Microporous and Mesoporous Materials 151 (2012) 26–33

Contents lists available at SciVerse ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier.com/locate/micromeso

Uniform mesoporous ZSM-5 single crystals catalyst with high resistance

to coke formation for methanol deoxygenation
Ali A. Rownaghi ⇑, Fateme Rezaei, Jonas Hedlund
Division of Chemical Technology, Luleå University of Technology, SE-971 87 Luleå, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: In this study, an efficient procedure for controllable synthesis of uniform mesoporous ZSM-5 single crys-
Received 14 August 2011 tals with various crystal architectures and high mesoporosity was developed. Compared with conven-
Received in revised form 14 October 2011 tional ZSM-5 catalyst, mesoporous ZSM-5 single crystals synthesized by this method exhibited
Accepted 13 November 2011
significantly higher external surface area and larger mesopore volume than conventional one. The cata-
Available online 20 November 2011
lytic performance of mesoporous zeolites were evaluated in the conversion of methanol to hydrocarbons
using a fixed-bed reactor operating at 370 °C, atmospheric pressure and weight hourly space velocities
Keywords:
(WHSV) of 0.16–6.58 h 1. By controlling the ZSM-5 synthesis procedure, the activity and stability of
Mesoporous ZSM-5 single crystals
Methanol dehydration
the ZSM-5 catalyst in the conversion of methanol to gasoline-range hydrocarbons can be favorably tuned.
Deactivation We also found that mesoporosity plays a crucial role in catalyst stability. Good correlation was observed
Catalyst lifetime between catalyst lifetime and mesoporosity. While the catalyst activity is related to the acid site density,
Methanol-to-gasoline the catalyst stability (deactivation rate) correlates with the measured surface ratio of BET and external
surface area (SBET/Smeso). The novel ZSM-5 catalyst exhibited improved stability due to the faster removal
of products with shorter diffusion path-length and lower coke formation. The obtained results indicated
that the novel mesoporous ZSM-5 catalyst containing relatively large pores (mostly mesopores) enhances
reaction yield towards gasoline-range hydrocarbons. It is therefore concluded that the relatively slow
deactivation rate (coke formation) of novel mesoporous ZSM-5 single crystals catalyst makes this zeolite
the preferred catalyst for the conversion of methanol to gasoline-range hydrocarbons at mild conditions.
Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction other chemicals [4–9]. In addition to Mobil’s methanol to gasoline

During the first global energy crisis in the early 1970s, alterna-
tives to fuels derived from crude oil became necessary. Methanol
to gasoline (MTG) process was developed by the Mobil company
in 1986 [1]. This process was not successfully commercialized
due to significant drop in crude oil prices in the 1990s. Recently,
with the problems associated with fossil resources use – namely,
greenhouse gas emission and looming petroleum crisis – biomass
to hydrocarbon process has received renewed interest as it will be-
come economically feasible. Biomass thermochemical-derived
intermediates contain a large fraction of light mixed oxygenated
compounds (less than C6) that are chemically and thermally
unstable and need to be deoxygenated to form compatible liquid
hydrocarbon molecules which could be a direct feed into current
petroleum refineries or that are direct replacements of diesel,
gasoline, or chemical products [2,3]. Furthermore, conversion of
methanol to hydrocarbons over solid acid catalysts (such as zeo-
lites) has drawn increasing attention in the petrochemical industry
as one of the efficient methods for manufacturing olefins, fuels and
⇑ Corresponding author. Tel.: +46 920 49 12 97; fax: +46 920 49 11 99.
E-mail addresses: ali.rownaghi@ltu.se, ali.rownaghi@gmail.com (A.A. Rownaghi).
(MTG) technology [1], several methanol to hydrocarbon (MTH)
processes are now being commercialized, including the Lurgi’s
methanol to propylene (MTP) [10], hydro/UOP’s methanol to ole-
fins (MTO) [11], and Topsøe integrated gasoline synthesis (TIGAS)
[12]. These processes are all examples of methanol to hydrocar-
bons using zeolite materials as catalysts. Hence, a novel route to
bring higher value on an energy content basis from renewable re-
sources can be achieved by MTH processes.
ZSM-5 zeolite with different SiO2/Al2O3 ratios has proven to be
one of the promising catalysts for conversion of small and large
oxygenates molecules to hydrocarbon products, which are similar
to gasoline-range molecules (C4–C10) of conventional petroleum
[13–15]. The activity of a ZSM-5 catalyst for the methanol dehydra-
tion is determined by the amount, the nature, and the accessibility
of its acid sites. In addition, it is well known that the structural de-
fects in ZSM-5 zeolite can strongly influence the catalytic activity
[16–18].
One of the main drawbacks of the methanol to hydrocarbon
process is related to the deactivation of the catalyst, due to coke
formation as a consequence of consecutive reactions between light
olefins and Brønsted or Lewis acid sites. This problem is due to the
limited diffusion of reactants/products within the zeolite materials

1387-1811/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.micromeso.2011.11.020
 A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33
[4,6]. To address this issue, mesoporous zeolites with faster diffu-
sion rate can be synthesized. In fact, higher resistance to coke poi-
soning and shorter diffusion path are the important advantages of
mesoporous zeolites for their successful applications in catalytic
reaction processes [19,20].
Several synthesis strategies for producing zeolite-based materi-
als with improved accessibility to the catalytically active sites lo-
cated in the microporous crystals have been pursued; they
include steaming, acid leaching treatments, carbon templating,
desilication, etc. [21–32]. Christensen et al. [21,22] reported the
preparation of mesoporous ZSM-5 crystals by carbon-templating
and catalytic evaluation in several reactions including gas phase
alkylation of benzene with ethylene and MTG. Another interesting
study published by the Kustova et al. [23,24] was the use of mes-
oporous iron-containing zeolite as catalysts for selective catalytic
reduction (SCR) of nitrogen oxide (NOx) with NH3. It was shown
that mesoporous Fe-ZSM-5 was more active than the conventional
ZSM-5 at all temperatures studied. Recently, base-treatment
(desilication) of zeolite either alone or in combination with dealu-
mination, has been performed to obtain similar advantages. Perez-
Ramırez et al. [25,26] synthesized a mesoporous ZSM-5 using the
post-synthetic steaming and alkali treatment and evaluated it in
the liquid phase benzene alkylation evidenced a higher activity
and selectivity to ethyl benzene. In a recent study published by
Kim et al. [28], a synthesis method capable of synthesizing sheets
of crystalline MFI was reported using an elegantly designed bifunc-
tional surfactant as the structure-directing agent. Such MFI config-
uration significantly increased the catalytic lifetime for MTG
reaction by more than five times compared to a conventional zeo-
lite with similar Si/Al ratio. Even though some of these approaches
are effective and result in improved catalytic performance in vari-
ous reactions, troublesome pre-preparation is still unavoidable in
most cases. Furthermore, treating zeolite crystals by leaching agent
(desilylation and dealumination) can cause a change in the SiO2/
Al2O3 ratio and nature of framework of zeolite [33,34]. Therefore,
a developed route is highly desired in terms of simplicity, repro-
ducibility and large-scale applicability.
Herein, we propose a facile and low cost route for synthesis of
high Al-content mesoporous ZSM-5 single crystals. To investigate
how the activity and, in particular, the deactivation relate to the
mesoporosity (structural properties) of ZSM-5 catalyst, we present
measured catalyst lifetime (activities and deactivation rates) and
hydrocarbon distribution in the conversion of methanol to hydro-
carbons of over a wide range of space time on novel mesoporous
ZSM-5 single crystals and conventional ZSM-5.
2. Experimental section
2.1. Catalyst preparation
Both ZSM-5 zeolites were synthesized from clear colloidal syn-
thesis mixtures using different synthesis procedures. Tetraethylor-
thosilicate (TEOS, 98%, Aldrich) and aluminum isopropoxide (AIP,
Aldrich) were used as Si and Al sources, respectively. Concentrated
tetrapropylammonium hydroxide (TPAOH, 40 wt.% aqueous solu-
tion, AppliChem) was also used as a template. The detailed synthe-
sis descriptions of ZSM-5 samples are as follows.
2.1.1. Synthesis of mesoporous ZSM-5 single crystals
Meso-ZSM-5 zeolite was synthesized by using a mixture with
the molar composition of 3TPAOH:0.5Na2O:0.5Al2O3:25SiO2:1500-
H2O:100EtOH:3PrOH according to following procedure. The
aluminum isopropoxide was added to an aqueous solution of
tetrapropylammonium hydroxide, water and NaOH. The mixture
was stirred at room temperature to obtain a clear solution. Then,
27
the clear aluminate solution was added to tetraethylorthosilicate
in the polypropylene bottle. In order to achieve a total hydrolysis
of the silicon species, the final solution was placed on shaker for
24 h at room temperature to hydrolyze TEOS and obtain a clear
solution. Hydrothermal treatment in a Teflon-lined autoclave for
24–30 h at 140–170 °C yielded white crystals that were further
purified by repeated centrifugation at 15,000 rpm for 30 min fol-
lowed by re-dispersion in water for four times. The purified prod-
uct was then freeze-dried. The calcined sample was obtained by
treatment of as-synthesized material in static air at 500 °C for
16 h (the calcination heating and cooling rate were 5 °C/min).
The SiO2/Al2O3 ratio was 50 on the basis of elemental analysis.
After air drying, mesoporous ZSM-5 crystals as small as 5 nm in
the form of mesoporous aggregates of approximately 1.95 lm in
length and an average pore diameter of 8 nm were obtained. This
structure is very open giving rise to very large surface areas, mes-
opore volume with a very narrow particle size distribution.
2.1.2. Synthesis of conventional ZSM-5 crystals
A conventional ZSM-5 (Con-ZSM-5) zeolite was synthesized
according to the procedure described elsewhere [35] and are omit-
ted here for brevity.
2.2. Catalyst characterization
2.2.1. X-ray diffraction (XRD)
X-ray powder diffraction (XRD) was performed to verify the
crystallinity of the zeolites. The diffraction patterns were recorded
on a Siemens diffractometer model D 5000 running in Bragg–
Brentano geometry employing CuKa (k = 1.54439 Å) radiation to
generate diffraction patterns from powder crystalline samples at
ambient temperature. The spectra were scanned at a rate of 2.0°/
min in the range 2h = 2–70°.
2.2.2. Surface area and porosity
Nitrogen adsorption/desorption isotherms at 196 °C were re-
corded using a Micromeritics ASAP 2010 instrument. Before the
measurements, samples were heated to 350 °C in a vacuum for at
least 12 h. The Brunauer–Emmett–Teller (BET) equation was used
to calculate the specific surface area (SBET) using adsorption data
at p/p0 = 0.05–0.25. The mesopore size distribution was derived
from the adsorption branch by using the Barrett–Joyner–Halenda
(BJH) method. The micropore volume was determined from a
t-plot.
2.2.3. Scanning electron microscopy (SEM)
The size, shape and size distribution of zeolite crystals were re-
corded using a FEI Magellan™ 400L instrument. Zeolite crystals
were deposited on a silicon wafer from dispersion and the sample
was then calcined in air at 500 °C. SEM images were recorded with-
out any coating. The average crystal length was determined by
measuring the length of 50 crystals for each sample.
2.2.4. Temperature-programmed desorption (TPD)
The acid site density of synthesized zeolites was determined by
temperature-programmed desorption of ammonia (NH3-TPD)
using ThermoFinnigan TPD/R/O 1100 Series instrument. The NH3-
TPD measurements were performed by pretreating the zeolites
(100 mg) by heating them to 773 K in NH3 flow (101 kPa,
25 cm3 min 1) and holding catalysts under that flow at 773 K for
1 h before cooling them to ambient temperature.
2.2.5. Inductively coupled plasma-atomic emission spectroscopy (ICP-
AES)
The quality of synthesized zeolites were checked with induc-
tively coupled plasma-atomic emission spectroscopy (ICP-AES) to
28 A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33

determine the bulk chemical compositions of the crystals using a
Perkin–Elmer Emission spectroscopy Model Plasma 1000.
2.3. Catalytic activity measurements
The catalytic measurements were conducted at atmospheric
pressure in a fixed-bed reactor connected to a Gas Chromatograph
(GC). The feed (N2) was saturated with methanol (99.9%) in a bub-
ble saturator at 30 °C. The saturated temperature was maintained
by a constant temperature bath. The inlet flow rate was adjusted
using a mass flow controller (MFC). The reaction was carried out
in a tubular reactor with an internal diameter of 17 mm and a total
length of 250 mm containing a standard mass of catalyst (i.e.,
0.50 g) at 370 °C with various ranges of weight hourly space veloc-
ity (WHSV) of 0.16–6.58 g g 1 h–1. For each test, the powder ZSM-5
catalyst was sieved to 250–300 lm and diluted with inert quartz
sand, then loaded to the reactor. The maximum temperature vari-
ation along the bed was ±1 °C. The catalyst and sand were mixed in
a 1:10 mass ratio. Soda glass beads and quartz glass were used as
catalysts beds during catalytic test. Soda glass beads and quartz
glass were placed on the top and bottom of the catalysts. A thermo-
couple was positioned in the center of the catalyst bed in order to
monitor the temperature. The catalyst was then pretreated in situ
at a heating rate of 5 °C/min under a flow of air (30 mL/min). When
the reactor temperature reached 500 °C, it was maintained at that
temperature for 4 h. The temperature was then decreased to the
♦
♦ catalysts were evaluated by NH3-TPD technique and the corre-
♦ ZSM-5
Intensity/ a.u.
reaction temperature in flowing air. To avoid possible condensa-
tion of hydrocarbons, the temperature of the effluent line was
constantly maintained at 100 °C. An on-line Varian 3800 Gas
Chromatograph with a flame ionization detector (GC-FID)
equipped with a capillary column (CP-Sil PONA CB fused silica
WCOT) analyzed the reactor exit gas with hydrogen as carrier
gas. The first sampling of the product gas was done 1 h after the
reactant feed started. During a typical run, subsequent samplings
and analyses were done with 1 h sampling periods. To simplify
the discussion, we defined the aromatics as the sum of C6–C10 aro-
matics, and the i-paraffins as the sum of C4–C10 i-paraffins.
3. Results and discussion
3.1. Catalyst characterization
Fig. 1 illustrates the X-ray powder diffraction (PXRD) patterns of
mesoporous and conventional ZSM-5 catalysts after calcination in
air for 16 h at 500 °C. The XRD patterns of both ZSM-5 samples
show well-resolved diffraction peaks, which are well matched to
MFI type structure, suggesting that their frameworks contain only
MFI type zeolite [36]. Table 1 lists the chemical composition of fi-
nal crystals in both mesoporous and conventional ZSM-5 samples.
From Table 1, it can be seen that the silicon-to-aluminum ratio
(SiO2/Al2O3) in the ZSM-5 samples determined by ICP-AES was
consistently 50 and did not change significantly at different syn-
thesis temperatures and time conditions. This ratio is in agreement
with previously reported data on catalytic properties of solid acid
ZSM-5 zeolite [7,9,28,32].
The acidic properties (content and strength of acid sites) of both
sponding profiles are displayed in Fig. 2. As shown in this figure,
three desorption steps were observed, leading to the assumption
that at least three types of acid sites exist. The Con-ZSM-5 catalyst

♦ contains three acid sites at 245 °C (weak acid strength), 265 °C

♦
♦ (moderate acid strength), and at 420 °C (strong acid strength). In
♦
the case of Meso-ZSM-5, two distinct desorption peaks were ob-
♦ ♦ ♦
served at 275 and 460 °C in NH3-TPD, which are usually ascribed
meso-ZSM-5
to NH3 desorption from moderate and strong acid sites. From Con-
ZSM-5 to Meso-ZSM-5, it is observed that the moderate peak shifts
to higher temperatures. This shift is accompanied by a gradual in-
crease in the intensity at intermediate temperatures (275 °C). The
Con-ZSM-5
relative values of total acidity of catalysts (i.e., the amount of
ammonia desorbing) correlates well with the amount of aluminum
20 40 60 in the framework for both catalysts, as estimated by ICP-AES, and a
2θ / Degree higher acidity is obtained for Meso-ZSM-5 catalyst. The tempera-
ture at which maximum amount of ammonia was desorbed by
Fig. 1. Powder XRD patterns of mesoporous ZSM-5 single crystals and conventional catalyst is also higher for Meso-ZSM-5 catalyst. Zhang et al.,
ZSM-5. 2h angles between 2° and 70° were collected.
reported that the NH3-TPD desorption at temperatures higher than

Table 1
Chemical and textural properties of ZSM-5 crystals.

Sample SiO2/Al2O3 Crystal lengthb Crystal widthb SBET Smeso SBET/Smeso Vtotal Vmicro Vmeso D
ratiosa (lm) (lm) (m2 g 1 c
) (m2 g 1 d
) (m2 g 1) (cm3 g 1 e
) (cm3 g 1 f
) (cm3 g 1 g
) (nm)h
Meso- 49.2 1.95 1.45 395 214 1.85 0.46 0.08 0.38 9.81
ZSM-5
Con-ZSM- 52.3 2.10 1.30 348 108 3.22 0.20 0.12 0.08 3.0
5
a
Determined by ICP-AES.
b
Estimated from SEM images.
c
Surface areas were obtained by the BET method using adsorption data in p/p0 ranging from 0.05 to 0.25.
d
Measured by t-plot method.
e
Total pore volumes were estimated from the adsorbed amount at p/p0 = 0.995.
f
Measured by t-plot method.
g
Vmeso = Vads,p/p0=99 Vmicro.
h
Average pore diameters were derived from the adsorption branches of the isotherms by the BJH method.
 A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33
Detector response/ a.u.
0 100 200 300 400
Temperature/ C
Fig. 2. NH3-TPD patterns of mesoporous ZSM-5 single crystals and conventional
ZSM-5 catalysts.
440 °C can be directly related to the density of Brønsted acid sites
[37]. In addition, desorption at lower temperatures may arise from
both Brønsted and Lewis acid sites. From the results presented in
Fig. 2, it can be observed that for Con-ZSM-5 catalyst the amount
of weak-medium acid sites is more than the strong acid sites.
Fig. 3 shows representative morphology of mesoporous ZSM-5
single crystals and conventional ZSM-5 crystals. The extreme high
resolution-SEM (XHR-SEM) images, shown in Fig. 3 clearly indicate
that the crystals obtained by different techniques have significantly
different crystal morphology, crystal size, and size distribution.
Fig. 3. XHR-SEM images of (a–c) mesoporous ZSM-5 single crystals (three different magnifications), and (d) conventional ZSM-5 crystals.
29
From each XHR-SEM image, approximately 50 ZSM-5 crystals were
Con-ZSM-5 measured and the average size was used as the crystal size (see Table
Meso-ZSM-5 1). As evident from Fig. 3a, Meso-ZSM-5 catalyst constitutes uniform
micro-rectangular crystals with about 1.95 lm  1.45 lm dimen-
sions. It is also noted that for this sample, the individual aggregates
are composed of closely packed nanocrystals, whose sizes vary from
2 to 5 nm. Aggregation processes of colloid chemistry derived by
intercrystal forces, rather than the formation of chemical bonds, is
more likely to be responsible for this morphology. In contrast,
Con-ZSM-5 displayed the typical twinned shape of micron-size
ZSM-5 crystals with the dimensions of 2.10 lm  1.30 lm along
the length of the envisioned twinns (Fig. 3d). Surface pitting was fol-
lowed by the attack at the twin boundaries, revealing the curved
junctions of the 90° intergrowth described and discussed in detail
500 600 700 elsewhere [38,39]. The preparation-dependent morphology ob-
served in the above ZSM-5 syntheses is closely related to the XRD re-
sults mentioned earlier.
The different morphologies of obtained zeolites were caused to
a large extent by different synthesis conditions even when the
same template was used. Therefore, ZSM-5 zeolites with various
sizes and morphologies can be synthesized with the same template
by carefully controlling the synthesis conditions. In fact, the
growth orientation of crystals varies largely with synthesis condi-
tions giving rise to production of crystals with different morpholo-
gies. The Meso-ZSM-5 single crystals (Fig. 3a) showed plough-land
roughs with some small furrow like on crystal surface. According
to SEM analysis and in accordance with X-ray diffraction patterns,
all obtained samples appear to be highly crystalline. It can be fur-
ther seen from high magnification images in Fig. 3b and c that the
Meso-ZSM-5 crystals have hierarchical rectangular architecture,
assembled by many stone pavement-like crystals of 2–5 nm in
30
300
250
Volume Adsorbed (cm 3/g STP)
Meso-ZSM-5
200 Con. ZSM-5
150
100
50
0 0.1 0.2 0.3 0.4 0.5
Relative Pressure (P/P 0)
Fig. 4. N2 adsorption/desorption isotherms of mesoporous ZSM-5 single crystals
and conventional ZSM-5 zeolites.
width. In addition, the stone pavement-like crystals are indeed
assembled layer-by-layer (see Fig. 3c). It is found that the obtained
mesostructured crystals (ca. 2–5 nm in diameter) consist of arrays
of parallel nanorods. The results obtained by both XHR-SEM and
ICP techniques are in agreement with that of XRD patterns. The
uniform mesoporous ZSM-5 single crystals are unique in the sense
that they contain both inter-connected micropores and mesopores
inside each individual single crystal.
The N2 adsorption data of the mesoporous ZSM-5 catalyst pre-
sented in Table 1 indicate the presence of mesopores that is rea-
sonably assigned to inter-crystalline voids between the stone
pavement-like crystals. Correspondingly, the average pore size
gives a diameter at about 10 nm. This indicates that ZSM-5 ob-
tained in this work has indeed a hierarchical structure (as also
shown by XHR-SEM). It is believed that the preparation of synthe-
sis mixture and hydrothermal reaction conditions are key factors
influencing crystal morphology, surface area (internal and exter-
nal), porosity and crystal size. Since inter-crystalline porosity is
developed, this simple method opens novel avenues for catalyst
design by improved diffusion characteristics, and thus a better uti-
lization of the zeolite crystal for catalytic applications. Based on
these results, the novel mesoporous ZSM-5 catalyst is potentially
a suitable candidate in various catalytic applications.
Fig. 4 shows different types of N2 adsorption/desorption iso-
therms corresponding to the prepared zeolites with SiO2/Al2O3 ra-
tio of 50. Meso-ZSM-5 sample exhibited a type-IV isotherm [40],
almost similar to that of Con-ZSM-5 sample at low-pressure re-
gions, but with significant increase in adsorption amounts with
pressure. Con-ZSM-5 sample, on the other hand, displayed a
type-I isotherm (plateau at relative pressure and without hystere-
sis), which is usually observed for microporous materials having
relatively low external surfaces [40]. Characteristic features of
type-IV isotherm are basically its hysteresis loop, which is associ-
ated with capillary condensation taking place in mesopores, and
the limiting uptake over a range of high p/p0 = 0.33. The Meso-
ZSM-5 showed a type-H4 hysteresis loop, which is often associated
with porous materials with narrow slit like pores. This finding is in
accordance with SEM images obtained from the same sample. The
isotherm corresponding to Meso-ZSM-5 crystal indicates that this
zeolite is mostly mesoporous as large amounts of nitrogen adsorp-
tion were observed at high p/p0. Meso-ZSM-5 possesses a total pore
volume of 0.46 cm3 g 1 measured at p/p0 = 0.995. The textural
properties of both zeolites (i.e., BET and mesoporous surface areas,
A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33
micro- and mesopore volumes and pore size) are summarized in
Table 1. The BET surface area obtained from the adsorption iso-
therm includes the contributions from both the internal and the
external surfaces. The Meso-ZSM-5 zeolite with higher surface area
and larger mesopore volume (Smeso = 214 m2 g 1 and
Vtotal = 0.46 cm3 g 1, see Table 1) displayed significantly higher
BET surface area of 395 m2 g 1. This value is very close to the value
reported by Abelló et al. for the same structure [41]. Kustova et al.
[42] synthesized a series of mesoporous samples consisting of a
ZSM-5 core and a silicalite shell and additional desilicated of mes-
oporous ZSM-5 samples resulted in mesopore volume of 0.39, 0.25
and 0.47 cm3 g 1, respectively. Kim et al. [28] reported the prepa-
ration of mesoporous ZSM-5 with mesopore volume of
0.32 cm3 g 1 via hydrothermal synthesis with the addition of orga-
nosilane surfactants. In another study, Christensen et al. [21,22,32]
compared the mesoporous ZSM-5 single crystals prepared by car-
0.6 0.7 0.8 0.9 1 bon-templating technique to a conventional ZSM-5 sample in
MTG reaction. Same authors reported the synthesis of mesoporous
ZSM-5 single crystals with mesopore volume of 0.31 cm3 g 1 in
fluoride media [43]. Mesopore volume of 0.82 cm3 g 1 for meso-
porous ZSM-5 was achieved by Park et al. [27] using carbon-tem-
plating and microwave heating techniques. The degree of
mesoporosity in Meso-ZSM-5 zeolite strongly depends on the mes-
opore volume as shown in Table 1. The N2-sorption data indicate
that the BET and mesopore surface areas of the zeolite samples de-
pend strongly on the synthesis mixture preparation method and
the hydrothermal treatment conditions.
In the case of conventional catalyst (Con-ZSM-5), the BET sur-
face area, mesopore surface area and total pore volume were esti-
mated to be 348 m2 g 1, 108 m2 g 1 and 0.20 cm3 g 1,
respectively. These values differ largely with those of novel
mesoporous zeolite. It is clear that the micropore volume (Vmicro)
of the Con-ZSM-5 catalyst is 1.5-fold higher than the micropore
volume of Meso-ZSM-5 catalyst (0.08 and 0.12 cm3 g 1, see Table
1). No significant dependence of micropore volume on the crystal
size or relative crystallinity of the samples was detected. The
Meso-ZSM-5 catalyst displays as 2-fold higher mesopore surface
area (214 m2 g 1) as Con-ZSM-5 catalyst (108 m2 g 1). Mesoporous
surface area of 75–244 m2 g 1 was achieved by Caicedo-Realpe and
Perez-Ramirez [33] using a two-step route comprising sodium alu-
minate and acid treatments of ZSM-5 zeolite. In addition, it can be
noted that, as mesopore surface area (Smeso) increases from 108 to
214 m2 g 1, mesopore volume (Vmeso) also enhances from 0.08 to
0.38 cm3 g 1 indicating a linear relation between Smeso and Vmeso.
Average pore diameters of crystals were obtained by applying
the BJH model from nitrogen adsorption data listed in Table 1.
The Con-ZSM-5 zeolite has smaller (3 nm) pores than Meso-
ZSM-5 (with an average pore size of 10 nm). The later pore size,
which is in the range anticipated for hydrocarbon cracking, is
unprecedented among all previously reported mesoporous forms
of MFI zeolite structure.
The characteristics of the final mesoporous zeolite are impor-
tant in terms of nature of framework, Si/Al ratio, and distribution
of metal species in the crystal volume. The mesoporosity is intro-
duced into the zeolite crystals directly during the crystallization
of the zeolite by having intercrystalline pores or voids. At the
beginning of synthesis, amorphous materials in the synthesis mix-
ture begin to form zeolite nanocrystals. These nanosized crystals
tend to aggregate and self-assemble into uniform-shaped polycrys-
talline particles. Constant stirring applied during the synthesis pro-
cess results in the formation of uniform ZSM-5 crystals. The
intercrystalline spaces in nanocrystals are directly accessible from
the outer surface of the zeolite. In this case, the condition that mes-
opores enhance the mass transport to/from the active sites in the
zeolite materials has been satisfied [32]. The highly crystalline
mesoporous zeolite synthesized in this work has in fact the crystal-
 A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33
100
Methanol Conversion (wt.%)
80
60
40
20
0 of catalysts lifetime was performed under similar SiO2/Al2O3 ratios
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Time on stream/ h
Fig. 5. Effect of mesoporosity on catalytic activity (deactivation curves); methanol
conversion vs time-on-stream for both catalysts (370 °C, 1 atm, WHSV = 1.05 g g 1 h).
line form of MFI zeolite with both micropores and mesopores in its
structure. Adjusting the composition of synthesis mixture and
hydrothermal synthesis time and temperature could easily modu-
late the composition and mesoporosity of zeolite. For instance, un-
der the conditions studied here, synthesis times ranging from 18 to
72 h and synthesis temperatures ranging from 120 to 170 °C are
optimum conditions for obtaining highly mesoporous single crys-
tals. These materials are an emerging and important class of cata-
lysts for a variety of reactions. Using this method to donate
uniform mesopores to ZSM-5 single crystals should significantly
extend its applications because such uniform single crystals with
additional mesoporosity lead to easier access to the active sites
as well as resistance to cock formation and consequently catalyst
deactivation.
3.2. Catalytic performance in methanol dehydration reaction
Fig. 5 shows an overview of the activities and deactivation
behavior (catalyst stability) of two catalysts described and
discussed in this contribution. As evident from this figure, the cor-
responding conversion rates are very close to each other during the
initial reaction period, implying similar extent of deactivation
Meso-ZSM-5 Single Crystal
Methanol
Reaction zone
Con-ZSM-5 Crystals
Methanol Unutilized part of
ZSM-5 crystals
Fig. 6. Combination of SEM-images and schematic, illustration of the diffusion pass rate of methanol through conventional and mesoporous ZSM-5 single crystals.
31
during the initial reaction time. However, as reaction proceeds, it
can be clearly seen that the mesoporous ZSM-5 catalyst is more
stable than the conventional ZSM-5 catalyst indicated by higher
methanol conversion rate. In the case of Con-ZSM-5 catalyst, the
methanol conversion started to decrease after an initial reaction
period of 10 h. Hereafter the catalytic lifetime is defined as the
time at which the catalytic conversion decreases by 50% (i.e., t1/2)
[44,45]. Thus, the t1/2 value obtained for conventional catalyst
(Con-ZSM-5) was 30 h (see Fig. 5). Unlike the conventional ZSM-
5 catalyst, mesoporous ZSM-5 catalyst is more stable (up to 95%)
over the same reaction time. It is worth noting that the comparison
because the catalyst physicochemical properties and lifetime are
more likely to change by Al-content [44–46]. These results show
that the catalytic lifetime is enhanced up to 2-fold for our meso-
porous ZSM-5 catalyst mainly due to the generation of mesoporos-
ity. Thus, we did succeed in producing a ZSM-5 catalyst with high
activity and stability.
The catalyst activity is determined by the acidity, whereas the
catalyst stability (deactivation behavior) is determined by the acid
site accessibility and diffusion rate, that is not correlated with the
Brønsted acidity of the ZSM-5 crystals. The ratio of the BET and
external surface areas for two catalysts are presented in Table 1.
As can be seen, a surprisingly clear correlation exists between
the deactivation rate (Fig. 5) and the SBET/Smeso ratio, indicating
that the higher mesopore surface area affects the catalyst stability
and deactivation rate. This leads to the conclusion that the param-
eter determining catalyst deactivity is related to the acid site
accessibility as well as the diffusion rate of reactants and products.
In general, the interest in mesoporous ZSM-5 is motivated by the
desire to improve the mass transport of reactants and products to
and from the active sites located inside the ZSM-5 micropores.
Schematic illustration of concentration profiles of a reagent
through mesoporous and conventional crystals are shown in
Fig. 6. For a catalytic reaction involving reactants and products
with relatively limited diffusion reaction, it is not possible to fully
utilize the entire ZSM-5 crystals unless intracrystalline mesopores
are introduced. As proposed by Bibby et al. for conventional ZSM-5
catalyst, the initial deposition of coke takes place inside the zeolite
Results
-higher activity and selectivity
-Light olefins
-Alkyl-aromatics
Results
-Lower activity and selectivity
-Wide-range hydrocarbons
-Coke
32 A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33
75
70
65
60
distribution, [wt%]
55
50
45
40
35
30
25
Hydrocarbon
20
15
10
5
0
-5
0.1
Fig. 7. Effect of space velocity on hydrocarbon distribution over mesoporous ZSM-5 single; selectivity vs space time (370 °C, 1 atm, methanol partial pressure: 0.21 atm).
channels [47,48] and results in an inactive or inaccessible zone in
the center of the crystals. Therefore, only the outer surface area of
the crystal is accessible to the reactant. In addition, the polyme-
thylated benzenes formed inside the center of the crystal (unuti-
lized part in Con-ZSM-5 in Fig. 6) do not reach the gas phase,
and hence a fast deactivation occurs. On the contrary, for meso-
porous ZSM-5 single crystals, the polymethylated benzenes accu-
mulation in the catalyst crystals is slower, because the transfer
to the gas phase is more efficient as a consequence of both a larger
total surface area and the shorter distance to be covered by diffu-
sion. Given that the diffusion becomes faster with an increasing
external surface area, it is found that crystals with a high external
surface area (Meso-ZSM-5) are expected to deactivate slow, which
agrees qualitatively with the observed trends in Fig. 5.
When the activity of two catalysts having similar Al-contents
were correlated with the acid content and external surface areas,
a good correlation was observed between catalytic activity and
strong acidic sites on external surface area (see Fig. 5 and Table
1). The partial destruction of the crystalline structure improves
the accessibility to the intra-crystalline active sites by large mole-
cules. This favorably affects the catalytic reactivity of mesoporous
ZSM-5 catalyst. Therefore, it is evident that introducing mesopo-
rosity in ZSM-5 structure can lead to a significant improvement
in the methanol dehydration reaction.
As earlier findings reported by Bibby et al., indicate, coke forma-
tion occurs much more rapidly in a microporous ZSM-5 catalyst
than in a mesoporous zeolite during the hydrocarbons reaction
[47,48]. Therefore, coke in ZSM-5 catalyst is preferentially gener-
ated inside the micropores prior to the deposition at the external
surface, demonstrating that external coking does not play a domi-
nant role when a significant amount of internal defects is present
at the same time. This can be verified in observing a significant dif-
ference in the color of the two catalysts after 30 h reaction.
Fig. 7 shows more complete collections of hydrocarbon distribu-
tion vs WHSV with five orders of magnitude. The corresponding data
are also listed in Table 2. The hydrocarbon products are very similar,
suggesting a common reaction pathway over various contact times.
The primary reaction is dehydration of methanol to dimethyl ether
and water [5]. Methanol and dimethyl ether further react to give
light olefins (secondary reaction), which then are transformed to
higher olefins, aromatics and paraffins. The detailed mechanism of
this process has been discussed elsewhere [49].
Dimethyl ether
Light paraffins (C1-C4)
Ligh olefins (C2-C3)
Other paraffins and olefins (C4+)
Alkyl aromatics (BTX)
1 10
WHSV, [g/g hr]
The results at the highest space velocity (WHSV = 6.58 g g 1 h)
show that the product stream contains a relatively large amount of
dimethyl ether (DME) formed by the reversible dehydration of
methanol. However, the presence of a small fraction of DME at high
space velocity implies that the contact time was not sufficient to
convert all DME to light olefins. Based on the data, the DME
forming reaction might be related as either parallel or sequential
to the major reaction steps leading to hydrocarbons. Running the
reaction at the highest velocity and lowest conversion (38% at
WHSV = 6.58 g g 1 h) gives rise to a 4% dimethyl ether,
hydrocarbon distribution of 35% primary hydrocarbons (paraffins
and olefins), and 40% alkyl aromatics (benzene, toluene, and
xylene). These results indicate that higher space velocity is required
for an efficient conversion of methanol to dimethyl ether over mes-
oporous ZSM-5 catalysts.
The light olefins formed during first steps then undergo conden-
sation and oligomerization over mesoporous zeolite catalyst, lead-
ing to formation of final alkyl aromatic products. The conversion of
methanol was completed between 0.16 and 1.05 WHSV, with the
higher formation of light olefins and substantial amounts of aromat-
ics. The highest selectivity for light olefins (17%) was reached at
370 °C at longest contact time, although the selectivity for light par-
affins was increased with enhanced space velocity. The existing
hydrocarbons exhibit a relatively narrow range of molecular
weights, terminating abruptly at about C6–C10, which correspond
to a boiling point range of gasoline. Alkyl aromatics were the main
products at various contact times. Essentially no hydrocarbons
above C10 were produced. These results imply that mesoporosity
not only plays a dominant role in stabilizing the lifetime of the cat-
alyst but also in increasing the selectivity to alkyl aromatics. As
shown in Table 2, the selectivity to alkyl aromatics exceeds 50%
while product distributions are very similar, in which light olefins
and aromatics are main products at low space velocity. Furthermore,
it can be seen from Fig. 7 that the selectivity to aromatics and light
olefins decreased gradually with increasing WHSV. On the other
hand, the selectivity to aliphatic hydrocarbons increased step-by-
step. At the same space velocity, the conversion of methanol dramat-
ically decreased from 100% (at low WHSV) to 38% (highest WHSV).
Through the choice of space velocity, the reaction can be tailored
to maximize alkyl aromatics (BTX), light olefins (ethylene and pro-
pylene), or liquefied petroleum gas (LPG). Moreover, due to shape-
selective constraints of ZSM-5 catalyst, this process yields a high
 A.A. Rownaghi et al. / Microporous and Mesoporous Materials 151 (2012) 26–33 33

Table 2
Effect of space velocity on methanol dehydration and hydrocarbon distribution over mesoporous ZSM-5 single crystals (370 °C, methanol partial pressure: 0.21 atm).

WHSV [g of liquid methanol/(g of Methanol conversion Hydrocarbon distribution (%)

catalyst/h)] (%)
Paraffins Olefins Dimethyl Paraffins and Olefins Alkyl Aromatics Othersa
(C1–C4) (C2–C3) ether (C4+) (BTX)
0.16 100 14.6 16.6 – 3.2 65.5 0.1
0.43 100 19.0 12.1 – 7.7 61.0 0.2
1.05 100 22.8 8.0 – 12.0 57.0 0.2
2.63 40 26.9 3.0 – 19.2 50.5 0.4
6.58 38 33.6 1.3 4.0 21.2 39.4 0.5
a
Including CO, CO2 and coke.

quality hydrocarbon and minimizes the production of branched [4] F.J. Keil, Micropor. Mesopor. Mater. 29 (1999) 49.
[5] C.D. Chang, A.J. Silvestrl, J. Catal. 47 (1977) 249.
products.
[6] M. Stocker, Micropor. Mesopor. Mater. 29 (1999) 3.
[7] A.A. Rownaghi, F. Rezaei, J. Hedlund, Catal. Comm. 14 (2011) 37.
4. Conclusions [8] A.T. Aguayo, D. Mier, A.G. Gayubo, M. Gamero, J. Bilbao, Ind. Eng. Chem. Res. 79
(2010) 12371.
[9] A.A. Rownaghi, J. Hedlund, Ind. Eng. Chem. Res. 50 (2011) 11872.
The preparation of uniform mesoporous MFI single crystals [10] H. Koempel, W. Liebner, Stud. Surf. Sci. Catal. 167 (2007) 261.
using a novel technique was reported in this study. The role of [11] J.Q. Chen, A. Bozzano, B. Glover, T. Fuglerud, S. Kvisle, Catal. Today 106 (2005)
103.
mesoporosity of ZSM-5 catalyst was investigated in conversion of [12] J. Topp-Jørgensen, Surf. Sci. Catal. 36 (1988) 293.
methanol to hydrocarbons at rather mild conditions, i.e., 370 °C [13] P. Li, W. Zhang, X. Han, X. Bao, Catal. Lett. 34 (2010) 124.
temperature and ambient pressure. The mesoporous ZSM-5 zeolite [14] W.J.H. Dehertog, G.F. Froment, Appl. Catal. 71 (1991) 153.
[15] V.R. Choudhary, V.S. Nayak, Zeolites 5 (1985) 325.
discussed in the present study exhibits several attractive features,
[16] S. Donk, A.H. Janssen, J.H. Bitter, K.P. Jong, Catal. Rev. 45 (2003) 297.
such as large BET and mesoporous surface area, controlled pore [17] F. Rezaei, P. Webley, Sep. Purif. Technol. 70 (2010) 243.
size, large mesopore volume and low cost and facile synthesis pro- [18] K. Wandelt, Surf. Sci. 387 (1991) 251.
[19] J. Maier, Angew. Chem. Int. Ed. 32 (1993) 528.
cedure. It was found that the deactivation rate correlates well with
[20] Y. Tao, H. Kanoh, L. Abrahms, K. Kaneko, Chem. Rev. 106 (2006) 896.
the BET and external surface area ratio. A ZSM-5 catalyst with a [21] C.H. Christensen, K. Johannsen, E. Törnqvist, I. Schmidt, H. Topsøe, Today 128
high SBET/Smeso ratio is expected to show a slow deactivation rate (2007) 117.
in the conversion of methanol to hydrocarbons. Introducing meso- [22] C.H. Christensen, I. Schmidt, C.H. Christensen, Catal. Comm. 5 (2004) 543.
[23] A.L. Kustov, K. Egeblad, M. Kustova, T.W. Hansen, C.H. Christensen, Top. Catal.
porosity to ZSM-5 crystals increases external surface area resulting 45 (2007) 159.
in a faster diffusion of the hydrocarbon products in the ZSM-5 [24] A.L. Kustov, T.W. Hansen, M. Kustova, C.H. Christensen, Appl. Catal. B 76 (2007)
channels, and a correspondingly slower deactivation rate. The pre- 311.
[25] J. Perez-Ramırez, S. Abello, A. Bonilla, J.C. Groen, Adv. Funct. Mater. 19 (2009)
sented data clearly show that apart from the commonly recognized 164.
deactivation of ZSM-5 due to coking at the external surface, the [26] J.C. Groen, L.A.A. Peffer, J. Perez-Ramırez, Micropor. Mesopor. Mater. 60 (2003)
deactivation via coking at internal pores of ZSM-5 plays an impor- 1.
[27] J.B. Koo, N. Jiang, S. Saravanamurugan, M. Bejblova, Z. Musilova, J. Cejka, S.E.
tant role. Deactivation of catalyst occurred much slower in the case Park, J. Catal. 276 (2010) 327.
of mesoporous ZSM-5 than conventional ZSM-5, mainly due to fas- [28] J. Kim, M. Choi, R. Ryoo, J. Catal. 269 (2010) 219.
ter diffusion of products from the zeolite channels, hence reducing [29] M. Rivallan, G. Ricchiardi, S. Bordiga, A. Zecchina, J. Catal. 264 (2009) 104.
[30] B. Louis, F. Ocampo, H.S. Yun, J.P. Tessonnier, Chem. Eng. J. 161 (2010) 397.
the formation of coke. This resulted in simultaneous improvement
[31] J. Cejka, S. Mintova, Catal. Rev. Sci. Eng. 49 (2007) 457.
of catalytic activity and selectivity towards alkyl aromatics. Cou- [32] M.S. Holm, E. Taarning, K. Egeblad, C.H. Christensen, Catal. Today 168 (2011) 3.
pled with existing synthesis gas-based biomethanol technologies, [33] R. Caicedo-Realpe, J. Perez-Ramirez, Micropor. Mesopor. Mater. 128 (2010) 91.
[34] M. Ogura, Catal. Surv. Asia 12 (2008) 16.
these results are very promising in developing a highly efficient
[35] J. Hedlund, O. Öhrmana, L.V. Msimang, E.V. Steenb, W. Böhringer, S. Sibya, K.
alternative process for practical applications in renewable fuel Möller, Chem. Eng. Sci. 59 (2004) 2647.
products and chemicals. These results suggest that the use of mes- [36] Ch. Baerlocher, L.B. McCusker, D.H. Olson, Atlas of Zeolite Framework Types,
oporous ZSM-5 catalyst prepared in this study, will significantly re- Elsevier, 2007.
[37] W. Zhang, E.C. Burckle, P.G. Smirniotis, Micropor. Mesopor. Mater. 33 (1999)
duce the coke formation and enhance production of alkyl 173.
aromatics from methanol at mild operation conditions. [38] J.R. Agger, N. Hanif, C.S. Cundy, A.P. Wade, S. Dennison, P.A. Rawlinson, M.W.
Anderson, J. Am. Chem. Soc. 125 (2003) 830.
[39] D.G. Hay, H. Jaeger, K.G. Wilshier, Zeolites 10 (1990) 571.
Acknowledgments [40] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T.
Siemieniewska, Pure Appl. Chem. 57 (1985) 603.
The Swedish Energy Agency, Knut and Alice Wallenberg foun- [41] S. Abello, A. Bonilla, J. Perez-Ramirez, Appl. Catal. A-Gen. 364 (2009) 191.
[42] M. Kustova, M.S. Holm, C.H. Christensen, Y.H. Pan, P. Beatoa, T.V.W. Janssensa,
dation is acknowledged for financially supporting the Magellan F. Joensen, J. Nerlov, Stud. Surf. Sci. Catal. 174 (2008) 117.
SEM instrument. [43] K. Egeblad, M. Kustova, S.K. Klitgaord, K. Zhu, C.H. Christensen, Micropor.
Mesopor. Mater. 101 (2007) 214.
[44] K.G. Ione, G.V. Echevskii, G.N. Nosyreva, J. Catal. 85 (1984) 287.
References
[45] S.M. Campbell, D.M. Bibby, J.M. Coddington, R.F. Howe, J. Catal. 161 (1996) 350.
[46] D.B. Luk’yanov, Zeolites 12 (1992) 287.
[1] S. Czernik, A.V. Bridgwater, Energ. Fuel. 18 (2004) 590. [47] D.M. Bibby, N.B. Milestone, J.E. Patterson, L.P. Aldridge, J. Catal. 97 (1986) 493.
[2] D. Mohan, C.U. Pittman, P.H. Steele, Energ. Fuel. 20 (2006) 848. [48] D.M. Bibby, C.G. Pope, J. Catal. 116 (1989) 407.
[3] J. Cobb, in: G. Connell (Ed.), New Zealand Synfuel: The Story of the World’s First [49] S.A. Tabak, S. Yurchak, Catal. Today 6 (1990) 307.
Natural Gas to Gasoline Plant, Cobb/Horwood Publications, Auckland, New
Zealand, 1995, p. 1.