Comparative study of methanol to olefins over ZSM-5, ZSM-11, ZSM-22 and EU-1: Dependence of catalytic performance on the zeolite framework structure

Article abstract of DOI:10.1166/jnn.2017.13986

The conversion of methanol to olefins (MTO) was comparatively conducted over four 10-ring zeolites, viz., ZSM-5, ZSM-11, ZSM-22 and EU-1, in both continuous-flow and pulse reactors, to reveal the relation between the catalytic performance and framework structure. The results illustrate that the performance of a zeolite catalyst in MTO is closely dependent on its framework structure. Among four zeolites, ZSM-5 with three dimensional channels and intersections exhibits the highest catalytic activity and stability in MTO. In comparison with ZSM-5 and ZSM-11, ZSM-22 and EU-1 are deactivated much more rapidly due to the blockage of the pore opening. Especially, the straight channels of ZSM-11 facilitate the orderly array of butene molecules in the channels, which may promote the hydrogen transfer reaction and lead to high selectivity to butane and C₅₊ products, in comparison with ZSM-5. The hydrocarbon pool mechanism is also applicable in ZSM-22, though less important; the 1D straight channels in ZSM-22 are easily blocked, leading to the quick deactivation. EU-1 zeolite exhibits quite high selectivity to propene due to its large 12-ring side-pockets in which polyalkylbenzenes are formed as the hydrocarbon pool species. The insights shown in this work are conducive to the development of better MTO catalysts and reaction processes.

Full text of DOI:10.1166/jnn.2017.13986

Article
Journal of
Nanoscience and Nanotechnology
Vol. 17, 3680–3688, 2017
Copyright © 2017 American Scientific Publishers
All rights reserved
www.aspbs.com/jnn
Printed in the United States of America
Comparative Study of Methanol to Olefins Over ZSM-5,
ZSM-11, ZSM-22 and EU-1: Dependence of Catalytic
Performance on the Zeolite Framework Structure
Buqin Jing1ꢀ2ꢀ3, Junfen Li1ꢀ∗, Zhikai Li1, Sen Wang1ꢀ3, Zhangfeng Qin1,
Weibin Fan1, and Jianguo Wang1ꢀ∗
1State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, P.O. Box 165,
Taiyuan, Shanxi 030001, P. R. China
2College of Chemistry and Environmental Engineering, Shanxi Datong University, Datong, Shanxi 037009, P. R. China
3University of Chinese Academy of Sciences, Beijing 100049, P. R. China
The conversion of methanol to olefins (MTO) was comparatively conducted over four 10-ring zeo-
lites, viz., ZSM-5, ZSM-11, ZSM-22 and EU-1, in both continuous-flow and pulse reactors, to reveal
the relation between the catalytic performance and framework structure. The results illustrate that
the performance of a zeolite catalyst in MTO is closely dependent on its framework structure.
Among four zeolites, ZSM-5 with three dimensional channels and intersections exhibits the highest
Delivered by Ingenta to: Freie Universitaet Berlin
catalytic activity and stability in MTO. In comparison with ZSM-5 and ZSM-11, ZSM-22 and EU-1
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
are deactivated much more rapidly due to the blockage of the pore opening. Especially, the straight
Copyright: American Scientific Publishers
channels of ZSM-11 facilitate the orderly array of butene molecules in the channels, which may
promote the hydrogen transfer reaction and lead to high selectivity to butane and C5+ products, in
comparison with ZSM-5. The hydrocarbon pool mechanism is also applicable in ZSM-22, though
less important; the 1D straight channels in ZSM-22 are easily blocked, leading to the quick deacti-
vation. EU-1 zeolite exhibits quite high selectivity to propene due to its large 12-ring side-pockets in
which polyalkylbenzenes are formed as the hydrocarbon pool species. The insights shown in this
work are conducive to the development of better MTO catalysts and reaction processes.
Keywords: Methanol to Olefins, ZSM-5, ZSM-22, ZSM-11, EU-1, Framework Structure, Pulse
Reaction.
1. INTRODUCTION
was an autocatalytic process through analyzing the kinetic
data;7 nowadays, the hydrocarbon pool (HCP) mechanism
has received wide recognition.8 By the HCP mechanism,
large olefinic compounds and/or cyclic organic species are
trapped in the zeolite cages or channels and act as a co-
catalyst; light olefins are then produced through the cat-
alytic reaction cycles involving the repeated addition of
methanol and DME to HCP and the elimination of olefins
from the HCP species.9,10
The performance of a zeolite catalyst in MTO as well
as the actual hydrocarbon pool (HCP) species is closely
related to its porous structure. Svelle and co-workers
reported that over H-ZSM-5 higher methylbenzenes were
virtually less reactive than that over SAPO-34;11 ethene
was formed via polymethylbenzenes, whereas propene
and higher alkenes were obtained mainly by alkene
With the increasingly growing demand for light olefins
and the depletion of crude oil sources, the conversion of
methanol to olefins (MTO) has become an important alter-
native to petrochemical processing to get light olefins, as
methanol can be easily produced via syngas from mul-
tifarious carbon sources.1–5 Besides H-ZSM-5 that has
been used as an efficient catalyst for the conversion of
methanol to hydrocarbons (MTH) since 1977, a large num-
ber of protonated zeolites or zeotype materials have been
tested as the catalyst in MTO.6 Meanwhile, great progress
has also been made in both industrial process develop-
ment and theoretical investigation on the MTO reaction
mechanism. Early, Chen and Reagan found that MTH
∗Authors to whom correspondence should be addressed.
3680
J. Nanosci. Nanotechnol. 2017, Vol. 17, No. 6
1533-4880/2017/17/3680/009
doi:10.1166/jnn.2017.13986

Jing et al.
Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
methylation and cracking. Such a dual-cycle concept was
also used by Olsbye and co-workers to rationalize the
product selectivity over different catalysts.12 However, it
is still a difficult task to give a clear description on the
relation between the performance in MTO and the frame-
work structure of a zeolite catalyst, due to the complexity
of the MTH reactions and the broad spectrum of products.
Besides, the operation conditions such as feed rate and
temperature13 as well as the characteristics of zeolites such
as acidity,14,15 pore structure and topology,16 and particle
size17 all have great impacts on the actual product yield
and catalytic lifetime in MTO.
A comparison of various zeolites on their catalytic per-
formance can help us to understand the performance-
structure relation of the zeolite catalysts in MTO. Dai
and co-workers found that among four silicoaluminophos-
phates zeolites (SAPO-34, SAPO-41, SAPO-11, and
SAPO-46), the adsorption of methanol, the formation
of HCP compounds as well as the catalyst deactiva-
tion behavior were strongly influenced by the amount of
Brønsted acid sites,18 whereas the diffusion of MTO prod-
ucts and then the product selectivity were controlled by
the framework structure. Li and co-workers illustrated that
for MTH over SAPO-34, ZSM-5 and ZSM-22, the reac-
tion route was related to their structure topology due to the
space confinement;19 over SAPO-34, the reactions were
effect of pore dimensionality and channel and cage size on
their catalytic performance in MTO was then clarified.
2. EXPERIMENTAL DETAILS
2.1. Catalyst Preparation
To synthesize ZSM-5, aqueous solution of sodium alumi-
nate was firstly added to tetrapropylammonium hydrox-
ide (TPAOH); seed crystals were then added slowly and
the solution was stirred for 20 min. After that, silica sol
(JN-40) was added to the mixture under stirring, to get
the synthesis gel with the composition of 1 Al2O3:59
SiO2:9 TPAOH:1.8 Na2O:1765 H2O. After stirring for
2 h, the gel was transferred into a Teflon-lined stain-
less steel autoclave and crystallized in an oven for 48 h
ꢀ
at 170 C. The final product was quenched, centrifuged,
washedꢀ repeatedly with distilled water, and then dried
at 100 C overnight. The organic template was removed
ꢀ
by calcination at 550 C for 10 h. The calcined sample
was further subjected to ion exchange thꢀree times with
NH4NO3 solution (1 mol L−1, 5 h at 80 C) and subse-
ꢀ
quent calcination at 550 C for 10 h, to obtain ZSM-5 in
H form (H-ZSM-5).
ZSM-11 was prepared using tetrabutylammonium
hydroxide (TBAOH, 40%) as the template. A clear solu-
tion of sodium aluminate and sodium hydroxide was added
to TBAOH. Silica sol (JN-30) was then slowly added and
dominated by the HCP mechanism, over ZSM-5, the reac-
Delivered by Ingenta to: Freie Universitaet Berlin
the mixture was stirred for 2 h to obtain a homogeneous
gel with the composition of 1 Al2O3:60 SiO2:18 TBAOH:6
Na2O:2353 H2O. After aging for 2 h at room temperature,
the gel was introduced into a Teflon-lined stainless steel
autoclave and then heated to 150 ꢀC and maintained at this
temperature for 120 h. The final product was quenched,
centrifuged, washed repeatedly with water and then dried
overnight. The dried samples were then subjected to a
series of calcination and ion-exchange processes to obtain
ZSM-11 in H form (H-ZSM-11), as described above for
ZSM-5.
ZSM-22 was synthesized by following a synthesis pro-
cedure from literature with minor modifications.3 An
aqueous solution of potassium hydroxide was added to
aluminum sulfate under stirring, in which diaminooctane
(DAO) was added and the mixture was stirred continu-
ously until the mixture became clear. After that, JN-30
was added under vigorous stirring. The resulting gel with
a composition of 1 Al2O3:60 SiO2:18 DAO:8 K2O:2353
H2O was transferred into a Teflon-lined stainless steel
autoclave for crystallization at 160 ꢀC for 6 day. The prod-
uct wasꢀrecovered by filtration, washed, dried, and calcined
at 600 C for 15 h; to get ZSM-22 in H form (H-ZSM-
22), it was then subjected to a series of calcination and
ion-exchange processes as mentioned above.
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
tions followed a dual-cycle model, whereas over ZSM-
Copyright: American Scientific Publishers
22, the olefin methylation-cracking route should be the
main reaction pathway. Bleken and co-workers compared
the catalytic performances of TNU-9, IM-5, ZSM-11 and
ZSM-5 in a fixed-bed reactor and found that four catalysts
gave similar product selectivity;20 however, TNU-9 and
IM-5 with larger cavities in channel intersection produced
more polyaromatics and were deactivated more rapidly,
in comparison with ZSM-5 and ZSM-11. Among four
1-dimensional 10-ring zeolites, viz., ZSM-22, ZSM-23,
ZSM-48 and EU-1, Teketel and co-workers found that EU-
1 gave the lowest selectivity to C5+ products but notable
amount of aromatics because of its 12-ring side pocket.21
All these suggest that a subtle variation in the porous struc-
ture of zeolite may have a dramatic influence on its cat-
alytic performance in MTO.
In this work, therefore, four 10-ring zeolites with differ-
ent porous structures, viz., ZSM-5, ZSM-11, ZSM-22 and
EU-1, were synthesized, characterized and used as the cat-
alysts in MTO. The synthesis conditions were delicately
controlled to obtain various kinds of zeolites with similar
particle size and acidity. Both continuous-flow and pulse
reactor were adopted to observe the reaction behaviors
in the induction period as well as in the steady state. In
addition, the adsorption isotherms of various reactant and
product molecules on these zeolites were measured and the
occluded organic species formed during MTO were also
analyzed. On the basis of these experimental results, the
EU-1 in alkaline media was prepared following the pro-
cedures reported by Shin and co-workers with hexam-
ethonium bromide (HMBr2) as a structure-directing agent
(SDA).22,23 In a typical synthesis, sodium aluminate and
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017
3681

Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
Jing et al.
an aqueous solution of sodium hydroxide were mixed and
stirred until a clear solution was obtained. SDA was then
added, stirring for 30 min. After that, JN-30 was added
and the mixture was manually stirred for 30 min. The
resultant gel with a composition of 1 Al2O3:59 SiO2:7
Na2O:6 HMBr2:1470 H2O was transferred into a Teflon-
lined stainless steel autoclave and crystallized for 7 days at
Solid-state magic-angle spinning nuclear magnetic reso-
nance (MAS NMR) spectra were recorded on a 400 MHz
spectrometer (Avance III, Bruker) equipped with a 2.5 mm
MAS NMR probe; the spinning speed was 15 kHz. 27Al
MAS NMR spectra were measured at a 27Al frequency
of 104.019 MHz with a ꢅ/8 rad pulse length of 2.0 ꢆs
and a recycle delay of 2 s. The 27Al chemical shifts were
reported relative to Al(H2O)36+ solution.
ꢀ
170 C. The final product was quenched, centrifuged, and
repeatedly washed with distilled water and dried at 100 ꢀC
Adsorption isotherms of ethene, propene, butene and
methanol on the zeolites were measured at 25 ꢀC on
a BELSORP-max instrument; prior to the measurement,
overnight. The organic template was removed by calcina-
ꢀ
tion at 600 C for 15 h. The calcined sample was then
ꢀ
ꢀ
20–30 mg of calcined sample was degassed at 300 C for
subjected to ion-exchange and then calcination at 600 C
8 h.
for 15 h to get EU-1 in H form (H-EU-1).
2.3. Catalyst Tests and Analytic Procedures
2.2. Catalyst Characterization
For the continuous-flow test, the reaction of methanol to
olefins (MTO) was performed in a fixed-bed reactor with
an inner diameter of 3.0 mm. The catalyst sample was
pressurized to wafers and then crushed and sieved to 40–60
mesh before use. In a typical run, 100 mg of the zeolite
catalyst was first pretreated in situ at 550 ꢀCꢀunder air flow
for 120 min and then cooled down to 300 C. After that,
methanol was then fed into the reactor through a pump,
The actual chemical composition of the calcined zeolites
was determined by inductively coupled plasma–atomic
emission spectroscopy (ICP–AES, Autoscan16, TJA).
Powder X-ray diffraction (XRD) patterns were collected
on a Rigaku MiniFlex II desktop X-ray diffractometer with
monochromated Cu Kꢀ radiation (ꢁ = 0ꢂ1542 nm, 40 kV
and 200 mA) in the range of 2ꢃ from 5ꢀ to 40ꢀ with a
scan speed of 5.0 min−1
.
with a weight hourly space velocity (WHSV) of 5.0 h−1
.
The scanning electron microscopy (SEM) image to char-
acterize the surface morphology of the as-prepared zeo-
lite sample was taken on a scanning electron microscope
(HITACHI S-4700); to get the SEM image, the zeo-
lite samples was covered with a thin layer of gold by
sputtering.
As ZSM-22 may be deactivated very fast at high temper-
ature, in this work, the MTO reactions were carried out
under atmospheric pressure and a relatively low tempera-
Delivered by Ingenta to: Freie Universitaet Berlin
ꢀ
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
ture of 300 C, to have a better study on the MTO product
Copyright: American Scientific Publishers
distribution.
For the pulse reaction, 100 mg of catalyst sample was
first pretreated by the same procedure as described above
and kept at 300 ꢀC. For each pulse, 1 ꢆL of methanol was
injected into the reactor.
The product stream was analyzed with a gas chromato-
graph (GC) equipped with a DB-Wax capillary column and
an Al2O3 packed column and a flame ionization detector
(FID). The conversion of methanol was referred to the per-
centage of methanol that was converted to hydrocarbons;
that is, DME was also considered as a reactant.
Nitrogen adsorption/desorption isotherms were mea-
ꢀ
sured at −195ꢂ8 C on a BELSORP-max gas adsorption
analyzer. Prior to the measurement, theꢀ zeolite sample
was degassed under high vacuum at 300 C for 8 h. The
surface area was calculated from the adsorption branch
in the range of relative pressure from 0.05 to 0.25 by
Brunauer–Emmett–Teller (BET) method and the pore size
distribution was derived from the adsorption branch by
Barrett–Joyner–Halenda (BJH) method. Pore volume was
estimated at a relative pressure of 0.99.
Temperature programmed desorption of NH3 (NH3-
TPD) was performed on a chemisorption analyzer
(Micromeritics AutoChem II 2920). Approximately
100 mg of the zeolite sample was used in each measure-
ment, which was first pretreated at 550 ꢀC for 2 h in
an argon stream (30 mL min−1) and then cooled down
to 120 ꢀC. Saturated adsorption of NH3 on the zeolite
sample was then achieved by introducing gaseous NH3
(5 vol% in argon, 30 mL min−1ꢄ into the sample tube
for 30 min. After that, the physically adsorbed NH3 was
removed by flushing the sample tube with an argon flow
2.4. Analysis of Occluded Organic Compounds in the
Zeolite Catalyst
The constituents of the occluded hydrocarbons in the zeo-
lite catalyst after the MTO reaction were analyzed by
gas chromatography–mass spectrometry (GC–MS). Typi-
cally, 20 mg of catalyst sample after reaction was carefully
dissolved in a screw-cap Teflon vial by using 10% HF
(1 mL). Dichloromethane (1 mL) with hexachloroethane
as the internal standard was used to extract the organic
phase,17,24 which was analyzed by a GC (SHIMADZU
2010 Plus) connected to an ultra mass-selective detector
(SHIMADZU QP2010) equipped with an RTx-5MS col-
umn (60 m×0.25 mm×0.25 ꢆm). The oven temperature
was programmed at 50 ꢀC for 10 min, from 50 and 250 ꢀC
ꢀ
(30 mL min−1ꢄ at 120 C for 2 h. To get the NH3-TPD
profile,ꢀthe zeolite sample was then heated up from 120
ꢀ
to 550 C at a ramp of 10 C min−1; the amount of NH3
released during the heating for desorption was measured
by a thermal conductivity detector (TCD).
ꢀ
ꢀ
with a heating rate of 3 C min−1, and then at 250 C for
3682
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017

Jing et al.
Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
a
b
c
(a) ZSM-5
(b) ZSM-11
d
10
20
30
40
2θ
Figure 2. XRD patterns of (a) ZSM-5, (b) ZSM-11, (c) ZSM-22 and
(d) EU-1 zeolites.
(c) ZSM-22
(d) EU-1
3.2. Characterization of the Synthesized Zeolites
Figure 1. Frameworks of four zeolites used in this work: (a) ZSM-5,
Figure 2 shows the powder XRD patterns of four zeolites
synthesized in this work. Typical diffraction lines corre-
sponding to the MFI, MEL, TON and EUO framework
structures are observed for ZSM-5, ZSM-11, ZSM-22 and
EU-1, respectively,20,22,25,26 indicating that they are in pure
phases with high crystallinity. The crystal morphologies
of the as-synthesized zeolite samples were also charac-
terized by SEM, as shown in Figure 3. ZSM-5 takes the
spherical form with a rugged surface and has a crystal
size of 120 nm. ZSM-11 consists of irregular crystals and
agglomerates, with a particle size of 80–100 nm. ZSM-
(b) ZSM-11, (c) ZSM-22 and (d) EU-1.
20 min. The compounds were identified according to the
NIST98 database.
3. RESULTS AND DISCUSSION
3.1. About the Topology of Four Zeolites
The topology characteristics of four 10-ring zeolites, viz.,
ZSM-5, ZSM-11, ZSM-22 and EU-1, are illustrated in
Figure 1 and summarized Table I. It is well known that
Delivered by Ingenta to: Freie Universitaet Berlin
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
ZSM-5 with MFI topology has channels with dimensions
22 is composed of log-shaped crystals with a crystal size
of 100–200 nm. EU-1 is present as intense agglomerate
with a crystal size of 100–200 nm. Hence, four zeolites
are different in their morphologies but with similar crys-
tal sizes. As given in Table II, ZSM-5 and EU-1 possess
surface areas of 518 and 451 m2 g−1, respectively, much
higher than the value of 179 m2 g−1 for ZSM-22 with
one-dimensional pore; their surface areas are in the same
range as those previously reported.21,24,27 On the other
Copyright: American Scientific Publishers
of 5.1 × 5.5 Å along the (100) direction and 5.3 × 5.6 Å
along the (010) direction; three-dimensional (3D) channel
intersections are created between the straight and sinu-
soidal channels and all intersections are equivalent. ZSM-
11 with MEL topology has channels with a dimension of
5.3 × 5.4 Å in two directions; 3D channel intersections
are created at the interface between two perpendicular but
slightly shifted straight channels, forming two types of
cavities: one has the same size as that in ZSM-5, whereas
the other is 30% greater than that in ZSM-5. On the other
hand, ZSM-22 with TON topology has slightly elliptical
channels with a dimension of 5.7 × 4.6 Å. Meanwhile,
EU-1 with EUO topology has channels with a dimension
of 5.4×4.1 Å and 12-ring side pockets that are 6.8×5.8 Å
in width and 8.1 Å in depth.
a
b
200 nm
200 nm
c
d
Table I. Topological structure characteristics of four zeolites with
10-ring channel used in this work.
Pore
Pore
Side
Zeolite Topology size (Å)
shape Dimensionality pocket (Å)
ZSM-5
MFI
5.1×5.5; Zigzag
5.3×5.6
3
None
200 nm
200 nm
ZSM-11 MEL
ZSM-22 TON
EU-1
5.3×5.4 Elliptical
5.7×4.6 Elliptical
5.4×4.1 Zigzag
3
1
3
None
None
6.8×5.8×8.1
Figure 3. SEM images of (a) ZSM-5, (b) ZSM-11, (c) ZSM-22 and
(d) EU-1 zeolites.
EUO
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017
3683

Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
Jing et al.
Table II. Textural properties of the zeolites synthesized in this work.
Si/Al
molar
ratio
Surface
area
External
surface
Micropore
volume
Crystal
size
(nm)
a
Zeolite
(m2 g−1
ꢄ
(m2 g−1
ꢄ
(cm3 g−1
ꢄ
b
ZSM-5
ZSM-11
ZSM-22
EU-1
30
22
26
26
518
320
179
451
160
36
79
0.156
0.148
0.037
0.175
120
80–100
100–200
100–200
c
38
d
hand, ZSM-11 exhibits a surface area of 320 m2 g−1, some
lower than the value of 400 m2 g−1 obtained by Bleken and
co-workers due to the difference in synthesis procedure.20
Figure 4 displays the 27Al MAS NMR spectra of the
calcined H-ZSM-5, H-ZSM-11, H-ZSM-22, and H-EU-1
zeolites. All zeolites show a relatively sharp resonance
at around 53 ppm, which was assigned to tetrahedral
aluminum species in the zeolite framework, whereas the
peaks at around 0 ppm representing the octahedral alu-
minum, i.e., the extra-framework aluminum species, are
relatively much weaker.16
100
200
300
400
500
Temperature (ºC)
Figure 5. NH3-TPD profiles of (a) ZSM-5, (b) ZSM-11, (c) ZSM-22
and (d) EU-1 zeolites.
Moreover, the adsorption sites as well as the structure of
alkene adsorption complex and its reactivity were related
to the pore size and structure of a zeolite catalyst.29 To
evaluate the interaction of various MTO products with the
zeolite catalysts, the adsorption capacities of four zeolites
towards ethene, propene, and butene were determined at
The NH3-TPD profiles of four zeolite catalysts are
shown in Figure 5; the shape of desorption profile reflects
the distribution of acid sites across strength.21 The amounts
of acid sites assorted by their strength are summarized
in Table III; the quantities of weak and strong acid sites
are the amounts of ammonia desorbed during NH3-TPD at
ꢀ
25 C, as shown in Figure 6(a–c).
For the alkene adsorbates, typical type-I adsorption
isotherms are observed on four zeolites; the adsorption
reaches equilibrium quickly at low pressure due to the
micropore filling, followed by a plateau at relatively higher
pressure.30 The amount of ethene adsorbed on four zeolites
Delivered by Ingenta to: Freie Universitaet Berlin
ꢀ
120–240 and 240–550 C, respectively. Figure 5 suggests
is in general lower than that of propene and butene, espe-
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
that ZSM-5, ZSM-22 and EU-1 exhibit similar density
for the strong acidic sites, whereas ZSM-11 has relatively
more strong acid sites.
Copyright: American Scientific Publishers
cially on ZSM-5 and ZSM-11, probably due to that ethene
molecule is much smaller and has no polarity. The adsorp-
tion capacity towards ethene on four zeolites follows the
order of ZSM-5 > EU-1 > ZSM-11 ≈ ZSM-22. However,
this order of adsorption capacity towards propene on four
zeolites turns to be ZSM-5 ≈ ZSM-11 > EU-1 > ZSM-
22, whereas adsorption capacity towards linear butene on
ZSM-11 is significantly higher than that on ZSM-5. In
comparison with the sinusoidal channel of ZSM-5, close
arrays of butene molecules may be more easily formed in
the straight channel of ZSM-11.
3.3. Adsorption Capacity of Four Zeolites Towards
Ethene, Propene, Butene, and Methanol
Alkene molecules entering the micropores of zeolites were
reported to be adsorbed onto OH groups (Si–OH–Al)
by forming a hydrogen bond via the ꢅ-electron of the
C
C bond (ꢅ-adsorption);28 however, diffusion may be
the rate-determining step due to the size-limited migration
of molecules on the internal walls of the micropores.29
As the adsorption of methanol on catalysts is the ini-
tial step in the MTO process, the adsorption behavior of
methanol on four zeolites was then considered, as shown
in Figure 6(d). Acidity and surface area are two key factors
that influence the adsorption of methanol on the zeolites.31
Figure 6(d) illustrate that methanol molecules can read-
ily enter the channels of four zeolites including ZSM-22.
d
c
Table III. NH3-TPD results of four zeolites prepared in this work.
b
a
Acid density (mmol g−1
Strong
ꢄ
Zeolite
Weak
Total
100
50
0
–50
ZSM-5
ZSM-11
ZSM-22
EU-1
0.134
0.209
0.245
0.068
0.382
0.536
0.418
0.416
0.516
0.745
0.663
0.484
δ (ppm)
Figure 4. 27Al MAS NMR spectra of (a) ZSM-5, (b) ZSM-11,
(c) ZSM-22 and (d) EU-1 zeolites.
3684
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017

Jing et al.
Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
a
b
a
b
c
d
100
80
60
40
20
120
100
80
60
40
20
0
ZSM-5
ZSM-11
ZSM-22
EU-1
C1-3
C5+
C4=
C4
0
100
c
d
C3=
C2=
80
60
40
20
0
0
40 80
0
40 80
0
40 80
0
5
10 15
p (kPa)
Figure 6. Adsorption isotherms of (a) ethene, (b) propene, (c) butene
and (d) methanol on four zeolites.
5
10 30 60 120
5
10 30 60 120
The adsorption capacity towards methanol on four zeo-
lites follows the sequence of ZSM-5 > ZSM-11 > EU-1 >
ZSM-22, which accords approximately with the order of
their surface area.
Time on stream (min)
Figure 8. Product selectivity for MTO at 300 ꢀC over (a) ZSM-5,
(b) ZSM-11, (c) ZSM-22 and (d) EU-1 zeolites.
The product selectivity in MTO is also greatly related
to the framework structure of zeolite catalysts, as shown
in Figure 8. Propene is the dominant product over all zeo-
lites, especially over ZSM-5. With ZSM-5 as the catalyst,
the selectivity to C2–C4 olefins is rather stable and remains
3.4. Catalytic Performance of Four Zeolites in MTO,
Continuous Flow Tests
Figure 7 shows the conversion of methanol over four zeo-
lite catalysts along with the time on stream. In general,
ZSM-5 and ZSM-11 exhibit much higher methanol con-
version and longer lifetime than ZSM-22 and EU-1. Espe-
Delivered by Ingenta to: Freie Universitaet Berlin
approximately 85% up to 120 min on stream; the selec-
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
tivity to propene over ZSM-5 is about 65%, the highest
cially, the conversion of methanol over ZSM-5 is relatively
stable with the time on stream and remains over 90% after
600 min on stream. ZSM-11 is deactivated slowly with the
time on steam and the conversion of methanol over ZSM-
11 is decreased to 65% after 600 min on stream. On the
contrary, the initial conversions of methanol over ZSM-
22 and EU-1 are only about 41% and 67%, respectively;
moreover, these two catalysts are deactivated rapidly and
the conversions of methanol are decreased almost to zero
in 120 min.
Copyright: American Scientific Publishers
obtained among four zeolites.
In contrast, the selectivity to C2–C4 olefins over ZSM-11
is about 75%, whereas the selectivity to butane over ZSM-
11 is much higher than that over other zeolites; such a phe-
nomenon is consistent with those observed by Bleken and
co-workers,20 which may be ascribed to the special chan-
nel structure and adsorption behavior of ZSM-11. As men-
tioned above, the adsorption capacity of ZSM-11 towards
butene is significantly higher that of ZSM-5, which may
give more chance to convert butene to butane through the
hydride transfer reaction. Correspondingly, the amount of
C5+ hydrocarbon in the products on ZSM-11 is also sig-
nificantly higher than that on ZSM-5, as the formation of
aromatics is promoted by the hydrogen transfer reactions.
A gradual increase in the selectivity to ethene is observed
over ZSM-11 with the time on stream from 5 to 120 min;
it is probably ascribed to the decrease of the free space in
the cages due to the gradual deposition of cokes, which
may suppress the formation of larger molecules, as sug-
gested by Chen and co-workers.32 As a result, the increase
of selectivity to ethene with the time on stream is in gen-
eral accompanied with the coke deposition and decrease
of methanol conversion.
100
a
80
b
60
40
ZSM-5
d
ZSM-11
20
0
ZSM-22
EU-1
c
0
120
240
360
480
600
Time on stream (min)
ZSM-22 and EU-1 exhibit poor stability in MTO, as
their low dimensional channels are subject to a rapid deac-
tivation, in accordance with those previously reported.2,33
The initial selectivity to C2–C4 olefins over ZSM-22 is
Figure 7. Methanol conversions for MTO over (a) ZSM-5, (b) ZSM-
11, (c) ZSM-22 and (d) EU-1 zeolites. Reaction conditions: atmospheric
pressure, 300 ꢀC, and methanol WHSV of 5 h−1
.
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017
3685

Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
Jing et al.
a
about 72%, but decreases significantly to about 40% in
120 min; especially, the selectivity to propene decreases
from 58% to 23%, whereas the selectivity to alka-
nes increases remarkably from 21% to 55% during the
same period, indicating that the hydride transfer reactions
become increasingly important. Meanwhile, the products
for MTO over ZSM-22 are almost free of C5+ fraction, as
previously reported by Teketel and co-workers.21
For MTO over EU-1, the initial selectivity to C2–C4
olefins is about 75% and decreases gradually to 60% in
120 min. Unexpectedly, some C5+ products appear after
30 min on stream. As EU-1 has a narrower channel system
than ZSM-22, the C5+ products observed should be formed
on the outer crystal surface acid sites rather than in the
10-ring pore of EU-1.21
b
100
80
60
40
20
C1-3
C5+
C4=
C4
0
100
c
d
C3=
C2=
80
60
40
20
0
1
2
3
4
5
1
2
3
4
5
3.5. Catalytic Results from Pulse Reaction Tests
To get a deep understanding on the catalytic behavior of
various zeolites in MTO, especially for ZSM-22 and EU-1
with very short lifetime, MTO reaction over four zeolites
Pulse number
ꢀ
Figure 10. Product selectivity for MTO in pulse reactor at 300 C over
(a) ZSM-5, (b) ZSM-11, (c) ZSM-22 and (d) EU-1 zeolites.
ꢀ
were tested in the pulse reaction system at 300 C with a
contact time of 36 ms, as shown in Figures 9 and 10.
Over ZSM-5, the conversion of methanol is only
0.7% for the first pulse of methanol injection; after five
injections, the conversion of methanol reaches 20.3%,
indicating the presence of an induction period. Notably,
its high density of strong acid sites, as given in Table III.
For all five pulses over ZSM-11, propene appears as the
main product, though the selectivity to propene is only
about 45%, much lower than that over ZSM-5. Meanwhile,
the selectivity to butane over ZSM-11 is remarkably higher
than that over ZSM-5, as the hydrogen transfer reaction
can be promoted by the high acid density, consistent with
the results obtained in the continuous flow reactor.
a variation is also observed in the product distribution
Delivered by Ingenta to: Freie Universitaet Berlin
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
with the consecutive methanol injections. For the first and
Copyright: American Scientific Publishers
second injections, methane and ethene are main products,
meaning that the hydrocarbon pool species are accumulat-
ing on the catalyst in the induction period;34 for the later
pulses, propene turns out to be the main product, as the
HCP species have been built in the channels.
Over ZSM-22, the conversion of methanol is only 0.1%
for the first injection and reaches 3.7% for the fifth injec-
tion, indicating the existence of an induction period for
MTO over ZSM-22. For the first injection, the selectivities
to ethene and methane are 56.3% and 43.7%, respectively,
indicating the accumulation of HCP species, similar to
that over ZSM-5; with further injections of methanol, the
selectivity to propene, butene, butane and C5+ is increased,
whereas the selectivity to ethene is decreased. However,
the conversion of methanol over ZSM-22 in the pulse reac-
tor is very low. Teketel and co-workers found that poly-
methylbenzenes may play a less important role in MTO
over ZSM-22 and the net methanol conversion proceeded
independently through the methylation and cracking of
alkenes.3 Because the olefins species are easily flushed out
of catalyst bed before the next injection in pulse reac-
tor, the olefins cycle is weakened and the conversion of
methanol is still very low for the fifth injection.
Over EU-1, the conversion of methanol is 8.6% for the
first methanol injection and increased to 17.3% for the fifth
injections. Among four zeolites, EU-1 gives the highest
selectivity to propene (about 68%), with a low selectiv-
ity to ethene (about 4%). Furthermore, EU-1 displays a
lower selectivity to C5+ products than ZSM-22. As 10-ring
EU-1 has the narrowest channels, the formation of C5+ is
then suppressed due to the diffusion limitations.21 Similar
Over ZSM-11, the conversion of methanol is 8.0% for
the first methanol injection and is increased gradually to
21% for the fifth pulse. The high initial methanol conver-
sion in the pulse reaction over ZSM-11 may be ascribed to
25
ZSM-5
ZSM-11
20
ZSM-22
EU-1
b
15
d
10
a
5
0
c
1
2
3
4
5
Pulse number
Figure 9. Methanol conversion for MTO in a pulse reactor obtained
with different pulses over (a) ZSM-5, (b) ZSM-11, (c) ZSM-22 and
(d) EU-1 zeolites. Reaction conditions: atmospheric pressure, 300 ꢀC,
Ar carrier gas (336 mL min−1ꢄ; l ꢆL of methanol was injected for each
pulse.
3686
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017

Jing et al.
Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
phenomena were also observed by Li and co-workers for
MTO over SAPO-34,19 where the selectivities to propene
and ethene over SAPO-34 for the first injection in the
induction period were 60.8% and 6.6%, respectively. As
the 10-ring channel of EU-1 is wider than the 8-ring pore
opening of SAPO-34 and 12-ring side pocket in EU-1 is
similar to the cages in SAPO-34,35 the selectivity to the
C4 product over EU-1 is higher than that over SAPO-34.19
gives more polymethylnaphthalenes (polyMNs). These
polyMNs probably locate at the intersection of ZSM-11
pores. Nevertheless, this is a good explanation for the
quick deactivation of ZSM-11 in MTO, in comparison with
ZSM-5. It should be noted that the difference of ZSM-11
and ZSM-5 in their stability for MTO may also be related
to the microstructure and interior orientation and distri-
bution of acid sites, which are dependent on the actual
synthetic procedures.37,38
3.6. Occluded Organic Species Formed on the Zeolite
Catalysts in MTO
In ZSM-22, the amount of occluded species is the small-
est among four zeolites, though both MBs and MNs are
observed; meanwhile, it is also deactivated most rapidly
(Fig. 7). Li and co-workers suggested that polymethyl-
benzenes and naphthalenes were responsible for the quick
deactivation of ZSM-22 by blocking the zeolite channel
openings.39 Current work shows that MTO over ZSM-22
proceeds mainly via the olefin-based cycle and the conver-
sion of methanol is very low in the pulse reaction tests. On
the other hand, the obvious induction period in the pulse
reaction tests demonstrates that the aromatic-based route
is also active, though less important. The diffusion limi-
tation of aromatic species in the 1D channel of ZSM-22
makes the hydrocarbon pool species easily transform into
coke precursor, leading to the quick deactivation.
Among the four zeolites, EU-1 yields the largest
amounts of occluded species, including a large number of
pentaMBs, hexaMBs and polyMNs. At the initial stage of
the reaction, certain large hydrocarbon pool species could
be formed and cracked in the 12-ring side pockets, leading
to the high selectivity to propene; however, they accumu-
late rapidly and convert to polyaromatics, which are dif-
ficult to diffuse out of the channel and thereby cause the
deactivation.
The components of the organic species trapped in the spent
catalysts for the continuous-flow MTO tests are shown in
Figure 11. Methylbenzenes (MBs) and methylnaphthalenes
(MNs) are observed in all zeolites, but they are differ-
ent in the specific distribution of aromatics. Over ZSM-5,
MBs are the main species, ranging from minor amounts
of xylenes, triMBs to larger amounts of tetra-, penta-,
and hexa-MBs, suggesting that MBs are important hydro-
carbon pool species in MTO. Some researchers claimed
that ZSM-5 could not provide sufficient spaces for the
formation of coke species greater than tetraMBs, but it
conflicts with our results.17,36 As MTO tests were con-
ꢀ
ducted in this work at a low temperature (300 C), we
may consider that methyl substituted benzene with methyl
number higher than four is more likely to decompose at
higher temperature but does exist at 300 ꢀC. Early research
Delivered by Ingenta to: Freie Universitaet Berlin
indicated that higher polymethylbenzenes (polyMBs) were
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
Copyright: American Scientific Publishers
more conducive to the formation of propene, whereas
lower polyMBs tended to produce ethene.11 In this work,
the selectivity to propene over ZSM-5 is as high as 65%,
which may support the existence of higher polyMBs in
ZSM-5.
The distribution of methyl substituted aromatics in
ZSM-11 is similar to that in ZSM-5, except that ZSM-11
4. CONCLUSIONS
MTO reactions were comparatively conducted over four
10-ring zeolites, viz., ZSM-5, ZSM-11, ZSM-22 and EU-1,
in both continuous-flow and pulse reactors. The results
illustrate that the performance of a zeolite catalyst in MTO
is strongly dependent on its framework structure.
Among four zeolites, ZSM-5 with three dimensional
channels and intersections exhibits the highest catalytic
activity and stability in MTO; the conversion of methanol
remains over 90% after 600 min on stream.
CH3
CH3
5
6
CH3
CH3
CH3
4
5
4
CH3
3
CH3
3
d
c
CH3
C2Cl6
2
The straight channels of ZSM-11 facilitate the orderly
array of butene molecules in the channels and give ZSM-
11 high adsorption capacity towards butene, which may
promote the hydrogen transfer reaction and lead to high
selectivity to butane and C5+ products and a fast deactiva-
tion, in comparison with ZSM-5.
b
a
10
20
30
40
50
60
70
80
90
Retention time (min)
The pulse reaction tests illustrate that the hydrocarbon
pool mechanism is also applicable in ZSM-22, though less
important; the 1D straight channels in ZSM-22 are easily
blocked, which lead to the quick deactivation.
Figure 11. GC-MS analysis of the retained organic species in (a) ZSM-
5, (b) ZSM-11, (c) ZSM-22 and (d) EU-1 zeolites after MTO for 120 min
at 300 ꢀC in continuous flow reactor. C2Cl6 is used as an internal
standard.
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017
3687

Comparative Study of MTO Over ZSM-5, ZSM-11, ZSM-22 and EU-1
Jing et al.
EU-1 zeolite exhibits quite high selectivity to propene
due to its large 12-ring side-pocket that can benefit the
formation of polyalkybenzenes served as the hydrocarbon
pool species; meanwhile, the small 10-ring pore open-
ing can restrict the diffusion of long chain olefins and
further decrease the production of other undesired prod-
ucts. The insights shown in this work are conducive to a
deeper understanding on the dependence of catalytic per-
formance in MTO on the zeolite framework structure as
well as the development of better MTO catalysts and reac-
tion processes.
16. Q. Wang, Z. M. Cui, C. Y. Cao, and W. G. Song, J. Phys. Chem. C
115, 24987 (2011).
17. B. P. C. Hereijgers, F. Bleken, M. H. Nilsen, S. Svelle, K. P. Lillerud,
M. Bjørgen, B. M. Weckhuysen, and U. Olsbye, J. Catal. 264, 77
(2009).
18. W. Dai, X. Wang, G. Wu, N. Guan, M. Hunger, and L. Li, ACS
Catal. 1, 292 (2011).
19. J. Li, Y. Wei, G. Liu, Y. Qi, P. Tian, B. Li, Y. He, and Z. Liu, Catal.
Today 171, 221 (2011).
20. F. Bleken, W. Skistad, K. Barbera, M. Kustova, S. Bordiga, P. Beato,
K. P. Lillerud, S. Svelle, and U. Olsbye, Phys. Chem. Chem. Phys.
13, 2539 (2011).
21. S. Teketel, W. Skistad, S. Benard, U. Olsbye, K. P. Lillerud, P. Beato,
and S. Svelle, ACS Catal. 2, 26 (2012).
22. J. Shin and S. B. Hong, Microporous Mesoporous Mater. 124, 227
(2009).
23. S. H. Lee, C. H. Shin, D. K. Yang, S. D. Ahn, I. S. Nam, and S. B.
Hong, Microporous Mesoporous Mater. 68, 97 (2004).
24. W. Song, R. E. Justice, C. A. Jones, V. H. Grassian, and S. C. Larsen,
Langmuir 20, 8301 (2004).
25. D. Verboekend, A. M. Chabaneix, K. Thomas, J. P. Gilson, and
J. Pérez-Ramírez, CrystEngComm. 13, 3408 (2011).
26. L. Zhang, H. Liu, X. Li, S. Xie, Y. Wang, W. Xin, S. Liu, and L. Xu,
Fuel. Process. Technol. 91, 449 (2010).
27. J. Li, Y. Qi, Z. Liu, G. Liu, and D. Zhang, Catal. Lett. 121, 303
(2007).
28. M. F. Zalazar,J. R. D. Duarte, and N. M. Peruchena, J. Phys.
Chem. A 113, 13797 (2009).
29. E. Yoda, J. N. Kondo, and K. Domen, J. Phys. Chem. B 109, 1464
(2005).
30. P. A. Russo, M. M. L. R. Carrott, P. J. M. Carrott, P. Matias, J. M.
Lopes, M. Guisnet, and F. R. Ribeiro, Microporous Mesoporous
Acknowledgments: The authors are grateful for the
financial supports by the National Natural Science Founda-
tion of China (21573270, 21273263, 21273264, 21227002,
U1510104), and Natural Science Foundation of Shanxi
Province of China (2015021003, 2013021007-3).
References and Notes
1. H. Yamazaki, H. Shima, H. Imai, T. Yokoi, T. Tatsumi, and J. N.
Kondo, J. Phys. Chem. C 116, 24091 (2012).
2. J. W. Park, J. Y. Lee, K. S. Kim, S. B. Hong, and G. Seo, Appl.
Catal. A 339, 36 (2008).
3. S. Teketel, U. Olsbye, K. P. Lillerud, P. Beato, and S. Svelle, Micro-
porous Mesoporous Mater. 136, 33 (2010).
4. S. Ilias and A. Bhan, ACS Catal. 3, 18 (2013).
5. M. Stocker, Microporous Mesoporous Mater. 29, 3 (1999).
6. C. D. Chang and A. J. Silvestri, J. Catal. 47, 249 (1977).
7. N. Y. Chen and W. J. Reagan, J. Catal. 59, 123 (1979).
8. M. D. Ivar and S. Kolboe, Catal. Lett. 20, 329 (1993).
Mater. 118, 473 (2009).
Delivered by Ingenta to: Freie Universitaet Berlin
31. D. Chen, H. P. Rebo, K. Moljord, and A. Holmen, Ind. Eng. Chem.
IP: 185.101.71.30 On: Fri, 03 Mar 2017 12:07:08
Res. 38, 4241 (1999).
Copyright: American Scientific Publishers
9. A. Bjornar and S. Kolboe, J. Am. Chem. Soc. 123, 8137 (2001).
32. D. Chen, A. Gronvold, K. Moljord, and A. Holmen, Ind. Eng. Chem.
Res. 46, 4116 (2007).
10. A. Bjornar and S. Kolboe, Catal. Lett. 71, 209 (2001).
11. S. Svelle, U. Olsbye, F. Joensen, and M. Bjorgen, J. Phys. Chem. C
111, 17981 (2007).
12. U. Olsbye, S. Svelle, M. Bjorgen, P. Beato, T. V. Janssens,
F. Joensen, S. Bordiga, and K. P. Lillerud, Angewandte Chemie.
51, 5810 (2012).
33. U. Olsbye, S. Svelle, K. P. Lillerud, Z. H. Wei, Y. Y. Chen, J. F. Li,
J. G. Wang, and W. B. Fan, Chem. Soc. Rev. 44, 7155 (2015).
34. L. Qi, Y. Wei, L. Xu, and Z. Liu, ACS Catal. 5, 3973 (2015).
35. H. Fu, W. G. Song, and J. F. Haw, Catal. Lett. 76, 89 (2001).
36. M. Guisnet and P. Magnoux, Appl. Catal. A 212, 83 (2001).
37. J. Deˇdecˇek, Z. Sobalík, and B. Wichterlová, Catal. Rev. 54, 135
(2012).
13. Z. Cui, Q. Liu, Z. Ma, S. Bian, and W. G. Song, J. Catal. 258, 83
(2008).
14. Q. Zhu, J. N. Kondo, R. Ohnuma, Y. Kubota, M. Yamaguchi, and
T. Tatsumi, Microporous Mesoporous Mater. 112, 153 (2008).
15. M. D. Ivar, M. Helle, A. Duncan, and W. Rune, Microporous Meso-
porous Mater. 29, 185 (1999).
38. Z. Sobalik, P. Sazama, J. Dedecek, and B. Wichterlová, Appl.
Catal. A 474, 178 (2014).
39. J. Li, Y. Wei, Y. Qi, P. Tian, B. Li, Y. He, F. Chang, X. Sun, and
Z. Liu, Catal. Today 164, 288 (2011).
Received: 24 January 2016. Accepted: 13 March 2016.
3688
J. Nanosci. Nanotechnol. 17, 3680–3688, 2017