Ammonia-basified 10 wt% Mo/HZSM-5 material with enhanced dispersion of Mo and performance for catalytic aromatization of methane

Article abstract of DOI:10.1016/j.apcata.2019.05.011

Mo/HZSM-5 is an excellent catalyst for nonoxidative aromatization of methane, but it tends to deactivate rapidly due to severe coking. To date much effort has been devoted to enhancing activity and stability of Mo-impregnated HZSM-5 zeolite with Mo loading of 6 wt% or lower in methane aromatization. For this aromatization reaction, this work first reports a significant improvement in the catalytic performance of 10 wt% Mo/HZSM-5 material after preparation via impregnating HZSM-5 zeolite with ammonia-basified molybdate aqueous solution and pretreatment with a CH₄/Ar/He (9:1:10, v/v) mixture. After 38 h of continuous reaction at 700 °C and 1500 mL g^?1 h^?1 (90 vol% CH₄/Ar), this 10 wt% Mo-impregnated catalyst is still able to give an aromatic yield of 7% and a CH₄ conversion of 9.1%. This enhanced catalytic performance of this 10 wt% Mo sample is attributed to the ammonia basification of impregnating solution in catalyst preparation, which leads to a larger amount of highly dispersed Mo species in the zeolitic channels and a less amount of Br?nsted acid sites remained on the Mo/HZSM-5 catalyst.

Full text of DOI:10.1016/j.apcata.2019.05.011

AppliedCatalysisA,General580(2019)111–120
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Ammonia-basified 10 wt% Mo/HZSM-5 material with enhanced dispersion
of Mo and performance for catalytic aromatization of methane
Pinglian Tan
Laboratory of Heterogeneous Catalysis, Anhui University of Technology, Hudong Road 59, Maanshan, Anhui, 243002, People’s Republic of China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Mo/HZSM-5 is an excellent catalyst for nonoxidative aromatization of methane, but it tends to deactivate rapidly
due to severe coking. To date much effort has been devoted to enhancing activity and stability of Mo-im-
pregnated HZSM-5 zeolite with Mo loading of 6 wt% or lower in methane aromatization. For this aromatization
reaction, this work first reports a significant improvement in the catalytic performance of 10 wt% Mo/HZSM-5
material after preparation via impregnating HZSM-5 zeolite with ammonia-basified molybdate aqueous solution
and pretreatment with a CH4/Ar/He (9:1:10, v/v) mixture. After 38 h of continuous reaction at 700 °C and
1500 mL g−1 h−1 (90 vol% CH4/Ar), this 10 wt% Mo-impregnated catalyst is still able to give an aromatic yield
of 7% and a CH4 conversion of 9.1%. This enhanced catalytic performance of this 10 wt% Mo sample is at-
tributed to the ammonia basification of impregnating solution in catalyst preparation, which leads to a larger
amount of highly dispersed Mo species in the zeolitic channels and a less amount of Brønsted acid sites remained
on the Mo/HZSM-5 catalyst.
Methane aromatization
10 wt% Mo/HZSM-5
Ammonia basification
Catalytic performance
Mo dispersion
Dealumination
1. Introduction
2–6 wt%, which were reported to be the most suitable loading for the
catalytic performance of this Mo-modified HZSM-5 zeolite [6,21–25].
Direct nonoxidative aromatization of methane, a very abundant
fossil fuel resource distributed widely around the globe, is a promising
process simultaneously producing COx-free hydrogen and value-added
aromatic hydrocarbons. Mo/HZSM-5 is an excellent catalyst that can
catalyze methane aromatization to near-equilibrium levels at much
lower temperatures, e. g., 700 °C, than 1000 °C, showing a great po-
tential for industrial application [1–4]. However, during the reaction,
this acidic catalyst tends to deactivate rapidly due to severe coking, and
CH4 conversion and formation rates of aromatics, such as benzene and
naphthalene, drastically decreases after a few hours on-stream [5].
Since the pioneering work of Wang et al. [5] much effort has been
devoted to enhancing the activity, especially the stability of Mo/HZSM-
5 catalyst in the methane dehydro-aromatization for the commercial
viability of this process [1–4]. For this Mo-based catalyst, calcination
temperature of 500 °C, zeolitic Si/Al ratio of 10–25, and impregnation
technique are gradually accepted as the most suitable essential pre-
paration conditions [6,7]. Further strategies to suppress the coke for-
mation and increase aromatic yield include adding promotor [8–15],
adjusting pH value of impregnating solution [16], modifying zeolitic
surface via dealumination/desilication [17,18] or silanation [19], and
optimizing zeolitic synthetize [20]. By far the most widely studied
catalysts are those based on Mo/HZSM-5 materials with Mo loadings of
However, enhancing the catalytic activity and stability of Mo/HZSM-5
with Mo loading of 10 wt% or higher in methane aromatization has
rarely been conducted because these high Mo loadings were shown in
multiple studies to have lower catalytic activity values for this ar-
omatization reaction.
Usually, a catalyst is calcined and/or preactivated at a temperature
slightly higher than reaction temperature for its better stability in a
catalytic reaction. In this case, the optimum calcination temperature of
Mo/HZSM-5 is, however, 200 °C lower than the reaction temperature of
methane aromatization. The large temperature difference between
them will promote the influence of pretreatment method on the cata-
lytic performance of the Mo catalyst. Previous activations of Mo/HZSM-
5 material prior to the aromatization reaction were mostly conducted in
oxygen/air [5,21] or inert gas such as helium [6,7,23,24] and argon
[25]. Recent research showed that the Mo/HZSM-5 sample could ex-
hibit better catalytic performance in this reaction after prereduction in
80 vol% CH4/He [26,27] and 10 vol% CH4/H2 [28,29] than pre-
activation in inert helium or oxidative air atmosphere. To avoid rapid
release of reaction heat and product water from MoO3 reduction by CH4
in this activation of catalyst, the author developed a pretreatment
method of Mo/HZSM-5 with a diluted CH4 with 55 vol% inert gases He/
Ar, which could remarkably improve the performance of Mo/HZSM-5
E-mail address: tanpinglian@sina.com.
https://doi.org/10.1016/j.apcata.2019.05.011
Received 25 January 2019; Received in revised form 5 May 2019; Accepted 6 May 2019
Availableonline07May2019
0926-860X/©2019ElsevierB.V.Allrightsreserved.

P. Tan
AppliedCatalysisA,General580(2019)111–120
with Mo loading of 10 wt% or higher for the catalytic aromatization of
methane [30]. On the other hand, it has been demonstrated that basi-
fication of ammonium heptamolybdate (AHM) aqueous solution with
ammonia in catalyst preparation was favorable for formation of tetra-
hedral Mo species rather than octahedral ones on surface of MoO3/
Al2O3 [31,32]. In this work, after preparation via impregnating HZSM-5
zeolite with NH3-basified AHM aqueous solution and pretreatment in
CH4/Ar/He (9:1:10, v/v), 10 wt% Mo-impregnated HZSM-5 zeolite
exhibited significantly enhanced performance in the catalytic dehydro-
aromatization of methane. For an insight into the novel catalytic be-
haviors of this Mo/HZSM-5 material, the effect of ammonia treatment/
basification on the dispersion of Mo species as well as the surface
structure of the zeolite were then investigated.
Disappearance rate of methane was calculated based on carbon balance,
and was expressed as nmol/(gcat. s).
2.3. Catalyst characterization
Nitrogen adsorption-desorption measurements of the parent and
Mo-impregnated HZSM-5 zeolites at −196 °C were conducted using a
Micromeritics automatic surface area and pore analyser (TriStar II
3020). Prior to the analysis, all the samples were outgassed under va-
cuum at 350 °C for 5 h. The specific surface area was evaluated using
the Brunauer-Emmett-Teller (BET) method, while the external surface
area, micropore surface area, and micropore volume were determined
according to the t-plot method.
Temperature-programmed desorption of ammonia (NH3-TPD) was
performed to determine the acidic properties of the parent and Mo-
modified HZSM-5 zeolites. A 0.1 g sample was first activated at 550 °C
for 40 min, and then cooled to 50 °C in helium. After exposure to a gas
mixture of ammonia/helium (2:23, v/v) for 40 min at 50 °C, the sample
was purged with pure helium at 100 °C for 1.0 h to remove the part of
NH3 physically adsorbed on the sample. The TPD measurements were
conducted within a temperature range of 100–650 °C at a heating rate
of 10 °C min−1. A thermal conductivity detector was used to monitor
the amount of desorbed ammonia in the reactor effluent.
The Mo content of Mo/HZSM-5 sample was determined by ICP-AES
(Optima 7300DV). The solutions for analysis were prepared by di-
gesting ca. 0.25 g of the Mo-loaded sample, dried at 120 °C for 6 h, in a
10 mL concentrated nitric acid, followed by adding 15 mL concentrated
hydrofluoric acid solution at 0 °C and then some deionized water up to
50 mL.
2. Experimental
2.1. Catalyst preparation
HZSM-5 zeolite with a molar Si/Al ratio of 25 was supplied by
Nankai University, People’s Republic of China. Both conventional Mo/
HZSM-5 and Mo/HZSM-5(ZA) samples were prepared by impregnation
of HZSM-5 zeolite with ammonium heptamolybdate tetrahydrate
(AHM, Sigma-Aldrich) aqueous solution. Before preparation of Mo/
HZSM-5(ZA), the zeolite (6 g) was impregnated by 100 mL basified
water with 6 mL 28% aqueous ammonia at room temperature in a
beaker whose mouth was closed with a sheet of polythene for 24 h and
then dried at 120 °C in an oven. Mo/HZSM-5(MA) sample was prepared
by impregnation of HZSM-5 zeolite with ammonia-basified AHM aqu-
eous solution. For preparing the Mo-based catalysts mentioned above,
each 6 g portion of original or ammonia-treated HZSM-5 zeolite was
impregnated with 100 mL of AHM aqueous solution or basified AHM
aqueous solution with 6 mL 28% aqueous ammonia, containing the
required amount of molybdenum, at room temperature for 24 h. The
impregnation process for Mo/HZSM-5(MA) was conducted in a closed
beaker with a sheet of polythene. The resulting materials were dried in
an oven at 120 °C and then calcined in a furnace at 500 °C in air for 6 h.
The freshly prepared Mo/HZSM-5 catalysts were pressed, crushed and
sieved to 20–40 mesh for catalytic reaction. An overview of the dif-
ferent preparations of the Mo-impregnated samples mentioned above
was provided in Table S1.
X-ray powder diffraction (XRD) patterns of the parent HZSM-5
zeolite and Mo/HZSM-5 catalysts were obtained on a diffractometer
(ARL X’TRA) using Cu Kα radiation at 40 mA and 40 kV. Powder dif-
fractograms were recorded in the scanning angle (2θ) range of 5–70° at
a scan rate of 5° min−1
.
X-ray photoelectron spectroscopy (XPS) measurement was per-
formed using a spectrometer (Thermo ESCALAB 250) equipped with a
monochromatic Al Kα X-ray source (1486.6 eV) operated at 15 kV and
150 W and a hemispherical energy analyzer. The base pressure inside
the analysis chamber was 1 × 10−9 Torr. The XPS measurement was
carried out at a detector take-off angle of 90°. The spectra were re-
corded with pass energy of 30 eV, X-ray spot size of 500 μm, and step
size of 0.05 eV. The Si 2p line at 103.4 eV was taken as a reference for
binding energy calibration. Near-surface atomic concentrations of the
Mo/HZSM-5 samples were calculated according to peak areas and
sensitivity factors.
2.2. Catalyst evaluation
Catalytic reaction was carried out in a continuous fixed-bed flow
reactor (i.d. 7 mm, quartz) system (Fig. S1) at 700 °C and atmospheric
pressure. Typically, 0.5 g catalyst, giving a catalyst-bed depth of ca.
1.5 cm, was heated in the reactor to 700 °C at a ramp rate of about
22 °C min−1. During this pretreatment, a gas mixture of CH4/Ar/He
(9:1:10, v/v) was introduced into the reactor at 25 mL min−1 to pre-
reduce the catalyst but also rapidly remove adsorbed and/or produced
water from the catalyst. As the temperature reached 700 °C, another gas
mixture of 90 vol% CH4 and 10 vol% Ar in place of the pretreatment gas
mixture was introduced at 12.5 mL min−1 into the reactor through a
mass flow controller (Brooks 5850E) and the catalyst test started. All
gases used in this test were UHP grade.
All reactants and products were analysed with two Agilent 5820
online gas chromatographs equipped with flame ionization (FID) and
thermal conductivity (TCD) detectors, where argon added in reactant
methane was used as an internal standard. Reactor outlet pipelines as
well as gas sampling valves were kept above 160 °C to prevent con-
densation of aromatic products. CH4 and higher hydrocarbon products,
C2H4, C2H6, C6H6, C7H8, and C10H8, were separated by a Porapak-P
packed column, and then analysed using the FID. H2, Ar, CO, CH4, and
CO2 were separated by an activated carbon packed column, and then
analysed using the TCD. Methane conversion and carbon-containing
product selectivity/yield were calculated on a carbon number basis.
Diffuse reflectance UV–vis spectroscopy (UV–vis DRS) was used to
assess Mo dispersion on Mo/HZSM-5 sample. UV–vis DRS spectra were
recorded at room temperature on a Shimadzu DUV-3700 UV–vis spec-
trophotometer using BaSO4 as standard.
Al magic-angle spinning nuclear magnetic resonance (27Al MAS
NMR) was carried out using a Bruker AVANCE III 400 WB spectrometer
at a magnetic field of 9.4 T and a H/X/Y CPMAS probe. The hydrated
sample was packed in a 4 mm zirconia rotor. The MAS sample rotation
speed was 15 kHz. The spectra were recorded at a resonance frequency
of 104.3 MHz with a relaxation delay of 1 s and 2048 scans. External Al
3+
(H2O)6
with a chemical shift of 0 ppm was used as a reference.
3. Results
3.1. CH4 aromatization over NH3-basified 2–16 wt% Mo/HZSM-5
Table 1, and Figs. 1 and 2 show methane conversion and yield/
selectivity of hydrocarbon products in methane dehydro-aromatization
at 700 °C and space velocity 1500 mL g−1 h−1 (90 vol% CH4/Ar) over
Mo-impregnated HZSM-5 zeolite after pretreatment in a diluted CH4
with 55 vol% Ar/He. In all cases, hydrocarbon products included
112

P. Tan
AppliedCatalysisA,General580(2019)111–120
Table 1
Catalytic activities of Mo/HZSM-5, Mo/HZSM-5(MA), and Mo/HZSM-5(ZA) samples with various Mo target loadings in methane aromatization at 700 °C and
1500 mL g−1 h−1a
.
Mo target loading (wt%)
CH4 Conversion (mol%)
Disappearance rate of CH4 (nmol/gcat. s)
Selectivity (%) of
C2H4
C2H6
C6H6
C7H8
C10H8
Cokeb
2
5.5
8.7
8.9
10.8
10.1
11.8
10.7
11.4
11.6
928
11.2
5.7
5.0
3.4
3.5
2.3
2.9
2.8
2.5
2.9
2.1
2.1
1.5
1.8
1.3
1.6
1.5
1.4
44.1
61.6
62.1
67.9
67.4
70.7
69.1
71.9
70.3
3.2
4.1
3.9
3.7
3.9
3.5
3.6
3.7
3.5
10.8
11.0
11.7
12.2
12.1
12.9
13.7
10.5
12.8
27.8
15.5
15.2
11.3
11.3
9.3
9.1
9.6
9.5
2 (MA)
6
6 (MA)
10
10 (MA)
10 (ZA)
16
1455
1493
1801
1682
1979
1791
1905
1935
16 (MA)
a
Data were recorded after 10 h of reaction time.
SCoke = 100%–∑Si (i, carbon-containing product detected).
b
mainly benzene and naphthalene, and a small amount of toluene,
ethylene and ethane (selectivity/yield ratio of C2H4/C2H6 > 1). For
conventional Mo/HZSM-5 sample, its catalytic activity and stability
were enhanced with increasing the targeted loading of Mo from 2 to
16 wt%. Compared with the conventional Mo/HZSM-5, corresponding
Mo/HZSM-5(MA), prepared by impregnating HZSM-5 zeolite with NH3-
basified molybdate aqueous solution, provided higher CH4 conversion
and selectivity to aromatics, and lower selectivity to C2-hydrocarbon
and selectivity/yield ratio of C2H4/C2H6 (Table 1), and showed better
catalytic stability in the aromatization of methane (Fig. 1). However,
the enhanced effect of NH3-basification on the catalytic performance of
the Mo catalyst gradually diminished with increasing the Mo target
loading from 2 to 16 wt% (Fig. 2a).
The catalytic activities of Mo/HZSM-5(MA) samples during a 10-h
test were shown in Fig. 1b. After 10 h of reaction, the 2 wt% Mo loaded
catalyst gave a CH4 conversion of 8.7%, an aromatic yield of 6.7%, and
a C2-hydrocarbon yield of 0.7%. The selectivity/yield ratio of C2H4/
C2H6 was 2.6. The CH4 conversion and aromatic yield increased with an
increase in Mo loading and reached maximum values of 11.8% and
10.3%, respectively, at the targeted Mo loading of 10 wt%. At the same
time, C2-hydrocarbon yield and selectivity/yield ratio of C2H4/C2H6
decreased to 0.4% and 1.8, respectively. By further increasing the tar-
geted loading of Mo to 16 wt%, both CH4 conversion and aromatic yield
slightly decreased, while C2-hydrocarbon yield as well as selectivity/
yield ratio of C2H4/C2H6 slightly increased. On the other hand, of those
Mo/HZSM-5(MA) catalysts investigated, the 10 wt% Mo-impregnated
sample exhibited the best catalytic stability in this aromatization re-
action. After 38 h of continuous reaction at 700 °C and space velocity
1500 mL g−1 h−1, this catalyst was still able to provide an aromatic
yield of 7% and a CH4 conversion of 9.1% (Fig. 2b). For comparison, the
aromatization of methane over 10 wt% Mo/HZSM-5(ZA) catalyst, pre-
pared by conventional impregnation of NH3-treated HZSM-5 zeolite
with unbasified molybdate aqueous solution, was also investigated
under the same reaction conditions. The result indicated that the am-
monia treatment of the zeolite before the Mo-impregnating process
could also improve the catalytic performance of the Mo sample, but its
effect on this performance was weaker than that of ammonia basifica-
tion of impregnating solution in catalyst preparation (Figs. 1 and 2a).
loading of the Mo-doped sample increased with increasing the targeted
Mo loading value to 10 wt%, at which the difference between nominal
and measured Mo loading of Mo/HZSM-5(MA) was, however, less than
those of conventional Mo/HZSM-5 and Mo/HZSM-5(ZA), implying that
there were more Mo species adsorbed or interacted with the zeolite on
the Mo/HZSM-5(MA). XRD patterns of HZSM-5 zeolite and Mo/HZSM-5
catalysts prepared via different methods are presented in Figs. 3b and
S2. All samples exhibited typical ZSM-5 zeolite structure. The zeolitic
crystallinity of conventional Mo catalysts decreased with increasing Mo
loading (Fig. S2a), indicating an interaction between Mo species and
the zeolite. After NH3 treatment/basification of the zeolite or the im-
pregnating solution in catalyst preparation, further decrease in the
zeolitic crystallinity of Mo/HZSM-5 was observed. Meanwhile, the in-
tensities of peaks at low angles of 8 and 9° decreased significantly re-
lative to those at angles of 22.5-25°, implying that there was a gradual
transformation from orthorhombic to monoclinic symmetry in the
zeolite entities [33,34]. Compared with Mo/HZSM-5(ZA), Mo/HZSM-
5(MA) showed less symmetry transformation and better zeolitic crys-
tallinity. On the other hand, orthorhombic α-MoO3 (2θ = 12.7, 25.7,
and 27.3°) (JCPDS 05-0508) crystallite was detected on all the Mo
samples investigated. Among the 10 wt% Mo-impregnated HZSM-5
zeolites, Mo/HZSM-5(MA) prepared with ammonia-basified impreg-
nating solution exhibited the weakest intensities of these signals of
orthorhombic α-MoO3 (Fig. S2b and d).
The surface areas and porous textural properties of HZSM-5, con-
ventional and ammonia-treated/basified Mo/HZSM-5 samples with Mo
loadings of 6 and 10 wt% were measured by nitrogen adsorption (Fig. 4
and Table 2). The parent zeolite exhibited a type-I isotherm with a
small identified hysteresis loop at high p/p0 = 0.44–1.0, which was
typical for microporous materials containing a small amount of meso
and macropores with a broad pore-size distribution centered around
approximately 80 nm (Figs. 4b and S3). Application of the t-plot
method showed that 79% of the surface area of the zeolite was con-
tributed by its micropores (Table 2). Introduction of 6 wt% Mo to the
HZSM-5 zeolite via conventional impregnation led to losses in surface
area (BET) and microporous volume, but increases in mesoporous vo-
lume and average pore size of the zeolite. The substantial increment of
the external surface area and mesoporous volume of the Mo-doped
sample demonstrated the dissolution of the framework and the forma-
tion of mesopores during the catalyst preparation, which was further
confirmed by a clear exhibition of mesopore at ca. 18 nm in the BJH
adsorption pore size distribution (Fig. 4b). Meanwhile, the decrement
of microporous volume more than the increment of the mesoporous
volume implied the migration of part of the loaded Mo species into the
micropores of the zeolite. All the surface area and porous properties
diminished with increasing Mo target loading from 6 to 10 wt%. The
decreased volume of mesopores with increasing Mo loading implied
that the migrated Mo species deposited in the mesopores in addition to
3.2. Mo content and structure analyses of Mo/HZSM-5
For understanding of this novel catalytic performance of 10 wt%
Mo/HZSM-5(MA), this work investigated physicochemical properties of
those 10 wt% Mo/HZSM-5, prepared by various methods, and disper-
sion of Mo species on the support zeolite and interaction between them.
For comparison, these features of conventional 6 wt% Mo/HZSM-5, the
most widely studied catalyst, was also included. Mo content analyses
(Fig. 3a) showed that difference between nominal and measured Mo
113

P. Tan
AppliedCatalysisA,General580(2019)111–120
Fig. 1. Catalytic aromatization of methane at 700 °C and space velocity 1500 mL g−1 h−1 (90 vol% CH4/Ar) over (1) conventional, and (2) NH3-treated/basified Mo/
HZSM-5 materials with various Mo target loadings: (a) CH4 conversions, (b) aromatic yields, (c) C2-hydrocarbon yields, and (d) selectivity/yield ratios of C2H4/C2H6.
the micropores of the zeolite.
respectively, than those of the conventional Mo/HZSM-5 (Table 2).
Meanwhile, for Mo/HZSM-5(MA), the pore size distribution derived
from the adsorption branch (Fig. 4b) clearly evidenced that the formed
mesopores showed a broad distribution centered around ca. 18 nm
(Fig. 4b). Compared with Mo/HZSM-5(MA), Mo/HZSM-5(ZA) exhibited
At Mo target loading of 10 wt%, the BET surface area and average
pore size of Mo/HZSM-5(MA) were similar to those of the con-
ventionally prepared Mo/HZSM-5, while the micro and mesoporous
volumes of Mo/HZSM-5(MA) were obviously less and more,
114

P. Tan
AppliedCatalysisA,General580(2019)111–120
Fig. 2. Catalytic activities of conventional and NH3-treated/basified Mo/HZSM-5 materials with various Mo target loadings after methane aromatization for 10 h (a),
and catalytic performance of 10 wt% Mo/HZSM-5(MA) material for methane aromatization (b). Note, reaction temperature, 700 °C; space velocity, 1500 mL g−1 h−1
(90 vol% CH4/Ar).
less BET surface area and microporous volume, and larger average pore
size. In addition, the isotherms of the latter showed a small hysteresis
loop at low p/p0 = 0.03−0.35 whose exhibition was attributed to
presence of a large amount of extra-framework Al species, probably
blocking micropores of the HZSM-5 zeolite [35,36].
or Mo/HZSM-5(ZA), Mo/HZSM-5(MA) exhibited high surface Si con-
centration and low surface Mo concentration, consistent with the ana-
lyses of the XPS spectra in the Mo 3d (Fig. 5c) and O 1s regions (Fig.
S4), implying less amount of Mo remained the external surface of the
zeolite after the Mo-doped catalyst preparation.
The dispersion of molybdenum oxides on the Mo/HZSM-5 was
further assessed by UV–vis spectroscopy. It was well known that the
absorption bands between 200 and 400 nm of MoO3 were attributed to
O2−→Mo6+ charge transfer (CT) transitions. The CT bands in the range
220–250 nm and around 280 nm have been commonly assigned to
tetrahedrally coordinated Mo-oxo species (MoO4) [32,37–39], while
those at higher wavelengths of 300–330 nm to Mo-oxo centers in oc-
tahedral geometry (MoO6) [32,38–41]. Furthermore, while the tetra-
hedral Mo-oxo species with CT bands at low wavelengths of
220–250 nm was reported to be isolated monomeric [38,40,41], Je-
zlorowski and Knözinger suggested that the absorption at the band
around 280 nm was possibly due to the formation of Mo-O-Mo con-
taining structures [38]. In this case, all the UV–vis spectra of 6 wt% and
10 wt% Mo-impregnated HZSM-5 zeolites investigated contained bands
at 230 (sharp) and 255 (broad) in addition to the band in the
300–330 nm range that might be assigned to octahedral Mo (Fig. 5d).
Since intense absorbance at 230 nm was not observed in the spectra of
both MoO3/SiO2, which exhibited three absorption bands at 220 (weak
shoulder), 260 (main), and 320 nm, respectively, [42], and MoO3/γ-
Al2O3 with absorption bands at 240, 280–290, and 320 nm [32],the
3.3. Dispersion of Mo species on HZSM-5
The near-surface compositions, which is mostly relevant to the ex-
ternal surface of the zeolite and perhaps the channels closest to it, of the
parent and Mo-modified HZSM-5 zeolites were measured by means of
XPS technique. The near-surface composition analyses (Fig. 5a and b) of
conventional Mo/HZSM-5 samples showed that with increasing Mo
loading to 10 wt%, there were an increase in surface Mo concentration
and decreases in surface Si and Al concentrations, but the increment
and decrement of them respectively gradually declined. Surface con-
centrations of Al and Si, as well as Al/Si ratio of the Mo-doped samples
lower than those of the parent HZSM-5 confirmed that part of Mo
species covered Si sites, and more preferentially, Al sites after catalyst
preparation. At Mo target loading of 10 wt%, the surface Al con-
centrations of Mo/HZSM-5(MZ) and Mo/HZSM-5(ZA) sample were
0.64% and 0.74%, respectively, which were both higher than that of the
conventional Mo/HZSM-5 one, demonstrating that ammonia-treatment
of HZSM-5 or impregnating solution in the catalyst preparation could
enhance the dealumination of the zeolite. Compared with Mo/HZSM-5
Fig. 3. Mo contents (a), and XRD patterns (b) of fresh 6 wt% and 10 wt% Mo-impregnated HZSM-5 zeolites prepared by various methods.
115

P. Tan
AppliedCatalysisA,General580(2019)111–120
nanoparticles [43] of Mo/HZSM-5 material.
At a high Si/Al ratio of 25, it was reported that the possibility to find
two Brønsted acid sites with an adequate distance diminished and the
2+
formation of monomeric dihapto Mo(=O)2
species located on the
zeolitic framework became difficult [43]. Based on Raman spectra
analyses and DFT calculations, Gao et al. [43] suggested this kind of
isolated tetrahedral Mo be monomeric monohapto Mo(=O)2(OH)+
species that anchored one framework Al atom via two oxygen bridges.
Meanwhile, the existence of Al atoms as double anchoring sites in the
arrangement Al-O-(Si-O)2-Al of HZSM-5 with Si/Al higher than 15 al-
2+
lowed the formation of dimeric Mo(=O)2 species in the channels of
the zeolite [43]. When the Si/Al ratio of the HZSM-5 zeolite reached 25
or 40, the Mo(=O)2(OH)+ species anchored on single Al-atom sites
became the dominant Mo dioxo species on the framework Al sites of the
Mo/HZSM-5 catalyst [43].
In this study, the UV–vis spectra (Fig. 5d) showed that with in-
creasing Mo loading from 6 to 10 wt% on the conventional Mo/HZSM-5
sample, the absorptions in the 245–260 nm and 300–330 nm ranges
became more intensive, indicating that the formations of tetrahedral
Mo species on Al or Si sites and octahedral Mo ones were both en-
hanced. At the Mo target loading of 10 wt%, the intensity exhibitions of
those signals associated with Mo species depended on the preparation
method of the Mo/HZSM-5 catalyst. Compared with the spectrum of the
conventional Mo/HZSM-5, that of Mo/HZSM-5(ZA) contained slightly
more intensive band at about 230 nm, and obviously less intensive ones
above 255 nm, implying that pretreatment of HZM-5 zeolite with am-
monia prior to impregnating process was unfavourable for the forma-
tions of octahedral Mo species. Among these 10 wt% Mo-doped samples
investigated, the Mo/HZSM-5(MA) possessed the most amount of tet-
rahedrally coordinated Mo-oxo species, as reflected in both the en-
hancement of the intensities in the bands rang 245–265 nm, and the
broadening in the signal around 230 nm, which indicated the coex-
istence of molybdenum species on normal or distorted Si-O−-Al of the
zeolite. After comparison of the distribution of Mo species measured by
UV–vis with catalytic activity of Mo/HZSM-5, these tetrahedral Mo-oxo
species were possibly the precursors of the active Mo phases for me-
thane activation.
3.4. Dealumination and surface acidic properties of Mo/HZSM-5
Fig. 4. N2 adsorption and desorption isotherms (a), and BJH adsorption pore
size distributions (b) of parent HZSM-5 zeolite, and fresh 6 wt% and 10 wt%
Mo-impregnated HZSM-5 zeolites prepared by various methods.
Modification of Mo and/or basic ammonia treatment will change
the nature and environment of Al species of the HZSM-5 zeolite, most of
which can be readily characterized with 27Al MAS NMR measurement
[44,45].As shown in Fig. 6a, the 27Al NMR spectrum of the parent
HZSM-5 zeolite contained a dominant feature at δ = 55 ppm (δ, che-
mical shift) associated with four-coordinated framework Al (F-Al)
atoms and a very small feature at δ =0 ppm associated with octahedral
extra-framework Al (EF-Al) species, such as [Al(OH)n]3−n and [Al
(H2O)6]3+ [44]. For the conventional samples, a loading of 6 wt% Mo
led to a decrease and broadening of the F-Al signal caused by the
proximity of cationic Mo-oxo complexes that replace the protons [45],
highly intensive band at this wavelength detected on Mo/HZSM-5
might be attributed to isolated tetrahedral Mo adsorbed on Si-O–Al of
the support zeolite. The broadened band at ca. 255 nm could be further
distinguished into two features, one at 250 nm and another at 260 nm.
While the latter was assigned to tetrahedral Mo oxide species on zeolitic
Si sites [42], the former with lower intensity than that at 230 nm might
be assigned to tetrahedral Mo species on extra-framework alumina
Table 2
Surface areas and pore properties of the parent HZSM-5 zeolite, and conventional and NH3-treated/basified Mo/HZSM-5 samples with various Mo target loadings.
Sample
SBET (m2/g)
Sexternala(m2/g)
Smicroporea(m/g)
Vtotalb(cm3/g)
Vmicroporea(cm3/g)
Vmesoporec(cm3/g)
Pore sized(Å)
HZSM-5
6 wt% Mo
10 wt% Mo
10 wt% Mo (ZA)
10 wt% Mo (MA)
309.4
277.8
265.2
219.2
260.5
65.6
98.7
96.7
86.0
105.5
243.8
179.1
168.5
133.2
155.0
0.153 (PD < 693.1 Å)
0.143 (PD < 708.7 Å)
0.136 (PD < 652.1 Å)
0.118 (PD < 678.1 Å)
0.134 (PD < 701.3 Å)
0.113
0.083
0.078
0.062
0.071
0.040
0.060
0.058
0.056
0.063
19.75
20.57
20.45
21.50
20.65
Note, S, surface area; V, volume.
a
Calculated by t-plot method.
b
Single point adsorption total pore volume of pores at p/p0 = 0.97; PD, pore diameter.
c
Vmesopore = Vads,p/p0 = 0.97 − Vmicropore
Adsorption average pore width (4 V/A by BET).
.
d
116

P. Tan
AppliedCatalysisA,General580(2019)111–120
Fig. 5. Near-surface atomic concentrations (a) and atomic ratios (b), XPS spectra in the Mo 3d region (c), and UV–vis spectra (d) of fresh 6 wt% and 10 wt% Mo-
impregnated HZSM-5 zeolites prepared by various methods.
and an increase in the intensity of the signal at δ =0 ppm. The peak
maximum of this resonance associated with F-Al gradually shifted to
high field (around 54 ppm). When the targeted loading of Mo increased
to 10 wt%, the main 27Al signal at the chemical shift of about 54 ppm
further broadened slightly, while its intensity was similar to that of the
conventionally prepared 6 wt% Mo/HZSM-5 (Figs. 6a and S5). Rahman
et al. [29] attributed the broadening of the main signal given by fra-
mework Al atoms due to increase in Mo loading from 6 to 10 wt% to the
exchange of protons and cationic Mo-oxo complex species. By curve-
fitting of the 27Al NMR spectra (Fig. S6), a new band at δ = 44.5 ppm,
which was possibly associated with the formation of slightly distorted
framework Al, could be distinguished from this broadened 27Al signal of
the 10 wt% Mo sample. At the same time, compared with the 6 wt% Mo
sample, the 10 wt% Mo one possessed slightly more amount of EF-Al
(Fig. S5), implying that extraction of framework Al by Mo species did
not strengthen markedly. For Mo/HZSM-5, since zeolitic dealumination
Fig. 6. 27Al MAS NMR spectra (a), and NH3-TPD profiles (b) of parent HZSM-5 zeolite, and fresh 6 wt% and 10 wt% Mo-impregnated HZSM-5 zeolites prepared by
various methods.
117

P. Tan
AppliedCatalysisA,General580(2019)111–120
was reported by multiple research groups as the key factor affecting the
structure of the support, this result reasonably explained that the XRD
patterns of the 6 and 10 wt% Mo-doped samples were similar (Figs. 3b
and S2). Curve fitting analysis (Fig. S6) of this Al spectrum showed that
ca. 10.6% of total Al detected was in the form of extra-framework en-
tities. In addition, a weak shoulder at δ = 30 ppm associated with dis-
torted four or five-coordinated Al species [27,33] became readily dis-
tinguished.
At the targeted loading of 10 wt% Mo, the intensity of the signal at δ
=0 ppm further increased after the HZSM-5 zeolite or impregnating
solution was treated or basified with ammonia in the catalyst pre-
paration process. Meanwhile, a new feature at a chemical shift of about
-14 ppm due to EF-Al species in the form of Al2(MoO4)3 [44,45] were
clearly observed. Compared with Mo/HZSM-5(ZA), Mo/HZSM-5(MA)
exhibited obviously low contents of Al2(MoO4)3 and distorted Al species
formed during the catalyst preparation. After curve fitting of the spectra
(Fig. S6), quantitative analyses showed that the total concentrations of
extra-framework Al on the Mo/HZSM-5(MA) and Mo/HZSM-5(ZA)
catalysts were about 19.7% and 22.4%, respectively.
NH3-TPD is a usual tool used to measure the surface acidity of a
pure or metal-modified zeolite. In this study, each NH3-TPD signal of
parent and Mo-doped HZSM-5 zeolites might be deconvoluted into
three distinct components (Fig. 6b). For the parent HZSM-5 zeolite, the
features at 240 °C (40.6%), 360 °C (19.6%), and 460 °C (39.8%) were
attributed to desorptions of chemisorbed NH3 on weak acid site, mod-
erate acid site possibly associated with extra framework Al species, and
Brønsted acid site, respectively [46]. With increasing Mo target loading
from 6 to 10 wt%, both the amount and strength of surface Brønsted
acid of the Mo/HZSM-5 catalyst decreased, as reflected in the reduction
of intensity and desorption temperature of the signal at 460 °C of the
NH3-TPD profiles (Fig. 6b). Ammonia treatment/basification of the
HZSM-5 zeolite or impregnating solution in this Mo catalyst prepara-
tion process resulted in further decline in the properties of the acids
including Brønsted ones. About 43%, 25%, and 21% of zeolitic surface
Brønsted acid sites were remained on Mo/HZSM-5, Mo/HZSM-5(MA),
and Mo/HZSM-5(ZA) catalysts with the targeted loading of 10 wt% Mo,
respectively.
MoO3, more dispersion of MoO3 became gradually difficult and the
increment and decrements in the surface concentrations of Mo, Al, and
Si respectively decreased. Since those exchanged Mo species inside the
zeolite pores are reported to be the main precursors of active phases
responsible for the activation of methane [2,4,23,49–51], this could
reasonably explain the enhanced catalytic activity of Mo/HZSM-5 with
suitably increasing Mo loading. On the other hand, although acid sites
are needed for the conversion of intermediate ethylene to aromatics,
excessive acidic sites would promote dehydrogenation of aromatic
products to carbonaceous materials, which would deposit on the active
sites. The increased dispersion of Mo species would gradually consume
extra acid sites of the support and, hence, enhanced the catalytic sta-
bility of the Mo/HZSM-5 material.
For the conventional Mo-impregnated HZSM-5 zeolite, the recent
work of the author’s [30] indicated that the catalytic behaviors of Mo/
HZSM-5 with Mo loading higher than 10 wt% in methane aromatization
were significantly improved after pretreatment of these Mo-based
samples with a CH4/Ar/He (9:1:10, v/v) gas mixture; a Mo loading of
16 wt% was found to provide the best catalytic activity and stability for
this aromatization reaction over Mo/HZSM-5 prepared via conven-
tional impregnation and calcination at 500 °C. Similarly, Rahman et al.
[29] reported that the catalytic performance of Mo/HZSM-5 with Mo
loading of 10 or higher were remarkably better than those of Mo/
HZSM-5 with Mo loading of 6 wt% or lower in methane aromatization
after this Mo-impregnated HZSM-5 zeolite was calcined at 500 °C, and
was then pretreated with a CH4/H2 gas mixture. Although the research
group of Hensen gave obviously different results [52], their conclusion
is not in contradiction to those of the author’s [30] and Khatib’s group
[29] because the Mo/HZSM-5 samples tested by Hensen et al. were
calcined at higher temperature (550 °C or higher) during the catalyst
preparation, which would lead to a stronger interaction between Mo
species and the support zeolite.
The result that catalytic performance of conventional Mo/HZSM-5
was enhanced with increasing Mo target loading to 16 wt% (Fig. 2a)
implied the possibility of further improving the catalytic performance
of the conventional Mo/HZSM-5 with Mo content less than 16 wt% via
enhancing dispersion of Mo and/or declining Brønsted acid sites. In this
case, although the dispersion of Mo species was not obviously improved
on Mo/HZSM-5(ZA) sample, near-surface atomic concentration and
27Al NMR measurement demonstrated that ammonia could extract
zeolitic framework Al species to form external-framework Al entities.
This dealumination was kinetically favourable for the formation of
Al2(MoO4)3 with MoO3 loaded on HZSM-5, which was confirmed by
27Al NMR of Mo/HZSM-5 (Fig. 6a). The consumption of some MoO3
loaded was probably responsible for the weakened intensities of these
signals of orthorhombic α-MoO3 exhibited in the XRD pattern of Mo/
HZSM-5(ZA) (Fig. 3b). On the other hand, this dealumination led to an
additional loss of Brønsted acid sites as well as an increase in the
average pore size of the zeolite. For this Mo/HZSM-5(ZA) with Mo
target loading of 10 wt% in methane aromatization, the loss of zeolitic
Brønsted acid sites caused by ammonia-treatment of HZSM-5 zeolite
was responsible for its improved stability, while the increase in the
average pore size of the zeolite for the increased selectivity to heavier
aromatic product naphthalene.
4. Discussion
Two kinds of molybdate ions, monomolybdate (MoO42−) and
6-
polymolybdate ions (Mo7O24 and Mo8O264-), are reported in ammo-
nium heptamolybdate (AHM) aqueous solution [47]. Their concentra-
tions depend on the pH of the AHM solution. After mixing with HZSM-5
zeolite without ammonia basification, the pH value of AHM solution
decreased from 6 to about 3. At this pH value, polymolybdate ions such
as Mo7O246- were the predominant molybdate ones [47]. Owing to their
large sizes, these Mo species distributed on the external surface after
impregnation. It was reported that the acidity of the zeolite was not
modified and relatively little Mo was present in the channels of ZSM-5
when the Mo-impregnated HZSM-5 material was treated at tempera-
tures as high as 400 °C [48]. After calcination at 500 °C under air, N2
adsorption measurement clearly demonstrated that part of Mo diffused
into the zeolitic channels. Since dealumination was not severe, the di-
minished acidic property of the zeolite after Mo modification could be
mainly attributed to Mo exchange with H+ of Si-(OH)-Al due to the
electronic attraction between the oxygen of MoO3 and the H+. Further
UV–vis measurement indicated that those exchanged Mo ions were
dominantly present as a kind of tetrahedral Mo-oxo species, which was
possibly in the form of [Mo(=O2)(OH)]+ anchoring on zeolitic
Brønsted acid site as suggested by Gao et al. [43] and Schwach et al.
[4]. With increasing Mo loading from 6 to 10 wt%, UV–vis and NH3-
TPD results showed more exchange of MoO3 with the H+ due to the
increased concentration difference of Mo between the external and
internal surfaces of the zeolite. However, owing to the strengthened
electronic repulsion of oxygen atoms of the dispersing and exchanged
When the impregnating solution was basified by ammonia, the ef-
fect of NH3 on the surface structure of the Mo/HZSM-5 material became
2−
more complex. Under the basic condition, MoO4
in place of poly-
6-
molybdate ions such as Mo7O24 became the predominant molybdate
ion in the AHM solution [47], which might readily enter the channels of
the zeolite due to its small size. After impregnation of the HZSM-5
+
zeolite with ammonia-basified AHM solution, NH4 would replace the
H+ of the zeolitic acid and be absorbed on the Brønsted acid site.
2−
Owing to the electronic attraction, the OH- and MoO4
trended to
approach cationic ammonium ions including the ones absorbed on
+
+
acidic sites to form loose ion-pairs, e. g., NH4ad.…OH- and NH4ad.
…
MoO42−…NH4+ad.(or NH4dissociative) (Scheme 1). On the other hand,
+
118

P. Tan
AppliedCatalysisA,General580(2019)111–120
Scheme 1. The effect of ammonia on the evolution of Mo species during the preparation of Mo/HZSM-5.
according to the NMR analysis (Fig. 6a), Mo/HZSM-5(MA) exhibited
less zeolitic dealumination than Mo/HZSM-5(ZA). Since they were
prepared using the same original chemicals, MoO42−, formed in the
NH3-basified AHM solution, should play an important role in dimin-
ishing the extraction of zeolitic framework Al of Mo/HZSM-5(MA). As
5. Conclusions
After pretreatment with a mixture of CH4/Ar/He, (9:1:10, v/v), Mo/
HZSM-5 material with Mo target loading of 10 wt%, prepared by im-
pregnation of HZSM-5 zeolite with NH3-basified molybdate aqueous
solution, exhibited significantly enhanced activity and stability in cat-
alysing methane to aromatic hydrocarbons in the absence of an added
oxidant. After 38 h of continuous reaction at 700 °C and space velocity
1500 mL g−1 h−1 (90 vol% CH4/Ar), this catalyst was still able to give
an aromatic yield of 7% and a methane conversion of 9.1%. This im-
proved catalytic performance was attributed to a larger amount of
highly dispersed Mo species in the zeolitic channels, and a less amount
of Brønsted acid sites on Mo/HZSM-5, which were caused by the ba-
sification of impregnating solution with ammonia in catalyst prepara-
tion. Pretreatment with a diluted CH4 with 55 vol% inert gases Ar/He
prior to the reaction is helpful for this Mo-doped catalyst to exhibit its
inherent catalytic behaviours via both prereduction of the catalyst with
CH4 and rapid removal of adsorbed/produced water from the catalyst.
Up to now, much effort has been devoted to enhancing catalytic activity
and stability of Mo/HZSM-5 material with Mo loading of 6 wt% or
lower. This work shows a new and highly efficient pathway to improve
the catalytic performance of Mo/HZSM-5 material for methane ar-
omatization via increasing Mo loading to 10 wt%, and provides an in-
sight into the effect of ammonia on the surface structure of Mo/HZSM-5
material.
2−
shown in Scheme 1, due to the electronic repulsion between MoO4
2−
+
and OH-, the MoO4
adsorbed on the NH4ad. might act as a check to
prevent the dissociative OH- from approaching the Al-OH-Si and ex-
tracting the framework Al. The decrease in dealumination made it
possible that more Brønsted acid sites were remained after impregna-
tion, increasing exchange of the H+ and Mo in the subsequent calcining
process of Mo/HZSM-5(MA).
According to the analyses of ICP-AES and XPS, the Mo/HZSM-
5(MA) possessed slightly more content of total Mo but remarkably less
surface Mo and O concentrations than the conventional Mo/HZSM-5
and Mo/HZSM-5(ZA). The results combined with BJH adsorption pore
size distribution (Fig. 4b) of the Mo catalyst implied that there was
more Mo entering the channels of the support after the preparation of
Mo/HZSM-5(MA), which originated from the adsorbed or dissociative
2−
MoO4
in the zeolitic pores in the impregnating process. The ad-
sorption measurement of perfluorotributyl amine showed that the
Brønsted acid sites of the HZSM-5 zeolite were mostly located inside the
zeolitic channels [24]. The spatial advantage of the Mo species in the
zeolitic channels made them easier for their exchange with the proton
of the Al−OH-Si. Thus, the increased amounts of Mo-oxo species in the
pores of HZSM-5 and Al−OH-Si remained on the Mo/HZSM-5(MA)
catalyst led to more exchanged Mo on Brønsted acid sites, the pre-
cursors of the active species, demonstrated by the UV–vis measure-
ments, which was responsible for the enhancement in the catalytic
activity of this ammonia-basified catalyst for methane activation.
Meanwhile, the enhancement in Mo migration into zeolitic channels
driven by NH3-basification partly counteracted the increment of the
pore volume due to the framework corrosion by ammonia. This part
counteracting led to the average pore size of Mo/HZSM-5(MA) slightly
larger that of conventional Mo/HZSM-5, but obviously less than that of
Mo/HZSM-5(ZA), and, consequently, the same evolutions in the se-
lectivity to heavy naphthalene as well as selectivity ratio of benzene/
naphthalene observed over these shape-selective catalysts in methane
aromatization. On the other hand, for Mo/HZSM-5(MA), more ex-
change of the proton with Mo and extraction of framework Al con-
sumed the extra acid sites of the conventional Mo/HZSM-5 with Mo
target loading of 10 wt% for methane aromatization and, consequently,
the catalytic stability of the NH3-basified Mo sample was significantly
improved in this aromatization reaction.
Acknowledgements
The author acknowledges financial support from Anhui University
of Technology.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2019.05.011.
References
[1] Z.R. Ismagilov, E.V. Matus, L.T. Tsikoza, Energy Environ. Sci. 1 (2008) 526–541,
https://doi.org/10.1039/B810981H.
[2] J.J. Spivey, G. Hutchings, Chem. Soc. Rev 43 (2014) 792–803, https://doi.org/10.
1039/c3cs60259a.
[3] C. Karakaya, R.J. Kee, Prog. Energy Combust. Sci. 55 (2016) 60–97, https://doi.
org/10.1016/j.pecs.2016.04.003.
[4] P. Schwach, X. Pan, X. Bao, Chem. Rev. 117 (2017) 8497–8520, https://doi.org/10.
1021/acs.chemrev.6b00715.
[5] L. Wang, L. Tao, M. Xie, G. Xu, J. Huang, Y. Xu, Catal. Lett. 21 (1993) 35–41,
119

P. Tan
AppliedCatalysisA,General580(2019)111–120
https://doi.org/10.1007/BF00767368.
org/10.1016/j.apcata.2018.03.023.
[6] S. Liu, L. Wang, R. Ohnishi, M. Ichikawa, J. Catal. 181 (1999) 175–188, https://doi.
[30] P. Tan, Catal. Commun. 103 (2018) 101–104, https://doi.org/10.1016/j.catcom.
org/10.1006/jcat.1998.2310.
2017.10.008.
[7] B.M. Weckhuysen, D. Wang, M.P. Rosynek, J.H. Lunsford, J. Catal. 175 (1998)
338–346, https://doi.org/10.1006/jcat.1998.2010.
[31] L. Wang, W.K. Hall, J. Catal. 66 (1980) 251–255, https://doi.org/10.1016/0021-
9517(80)90029-9.
[8] S. Liu, Q. Dong, R. Ohnishi, M. Ichikawa, Chem. Commun. (1997) 1455–1456,
[32] G. Xiong, Z. Feng, J. Li, Q. Yang, P. Ying, Q. Xin, C. Li, J. Phys. Chem. B 104 (2000)
https://doi.org/10.1039/A702731A.
3581–3588, https://doi.org/10.1021/jp993609z.
[9] S. Li, C. Zhang, Q. Kan, D. Wang, T. Wu, L. Lin, Appl. Catal. A 187 (1999) 199–206,
https://doi.org/10.1016/S0926-860X(99)00231-8.
[10] B. Liu, Y. Yang, A. Sayari, Appl. Catal. A 214 (2001) 95–102, https://doi.org/10.
1016/S0926-860X(01)00470-7.
[33] T.C. Hoff, R. Thilakalatne, D.W. Gardner, R.C. Brown, J.-P. Tessonnier, J. Phys.
Chem. C 120 (36) (2016) 20103–20113, https://doi.org/10.1021/acs.jpcc.
6b04671.
[34] H. van Koningsveld, J.C. Jansen, H. van Bekkum, Zeolites 10 (1990) 235–242,
[11] J. Zhao, X. Wang, T. Zhang, J. Tang, L. Ren, L. Li, L. Lin, Chin. J. Catal. 23 (2002)
197–198 http://en.cnki.com.cn/Article_en/CJFDTOTAL-CHUA200203001.htm.
[12] Y. Zhang, D. Wang, J. Fei, X. Zheng, Aust. J. Chem. 55 (2002) 531–534, https://doi.
org/10.1071/CH01170.
[13] A.V. Vosmerikov, V.I. Zaikovskii, L.L. Korobitsyna, G.V. Echevskii, V.V. Kozlov,
Y.E. Barbashin, S.P. Zhuravkov, Kinet. Catal. 50 (2009) 725–733, https://doi.org/
10.1134/S0023158409050140.
https://doi.org/10.1016/0144-2449(94)90134-1.
[35] J.C. Groen, L.A.A. Peffer, J. Pérez-Ramírez, Micropor. Mesopor. Mat. 60 (2003)
1–17, https://doi.org/10.1016/S1387-1811(03)00339-1.
[36] Z. Hu, H. Zhang, L. Wang, H. Zhang, Y. Zhang, H. Xu, W. Shen, Y. Tang, Catal. Sci.
Technol. 4 (2014) 2891–2895, https://doi.org/10.1039/C4CY00376D.
[37] R.K. Rana, B. Visvanathan, Catal. Lett. 52 (1998) 25–29, https://doi.org/10.1023/
A:1019019403375.
[14] M.W. Ngobeni, A.F. Carley, M.S. Scurrell, C.R. Nicolaides, J. Mol. Catal. A 305
(2009) 40–46, https://doi.org/10.1016/j.molcata.2008.10.047.
[15] A.K. Aboul-Gheit, A.E. Awadallah, A.A. Aboul-Enein, A.L.H. Mahmoud, Fuel 90
(2011) 3040–3046, https://doi.org/10.1016/j.fuel.2011.05.010.
[16] P.L. Tan, C.T. Au, S.Y. Lai, Appl. Catal. A 324 (2007) 36–41, https://doi.org/10.
1016/j.apcata.2007.03.002.
[38] H. Jezlorowski, H. Knözinger, J. Phys. Chem. 83 (1979) 1166–1173, https://doi.
org/10.1021/j100472a012.
[39] A. Martinez, E. Peris, Appl. Catal. A 515 (2016) 32–44, https://doi.org/10.1016/j.
apcata.2016.01.044.
[40] C.C. Williams, J.G. Ekerdt, J.M. Jehng, F.D. Hardcastle, A.M. Turek, I.E. Wachs, J.
Phys. Chem. 95 (1991) 8781–8791, https://doi.org/10.1021/j100175a067.
[41] E.L. Lee, I.E. Wachs, J. Phys. Chem. C 111 (2007) 14410–14425, https://doi.org/
10.1021/jp0735482.
[17] Y. Song, C. Sun, W. Shen, L. Lin, Catal. Lett. 109 (2006) 21–24, https://doi.org/10.
1007/s10562-006-0066-2.
[18] Y. Li, L. Liu, X. Huang, X. Liu, W. Shen, Y. Xu, X. Bao, Catal. Commun. 8 (2007)
1567–1572, https://doi.org/10.1016/j.catcom.2007.01.013.
[19] W. Ding, G.D. Meitzner, E. Iglesia, J. Catal. 206 (2002) 14–22, https://doi.org/10.
1006/jcat.2001.3457.
[20] N. Chu, J. Yang, J. Wang, S. Yu, J. Lu, Y. Zhang, D. Yin, Catal. Commun. 11 (2010)
513–517, https://doi.org/10.1016/j.catcom.2009.12.004.
[42] M. Šustek, B. Horváth, I. Vávra, M. Gál, E. Dobročka, M. Hronec, Chin. J. Catal. 36
(2015) 1900–1909, https://doi.org/10.1016/S1872-2067(15)60961-5.
[43] J. Gao, Y. Zheng, J.M. Jehng, Y. Tang, I.E. Wachs, S.G. Podkolzin, Science 348
(2015) 686–690, https://doi.org/10.1126/science.aaa7048.
[44] D. Ma, Y. Lu, L. Su, Z. Xu, Z. Tian, Y. Xu, L. Lin, X. Bao, J. Phys. Chem. B 106 (2002)
8524–8530, https://doi.org/10.1021/jp020166h.
[21] Y. Xu, W. Liu, S. Wong, L. Wang, X. Guo, Catal. Lett. 40 (1996) 207–214, https://
[45] J.-P. Tessonnier, B. Louis, S. Rigolet, M.J. Ledoux, C. Pham-Huu, Appl. Catal. A 336
(2008) 79–88, https://doi.org/10.1016/j.apcata.2007.08.026.
[46] P. Tan, J. Catal. 338 (2016) 21–29, https://doi.org/10.1016/j.jcat.2016.01.027.
[47] F. Barath, M. Turki, V. Keller, G. Maire, J. Catal. 185 (1999) 1–11, https://doi.org/
10.1006/jcat.1999.2478.
doi.org/10.1007/BF00815284.
[22] W. Liu, Y. Xu, J. Catal. 185 (1999) 386–392, https://doi.org/10.1006/jcat.1999.
2515.
[23] E.V. Matus, I.Z. Ismagilov, O.B. Sukhova, V.I. Zaikovskii, L.T. Tsikoza,
Z.R. Ismagilov, J.A. Moulijn, Ind. Eng. Chem. Res. 46 (2007) 4063–4074, https://
doi.org/10.1021/ie0609564.
[48] D. Wang, J.H. Lunsford, M.P. Rosynek, J. Catal. 169 (1997) 347–358, https://doi.
org/10.1006/jcat.1997.1712.
[24] W. Zhang, D. Ma, X. Han, X. Liu, X. Bao, X. Guo, X. Wang, J. Catal. 188 (1999)
[49] R.W. Borry III, Y.H. Kim, A. Huffsmith, J.A. Reimer, E. Iglesia, J. Phys. Chem. B 103
(1999) 5787–5796, https://doi.org/10.1021/jp990866v.
393–402, https://doi.org/10.1006/jcat.1999.2670.
[25] D. Ma, Y. Shu, X. Bao, Y. Xu, J. Catal. 189 (2000) 314–325, https://doi.org/10.
1006/jcat.1999.2704.
[50] Y. Wu, L. Emdadi, S.C. Oh, M. Sakbodin, D. Liu, J. Catal. 323 (2015) 100–111,
https://doi.org/10.1016/j.jcat.2014.12.022.
[26] C.H.L. Tempelman, X. Zhu, E.J.M. Hensen, Chin. J. Catal. 36 (2015) 829–837,
https://doi.org/10.1016/S1872-2067(14)60301-6.
[27] C.H.L. Tempelman, E.J.M. Hensen, Appl. Catal. B 176–177 (2015) 731–739,
https://doi.org/10.1016/j.apcatb.2015.04.052.
[51] N. Kosinov, F.J.A.G. Coumans, E.A. Uslamin, A.S.G. Wijpkema, B. Mezari,
E.J.M. Hensen, ACS Catal. 7 (1) (2017) 520–529, https://doi.org/10.1021/acscatal.
6b02497.
[52] N. Kosinov, F.J.A.G. Coumans, G. Li, E. Uslamin, B. Mezari, A.S.G. Wijpkema,
E.A. Pidko, E.J.M. Hensen, J. Catal. 346 (2017) 125–133, https://doi.org/10.1016/
j.jcat.2016.12.006.
[28] A. Sridhar, M. Rahman, S.J. Khatib, Chem. Cat. Chem 10 (2018) 2571–2583,
https://doi.org/10.1002/cctc.201800002.
[29] M. Rahman, A. Sridhar, S.J. Khatib, Appl. Catal. A 558 (2018) 67–80, https://doi.
120