Effect of Desilication on the Performance of Hierarchical ZSM-11 Catalysts for Alkylation of Benzene with Dimethyl Ether

Article abstract of DOI:10.1007/s10562-015-1589-1

Hierarchical ZSM-11 zeolites (Z-xat) were prepared by treating the pristine ZSM-11 zeolite in a solution of NaOH with different concentrations (C _NaOH). It was found the porosity and acidity (concentration, strength, distribution and accessibil

Full text of DOI:10.1007/s10562-015-1589-1

Catal Lett
DOI 10.1007/s10562-015-1589-1
Effect of Desilication on the Performance of Hierarchical ZSM-11
Catalysts for Alkylation of Benzene with Dimethyl Ether
Hui Liu1,2 Shenglin Liu1 Sujuan Xie1 Chao Song1,2 Wenjie Xin1
Longya Xu1
•
•
•
•
•
Received: 9 April 2015 / Accepted: 12 July 2015
Ó Springer Science+Business Media New York 2015
Abstract Hierarchical ZSM-11 zeolites (Z-xat) were
prepared by treating the pristine ZSM-11 zeolite in a
solution of NaOH with different concentrations (CNaOH). It
was found the porosity and acidity (concentration, strength,
distribution and accessibility factor of Brønsted and Lewis
acid sites) of modified ZSM-11 zeolite changed obviously
with CNaOH. When used for alkylation of benzene with
dimethyl ether, the modified Z-xat exhibited better reaction
stability than the unmodified HZSM-11 zeolite. Typically,
the reaction stability of Z-xat zeolite catalysts was directly
related to the mesopore volume, besides the weak-medium
acid concentrations determined by Py-IR technique.
Graphical Abstract
Keywords Hierarchical ZSM-11 Á Alkaline treatment Á
Brønsted and Lewis acid sites Á Accessibility factor Á
Alkylation of benzene with DME
1 Introduction
ZSM-11 zeolite (MEL topology) possesses two straight
intersecting channels, which is similar with the ZSM-5
(MFI topology) consisting of intersectional straight and
sinusoidal channels [1–3]. In spite of the similarity, few
studies on ZSM-11 have been reported, compared with the
widely studied ZSM-5 zeolite [4].
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-015-1589-1) contains supplementary
material, which is available to authorized users.
& Longya Xu
lyxu@dicp.ac.cn; liuhui@dicp.ac.cn
As far as the porosity of zeolite, the sole presence
micropores may inevitably impose diffusion limitations or
confinement effects [5] on the shape-selective reactions
and adsorption rates. Enormous alternative strategies have
been attempted to circumvent this problem [6, 7], and
1
State Key Laboratory of Catalysis, Dalian Institute of
Chemical Physics, Chinese Academy of Sciences,
Dalian 116023, Liaoning, China
2
University of Chinese Academy of Sciences, Beijing 100049,
China
123

H. Liu et al.
desilication in alkaline aqueous solution has become the
most promising method to introduce hierarchical micro-
mesoporous system [8–11], due to the cheap and facile
operation and efficient mesopores production.
ammonia temperature-programmed desorption as well as
pyridine-adsorption IR and Pn-IR measurements. The acid
sites located on external surfaces and in pore mouths
monitored by Pn-IR technique and the corresponding
accessibility factor were first reported for the modified
ZSM-11 zeolites. The obtained hierarchical ZSM-11 zeo-
lites were tested in the alkylation of benzene with DME.
The results suggested that the alkaline treatment was
helpful for improving the reaction performance of the
desilicated ZSM-11 catalyst zeolite. Besides, the difference
in the reaction stability was relevant to the porosity and
acidity of the as prepared hierarchical ZSM-11 zeolites.
Except for the porosity, acidity is another significant
feature for zeolite, due to its influence on catalytic and
reaction performance on catalysts [12, 13]. Zeolites contain
many different kinds of acid sites in terms of their types
and strengths, as well as their locations in the zeolite pore
system. In addition to conventionally used NH3-TPD and
pyridine-adsorption IR techniques, recently the acid sites
located on the external surface have been investigated with
the IR technique of lutidine and collidine [14–16] and 2,6-
di-tert-butylpyridine [16–18] adsorption, however, these
studies were only limited to the Brønsted acid sites. And
only non-quantitative accessibility of Lewis acid sites on
external surface was obtained by indirect method [19].
Fortunately, the IR technique of pivalonitrile adsorption
(Pn-IR) has been developed and used to characterize both
Brønsted and Lewis acid sites located on external surfaces
and in pore mouths [20–22]. However, such a kind of IR
technique was only reported for HZSM-5 zeolites, and no
any work for other kinds of zeolites has been reported till
now.
2 Experimental
2.1 Synthesis of ZSM-11 Zeolite and Treated
with Pure NaOH Solution
ZSM-11 of Si/Al2 of 53 (particle size of 300–400 nm) was
synthesized in laboratory based on our previous work [1].
Prior to the treatment, the as-synthesized zeolite was cal-
cined in static air at 540 °C for 4 h in order to remove the
template. The calcined sample was desilicated by NaOH
solution as follows. ZSM-11 zeolite powder of 8 g was
treated in aqueous NaOH solutions with the concentration
ranging from 0.1 to 1.0 M of 280 mL under stirring at
75 °C for 15 min. The suspensions were then quenched in
cold water and filtered, and the isolated solids were washed
with deionized water until it had a neutral PH. Then, all the
samples were converted to the protonic form by conven-
tional ion exchange in aqueous NH4NO3 solution (0.8 M,
80 °C, 1 h, 20 cm3 per gram of zeolite for 3 times), fol-
lowed by calcination (air condition, 510 °C, 3 h). The
name of the resultant samples was denoted as Z-xat, where
‘‘x’’ related to the concentration of NaOH solution applied.
Accordingly, the unmodified sample was denoted as
HZSM-11.
ZSM-11 zeolite attracts more and more attention in
many fields recently [1–3]. Nevertheless, little information
is available regarding the effect of ZSM-11 by alkaline
treatment (NaOH). Moreover, how the NaOH concentra-
tion (denoted as CNaOH) exactly affects the mesopores
production and acidic properties for ZSM-11 zeolite are
reported rarely up to now. Despite plenty of acidity and
porosity investigation for different hierarchical zeolites
[16, 23, 24], there is little systematic report about the types,
strength, distribution (B/L ratio) and accessibility factor
(AF) of different acid sites variation with the concentration
of alkaline.
Recently, the alkylation of benzene has been received
much more interest since the aromatic hydrocarbon prod-
ucts are highly needed by the market around the world. In
spite of this situation, alkylation of benzene with dimethyl
2.2 Catalyst Characterization
´
ether (DME) has been reported only by Szechenyi et al.
[25]. Recently, we investigated the alkylation of benzene
with DME over Zn loaded [26] and AlPO bounded [3]
ZSM-11 zeolite catalysts, in which we explored the effect
of Zn and P modification on the zeolite catalytic perfor-
mance. As an extension, this work systematically studied
the pristine ZSM-11 zeolite materials treated with NaOH
solution with different concentrations.
X-ray diffraction (XRD) patterns were recorded on an X’
Pert PRO X-ray diffractometer using Cu Ka radiation
(40 kV, 40 mA), with a scanning range of 2h from 5° to
50° at a scanning rate of 10o min-1. The relative crys-
tallinity is calculated based on the relative intensities of the
reflection at 2h = 7.9 0.1o, 8.8 0.1o, 23.0 0.1o,
23.8 0.1o and 45.0 0.1o, taking the crystallinity of the
pristine HZSM-11 zeolite to be 100 %.
The changes of structural properties and acidities for the
as modified materials were characterized by many tech-
niques, such as X-ray diffraction, X-ray fluorescence, N2
and Benzene adsorption–desorption isotherms measure-
ments, scanning and transmission electron microscopy,
X-ray fluoerescence (XRF) experiments were conducted
on a Philips Magix 601X X-ray fluorescence spectrometer,
and IQ? quantitative software package was employed for
the analysis.
123

Effect of Desilication on the Performance of Hierarchical ZSM-11 Catalysts for Alkylation of…
N2 adsorption–desorption isotherms were performed on
Micromeritics ASAP-2020 system at liquid nitrogen tem-
perature of -196 °C. Prior to analysis, each sample was
degassed at 350 °C for 10 h.
Lewis (AFL) acid sites. They were initially defined in Ref.
[20] and calculated according to Eqs. (1) and (2):
AFB ¼ CBÀPn=CBÀPyÀ150
AFL ¼ CLÀPn=CLÀPyÀ150
ð1Þ
ð2Þ
Ammonia temperature-programmed desorption (NH3-
TPD) was carried out in a stainless U-shaped micro-reactor
(i.d. = 4 mm). Typically, 0.1400 g sample was loaded in
the reactor and pretreated at 605 °C in flowing He with a
flowrate of 25 mL min-1 for 0.5 h. Then it was cooled
down to 150 °C and saturated with NH3 gas. The reactor
temperature was raised from 150 to 605 °C at a heating rate
of 19.8 °C min-1 and maintained at 605 °C for 0.5 h. The
concentration of NH3 in the exit gas was monitored con-
tinuously by a gas chromatograph equipped with a TCD.
Scanning electron microscopy (SEM) was taken using a
JSM-7800F field-emission microscope at an acceleration
voltage of 30 kV. Transmission electron microscopy
(TEM) images were obtained with a JEM-2100 microscope
at an acceleration voltage of 200 kV.
where CB-Pn and CL-Pn represented the concentrations of
Brønsted and Lewis acid sites determined by Pn-IR tech-
nique respectively, and CB-Py-150 and CL-Py-150 represented
the total concentrations of Brønsted and Lewis acid sites
determined by Py-IR technique respectively. All the values
were shown in Table 3.
The isotherms of benzene adsorption–desorption were
measured on an intelligent gravimetric analyzer (IGA-100)
from Hiden Isochema Ltd. This measurement was applied
to determine the diffusion properties of the prepared sam-
ples, in line with the procedure given in Ref. [26].
2.3 Catalytic Evaluation
Pyridine adsorption-IR (Py-IR) was carried out on a
Vertex 70 IR spectrometer (Bruker Corp.). Self-supporting
wafers (13 mm diameter) were made from ca. 10 mg
zeolite sample and the IR cell can hold the wafer. The
pretreatment of the fresh samples was conducted as fol-
lows. The cell containing a zeolite wafer was evacuated
while slowly increasing the temperature from r.t. to 450 °C
and was finally evacuated at 450 °C for 1 h. It was then
cooled to r.t. to record the background spectra of zeolite.
Infrared spectra were measured at 4 cm-1 resolution. After
that, the sample was saturated with Py and the excess of Py
was removed under vacuum at 150 °C for 0.5 h, after
cooling to r.t. it was followed by IR measurements to
quantify the amount of total acid sites. Then the excess of
Py on the sample was further removed under vacuum at
450 °C for 0.5 h, after cooling to r.t., it was followed by IR
measurements to quantify the amount of strong acid sites.
The hierarchical zeolite samples were tested in the alkylation
of benzene with DME in a stainless fixed bed reactor with
320 mm in length and 12 mm in diameter. All the samples
were pressed, crushed and sieved into 0.38–0.85 mm particles
before they were loaded into the reactor. 1 g catalyst was
located in the middle part of a fixed-bed stainless reactor.
Unless specified, the reaction was performed under the con-
ditions of 450 °C, 0.1 MPa, n(benzene)/n(DME) = 2, and
the DME weight hourly space velocity (WHSV) of 8 h-1
.
Catalyst was activated at 500 °C for 1 h in a nitrogen flow
before the reaction. DME and benzene were introduced after
the temperature was stable at 450 °C. Both gas and liquid
products were collected and analyzed by a GC (Agilent
7890A) equipped with a FID and a PONA capillary column
(50 m 9 0.2 mm 9 0.5 lm). Conversion and selectivity
were calculated according to Eqs. (3) and (4):
Amount of benzene ðDMEÞ % in the feed À amount of benzene ðDMEÞ % in the products
Amount of benzene ðDMEÞ % in the feed
Conversion ð%Þ ¼
 100 ð3Þ
The bands at ca. 1540 cm-1 and ca. 1450 cm-1 were
integrated to determine the concentrations of Brønsted and
Lewis acid sites, respectively. In the case of pivalonitile
adsorption-IR (Pn-IR), Pn was sorbed at r.t. to determine
the amount of acid sites located on external surface,
mesopore surface and in pore mouths of the zeolite sam-
ples. The accessibility factors (AF) fell into two kinds of
factors: the accessibility factors of Brønsted (AFB) and
Amount of i % in the products
P
Selectivity ðiÞ % ¼
 100
ð4Þ
Amount of i % in the products
where i represented one of the gas or liquid products. All
the reaction results presented in this work are expressed in
weight percent.
123

H. Liu et al.
silicon [28, 30] or more extensive desilication than dealu-
mination in NaOH solution [22, 31].
3 Results and Discussion
The isotherms from the N2 adsorption–desorption mea-
surements of Z-xat samples are shown in Fig. 1a, which
indicate the changes in the pore structure occurred after the
alkaline treatment of the investigated ZSM-11 materials.
The total adsorbed amount of N2 increased and the shapes
of the isotherms themselves had changed for the desilicated
samples, compared to the pristine zeolite. The HZSM-11
and Z-0.1at samples showed the typical type I isotherm,
with a high nitrogen uptake at low relative pressure being
characteristic for pure microporous materials. The fabri-
cation of mesopores by treatment of the ZSM-11 zeolite
with alkaline medium resulted in a higher adsorption
capacity in the range of mesopores. Definitely, the
adsorption isotherms for zeolites Z-0.2at to Z-0.7at showed
the type I ? IV isotherm with enhanced uptake at
P/P0 = 0.4–1.0, indicating that there are internal voids
with irregular shape between particles with a broad size
distribution [32]. The parallel adsorption and desorption
branches also verified that there are open and channel-like
mesopores for these alkaline treated samples [33]. More-
over, Z-1.0at showed only little N2 adsorption due to the
complete structure collapse. Obviously, the results of N2
adsorption–desorption measurement were in good agree-
ment with the XRD results.
3.1 Characterization of HZSM-11 and Z-xat
Samples
The XRD patterns (Fig. S1) of all studied zeolites depict
the typical diffraction patterns of ZSM-11 zeolite [27]. It
has been reported that the CNaOH is found to be the most
dominant factor affecting the desilication process [28].
Similarly here, the CNaOH played an important role on the
crystallinity of Z-xat samples. As the CNaOH increased, the
intensity of the XRD peaks decreased progressively. When
alkaline treatment was carried out in diluted NaOH solu-
tion with CNaOH ranging from 0.1 to 0.3 M, zeolite samples
maintained their good crystallinity. However, if the alka-
line treatment performed in medium NaOH solution with
the CNaOH around 0.5 M, the treated zeolite suffered from
partial destruction of its zeolite framework. If the alkaline
treatment was done with the more concentrated NaOH
solution ranging from 0.6 to 0.7 M, the treated zeolite
could be resulted in severe structure destruction. Moreover,
when the CNaOH of 1.0 M was used, the framework struc-
ture of the treated zeolite was completely collapsed.
The value of relative crystallinity (RC) of Z-xat mate-
rials are listed in Table 1. The RC value of Z-0.1at
increased slightly to 102 %. This could be attributed to the
purification by cleaning the amorphous particles on the
zeolite surface with diluted NaOH solution [29]. Increasing
the CNaOH from 0.2 to 1.0 M, the RC value of the treated
zeolites monotonously decreased from 93 to 1 %. Corre-
spondingly, the Si/Al2 ratios of Z-xat zeolites measured by
XRF showed a similar tendency and the values mono-
tonously decreased from 53 to 5 (Table 1), obviously due
to the essence of alkaline treatment is the rational use of the
intrinsic nature of NaOH to selectively remove framework
The BJH pore size distribution (PSD) derived from the
N2 adsorption branches of Z-xat is presented in Fig. 1b.
The HZSM-11 zeolite showed little mesopore. A kind of
broad PSD from 10 to 100 nm could be observed over
Z-0.1at, Z-0.2at and Z-0.3at, and the corresponding band
intensity increased with the CNaOH. Noticeably, there was a
broad band with high intensity centered at ca. 30 nm in the
PSDs from Z-0.5at to Z-0.7at samples, indicating that
mesopores formed after being treated with NaOH solution.
Table 1 Textural and physiochemical data of HZSM-11 and Z-xat samples
Samples
CNaOH (mol L-1
)
Si/Ala2
RCb
Textural data
SBET (m2 g-1
)
Sext (m2 g-1
)
Vmicro (cm3 g-1
)
Vmeso (cm3 g-1
)
HZSM-11
Z-0.1at
Z-0.2at
Z-0.3at
Z-0.5at
Z-0.55at
Z-0.6at
Z-0.7at
a
–
53
45
36
29
16
15
12
9
100
102
93
70
38
28
19
4
371
427
476
534
503
447
394
330
112
163
224
277
306
271
255
228
0.12
0.12
0.11
0.12
0.09
0.08
0.06
0.04
0.30
0.47
0.63
0.88
1.14
1.07
1.01
0.98
0.1
0.2
0.3
0.5
0.55
0.6
0.7
Si/Al2: molar ratio detected by XRF
b
RC: relative crystallinity (%)
123

Effect of Desilication on the Performance of Hierarchical ZSM-11 Catalysts for Alkylation of…
35]. As a reactant of the test reaction, benzene molecule
was used as the adsorbent in the IGA experiment as
reported in our previous work [26]. Here, similar experi-
ment is applied on four typical samples and the results are
presented in Fig. S2. The adsorption capacity for the tested
samples followed the sequences of Z-0.5at [ Z-0.3at [ Z-
0.1at [ HZSM-11, which was in parallel with the variation
tendency of Sext and Vmeso (see Table 1). The results further
verified that the diffusion property of ZSM-11 zeolite
samples could be greatly improved after being treated with
NaOH solution. In general, the higher the capacity of
benzene adsorption is, the better the diffusion property of
Z-xat samples can be [26].
3.2 Morphology Characterization of Z-xat Samples
SEM images of Z-xat samples are presented in Fig. 2,
which reveals the shapes and particle surface of partial
material. For HZSM-11 spherical aggregation with particle
diameters in a range of 300–400 nm could be observed,
and the crystalline grains showed irregular spheroidicity
with smooth surface. The sample of Z-0.1at showed similar
morphology with the HZSM-11, but the crystalline grains
were slightly eroded. In the case of Z-0.3at, this spherical
aggregation became loose, and holes appeared on the
grains surface. For Z-0.5at, this aggregation almost disap-
peared, and the grains were significantly destroyed.
TEM images of Z-xat samples are also exhibited in
Fig. 2. The HZSM-11 showed well-defined micropores.
The Z-0.1at sample was similar to the HZSM-11, with no
indication of mesopores in crystalline region. In fact, dis-
tinct opening pores with wide PSD (10–100 nm) were
uniformly distributed in Z-0.3at and Z-0.5at, while the
sample Z-0.5at exhibited more loose morphology. These
were in accordance with the BJH PSD measurement
(Fig. 1b). Such a heavily corrosive surface in Z-0.5at could
also be observed over NaOH treated ZSM-5 zeolite [33].
Mesopore production in Z-xat was correlated with the cases
of ZSM-5 [36] and MOR [37] zeolites. However, the
alkaline treated beta zeolite [38] exhibited distinct alter-
ation particularly at the edges of the crystal, contrast to the
pore production throughout the whole ZSM-11 zeolite
crystal in present work, which may be attributed to the
different crystal size. Besides, when ZSM-5 zeolite was
leached in NaOH solution under standard conditions
(0.2 M, 65 °C, 30 min), the TEM images showed meso-
pores centered ca. 10 nm [39].
Fig. 1 Nitrogen adsorption–desorption isotherms at -196 °C (a) and
BJH pore size distribution plots derived from the adsorption branch
(b) for HZSM-11 and Z-xat samples
Besides, the band was in the wide-spread form, which
could be attributed to that there were mesopores between
particles with a broad size distribution.
The data related N2 adsorption–desorption isotherms are
summarized in Table 1. Obviously, the NaOH treatment
could greatly increase both surface area and mesopore
volume (Vmeso). The largest external surface area (Sext) of
306 m2 g-1 and Vmeso of 1.14 cm3 g-1 could be obtained
on the Z-0.5at sample. On the other hand, its microporous
volume (Vmicro) decreased by 25 % as compared with the
HZSM-11. As reported in Ref. [29], aggravating the
severity of the alkaline treatment condition modified the
damage of the microporosity and an increase in meso-
porosity. The present results also showed that the CNaOH
was a crucial factor. If the zeolite was treated with the
CNaOH higher than 0.6 M (in the range of 0.6–1.0 M),
obvious decrease in both Sext and Vmeso could be observed,
indicating that some part of zeolite structure was collapsed.
IGA technique is a useful tool for characterization of
structure properties by applying different adsorbents [34,
3.3 Acidity Characterization of HZSM-11 and Z-xat
Samples by NH3-TPD and FT-IR
The acidities of the HZSM-11 and Z-xat materials were
investigated by NH3-TPD technique. The NH3-TPD profile
123

H. Liu et al.
Fig. 2 SEM and TEM images of HZSM-11 and Z-xat samples
of the HZSM-11 sample can be deconvoluted to three
desorption bands in the tested temperature range. Typi-
cally, the desorption peak temperatures for HZSM-11 are
centered at ca. 239, 349 and 440 °C corresponding to the
weak, medium and strong acid sites, respectively
(Fig. S3a). NH3-TPD profiles of HZSM-11 and Z-xat
samples and corresponding data obtained from the profiles
are presented in Fig. S3b and Table 2, respectively. The
total acid concentration (denoted as CT) of Z-0.1at
decreased a little compared with the HZSM-11, which may
be attributed to the purification by cleaning the amorphous
particles on the zeolite surface with diluted NaOH solution
[29], leading to the significant decrease in weak acid sites.
With the increase of CNaOH, the CT would gradually
increase. And the samples of Z-0.5at, Z-0.55at and Z-0.6at
showed the highest CT of ca. 550 lmol g-1. If the CNaOH
was higher than 0.6 M, due to the partially collapse of
zeolite structure, the CT and strong acid concentration
decreased simultaneously.
We should note that except for Z-0.1at and Z-0.7at all
the Z-xat samples exhibited a higher CT compared to the
HZSM-11 zeolite, which was in agreement with Groen
et al. [10] who reported that zeolite treatment with diluted
bases would result in the increasing concentration of acid
123

Effect of Desilication on the Performance of Hierarchical ZSM-11 Catalysts for Alkylation of…
sites determined by NH3-TPD technique. Besides, since
presence of relatively higher Al in the zeolitic framework
on account of selective desilication, the CT increased in
alkaline treated samples [28]. Generally, the NH3 desorp-
tion peak temperatures of weak acid sites remained at ca.
240 °C, and those of medium and strong acid sites was ca.
300–320 and ca. 420 °C respectively on Z-xat (x [0),
which were different from those of the HZSM-11.
Table 2 Characterization results of HZSM-11 and Z-xat zeolites by
NH3-TPD technique
Samples
Concentration of acid sites (lmol g-1
)
Total
Weak
Medium
Strong
HZSM-11
Z-0.1at
Z-0.2at
Z-0.3at
Z-0.5at
Z-0.55at
Z-0.6at
Z-0.7at
a
466
448
468
530
543
547
549
460
138 (240)a
107 (232)
118 (238)
140 (240)
159 (244)
174 (240)
184 (244)
161 (244)
94 (349)
234 (440)
244 (423)
246 (421)
268 (420)
179 (423)
184 (422)
190 (428)
136 (424)
97 (302)
104 (300)
122 (302)
205 (310)
189 (312)
175 (320)
163 (320)
The acidities of HZSM-11 and Z-xat materials were
further investigated by Py-IR technique. The corresponding
spectra and results are shown in Fig. S4 and Table 3,
respectively. Both the concentrations of the total (Total-Py-
150) and strong (Strong-Py-450) acid sites, whether they are
Brønsted or Lewis ones, first increased until reaching the
highest value on Z-0.3at zeolites and then decreased
smoothly from Z-0.3at to Z-0.6at and finally decreased
steeply from Z-0.6at to Z-0.7at. Nevertheless, all the Z-xat
zeolites, except for Z-0.7at, exhibited increased total
Brønsted (CB-Py-150) and Lewis (CL-Py-150) acid concen-
trations (per gram) compared with the HZSM-11. This was
in well agreement with the recent results reported by
Sadowska et al. [21, 22], who also reported the notable
increase in concentrations of Brønsted and Lewis acid sites,
while contracted with the results reported by Milina et al.
[16].
Data presented in parentheses were the desorption peak tempera-
tures (°C) of NH3-TPD profiles
samples obtained by Py-IR technique (Table 3) were lower
compared to that determined by TPD of NH3 (Table 2).
The possible reason is that part of the acid sites is inac-
cessible to bulky pyridine molecules but can be reached by
NH3 molecules [18, 40, 41], or that the possible formation
of NH4?nNH3 association [42] during NH3-TPD mea-
surement leads to a higher acid concentration.
IR technique of Pn adsorption (Pn-IR) was applied to
determine both the Brønsted and the Lewis acid sites sit-
uated on external surfaces and and in pore mouths (here-
after collectively called external surface) of HZSM-11 and
hierarchical Z-xat materials, which followed the Ref. [20].
The normalized and deconvoluted spectrum of Pn-IR on
the Z-0.5at sample (Fig. 3a) showed three bands centered
at 2248, 2277, and 2305 cm-1, respectively, which were
well consistent with the results reported in Ref. [20]. Pn-IR
spectra of HZSM-11 and Z-xat samples are shown in
Fig. 3b. The corresponding Brønsted (CB-Pn) and Lewis
(CL-Pn) acid concentrations listed in Table 3 were calcu-
lated based on the extinction coefficients following the Ref.
[20].
The ratio of total Brønsted to Lewis acid concentration
(B/L-Py-150) of the HZSM-11 was 1.42. While Z-0.1at
sample showed a reduced B/L-Py-150 value of 1.32. When
HZSM-11 zeolite was treated with a more concentrated
NaOH solution, Z-xat (x is in the range of 0.2–0.7) mate-
rials would present a similar B/L-Py-150 value of ca. 1.00,
which was inconsistent with previous reports [16, 22, 31].
While the ratio of strong Brønsted to Lewis acid con-
centration (B/L-Py-450) on HZSM-11 and Z-xat samples
decreased faster than B/L-Py-150 with the increase of CNaOH
.
For HZSM-11 the B/L-Py-450 value was 4.63, which was
significantly higher than those of Z-xat samples, and the
value decreased monotonously from 2.65 to 0.77 with
C
NaOH. In the case of the weak-medium acid concentration
The HZSM-11 zeolite contained little acid sites acces-
sible to hindered Pn molecules (1 lmol g-1 for CB-Pn and
5 lmol g-1 for CL-Pn, respectively). Both the CB-Pn and
CL-Pn of the Z-xat materials first increased and then
decreased with the CNaOH, and the sample of Z-0.3at
exhibited the highest value (180 lmol g-1 for CB-Pn and 34
lmol g-1 for CL-Pn, respectively) among the tested sam-
ples. Moreover, the B/L ratios determined by Pn-IR (B/L-
Pn) of all the Z-xat samples (x [ 0) were remarkably higher
than that of the HZSM-11, implying that there was an
obvious change in external acid sites distribution after
treatment with NaOH solution. Besides, for all Z-xat
(x [ 0) samples, the B/L-Pn value was higher than the
corresponding B/L-Py-150 value, indicating that after the
alkaline treatment the distribution and concentrations of
(WM-Py-150–450), it first increased and then decreased,
reaching the highest value at CNaOH of 0.5 M. Meanwhile,
the ratio of weak-medium Brønsted to Lewis acid con-
centration (B/L-Py-WM) of Z-xat increased from 0.16 to 1.48
when the CNaOH increased from 0 to 0.6 M.
In summary, alkaline treatment resulted in the decrease
in Strong-Py-450 and the increase WM-Py-150–450, which was
in well agreement with the results reported in Ref. [36].
Besides, the acid sites distributions (B/L-Py-150, B/L-Py-450
and B/L-Py-WM) showed different tendency with CNaOH
.
It should be noticed that the resultant acidities of
HZSM-11 and Z-xat materials obtained by Py-IR technique
showed some differences from those obtained by NH3-TPD
(Table 2). The total acid concentrations of the tested
123

H. Liu et al.
Table 3 Characterization results of HZSM-11 and Z-xat zeolites by FT-IR techniques
Samples
Concentration of acid sites (lmol g-1
)
Determined with pyridine
Determined with
pivalonitrilef
Totala-Py-150
WMd-Py-150–450
Stronge-Py-450
CBb -Py-150 CcL-Py-150 B/L-Py-150 CB-Py-WM CL-Py-WM B/L-Py-WM CB-Py-450 CL-Py-450 B/L-Py-450 CB-Pn CL-Pn B/L-Pn
HZSM-11 151
106
131
136
214
202
176
153
93
1.42
1.32
1.15
1.10
0.94
1.01
1.16
0.99
12
51
53
81
95
97
99
59
76
0.16
0.60
0.62
0.79
0.90
1.08
1.47
1.18
139
122
104
155
95
30
46
51
112
96
86
86
43
4.63
2.65
2.04
1.38
0.99
0.93
0.92
0.77
1
5
0.2
5.6
3.5
5.2
4.3
Z-0.1at
Z-0.2at
Z-0.3at
Z-0.5at
Z-0.55at
Z-0.6at
Z-0.7at
a
173
157
236
190
177
178
92
85
90
106
180
98
–
16
30
34
23
–
85
102
106
90
80
67
79
69
–
16
–
4.3
50
33
Total-Py-150: total acid concentration determined by Py-IR technique at 150 °C
CB: Brønsted acid concentration
b
c
d
e
f
CL: Lewis acid concentration
WM-py-150–450: weak-medium acid concentration was defined as the difference between the total and strong acid concentration
Strong-py-450: strong acid concentration determined by Py-IR technique at 450 °C
The extinction coefficients of the IR bands of pivalonitrile interacting with both Brønsted (0.11 cm2 lmol-1) and Lewis (0.16 cm2 lmol-1
)
acid sites were reported in Ref. [20]
acid sites on the external surface of Z-xat were quite dif-
ferent from that in the bulk.
zeolite catalysts. Here, we take Z-0.5at zeolite catalyst as
an example (Fig. 5). The initial conversions (the conver-
sions at the reaction time on stream (TOS) of 2 h) of
benzene and DME were ca. 48 and 100 %, respectively.
Besides, their variation with TOS is shown in Figs. 6a, b.
During the 73 h TOS, the selectivities of dry gas (\1 %)
and LPG (\2 %) over Z-0.5at were still very low, and the
selectivity of the liquid products was as high as ca. 97 %.
As shown in Fig. 5, when running for the 73 h, the
selectivity of toluene increased from 46 to 62 %. The
selectiveties of xylene and trimethylbenzene first kept
constant (in TOS range of 0–58 h) and then decreased. In
the case of C5–6, the selectivity first increased a little and
then decreased slowly. Meanwhile the selectivities of
ethylbenzene, propylbenzene and ethyltoluene decreased
with TOS.
The diagrams representing the dependence of AFB and
AFL of the HZSM-11 and Z-xat samples on the CNaOH are
presented in Fig. 4. For all samples, the value of AFB was
higher than that of AFL on the whole. The HZSM-11
zeolite contained little Brønsted and Lewis acid sites
accessible to Pn molecules, leading to the small AFB value
of 0.6 % and AFL value of 4.7 %. For the Z-xat zeolites,
with the increase of CNaOH, both the AFB and AFL values
first increased and then decreased. Besides, when CNaOH
was less than 0.5 M, the AF values basically correlated
with Sext and Vmeso of materials since their crystallinities
were preserved well. However, for the materials treated in
concentrated NaOH solution (0.5–1.0 M), the AF values
did not correlate with the Sext and Vmeso, somehow different
from the results reported in Ref. [20]. This may because
that the resultant materials treated in more concentrated
NaOH solution are partially amorphous, which changes the
concentration and distribution of acid sites over the
zeolites.
Based on the above results, a possible reaction route is
illustrated in Scheme S1. Firstly, a proton from the zeolite
attacked DME to form methyl and methoxy species [25]. A
methoxy cation then attacked a benzene molecule to form
toluene. The xylene and trimethylbenzene formed in the
similar manner: a methoxy cation attacked a toluene or
xylene. Therein, the toluene and xylene were the main
liquid products, similar to the case of ZSM-5 [25]. On the
other hand, DME could be converted to hydrocarbones
(CxHy) and aromatics after being activated [44]. Also the
formed CxHy could be activated and then attacked benzene
or other aromatic molecules to produce alkyl-aromatics,
such as ethylbenzene, propylbenzene and ethyltoluene,
3.4 Reaction Performance
The reaction performance of alkylation of benzene with
DME on HZSM-11 and Z-xat zeolite catalysts were
investigated. Since the similar zeolite topology results in
similar product distribution [43], almost the same product
distributions can be obtained on HZSM-11 and Z-xat
123

Effect of Desilication on the Performance of Hierarchical ZSM-11 Catalysts for Alkylation of…
Fig. 4 Plots of AFB and AFL versus alkaline concentration
zeolite with pure micropores exhibited a dramatic decay in
conversions of benzene and DME after 12 h on stream. The
stability of Z-xat catalysts first increased and then
decreased with the increase of CNaOH. Therein, Z-0.5at
zeolite catalyst showed the best performance, which
maintained its stable conversion for ca. 50 h.
3.5 Relevance of Reaction Stability and Porosity
as Well as Acidity of HZSM-11 and Z-xat
Catalysts
Although the initial DME and benzene conversions were
similar for all the investigated ZSM-11 catalysts, their
reaction stabilities showed remarkable difference. For
convenience, two indexes are used to measure the reaction
stability: one is TD which is defined as the TOS when the
DME conversion decreased to 95 %, and the other is TB
which means the TOS when the benzene conversion
decreased to 34 %.
Fig. 3 Pn-IR spectra recorded at r.t. on Z-0.5at sample (a) and
HZSM-11 and Z-xat samples (b)
although these products took only a small part. With the
reaction progress, the specific selectivities of products
would vary with TOS as shown in Fig. 5.
As reported in Ref. [45], the NaOH treatment of the
parent zeolite causes the main effect on the stability in the
conversion of methanol. Similarly in our present study, the
alkylation stability of benzene with DME over HZSM-11
and Z-xat zeolite catalysts is compared and the results are
shown in Fig. 6 (a, based on DME conversion; b, based on
benzene conversion). The scattering of benzene conversion
data is very likely due to the fluctuation in the evaporation
of benzene [25]. Except for Z-1.0at with much poor reac-
tion performance, the initial DME conversion of 100 %
and initial benzene conversion of ca. 48 % were similar for
all Z-xat zeolite catalysts. When the conversion of benzene
began to decrease, the ratio of DME conversion to benzene
conversion became higher and higher with the reaction
progress, suggesting the reactivity of DME is higher than
that of benzene in the feed. As expected, the HZSM-11
The influences of the textural properties and acidities
over HZSM-11 and Z-xat catalysts on the reaction stability
of alkylation of benzene with DME were further investi-
gated. As shown in Fig. 7, it was interesting to notice that
the change of Vmeso was in line with that of reaction sta-
bility. Increasing CNaOH from 0 to 0.5 M, the Vmeso and the
reaction stability increased together. With CNaOH further
increasing to 0.7 M, the Vmeso and the stability decreased
consistently. Therein, Z-0.5at showed the highest Vmeso and
the best stability with TD of 57 h and TB of 64 h. This
could be attributed to the better diffusion capability of the
catalyst caused by the generation of open mesopores,
which facilitated the diffusion of coke precursors [9]. The
Vmicro first maintained at ca. 0.12 cm3 g-1 and then
decreased with CNaOH, while the Vmeso first increased and
then decreased. For Z-0.5at the former was 0.09 cm3 g-1
123

H. Liu et al.
Fig. 6 Reaction stability of HZSM-11 and Z-xat zeolite catalysts
correlation between these two parameters with catalytic
performance.
To evaluate the influence of acidity on reaction perfor-
mance, NH3-TPD, Py-IR and Pn-IR techniques were
applied as well. In Sect. 3.3, we described and discussed
the properties of acidity, including their concentration,
intensity, distribution (B/L-Py-150, B/L-Py-450, B/L-Py-WM and
B/L-Pn) and AF. Relating these properties with the reaction
stability, it was interesting to find that the WM-Py-150–450
was well accordant with the stability (Fig. 7).
As shown in Fig. 7, the mesoporosity was positively
related to the reaction stability. Generally speaking,
coke deposition is the reason for zeolite catalyst deac-
tivation [47]. The origin of superior stability can be
explained in terms of the facile diffusion of coke pre-
cursors from the micropores to the surfaces of the
mesopore walls owing to the large external surface area
and the short diffusion path length [9]. It should be
noticed that we did not investigate the coke of the used
samples in present work.
Fig. 5 Products selectivities as a function of TOS over Z-0.5at
and the latter was 1.14 cm3 g-1. Therefore, the suitable
balance between mesoporosity and microporosity is the
essential factor to improve the stability [8].
In the case of the NaOH treated ZSM-11 materials the
additional porosity was not the only reason for the
improved reaction stability. As reported by Refs. [9, 12, 36,
46], the changes in both acidic properties and porosity
influenced the catalytic properties of the catalysts, never-
theless the authors did not analyze specifically the
The reaction stability can be promoted by reducing the
number of strong acid sites and meanwhile increasing the
123

Effect of Desilication on the Performance of Hierarchical ZSM-11 Catalysts for Alkylation of…
Fig. 7 Variation of stability,
pore volumes and WM-Py-
150–450 with CNaOH of HZSM-11
and Z-xat catalysts
mesoporosity [9]. As shown in Table 3, CB-Py-450 of Z-xat
samples were lower than the HZSM-11 on the whole,
therefore, the decreased CB-Py-450 may be beneficial to the
improved reaction stability. However these two parameters
did not show clear linear relation.
CNaOH could be attributed to the formation of mesopores.
Most likely, these acid sites or these amorphous silica-
alumina materials were mainly situated on the mesopores,
leading to the consistence of WM-Py-150–450 with Vmeso
.
Therefore, the reaction stability of the studied zeolites
depended on the combination of WM-Py-150–450 with Vmeso
(Fig. 7).
DME-to-olefins reaction is favored by the weak acidity
because the Brønsted acid sites have the ability to bind the
methoxy species, and the weak acid intensities are bene-
ficial to reduce hydrogen transfer reaction rate, then to
lower the formation rate of coke and heavy hydrocarbons
[48, 49]. Besides, the weak acid site is responsible for the
reaction of aromatization, whereas the strong acid center is
more likely to be the deactivation center in the methanol
aromatization reaction over ZSM-5 zeolite [50]. It was also
reported that from the NH3-TPD profile of the deactivated
hierarchical zeolite, the strong acid sites, which are
responsible for catalyzing the MTH reaction, are not
available for NH3 adsorption anymore. While a large
number of weak acid sites are still able to adsorb NH3 [51].
Similarly in our present work, the strong acid sites were
poisoned by the deposition of coke soon after the beginning
of the reaction, and the remaining weak-medium acid sites
turned to be the major active centers. This may be the
reason that the WM-Py-150–450 could be positively related
with the reaction stability (Fig. 7).
4 Conclusion
ZSM-11 zeolite has been modified with NaOH, resulting in
a series of samples (Z-xat) with different porosity and
acidity. Textural properties for ZSM-11 zeolite showed that
at an appropriate NaOH concentration, external surface
area and mesopore volume were found to be observably
improved.
Acidic properties of the alkaline treated samples were
investigated in detail. NH3-TPD measurement revealed that
the total acid concentrations of Z-xat increased as a whole
compared with the HZSM-11 except for Z-0.1at and
Z-0.7at, and the value first increased and then decreased
with CNaOH. Py-IR technique showed that with the increase
of CNaOH, the ratio of total Brønsted to Lewis acid con-
centration of Z-xat first decreased a little and then kept
around 1.00, while that of the strong Brønsted to Lewis
acid concentration reduced regularly. IR technique of
pivalonitrile adsorption indicated that NaOH treatment
changed the B/L value on the external surface and this
value was of obvious difference from that in the bulk. The
AF of acid sites for Z-xat samples revealed that the AFB
It was believed that during NaOH treatment of ZSM-5
zeolite some protonic sites situated on amorphous silica-
alumina material were of weaker acid strength than typical
zeolitic Si–OH–Al groups (strong acid sites) [24]. This was
similar with the case of weak-medium acid sites in our
present work. The changes in the WM-Py-150–450 with
123

H. Liu et al.
´
21. Sadowska K, Gora-Marek K, Datka J (2012) Vib Spectrosc
63:418
was higher than the AFL as a whole, besides the evolution
of both values with CNaOH showed similar regulation.
The reaction stability of alkylation of benzene with
DME over the HZSM-11 and Z-xat zeolite catalysts was
closely linked with both porosity and acidity. The superior
reaction stability of the alkaline treated ZSM-11 zeolite
catalysts resulted from the interplay of the improved dif-
fusion property by mesoporosity and appropriate weak-
medium acidity determined by Py-IR technique.
´
´
22. Sadowska K, Gora-Marek K, Drozdek M, Kustrowski P, Datka J,
Triguero JM, Rey F (2013) Microporous Mesoporous Mater
168:195
´
ˇ ´
23. Rac V, Rakic V, Stosic D, Otman O, Auroux A (2014) Microp-
orous Mesoporous Mater 194:126
24. Gil B, Mokrzycki Ł, Sulikowski B, Olejniczak Z, Walas S (2010)
Catal Today 152:24
´
25. Szechenyi A, Solymosi F (2008) Catal Lett 127:13
26. Liu H, Wei HJ, Xin WJ, Song C, Xie SJ, Liu ZN, Liu SL, Xu LY
(2014) J Energy Chem 23:617
27. Yu QJ, Chen J, Zhang Q, Li CY, Cui QK (2014) Mater Lett
120:97
28. Wei XT, Smirniotis PG (2006) Microporous Mesoporous Mater
Acknowledgments The guidance and help of Professor Yide Xu is
greatly appreciated.
97:97
29. Liu KF, Xie SJ, Xu GL, Li YN, Liu SL, Xu LY (2010) Appl Catal
A Gen 383:102
30. Qin ZX, Shen BJ, Gao XH, Lin F, Wang BJ, Xu CM (2011) J
Catal 278:266
References
´
31. Sadowska K, Wach A, Olejniczak Z, Kustrowski P, Datka J
(2013) Microporous Mesoporous Mater 167:82
32. Kruk M, Jaroniec M (2001) Chem Mater 13:3169
1. Zhang L, Liu HJ, Li XJ, Xie SJ, Wang YZ, Xin WJ, Liu SL, Xu
LY (2010) Fuel Process Technol 91:449
2. Xie PF, Ma Z, Zhou HB, Huang CY, Yue YH, Shen W, Xu HL,
Hua WM, Gao Z (2014) Microporous Mesoporous Mater 191:112
3. Liu H, Wei HJ, Xin WJ, Liu SL, Xie SJ, Xu LY (2014) Acta Pet
Sin (Pet Process) 30:115
´
´
33. Milina M, Mitchell S, Crivelli P, Cooke D, Perez-Ramırez J
(2014) Nature Commun. doi:10.1038/ncomms4922
34. Zhao L, Shen BJ, Gao JS, Xu CM (2008) J Catal 258:228
35. Yue YY, Liu HY, Yuan P, Li TS, Yu CZ, Bi H, Bao XJ (2014) J
Catal 319:200
4. Jury FA, Polaert I, Estel L, Pierella LB (2014) Microporous
Mesoporous Mater 198:22
36. Svelle S, Sommer L, Barbera K, Vennestrøm PNR, Olsbye U,
Lillerud KP, Bordiga S, Pan Y-H, Beato P (2011) Catal Today
168:38
5. Derouane EG, Andre J-M, Lucas AA (1988) J Catal 110:58
6. Perez-Ramirez J, Christensen CH, Egeblad K, Christensen CH,
Groen JC (2008) Chem Soc Rev 37:2530
37. Ordomsky VV, Ivanova II, Knyazeva EE, Yuschenko VV, Zai-
kovskii VI (2012) J Catal 295:207
´
´
7. Verboekend D, Perez-Ramırez J (2011) Catal Sci Technol 1:879
8. Li YN, Liu SL, Zhang ZK, Xie SJ, Zhu XX, Xu LY (2008) Appl
Catal A Gen 338:100
´
´ ´
38. Groen JC, Abello S, Villaescusa LA, Perez-Ramırez J (2008)
Microporous Mesoporous Mater 114:93
39. Perez-Ramirez J, Verboekend D, Bonilla A, Abello S (2009) Adv
Funct Mater 19:3972
40. Awayssa O, Al-Yassir N, Aitani A, Al-Khattaf S (2014) Appl
Catal A Gen 477:172
9. Kim J, Choi M, Ryoo R (2010) J Catal 269:219
´ ´
10. Groen JC, Moulijn JA, Perez-Ramırez J (2005) Microporous
Mesoporous Mater 87:153
11. Wang LY, Wang Y, Wang AJ, Li X, Zhou F, Hu YK (2013)
Microporous Mesoporous Mater 180:242
12. Schmidt F, Lohe MR, Bu¨chner B, Giordanino F, Bonino F,
Kaskel S (2013) Microporous Mesoporous Mater 165:148
13. Liu J, Zhang C, Shen Z, Hua W, Tang Y, Shen W, Yue Y, Xu H
(2009) Catal Commun 10:1506
´
´
41. Gora-Marek K, Derewinski M, Sarv P, Datka J (2005) Catal
Today 101:131
´
42. Lonyi F, Valyon J (2001) Microporous Mesoporous Mater 47:293
43. Jae J, Tompsett GA, Foster AJ, Hammond KD, Auerbach SM,
Lobo RF, Huber GW (2011) J Catal 279:257
´
14. Thibault-Starzyk F, Stan I, Abello S, Bonilla A, Thomas K,
44. Kecskemeti A, Barthos R, Solymosi F (2008) J Catal 258:111
45. Lietz G, Schnabel KH, Peuker C, Gross T, Storek W, Volter J
(1994) J Catal 148:562
46. Bjørgen M, Joensen F, Holm MS, Olsbye U, Lillerud K-P, Svelle
S (2008) Appl Catal A Gen 345:43
47. Mei CS, Wen PY, Liu ZC, Liu HX, Wang YD, Yang WM, Xie
ZK, Hua WM, Gao Z (2008) J Catal 258:243
48. Zhao TS, Takemoto T, Tsubaki N (2006) Catal Commun 7:647
49. Ahmed MHM, Muraza O, Al Amer AM, Sugiura Y, Nishiyama N
(2015) Microporous Mesoporous Mater 207:9
´
´
Fernandez C, Gilson J-P, Perez-Ramırez J (2009) J Catal 264:11
15. Nesterenko NS, Thibault-Starzyk F, Montouilliout V, Yush-
chenko VV, Fernandez C, Gilson JP, Fajula F, Ivanova II (2006)
Kinet Catal 47:40
16. Milina M, Mitchell S, Michels N-L, Kenvin J, Perez-Ramirez J
(2013) J Catal 308:398
17. Gora-Marek K, Tarach K, Choi M (2014) J Phys Chem C
118:12266
18. Gora-Marek K, Tarach K, Tekla J, Olejniczak Z, Kustrowski P,
Liu L, Martinez-Triguero J, Rey F (2014) J Phys Chem C
118:28043
19. Holm MS, Svelle S, Joensen F, Beato P, Christensen CH, Bordiga
S, Bjørgen M (2009) Appl Catal A Gen 356:23
´
´
´
50. Yang FK, Wang SM, Xia YM, Bi WH, Wang W (2011) Pet Sci
Technol 29:1675
51. Schmidt F, Hoffmann C, Giordanino F, Bordiga S, Simon P,
Carrillo-Cabrera W, Kaskel S (2013) J Catal 307:238
´
20. Sadowska K, Gora-Marek K, Datka J (2013) J Phys Chem C
117:9237
123