Centrifugation-free and high yield synthesis of nanosized H-ZSM-5 and its structure-guided aromatization of methanol to 1,2,4-trimethylbenzene

Article abstract of DOI:10.1039/c4ta04444d

Nanosized H-ZSM-5 has been proven to be an efficient way to improve mass transport properties with shape selectivity in many catalytic reactions. Generally, the synthesis of very fine nanosized H-ZSM-5 always suffers from low product yield and requires a complicated centrifugal separation process, both of which severely hinder its large-scale preparation and industrial applications. Herein, we report a centrifugation-free and high yield synthesis route for hierarchically nanosized ZSM-5 with a wide Si/Al ratio range by a combination of pre-aging process and steam-assisted conversion method using alkalis-free powder as the ZSM-5 precursor. This facile route not only avoids the energy-intensive centrifugal separation and ion-exchange process, but also significantly increases the crystallization efficiency along with a high yield. The obtained nanosized ZSM-5 possesses an ultrafine uniform size, high surface area, high total pore volumes, tunable Si/Al molar ratio, and high crystallinity. As a result, the nanosized ZSM-5 shows excellent catalytic performance when used in the catalytic conversion of methanol to aromatics. Notably, the nanosized ZSM-5 with a Si/Al_th of 60 (NZS-60) shows an almost 25-fold longer catalytic lifetime, as well as up to 16% higher total aromatic selectivity when compared with conventional ZSM-5. Furthermore, the selectivity of 1,2,4-trimethylbenzene over this catalyst can be up to 44% in all products and 64% in aromatics products. Characterization results of the spent samples reveal that the most-improved catalytic performance and high selectivity of 1,2,4-trimethylbenzene over the nanosized ZSM-5 could be attributed to its small crystal size and hierarchical structure, which not only prevents the deposition of polyaromatic hydrocarbon in the microspores, but also sharply increases the reaction efficiency of bulky intermediate products on the surface of the catalyst.

Full text of DOI:10.1039/c4ta04444d

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Centrifugation-free and high yield synthesis of
nanosized H-ZSM-5 and its structure-guided
aromatization of methanol to 1,2,4-
trimethylbenzene†
Cite this: J. Mater. Chem. A, 2014, 2,
19797
a
Kui Shen,ab Weizhong Qian, Ning Wang,a Chang Sua and Fei Weia
*
Nanosized H-ZSM-5 has been proven to be an efficient way to improve mass transport properties with
shape selectivity in many catalytic reactions. Generally, the synthesis of very fine nanosized H-ZSM-5
always suffers from low product yield and requires a complicated centrifugal separation process, both of
which severely hinder its large-scale preparation and industrial applications. Herein, we report a
centrifugation-free and high yield synthesis route for hierarchically nanosized ZSM-5 with a wide Si/Al
ratio range by a combination of pre-aging process and steam-assisted conversion method using alkalis-
free powder as the ZSM-5 precursor. This facile route not only avoids the energy-intensive centrifugal
separation and ion-exchange process, but also significantly increases the crystallization efficiency along
with a high yield. The obtained nanosized ZSM-5 possesses an ultrafine uniform size, high surface area,
high total pore volumes, tunable Si/Al molar ratio, and high crystallinity. As a result, the nanosized ZSM-5
shows excellent catalytic performance when used in the catalytic conversion of methanol to aromatics.
Notably, the nanosized ZSM-5 with a Si/Alth of 60 (NZS-60) shows an almost 25-fold longer catalytic
lifetime, as well as up to 16% higher total aromatic selectivity when compared with conventional ZSM-5.
Furthermore, the selectivity of 1,2,4-trimethylbenzene over this catalyst can be up to 44% in all products
and 64% in aromatics products. Characterization results of the spent samples reveal that the most-
improved catalytic performance and high selectivity of 1,2,4-trimethylbenzene over the nanosized ZSM-5
could be attributed to its small crystal size and hierarchical structure, which not only prevents the
deposition of polyaromatic hydrocarbon in the microspores, but also sharply increases the reaction
efficiency of bulky intermediate products on the surface of the catalyst.
Received 28th August 2014
Accepted 6th October 2014
DOI: 10.1039/c4ta04444d
www.rsc.org/MaterialsA
channels, high surface area, tunable acidity, and high hydro-
thermal stability.2,5,7,13–15 The ZSM-5 zeolite has unique molec-
ular-sized micropores with sinusoidal channels of 0.55 nm ꢀ
1. Introduction
The limited availability of crude oil stimulates the study of new
processes using other cheap or renewable resources to prepare
aromatics,1–7 which are one of the most important raw materials
for chemical industry and are basically produced by petro-
chemical processes such as reforming and cracking using crude
oil.5,8,9 Among the alternative processes, the catalytic conversion
of methanol to aromatics (MTA) is a promising technology for
the production of high quality aromatics, and thus has attracted
extensive attention.10–12 ZSM-5 zeolite with a MFI structure has
been proved to be a highly effective catalyst for the aromatiza-
tion of a wide range of feedstocks, because of its well-dened
0.51 nm in [100], and straight channels of 0.56 nm ꢀ 0.53 nm in
[010], and thus shows excellent shape selectivity in many
aromatization reactions.2,5,7,16,17 However, these denite micro-
pores could cause severe diffusion resistance of larger mole-
cules in the zeolite, and thus oen result in quick deactivation
by the deposition of carbonaceous residues that block the
reactants from accessing the active sites.13,14,18–20 Nanosized
ZSM-5 zeolite has a short channel in all a, b, c-axis directions
and allows the quick diffusion of reactants or intermediate
products, and consequently exhibits a much longer lifetime in
high temperature catalytic processes, compared to conventional
microsized ZSM-5. However, the annual amount of nanosized
ZSM-5 produced to date is far lower than that of microsized
ZSM-5, due to the following apparent engineering challenges.
First, the traditional synthesis routes for nanosized ZSM-5,
mainly concentrating on the conned space synthesis21–26 and
clear solutions synthesis,27–30 are normally conducted in
aBeijing Key Laboratory of Green Chemical Reaction Engineering and Technology,
Department of Chemical Engineering, Tsinghua University, Beijing 100084, P.R.
China. E-mail: qianwz@tsinghua.edu.cn
bSchool of Chemistry and Chemical Engineering, South China University of
Technology, Guangzhou 510640, P.R. China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ta04444d
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 19797–19808 | 19797

View Article Online
Journal of Materials Chemistry A
Paper
aqueous solutions with NaOH as the mineralizer.21,22 Thus, the 1.43 g Al(NO3)3$9H2O, and 0–0.75 g NH4NO3 was slowly added
as-prepared nanosized Na-ZSM-5 in crystallization liquids to the mixture. The resulting mixture was further stirred for
ꢂ
always requires repeated energy-intensive centrifugal separa- about 1 h, and then heated at 60 C to complete dryness. The
tion and an ion-exchange process to obtain the nal active H- resulting dry precursor lumps were coarsely crushed. Then, 5 g
form nanosized ZSM-5. The signicant energy consumption of precursor powder was transferred into a 15 mL Teon cup,
and long-time operation seriously retard its commercial appli- which itself was placed into a 70 mL Teon liner with a support.
cation. Second, the hydrothermal synthesis of nanosized ZSM-5 2 g of distilled water was poured at the bottom of the liner
always suffers from low product yield, and thus results in without contacting the dry gel in the inner Teon cup. The
stringent environmental regulations due to the uncrystallized charged liner assembly was placed into an 80 mL steel reactor
aluminosilicate and serious loss of centrifugal separation in the and transferred into a preheated oven at 180 ꢂC and heated for
remaining synthesis solution.27,29,31,32 It is, therefore, highly 24 h. The crystallization reaction was subsequently quenched by
desirable to develop a novel route for nanosized H-ZSM-5 in cold water. The obtained powder was directly dried at 110 ꢂC
terms of simplicity, reproducibility, and large-scale applicability overnight, without centrifugal separation and ion-exchange
for the MTA process.
steps, and subsequently calcined in air at 550 ꢂC for 5 h to
Herein, we report a centrifugation-free and high yield route remove the template (TPAOH) to obtain the nal H-form
for nanosized H-ZSM-5 with excellent structural properties by a nanosized ZSM-5.
combination of pre-aging process and steam-assisted conver-
For comparison, three conventional Na-ZSM-5 with Si/Alth of
sion (SAC) method using alkalis-free powder as the ZSM-5 30, 60, and 90 (Fig. S2†) were also synthesized according to the
precursor. The thorough characterization and systematic study method reported by JC Groen et al.,33 but with a slight modi-
of the synthesis conditions indicate that the nanosized ZSM-5 cation. The corresponding H-ZSM-5 (denoted as CZS-30, CZS-60
with an ultrane size (30–60 nm), wide Si/Al ratio range, high and CZS-90, respectively) were prepared through ion-exchange
surface area (above 440 m2 gꢁ1), high total pore volumes (above three times and via calcination with conventional Na-ZSM-5 as
0.7 cm3 gꢁ1), and high crystallinity can be obtained by a two- the starting material. As detailed, the obtained conventional Na-
step approach with an aging time of 72 h in 90 ꢂC and an ZSM-5 was rst converted to the ammonium form by cation-
NH4NO3/SiO2 molar ratio of 0.017–0.34. Then, these obtained exchange in a NH4NO3 solution. Five grams of Na-ZSM-5 was
nanosized ZSM-5 catalysts with different densities of acid sites dispersed in 100 mL NH4NO3 solution (1 M) and vigorously
(Si/Alth ¼ 30, 60 and 90) were applied as a solid acid catalyst for stirred for 6 h. The solid products were collected by ltering
the catalytic conversion of methanol to aromatics. The reaction separation. The exchange process was performed repeatedly for
results indicate that the nanosized ZSM-5 has a much better 3 times to complete the exchange reaction, and NH4-ZSM-5 was
catalytic stability and higher selectivity of 1,2,4-trime- formed. The nal C-ZSM-5 product was obtained by calcining
thylbenzene compared to conventional ZSM-5. To determine the NH4-ZSM-5 powder at 550 ꢂC for 5 h. In order to investigate
the reason, we thoroughly studied the deactivated catalysts and the necessity of the cation-exchange process, the N-ZSM-5
monitored the quantity and location of coke formation during obtained directly by the direct synthesis route was also ion-
the reaction.
exchanged according to the process above, except the use of
centrifugal separation with a high speed of 12 000 rpm for
30 min.
2. Experimental section
2.1. Centrifugation-free and high yield synthesis of
nanosized ZSM-5
2.2. Characterization of the fresh ZSM-5 samples
The reactants used in the synthesis were tetraethylorthosilicate X-ray diffractions (XRD) were recorded with a Rigaku D/Max-RB
(TEOS, Beijing Modern Eastern Fine Chemicals Co., Ltd), tet- diffractometer using Cu Ka Radiation at 40 kV and 120 mA.
rapropylammonium hydroxide (TPAOH, 25% in water, Zhejiang Scanning electron microscopy (SEM) images were performed by
Ken Te Chemical Co., Ltd.), Aluminium nitrate nonahydrate (Al a high-resolution scanning electron microscope (JEOL, JSM-
(NO3)3$9H2O, Beijing Chemical Works) and ammonium nitrate 7401) at 3.0 kV. TEM experiments were performed on a high-
(NH4NO3, Beijing Chemical Works).
resolution transmission electron microscope (JEOL, JEM-2010,
In this study, three samples with Si/Alth molar ratios of 30, 60 exited at 120 kV). Brunauer–Emmett–Teller (BET) surface area
and 90 (denoted as NZS-30, NZS-60 and NZS-90) were synthe- were recorded in a Quantachrome automated surface area and
sized using this facile synthesis route to certify the tunable Si/Al porosity analyzer with N2 as the adsorption gas. NH3-TPD was
of our route to cater the strict operation condition of the MTA recorded in
a Quantachrome automated chemisorption
reaction. In a typical synthesis of nanosized ZSM-5, 24.3 g TEOS, analyzer from room temperature to 850 ꢂC with a ramp of 10 ꢂC
used as the silica source, was added into 25.8 g TPAOH with minꢁ1. The Si/Al of the zeolite was obtained by an inductively
continuous stirring. The molar composition of the gel was coupled plasma optical emission spectrometer (ICP-OES, IRIS
92SiO2 : 25TPAOH : 479H2O : 368EtOH. The presence of Intrepid II XSP). NMR spectra were obtained on a Varian
ethanol (EtOH) in the mixed gel was due to the hydrolysis of Innityplus-400 spectrometer. Brønsted acid sites and Lewis
TEOS with H2O. Aer homogenizing at room temperature for 1 acid sites were determined by pyridine adsorption. The samples
h, the mixture was hydrothermally treated at 90 ꢂC for 72 h. were rst dried, in situ, by heating to 723 K under vacuum, and
Aer cooling down, the solution containing a little water, 0.58– then were cooled to 323 K. At this temperature, the samples
19798 | J. Mater. Chem. A, 2014, 2, 19797–19808
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
were exposed to pyridine vapor using an equilibration time of 30
min. Aer the physically adsorbed pyridine molecules were
removed by outgassing at 423 K for 1 h, IR spectra were collected
at 423 K with 4 cmꢁ1 resolution using a Nicolet FTIR
spectrometer.
3. Results and discussion
3.1. Characterization of the fresh ZSM-5 samples
The preparation of N-ZSM-5 is also shown schematically in
Fig. 1. The steam assisted conversion (SAC) method is a very
efficient method for preparing zeolites.38–41 In our route using
the SAC method, a little amount of water necessary for crystal-
lization is always kept separate from the precursor powder
during the crystallization reactions. Obviously, aer the crys-
tallization process, the products are in the solid phase, free of
alkali ions, which could be directly converted to the H-form
nanosized ZSM-5 for acid catalytic applications only via calci-
nation. In addition, the avoidance of solvents in the synthesis
and the repeated wash processes for removing NaOH not only
signicantly reduces the waste production, but also signi-
cantly enhances its crystallization efficiency along with a high
yield.
Fig. 2 shows the representative low and high magnication
SEM images of the nanosized ZSM-5 prepared by our route. The
SEM images clearly show that the N-ZSM-5 is composed of many
spherical crystallites with uniform sizes of about 50 nm. To
further conrm the particle sizes and the structure, the nano-
sized ZSM-5 were studied by transmission electron microscopy
(TEM). The TEM images (Fig. 3a–c) reveal an abundant meso-
porous network through the assembly of these uniform nano-
sized ZSM-5 crystals. Analysis with HR-TEM images in Fig. 3d–f
elucidates that all the samples are completely crystalline, as the
parallel lattice fringes can be clearly observed and are clearly
spread throughout the entire crystals. Individual particles can
be recognized with average sizes of about 50 nm. Obviously, the
Si/Al of the ZSM-5 precursor hardly affects the size and
morphology of the nal products, which is different from the
general pattern when using hydrothermal methods.42 The
absence of a sponge-like material in the SEM and TEM pictures
indicates the absence of amorphous material. Since no silica
and aluminium source are washed away in the entire synthetic
route, the yield of these nanosized ZSM-5 is nearly 100%, in
stark contrast to the low yields of nanoparticles usually
obtained from the conventional hydrothermal route.27,29,42,43
The stacking of these nanosized crystals produces the
substantial mesoporosity, leading to a different pore structure
from the conventional ZSM-5 (Fig. S2†). Nitrogen adsorption/
desorption isotherms and BJH pore size distribution are shown
2.3. Catalytic conversion of methanol to aromatics
Before the catalytic test, all the samples were modied with 2
wt% Zn by traditional incipient wetness impregnation using
aqueous solutions of Zn (NO3)2$6H2O. Then the Zn-modied
samples were dried overnight and calcinated in air at 823 K for
5 h.
The MTA reactions were performed at the conditions of 1
atm, 748 K and WHSV ¼ 0.75 hꢁ1 (under N2 ow (10 mL minꢁ1))
in a conventional xed bed stainless steel reactor (13.0 mm i.d.)
equipped with a thermocouple in the middle of the catalyst bed.
0.70 g of catalyst was placed in the xed bed reactor. The ow
rate of the pure methanol and N2 was controlled using a dual
micro-plunger pump and the mass ow controllers, respec-
tively. The products (hydrocarbons) were analyzed using two
ame ionization detectors (FIDs). The conversion of methanol
and the selectivity of the different products were calculated
(carbon base) by considering the oxygenates (methanol and
dimethylether) as the unconverted reactant.
2.4. Analysis of partially deactivated catalysts
The partially deactivated catalysts aer the catalytic tests were
characterized by TGA-DSC, nitrogen adsorption, TEM and GC-
MS to investigate the factors resulting in the superior catalytic
performance of the nanosized ZSM-5. TGA-DSC was performed
using a thermo-gravimetric analyzer (TGA/DSC-1) in the range
30–900 ꢂC at a heating rate of 10 ꢂC minꢁ1 in air. The coke
species formed on the deactivated catalysts aer reactions were
also characterized by gas chromatography-mass spectroscopy
(GC-MS). The coke species were extracted by the following
method developed by Guisnet34 with a slight modication. To
obtain the relative concentrations of the organics formed over
different samples, the exactly same amount (20 mg) of the
deactivated catalyst was always dissolved in 1 mL of 15% HF
solution. Organic species were then extracted with 0.8 mL
C(CH3)2Cl2 and analyzed using GC-MS (MSD 5973, EI 70 eV)
with a HP-5MS column (30 m, 0.25 mm i.d.), and the graphite
carbon in the water phase was characterized by TEM
experiments.
in Fig. 4. The CZS-30 shows
a typical type-I isotherm
Furthermore, the location and the amount of coke in Zn/
NZS-60 were analyzed as a function of the reaction time, using
the previously reported methods.35–37 The total amount of coke
(external + internal) within the partially deactivated Zn/NZS-60
at different reaction times was obtained using TGA analysis.
The amount of the internal coke was estimated from the
decrease in the micropore volume (determined by the N2
adsorption measurements) relative to the pristine sample, by
assuming a coke density of 1.22 g cmꢁ3. The external coke was
calculated by subtracting the internal coke content from the
total coke content.
Fig. 1 Schematic diagram of centrifugation-free synthetic procedures
for nanosized ZSM-5 with tunable Si/Al molar ratios.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 19797–19808 | 19799

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 2 SEM images of (a) NZS-30, (b) NZS-60, and (c) NZS-90, and the high-magnification SEM images of (d) NZS-30, (e) NZS-60, and (f) NZS-90,
which are taken from the part surrounded with a white frame in (a), (b), and (c).
corresponding to the solely microporous structure. However, all
isotherms of the nanosized H-ZSM-5 exhibit a typical adsorp-
tion curve of type I plus type IV with an apparent enhanced
uptake and an obvious hysteresis loop in the P/Po range of 0.8 to
1, indicating its remarkable micro-/mesoporous structure.24,44,45
Correspondingly, pore size analysis by the BJH method using
the adsorption branch of the isotherm (Fig. 4b) shows that all
the nanosized H-ZSM-5 possess a broad pore distribution with a
mean pore size of about 20 nm. As a result, the nanosized H-
ZSM-5 have a much higher BET area and a larger mesopore
Fig. 4 (a) Nitrogen adsorption/desorption isotherms and (b) BJH pore
size distribution curves calculated from the adsorption branch of the
volume than the CZS-30. As summarized in Table 1, the BET isotherm (inset) for NZS-30, NZS-60, NZS-90, and CZS-30.
surface area increases from 388 m2 gꢁ1 for the CZS-30 up to
441.5–484.0 m2 gꢁ1 for nanosized ZSM-5. The total pore volume
increases from 0.19 cm3 gꢁ1 to 0.72–0.76 cm3 gꢁ1 and the
structured ZSM-5 without any impurities or amorphous
mesopore volume increases from 0.04 cm3 gꢁ1 to more than
0.58–0.62 cm3 gꢁ1. It is also worth noting that the textural
properties do not noticeably change while increasing the Si/Al of
the nanosized H-ZSM-5.
phase.30,46 The XRD line widths of all the nanosized H-ZSM-5 are
much broader than those for CZS-30 due to their smaller crystal
size.47 Fig. 5b shows the UV Raman spectra of the samples
excited at 325 nm. All the samples show the characteristic bands
of the MFI structure with a wide band at 380 cmꢁ1 (Fig. 5b),
which is associated with the framework symmetric stretching
vibration of a ve-membered building unit in the ZSM-5
The XRD patterns of all the samples (Fig. 5a) show ve well-
resolved peaks at 7.98ꢂ, 8.82ꢂ, 23.18ꢂ, 24.02ꢂ and 24.46ꢂ, which
are in good agreement with the highly crystalline MFI-
Fig. 3 Low-magnification TEM images of (a) NZS-30, (b) NZS-60, and (c) NZS-90, and the high-magnification TEM images of (d) NZS-30, (e)
NZS-60, and (f) NZS-90, which are taken from the part surrounded with a white frame in (a), (b), and (c).
19800 | J. Mater. Chem. A, 2014, 2, 19797–19808
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Table 1 Textural properties and compositions of various samples
Si/Alth
Si/AlICP
BET surface
area
Mesopore surface
areac
Total pore
volume
Micropore
volumec
Mesopore
volumed
Sample
name
atomic ratioa
atomic ratiob
(mol molꢁ1
)
(mol molꢁ1
)
(m2 gꢁ1
)
(cm3 gꢁ1
)
(cm3 gꢁ1
)
(cm3 gꢁ1
)
(cm3 gꢁ1
)
CZS-30
NZS-30
NZS-60
NZS-90
30
30
60
90
24.1
28.6
56.2
84.4
388.2
484.0
441.5
446.7
33.1
118.6
100.4
107.7
0.19
0.73
0.72
0.76
0.15
0.15
0.14
0.14
0.04
0.58
0.58
0.62
a
b
c
The theoretical Si/Al based on the molar composition of the initial gel. Measured by ICP-OES. t-plot method. d Vmeso ¼ Vtot ꢁ Vmicro
.
zeolites.32,48,49 Generally, the catalytically active acid sites of
Furthermore, the different textural properties brought about
ZSM-5 are always correlated to the presence of intra-framework by the crystal sizes might induce remarkable changes of surface
aluminium atoms. Thus, successful incorporation of properties, which play a crucial role in the catalytic performance
aluminium atoms into the silica framework of nanosized ZSM-5 of the zeolite-based catalysts. The infrared spectra of the dehy-
with a wide Si/Alth is veried by 27Al MAS NMR spectroscopy. As drated form of CZS-30 (Fig. 5d) show no obvious peaks because
shown in Fig. 5c, the 27Al MAS-NMR spectra of all the samples of the broad background peak. In sharp contrast, the infrared
show only one sharp aluminium peak at 52 ppm, indicating that spectra of three nanosized ZSM-5 in the region of the OH
all the aluminium atoms are incorporated into the ZSM-5 stretching vibrations show a major band at 3745 cmꢁ1, attrib-
framework with a tetrahedral coordination environment.50 In uted to silanol groups, which is in agreement with its decreased
addition, the 29Si MAS NMR spectra of all the samples (Fig. S3†) crystal sizes and increased external surface area. It is reported
show similar features with a sharp peak at ꢁ112 ppm and a that this type of Si-related defective site is very effective for
small peak at ꢁ105 ppm. The sharp peak can be attributed to increasing the yields of aromatics and paraffins for the MTH
the Si(0Al) contribution, and the small peak to the Si(1Al) reaction.52 Besides, additional spectral components of lower
contribution,51 which further conrm the presence of frame- intensity are found over nanosized ZSM-5, which could be
work aluminium in the nanosized ZSM-5. Thus, nanosized attributed to: (1) perturbed silanols mainly located inside the
ZSM-5 with a wide range of Si/Al can be prepared directly by this zeolite crystals (3727 cmꢁ1), (2) Al(OH)Si groups corresponding
facile direct synthesis route.
to strongly acidic Brønsted sites (3612 cmꢁ1), and (3) a broad
Fig. 5 (a) XRD patterns, (b) UV-Raman spectra, and (c) 27Al-NMR spectra of the NZS-30, NZS-60, NZS-90, and CZS-30. (d) FT-IR spectra of the
dehydrated NZS-30, NZS-60, NZS-90, and CZS-30.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 19797–19808 | 19801

View Article Online
Journal of Materials Chemistry A
Paper
band corresponding to silanols interacting through hydrogen
bonding (3500 cmꢁ1). As for NZS-60 and NZS-90, the intensity of
the bands at 3745 cmꢁ1 and 3500 cmꢁ1 are stronger than NZS-
30, suggesting that the Si/Al of nanosized ZSM-5 could inuence
its surface characteristics signicantly. There are hardly any
peaks at about 3660 cmꢁ1 ascribed to OH groups bound to extra-
framework and/or perturbed framework Al atoms,52,53 indi-
cating the successful incorporation of all the aluminium atoms
into the silica framework of the samples.
In order to certify the feasibility of our synthesis route, we
compared the acidic properties of pristine nanosized ZSM-5 and
the samples aer 3 times of ion-exchange. As shown in Fig. 6,
minor differences are observed in both the TPD curves (Fig. 6a)
and in the pyridine-absorbed FTIR spectra (Fig. 6b) between the
three couples of nanosized ZSM-5, indicating that an energy-
intensive centrifugal separation and ion-exchange process are
indeed unnecessary in our route. In addition, Fig. 6a shows that
the acid strength is declined sharply with the increase in Si/Al
for the nanosized H-ZSM-5, and that the acidity of NZS-30 is also
comparable with CZS-30 because of their similar Si/AlICP (Table
1). It is also noteworthy that the amount of L acid reduces
rapidly with the increase in Si/Al of the nanosized H-ZSM-5
(Fig. 6b). Moreover, the intensity of the hydrogen-bonded PyH+
peak over nanosized ZSM-5 is much stronger than in CZS-30,
due to the enhanced adsorption capacity caused by its large
mesopore volume.54
Fig. 7 Investigation into the crystallization process of NZS-60. (a) SEM
image of the product crystallized at 180 ꢂC for 180 min, (b) XRD
patterns, (c) Nitrogen adsorption/desorption isotherms, and (d) BJH
pore size distribution calculated from the adsorption branch of the
isotherm for the samples crystallized at various crystallization times.
3.2. Evolution of the nanosized ZSM-5 with crystallization
time
important role in shortening the nucleation stage, which is the
most time consuming step in the crystallization of ZSM-5.49
Interestingly, further increasing the crystallization time leads to
the slight reduction of crystallinity, possibly due to the
destructive effect of the template on the framework of the
product. It is well reported that the TPAOH template plays the
role of a scaffold forming a mesostructure during the crystalli-
zation process.55 This is conrmed by the N2 adsorption
isotherm and the corresponding pore size distribution. As
shown in Fig. 7c, the isotherm of the sample with 0 h crystal-
lization has a near linear uptake over the P/Po range from 0 to
0.4, which corresponds to micropores or small mesopores
To study the crystallization mechanism of NZS-60, a kinetic run
was carried out by varying the synthesis time from 0 h to 24 h at
180 ꢂC. Fig. 7b shows the XRD patterns of the samples at
different reaction times. Before stream-assisted crystallization,
the sample exhibited an extremely faint XRD pattern of a MFI
structure. With the steam-assisted conversion at 180 ꢂC, the
intensity of the characteristic MFI-structured peaks increased
rapidly and drew the maximum as crystallization proceeded up
to 3 h, which indicated that the amorphous structure of the
starting powder could translate into a pure ZSM-5 phase just
within 3 h (Fig. 7a). Obviously, the presence of seeds plays an
Fig. 6 (a) NH3-TPD adsorption spectra of the NZS-30, NZS-60, NZS-90, and CZS-30. (b) FTIR spectra of the dehydrated NZS-30, NZS-60, NZS-
90, and CZS-30.
19802 | J. Mater. Chem. A, 2014, 2, 19797–19808
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Table 2 Textural properties and crystallinity of the NZS-60 crystallized at various times
Crystallization
time (min)
Crystallinitya
(%)
BET surface area
Mesopore surface areab
Total pore volume
Micropore volumeb
Mesopore volumec
(m2 gꢁ1
)
(cm3 gꢁ1
)
(cm3 gꢁ1
)
(cm3 gꢁ1
)
(cm3 gꢁ1
)
d
d
d
0
40
70
120
180
240
1440
0.04
19.37
54.46
75.57
100.00
99.87
95.10
1081.5
854.0
593.9
519.9
472.8
466.4
441.5
—
0.58
0.84
0.88
0.93
0.86
0.77
0.72
—
—
547.4
189.3
137.0
119.4
115.6
100.4
0.14
0.16
0.17
0.15
0.15
0.14
0.70
0.72
0.76
0.71
0.62
0.58
a
Calculated via XRD result by assuming the maximum crystallinity of the product aer 180 min of crystallization time is 100%. b t-plot method.
c
d
Vmeso ¼ Vtot ꢁ Vmicro
.
t-plot is not applicable because of the continuous distribution of pore sizes around 3 nm.
below 3 nm. Upon increasing the crystallization time from 0 h to investigated, but with other conditions remaining unchanged.
180 min, the uptakes of the isotherms move toward high P/Po. Fig. 9 shows the SEM images of the products synthesized with
Correspondingly, the average pore size increases from less than the pre-aging temperature of 90 ꢂC for different times, which
3 nm in the starting dry gel to about 11 nm in the sample at 180 indicate that the crystal size is decreased with the increase in
min crystallization time. A longer crystallization treatment has pre-aging time. The average crystal sizes of the products at pre-
little effect on the mesoporous structure, since no obvious aging times of 0, 24, and 72 h were found to be about 160 nm,
changes can be observed from the pore size distribution curves. 100 nm, and 50 nm, respectively. A further extension of the
Table 2 summarizes the textural properties and crystallinity of aging time aer 72 h had little effect on the crystalline phases
the samples at various crystallization times. The external and crystal size, since no obvious changes could be observed
surface area decreases with crystallization time, but remains from the SEM image.
very high, even aer reaching 100% crystallization. However,
both the total pore volume and the mesopore volume increase
within 120 min crystallization time, and nally achieve stability,
indicating the high efficiency of our route for obtaining meso/
3.4. Catalytic performances of the nanosized ZSM-5 with
different Si/Al
The methanol to aromatics reaction (MTA) is an important
process for the non-oil route to aromatics.10–12,57 It is respected
that the meso/macroporous structure and the short diffusion
length of nanosized ZSM-5 are favorable for the MTA reaction
with high stability. Therefore, the catalytic performances over
2% Zn-modied nanosized (denoted as Zn/NZS-30, Zn/NZS-60,
and Zn/NZS-90) and 2% Zn-modied conventional ZSM-5
micropore zeolites (Table 2).
3.3. Effects of gel composition and aging conditions
In our preliminary experiments, we discovered that the crystal
size of the products was sensitive to the concentrations of NH4
+
.
Therefore, at an aging temperature of 90 ꢂC for 72 h and a
constant Si/Alth of 60, the inuence of NH4+/SiO2 was investi-
gated, and the SEM images of products prepared with different
NH4NO3/SiO2 are shown in Fig. 8. When there is no NH4NO3
added into the synthesis mixture, the size distribution of the
product is broad and most of the ZSM-5 crystals are greater than
100 nm. Products with a uniform diameter of about 50 nm
could be obtained with a NH4/SiO2 ratio of 0.017–0.034 (Fig. 8b
and c). This can be explained by the fact that aluminium
+
incorporation is achieved more easily when the smaller NH4
are available as charge compensators, compared with the situ-
ation when only the larger TPA+ are present. Further increasing
+
the NH4 to 0.067, a sample with a diameter of about 300 nm
and a smaller BET surface area was obtained. The increased
particle size and the reduced crystallinity of the ZSM-5 crystals
obtained in this run were probably due to the excessively
+
decreased alkalinity, caused by the hydrolysis of plentiful NH4
.
In addition, our synthesis route includes the ‘‘pre-aging
process’’ before stream-assisted crystallization, which is
considered to produce abundant ZSM-5 nuclei (Fig. S1†),56
leading to the formation of small ZSM-5. Hence, the effect of the
Fig. 8 SEM images of the NZS-60 prepared with an NH4NO3/SiO2
molar ratio of (a) 0, (b) 0.017, (c) 0.034, and (d) 0.067, indicating that a
trace amount of NH4NO3 addition to the precursor gel markedly
pre-aging time on the crystal size of the product was also reduces the particle sizes of the NZS-60.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 19797–19808 | 19803

View Article Online
Journal of Materials Chemistry A
Paper
that of Zn/CZS-60 (t1/2 z 3 h). In addition, the detailed initial
product selectivity is also given in Tables S2 and S3.† Notably,
the most stable Zn/NZS-60 shows a high total aromatics selec-
tivity of 68%, with an absolute selectivity of 16% higher than the
CZS-60 (52%). More importantly, the selectivity of 1,2,4-trime-
thylbenzene over this catalyst can be up to 44% for all products
and 64% for aromatics products (Table 4). Compared with the
previously reported results in the process of MTA,13 the high
selectivity of 1,2,4-trimethylbenzene over nanosized Zn/ZSM-5
could signicantly reduce the cost and energy consumption of
the subsequent separation and purication process to obtain
high purity 1,2,4-trimethylbenzene. Since the acid strength,
acid amount, Si/AlICP ratio, and micropore volume are all
similar for Zn/NZS-60 and Zn/CZS-60 (Fig. 6 and Table 1), the
superior catalytic stability and higher 1,2,4-trimethylbenzene
selectivity over the former can be directly attributed to the
unique mesoporous structure and short diffusion length of the
nanosized ZSM-5. It is well documented that the catalyst deac-
tivation of ZSM-5 is mainly caused by coke formation on the
micropore mouths and in the micropores, which are produced
via secondary reactions.37,58 Hence, micropore blockage could
occur very slowly over the nanosized ZSM-5 because of its
unique mesoporous structure and the abundant pore mouths at
the surface of the nanosized ZSM-5, both of which not only favor
the coke precursor diffusion, but also sharply increase the
evacuation of aromatics and make the isomerization reaction of
trimethylbenzene dynamically balanced, leading to the high
selectivity of 1,2,4-trimethylbenzene, which is the most ther-
modynamically stable product when the isomerization reaction
of trimethylbenzene reaches thermodynamic equilibrium. Due
to the enlarged external surface area and short diffusion path
lengths, it is feasible that the thermodynamically stable 1,2,4-
trimethylbenzene formed inside the micropores could readily
migrate to the external surface. In the case of solely micropo-
rous Zn/CZS-60, the diffusion of this bulk product from the
micropores to the external surface would be relatively slow,
leading to a lower aromatics selectivity and shorter lifetime in
the MTA process.
Fig. 9 SEM images of the NZS-60 prepared with an aging time of (a)
0 h, (b) 36 h, (c) 72 h, and (d) 144 h, indicating that the particle size of
NZS-60 is decreased with an increase in the pre-aging time.
(denoted as Zn/CZS-30, Zn/CZS-60, and Zn/CZS-90) in the MTA
reaction were compared in a xed bed stainless steel reactor.
Fig. 10 shows an overview of the activities and stabilities of the
three groups between nanosized and conventional ZSM-5
described in this contribution. It is worth noting that the
comparison of catalysts performance was performed under the
same Si/Alth ratios, because the catalytic properties are more
likely to change by Al content. Under such a strict operation
condition, all the catalysts showed almost the similar high
initial methanol conversions of above 95% (Table 3). However,
as the reaction proceeded, the activities of Zn/CZS-30, Zn/CZS-
60, and Zn/CZS-90 all declined gradually within just a few hours.
In sharp contrast, all the nanosized ZSM-5 showed a more
superior catalytic stability and a higher total aromatics selec-
tivity than the corresponding conventional Zn/ZSM-5 with the
same Si/Alth. It is noted that the Si/Alth ratios also have a
remarkable effect on the catalytic performance of nanosized Zn/
ZSM-5 catalysts. Among these catalysts, the Zn/NZS-60 with Si/
Alth ¼ 60 gives the slowest deactivation with a stable total
aromatics selectivity and long catalytic lifetime (dened as the
time at which the catalytic conversion decreases by 50%, which
is denoted as t1/2) of more than 75 h, which is almost 25-fold
In order to verify this point, we conducted a series of char-
acterization tests over the spent Zn/NZS-60 and Zn/CZS-60,
including TG/DSC, GC-MS, and TEM. The TG/DTA proles are
shown in Fig. 11. It is obvious that the total weight loss over the
spent Zn/NZS-60 is obviously much higher than that over
Fig. 10 Methanol conversions and overall aromatics selectivity with various Zn/ZSM-5 samples as a function of time-on-stream (reaction
temperature: 475 ꢂC; reaction pressure: atmospheric pressure, WHSV: 0.75 hꢁ1).
19804 | J. Mater. Chem. A, 2014, 2, 19797–19808
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Table 3 Methanol conversion and product selectivity of MTA reactions over various catalysts used in the present study measured after the
reactions were stable
Selectivity, (%)
Conversion of
methanol (%)
Total aromatics
(%)
Catalystsa
C1–C5 alkanes
C2–C5 olens
C6 ben.
C7 ben.
C8 ben.
C9+ ben.
Zn/CZS-30
Zn/CZS-60
Zn/CZS-90
Zn/NZS-30
Zn/NZS-60
Zn/NZS-90
99.03
96.80
95.54
99.47
99.37
99.41
20.83
16.77
16.02
15.06
24.36
18.12
28.95
31.23
27.43
22.80
7.69
1.76
0.64
0.81
0.48
0.32
0.50
11.19
5.29
7.19
3.71
2.07
3.20
36.09
34.57
31.88
25.88
14.60
14.62
1.16
11.49
16.65
32.07
50.95
52.21
50.22
51.99
56.54
62.15
67.94
70.55
11.32
a
The sampling time (Zn/CZS-30, Zn/CZS-60, Zn/CZS-90: 1.2 h, Zn/NZS-60: 3.6 h, and Zn/NZS-90: 10.2 h).
conventional ZSM-5, indicating its much higher capacity for dissolution and extraction. As shown in Fig. 12a and b, only a
holding coke. However, as the reaction time of Zn/NZS-60 is tiny amount of liberating heavy hydrocarbons could be found in
much longer than the corresponding Zn/CZS-60, the spent Zn/ the extract phase of Zn/NZS-60, and the main phase of deposits
ꢁ1
NZS-60 has a slow coke formation rate of 3.78 mg gcat hꢁ1
,
over this sample was graphite carbon, which could not be
much lower than that of 26.33 mg gcatꢁ1 hꢁ1 for spent Zn/CZS- extracted by CH2Cl2 (Fig. 12a). However, signicant amounts of
60. As shown in Fig. 11b, the DTA proles of the spent Zn/CZS- polyaromatic hydrocarbon are found over conventional ZSM-5,
60 has a broad exothermic peak, implying that the coke in this which could occlude the pore channel and lead to the quick
catalyst is composed of various carbon species deposited on deactivation of the catalysts (Fig. 12b).
active sites. However, there is almost only one sharp exothermic
Furthermore, in order to understand how the particle size
peak around 480 ꢂC for NZS-60, indicating that the carbon of ZSM-5 is related to the catalytic performance, the location
species formed over this catalyst is mainly a relatively single and the amount of coke in Zn/NZS-60 were analyzed as a
coke species, which is also conrmed by the GC-MS analysis of function of the reaction time using the characterization
the retained organic species in spent catalysts by 15% HF method described in Section 2.4. As the results in Fig. 12c
Table 4 The distribution of trimethylbenzene selectivity over various catalysts used in the present study measured after the reactions were stable
Selectivity in products, (%)
Selectivity in aromatics, (%)
Catalystsa
1,3,5-Tri-ben
1,2,4-Tri-ben
1,2,3-Tri-ben
1,3,5-Tri-ben
1,2,4-Tri-ben
1,2,3-Tri-ben
Zn/CZS-30
Zn/CZS-60
Zn/CZS-90
Zn/NZS-30
Zn/NZS-60
Zn/NZS-90
—
—
0.53
2.38
3.90
4.53
1.16
10.92
15.52
27.66
44.12
44.40
—
—
0.58
2.02
2.92
3.57
—
—
0.94
3.83
5.74
6.42
2.31
21.00
27.45
44.51
64.94
62.93
—
—
1.03
3.25
4.30
5.06
a
The sampling time (Zn/CZS-30, Zn/CZS-60, Zn/CZS-90: 1.2 h, Zn/NZS-60: 3.6 h, and Zn/NZS-90: 10.2 h).
Fig. 11 (a) TG and (b) DSC profiles of the spent Zn/CZS-60 and Zn/NZS-60. Oxidation gas: 20% O2/N2, heating rate 10 ꢂC minꢁ1
.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 19797–19808 | 19805

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 12 (a) Photographs of the spent catalysts after being dissolved in 15% HF and extracted by CCl4. (b) GC-MS chromatograms of the organic
species in spent catalysts extracted by CH2Cl2. (c) Coke formation over Zn/NZS-60 with various time-on-stream during the MTA reaction. (d) and
(e) TEM images of the graphite carbon depositing on the outer surface of Zn/NZS-60 after the inner Zn/NZS-60 is dissolved by 15% HF solution.
(f) The corresponding EDX spectrum of the graphite carbon.
show, the coke in Zn/NZS-60 is preferentially deposited at the
stacking pores of nanosized ZSM-5 prior to the generation
4. Conclusions
inside the straight and sinusoidal channels with a much
slower coking rate than Zn/CZS-60. The coke in the stacking
pores is not completely blocked, and the reactants and prod-
ucts could still move to the active sites and escape from the
micropores. The TEM images and the corresponding EDX
analysis of the graphite carbon also verify the above results. As
shown in Fig. 12d and e, the graphite carbon, as deposited on
the outer surface of Zn/NZS-60, forms hollow carbon nano-
spheres aer the inner Zn/NZS-60 is dissolved by 15% HF
solution. However, 89% of the coke is formed inside the
micropore for the Zn/CZS-60 with solely micropores, which is
the main cause of the sudden drop in catalytic activity.
Therefore, we, for the rst time, validate directly that the coke
over nanosized ZSM-5 is preferentially deposited on the
outside surface prior to the intracrystalline micropores. Above
all, compared with conventional ZSM-5, the nanosized ZSM-5
has the following obvious advantages when used in the MTA
reaction: (1) short diffusion length, which could enhance the
diffusibility of reactants and products and prevent the depo-
sition of polyaromatic hydrocarbons in the microspores; (2)
substantial mesoporosity, which endows the nanosized ZSM-5
with a much higher capacity for holding coke than the
conventional one; (3) high external surface, which sharply
increases the reaction efficiency of bulky intermediate prod-
ucts on the surface of the catalyst, leading to a high selectivity
of 1,2,4-trimethylbenzene; (4) abundant surface groups (e.g.,
silanol groups, perturbed silanols, and Al(OH)Si groups) could
provide more active sites for the MTA reaction.
The present work reports a centrifugation-free synthesis route
for hierarchical nanosized ZSM-5 with 100% yield by steam-
assisted conversion using a solid alkalis-free powder as the
ZSM-5 precursor. The obtained nanosized ZSM-5 possesses
excellent properties, such as ultrane uniform size, robust
thermal stability, high surface area, high total pore volumes,
tunable Si/Al molar ratio, and high crystallinity, and shows
excellent catalytic performance when used in the catalytic
conversion of methanol to aromatics. The Zn/NZS-60 shows
an almost 25-fold catalyst lifetime and up to 16% higher total
aromatics selectivity compared with conventional microsized
Zn/CZS-60. More importantly, the selectivity of 1,2,4-trime-
thylbenzene over this catalyst can be up to 44% for all prod-
ucts and 64% for aromatics products. Analysis of the partially
deactivated catalysts revealed that the superior catalytic
stability and higher 1,2,4-trimethylbenzene selectivity over
the nanosized Zn/ZSM-5 could be directly attributed to its
small crystal size and mesoporous structure, which prevent
the deposition of polyaromatic hydrocarbon in the micro-
pores, but also sharply increase the reaction efficiency of the
bulky molecule. The nanosized ZSM-5 prepared by this
route shows immense potential for large scale industrial
applications for the MTA process with a high selectivity of
1,2,4-trimethylbenzene.
Conflict of interest
The authors declare no competing nancial interest.
19806 | J. Mater. Chem. A, 2014, 2, 19797–19808
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
27 R. Van Grieken, J. Sotelo, J. Menendez and J. Melero,
Microporous Mesoporous Mater., 2000, 39, 135.
28 S. Lee, C. S. Carr and D. F. Shantz, Langmuir, 2005, 21, 12031.
29 W. Song, V. Grassian and S. Larsen, Chem. Commun., 2005,
2951.
Acknowledgements
This work was supported by NSFC program of 21376135 and
NSFC key program of 51236004.
30 Y.-Q. Deng, S.-F. Yin and C.-T. Au, Ind. Eng. Chem. Res., 2012,
51, 9492.
31 J. Aguado, D. Serrano and J. Rodriguez, Microporous
Mesoporous Mater., 2004, 75, 41.
32 L. Ren, Q. Wu, C. Yang, L. Zhu, C. Li, P. Zhang, H. Zhang,
X. Meng and F.-S. Xiao, J. Am. Chem. Soc., 2012, 134, 15173.
33 J. C. Groen, T. Bach, U. Ziese, A. M. Paulaime-van Donk,
References
1 T. R. Carlson, Y. T. Cheng, J. Jae and G. W. Huber, Energy
Environ. Sci., 2011, 4, 145.
2 B. Valle, A. G. Gayubo, A. T. Aguayo, M. Olazar and J. Bilbao,
Energy Fuels, 2010, 24, 2060.
´
´
3 T. R. Carlson, J. Jae, Y.-C. Lin, G. A. Tompsett and
G. W. Huber, J. Catal., 2010, 270, 110.
K. P. de Jong, J. A. Moulijn and J. Perez-Ramırez, J. Am.
Chem. Soc., 2005, 127, 10792.
4 X. Guo, G. Fang, G. Li, H. Ma, H. Fan, L. Yu, C. Ma, X. Wu, 34 M. Guisnet and P. Magnoux, Appl. Catal., 1989, 54, 1.
D. Deng and M. Wei, Science, 2014, 344, 616.
5 L. Yu, S. Huang, S. Zhang, Z. Liu, W. Xin, S. Xie and L. Xu,
ACS Catal., 2012, 2, 1203.
6 Y. T. Cheng and G. W. Huber, ACS Catal., 2011, 1, 611.
7 Y. T. Cheng, Z. Wang, C. J. Gilbert, W. Fan and G. W. Huber,
Angew. Chem., Int. Ed., 2012, 51, 11097.
8 T. F. Degnan Jr, Top. Catal., 2000, 13, 349.
9 C. Marcilly, J. Catal., 2003, 216, 47.
10 K. Shen, W. Qian, N. Wang, J. Zhang and F. Wei, J. Mater.
Chem. A, 2013, 1, 3272.
35 D. Bibby, N. Milestone, J. Patterson and L. Aldridge, J. Catal.,
1986, 97, 493.
36 F. L. Bleken, K. Barbera, F. Bonino, U. Olsbye, K. P. Lillerud,
S. Bordiga, P. Beato, T. V. Janssens and S. Svelle, J. Catal.,
2013, 307, 62.
37 M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki and R. Ryoo,
Nature, 2009, 461, 246.
38 S. P. Naik, A. S. Chiang and R. Thompson, J. Phys. Chem. B,
2003, 107, 7006.
39 J. Yao, H. Wang, S. P. Ringer, K.-Y. Chan, L. Zhang and N. Xu,
Microporous Mesoporous Mater., 2005, 85, 267.
40 K. Zhu, J. Sun, J. Liu, L. Wang, H. Wan, J. Hu, Y. Wang,
C. H. Peden and Z. Nie, ACS Catal., 2011, 1, 682.
´
´
¨
11 R. Barthos, T. Bansagi, T. Suli Zakar and F. Solymosi, J.
Catal., 2007, 247, 368.
12 Y. Ni, A. Sun, X. Wu, G. Hai, J. Hu, T. Li and G. Li,
Microporous Mesoporous Mater., 2011, 143, 435.
13 C. S. Cundy and P. A. Cox, Chem. Rev., 2003, 103, 663.
14 A. Corma, Chem. Rev., 1997, 97, 2373.
´
´
´
41 N. Ren, B. Subotic, J. Bronic, Y. Tang, M. Dutour Sikiric,
ˇ ´
ˇ ´
´
´
T. Misic, V. Svetlicic, S. Bosnar and T. Antonic Jelic, Chem.
Mater., 2012, 24, 1726.
15 F. Pan, X. Lu, Y. Wang, S. Chen, T. Wang and Y. Yan, 42 V. B. Mortola, A. P. Ferreira, J. M. Fedeyko, C. Downing,
Microporous Mesoporous Mater., 2014, 184, 134.
16 D. Van Vu, M. Miyamoto, N. Nishiyama, Y. Egashira and
K. Ueyama, J. Catal., 2006, 243, 389.
J. M. Bueno, M. C. Kung and H. H. Kung, J. Mater. Chem.,
2010, 20, 7517.
43 S. C. Larsen, J. Phys. Chem. C, 2007, 111, 18464.
17 M. B. Roeffaers, R. Ameloot, M. Baruah, H. Uji-i, M. Bulut, 44 X. Liu, B. Tian, C. Yu, F. Gao, S. Xie, B. Tu, R. Che, L. M. Peng
¨
G. De Cremer, U. Muller, P. A. Jacobs, J. Hoens and
and D. Zhao, Angew. Chem., Int. Ed., 2002, 41, 3876.
45 M. Imperor-Clerc, P. Davidson and A. Davidson, J. Am. Chem.
Soc., 2000, 122, 11925.
46 F. Liu, T. Willhammar, L. Wang, L. Zhu, Q. Sun, X. Meng,
W. Carrillo-Cabrera, X. Zou and F.-S. Xiao, J. Am. Chem.
Soc., 2012, 134, 4557.
B. F. Sels, J. Am. Chem. Soc., 2008, 130, 5763.
18 V. Valtchev, E. Balanzat, V. Mavrodinova, I. Diaz, J. El Fallah
and J.-M. Goupil, J. Am. Chem. Soc., 2011, 133, 18950.
19 H. Jin, M. B. Ansari and S.-E. Park, Chem. Commun., 2011, 47,
7482.
20 S. I. Cho, S. D. Choi, J. H. Kim and G. J. Kim, Adv. Funct. 47 J. Kim, M. Choi and R. Ryoo, J. Catal., 2010, 269, 219.
Mater., 2004, 14, 49.
48 Y. Yu, G. Xiong, C. Li and F.-S. Xiao, J. Catal., 2000, 194, 487.
49 K. Shen, W. Qian, N. Wang, C. Su and F. Wei, J. Am. Chem.
Soc., 2013, 135, 15322.
50 T.-O. Do, A. Nossov, M.-A. Springuel-Huet, C. Schneider,
J. L. Bretherton, C. A. Fyfe and S. Kaliaguine, J. Am. Chem.
Soc., 2004, 126, 14324.
51 C. Fyfe, G. Gobbi, G. Kennedy, J. Graham, R. Ozubko,
W. Murphy, A. Bothner-By, J. Dadok and A. Chesnick,
Zeolites, 1985, 5, 179.
21 C. H. Jacobsen, Chem. Commun., 1999, 673.
22 I. Schmidt, C. Madsen and C. J. Jacobsen, Inorg. Chem., 2000,
39, 2279.
23 P.-S. Lee, X. Zhang, J. A. Stoeger, A. Malek, W. Fan, S. Kumar,
W. C. Yoo, S. Al Hashimi, R. L. Penn and A. Stein, J. Am.
Chem. Soc., 2010, 133, 493.
24 H. Chen, J. Wydra, X. Zhang, P.-S. Lee, Z. Wang, W. Fan and
M. Tsapatsis, J. Am. Chem. Soc., 2011, 133, 12390.
25 W. C. Yoo, S. Kumar, R. L. Penn, M. Tsapatsis and A. Stein, J. 52 P. Sazama, B. Wichterlova, J. Dedecek, Z. Tvaruzkova,
Am. Chem. Soc., 2009, 131, 12377.
Z. Musilova, L. Palumbo, S. Sklenak and O. Gonsiorova,
26 W. Fan, M. A. Snyder, S. Kumar, P.-S. Lee, W. C. Yoo,
Microporous Mesoporous Mater., 2011, 143, 87.
A. V. McCormick, R. L. Penn, A. Stein and M. Tsapatsis, 53 S. Zheng, H. R. Heydenrych, A. Jentys and J. A. Lercher, J.
Nat. Mater., 2008, 7, 984.
Phys. Chem. B, 2002, 106, 9552.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 19797–19808 | 19807

View Article Online
Journal of Materials Chemistry A
Paper
54 W. Song, R. Justice, C. Jones, V. Grassian and S. Larsen, 57 M. Conte, J. A. Lopez-Sanchez, Q. He, D. J. Morgan,
Langmuir, 2004, 20, 8301.
55 J. Wang, J. C. Groen, W. Yue, W. Zhou and M.-O. Coppens,
Chem. Commun., 2007, 4653.
56 H. Mochizuki, T. Yokoi, H. Imai, R. Watanabe, S. Namba,
J. N. Kondo and T. Tatsumi, Microporous Mesoporous
Mater., 2011, 145, 165.
Y. Ryabenkova, J. K. Bartley, A. F. Carley, S. H. Taylor,
C. J. Kiely and K. Khalid, Environ. Sci. Technol., 2012, 2, 105.
58 L. Narasimhan, J. Thybaut, G. Marin, J. Denayer, G. Baron,
J. Martens and P. Jacobs, Chem. Eng. Sci., 2004, 59, 4765.
19808 | J. Mater. Chem. A, 2014, 2, 19797–19808
This journal is © The Royal Society of Chemistry 2014