
Catal., 2019, 9, 2880–2892.
21 T. Yokoi, H. Mochizuki, S. Namba, J. N. Kondo and T.
Tatsumi, J. Phys. Chem. C, 2015, 119, 15303–15315.
22 P. Sazama, J. Dědeček, V. Gábová, B. Wichterlová, G. Spoto
and S. Bordig, J. Catal., 2008, 254, 180–189.
23 A. Janda and A. T. Bell, J. Am. Chem. Soc., 2013, 135,
19193–19207.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
24 J. Dědeček, D. Kaucký, B. Wichterlová and O. Gonsiorová,
Phys. Chem. Chem. Phys., 2002, 4, 5406–5413.
25 J. Dědeček, D. Kaucký and B. Wichterlová, Microporous
Mesoporous Mater., 2000, 35–36, 483–494.
26 L. Guo, X. Bao, Y. Fan, G. Shi, H. Liu and D. Bai, J. Catal.,
2012, 294, 161–171.
The authors are grateful to Synfuels China Co., Ltd. and
Beijing Key Laboratory of Coal to Cleaning Liquid Fuels for
financial support. This project is also supported by the
National Natural Science Foundation of China (No.
21902171).
27 M. Guisnet, P. Ayrault and J. Datka, Pol. J. Chem., 1997, 71,
1455–1461.
References
1 S. Liu, J. Ren, S. Zhu, H. Zhang, E. Lv, J. Xu and Y.-W. Li,
J. Catal., 2015, 330, 485–496.
28 J. Li, S. Liu, H. Zhang, E. Lü, P. Ren and J. Ren, Chin. J.
Catal., 2016, 37, 308–315.
2 G. Henrici-Olivé and S. Olivé, Angew. Chem., Int. Ed. Engl.,
1976, 15, 136–141.
3 M. E. Dry, Appl. Catal., A, 1996, 138, 319–344.
4 X. Su, G. Wang, X. Bai, W. Wu, L. Xiao, Y. Fang and J.
Zhang, Chem. Eng. J., 2016, 293, 365–375.
29 H. Mochizuki, T. Yokoi, H. Imai, R. Watanabe, S. Namba,
J. N. Kondo and T. Tatsumi, Microporous Mesoporous Mater.,
2011, 145, 165–171.
30 Y. Liu, Y. Li and W. Yang, J. Am. Chem. Soc., 2010, 132,
1768–1769.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 7034–7044 | 7043

View Article Online
Paper
Catalysis Science & Technology
31 W. Chaikittisilp, M. E. Davis and T. Okubo, Chem. Mater.,
2007, 19, 4120–4122.
43 G. Bagnasco, J. Catal., 1996, 159, 249–252.
44 R. B. Borade, S. G. Hedge, S. B. Kulkarni and P. Ratnasamy,
Appl. Catal., 1984, 13, 27–38.
45 D. Zhou, N. He, Y. Wang, G. Yang, X. Liu and X. Bao, J. Mol.
Struct.: THEOCHEM, 2005, 756, 39–46.
46 Y. Li, S. Liu, S. Xi and L. Xu, Appl. Catal., A, 2009, 360, 8–16.
47 M. W. Anderson and J. Klinowski, Nature, 1989, 339,
200–203.
32 J. A. Ke and I. Wang, Mater. Chem. Phys., 2001, 68, 157–165.
33 Y. Jia, J. Wang, K. Zhang, W. Feng, S. Liu, C. Ding and P.
Liu, Microporous Mesoporous Mater., 2017, 247, 103–115.
34 A. Vinu, D. P. Sawant, K. Ariga, M. Hartmann and S. B.
Halligudi, Microporous Mesoporous Mater., 2005, 80,
195–203.
35 S. Liu, J. Ren, H. Zhang, E. Lv, Y. Yang and Y.-W. Li,
J. Catal., 2016, 335, 11–23.
48 H. P. Röger, K. P. Möller and C. T. O'Connor, Microporous
Mater., 1997, 8, 151–157.
36 S. Sklenak, J. Dědeček, C. Li, B. Wichterlová, V. Gábová, M.
Sierka and J. Sauer, Angew. Chem., Int. Ed., 2007, 46, 7286–7289.
37 J. Dědeček, S. Sklenak, C. Li, B. Wichterlová, V. Gábová, J.
Brus, M. Sierka and J. Sauer, J. Phys. Chem. C, 2009, 113,
1447–1458.
49 M. Bjorgen, S. Svelle, F. Joensen, J. Nerlov, S. Kolboe, F.
Bonino, L. Palumbo, S. Bordiga and U. Olsbye, J. Catal.,
2007, 249, 195–207.
50 F. A. Twaiq, N. A. M. Zabidi and S. Bhatia, Ind. Eng. Chem.
Res., 1999, 38, 3230–3237.
38 V. Gábová, J. Dědeček and J. Čejka, Chem. Commun.,
2003, 1196–1197.
51 J. S. Buchanan, J. G. Santiesteban and W. O. Haag, J. Catal.,
1996, 158, 279–287.
39 D. Li, F. Li, J. Ren and Y. Sun, Appl. Catal., A, 2003, 241,
15–24.
52 V. R. Choudhary, D. Panjala and S. Banerjee, Appl. Catal., A,
2002, 231, 243–251.
40 S. Wang, Y. Chen, Z. Wei, Z. Qin, H. Ma, M. Dong, J. Li, W.
Fan and J. Wang, J. Phys. Chem. C, 2015, 119, 28482–28498.
41 L. H. Ong, M. Dömök, R. Olindo, A. C. V. Veen and J. A.
Lercher, Microporous Mesoporous Mater., 2012, 164, 9–20.
42 M. Niwa, N. Katada, M. Sawa and Y. Murakami, J. Phys.
Chem., 1995, 99, 8812–8816.
53 A. Corma, F. V. Melo, L. Sauvanau and F. J. Ortega, Appl.
Catal., A, 2004, 265, 195–206.
54 G. Bellussi, G. Pazzuconi, C. Perego, G. Girotti and G.
Terzoni, J. Catal., 1995, 157, 227–234.
55 I. Craciun, M. F. Reyniers and G. B. Marin, J. Catal.,
2012, 294, 136–150.
7044 | Catal. Sci. Technol., 2019, 9, 7034–7044
This journal is © The Royal Society of Chemistry 2019