Synthesis, characterization and isomerization performance of micro/mesoporous materials based on H-ZSM-22 zeolite

Article abstract of DOI:10.1016/j.jcat.2015.12.009

Micro/mesoporous materials with different mesoporosities were prepared through recrystallization of H-ZSM-22 zeolite in alkaline solution with cetyltrimethylammonium bromide template (CTAB). The structure, morphology, pore properties, acidity and isomerization performance of the catalysts by using the resulting materials were characterized and assessed. The dissolution and recrystallization procedure introduced the well-developed mesoporous structure of MCM-41 type with the meso-scale channels of about 3 nm in size on the outer surfaces of the microporous H-ZSM-22 zeolites, forming the micro/mesoporous materials, which possessed increased weak B acid sites at the pore mouths and a reduced amount of total acid sites. It is shown that the presence of well-developed mesopores could remarkably improve the selectivity to multi-branched products and suppress the side cracking reactions in n-dodecane isomerization. The micro/mesoporous Pt/ZSM-22/MCM-41 bifunctional catalyst with suitable recrystallization degree exhibits high isomerization selectivity under high conversion in long-chain n-alkane isomerization compared to the original microporous Pt/H-ZSM-22 catalyst.

Full text of DOI:10.1016/j.jcat.2015.12.009

Journal of Catalysis 335 (2016) 11–23
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Synthesis, characterization and isomerization performance of
micro/mesoporous materials based on H-ZSM-22 zeolite
Suyao Liu a,c, Jie Ren a,b, , Huaike Zhang a,b, Enjing Lv , Yong Yang a,b, Yong-Wang Li a,b,
⇑
b
⇑
a
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China
National Energy Research Center for Clean Fuels, Synfuels China Co., Ltd, Beijing 101400, PR China
b
c
University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Micro/mesoporous materials with different mesoporosities were prepared through recrystallization of H-
ZSM-22 zeolite in alkaline solution with cetyltrimethylammonium bromide template (CTAB). The struc-
ture, morphology, pore properties, acidity and isomerization performance of the catalysts by using the
resulting materials were characterized and assessed. The dissolution and recrystallization procedure
introduced the well-developed mesoporous structure of MCM-41 type with the meso-scale channels of
about 3 nm in size on the outer surfaces of the microporous H-ZSM-22 zeolites, forming the micro/
mesoporous materials, which possessed increased weak B acid sites at the pore mouths and a reduced
amount of total acid sites. It is shown that the presence of well-developed mesopores could remarkably
improve the selectivity to multi-branched products and suppress the side cracking reactions in
n-dodecane isomerization. The micro/mesoporous Pt/ZSM-22/MCM-41 bifunctional catalyst with
suitable recrystallization degree exhibits high isomerization selectivity under high conversion in long-
chain n-alkane isomerization compared to the original microporous Pt/H-ZSM-22 catalyst.
Ó 2015 Elsevier Inc. All rights reserved.
Received 3 August 2015
Revised 9 November 2015
Accepted 14 December 2015
Keywords:
Micro/mesoporous
Recrystallization
H-ZSM-22
Isomerization
Multi-branch selectivity
1
. Introduction
Skeletal isomerization of long-chain n-alkanes to their branched
It has been proposed that both the narrow one directional
tubular micropores and the strong acid sites at the pore mouth
of ZSM-22 lead to the lengthy diffusion pathways between the
acidic sites and the metal sites seem to be the major reason that
the isomerization products are predominantly of mono-branched
components, while the multi-branched isomers are of essential
importance for desired properties of middle distillate isomeriza-
tion [15,16]. However, it is difficult to find the direct evidence on
this assumption. Nevertheless, introducing mesoporous structure
on the basis of the microporous zeolite crystals has been expected
to be a useful way to overcome the inherent diffusion limitations
[17,18] and optimize the acid site distribution [19].
Several approaches have successfully been developed for the
preparation of hierarchical porous zeolites, namely the post-
treatment method [20,21], the hard-templating method and the
soft-templating method [22]. Ivanova’s group demonstrated a con-
trollable approach, including simultaneous dissolution and re-
deposition of the dissolved species on the surface of the parent
materials in the presence of long-chain alkylammonium surfac-
tants [23,24], which would permit a better control of mesopore
size [25]. Their work has highlighted the new perspectives to for-
mulate the controllable micro/mesoporous materials [26,27].
Inspired by this synthesis concept, a few micro/mesoporous mate-
rials have been prepared with remarkable performance proved for
isomers nowadays is one of the most demanded processes that
have been widely used to improve the cold-flow properties of
diesel and lube base oil [1,2], especially in the industrial processes
of coal/natural gas to liquids (CTL/GTL) via Fischer–Tropsch syn-
thesis [3–7]. For the isomerization of synthetic middle distillates,
developing catalysts with high selectivity to various isomers and
low to cracking products is the important subject of many investi-
gations [8–10].
For the isomerization of the synthetic middle distillates com-
posing mainly n-alkanes free from contaminate (sulfur and nitro-
gen), bifunctional catalysts with platinum on zeolite supports
have been evaluated [11,12] as potential candidates for industrial
applications. It has been found that the platinum on ZSM-22
molecular sieve with one-dimensional 10-membered ring chan-
nels of 0.45 Â 0.54 nm is one of the successful n-alkane isomeriza-
tion catalysts owing to the unique shape-selectivity [13,14].
⇑
Corresponding authors at: State Key Laboratory of Coal Conversion, Institute of
Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China.
E-mail addresses: renjie@sxicc.ac.cn (J. Ren), ywl@sxicc.ac.cn (Y.-W. Li).
http://dx.doi.org/10.1016/j.jcat.2015.12.009
0
021-9517/Ó 2015 Elsevier Inc. All rights reserved.

1
2
S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
several catalytic reactions [24]. For example, micro/mesoporous
MOR/MCM-41 materials with well-distributed mesopores were
prepared and proved to be of high performance in cumene dispro-
to the assembling of the dissolved species into mesoporous phase
around the parent zeolite structures. The degree of destruction of
the parent H-ZSM-22 was adjusted by varying the NaOH concen-
tration in the aqueous solution from 0.15, 0.30, 0.45 to 0.90 M,
and the prepared materials were denoted as RZEO-1, RZEO-2,
RZEO-3, and RZEO-4, respectively. The products were washed with
de-ionized water to neutral and dried at 120 °C for 12 h. All sam-
ples were calcined to remove the surfactant at 550 °C for 8 h. The
micro/mesoporous ZSM-22 materials were subjected to ion-
exchange with aqueous solution of NH NO . Then, the materials
4 3
were washed, dried and calcined in the air flow at 550 °C for 4 h.
The ion-exchange steps were repeated for 3 times.
For comparison, the desilicated samples were prepared by stir-
ring the K-ZSM-22 zeolites in the aqueous solution of 0.3 M NaOH
at 80 °C for 0.5 h without CTAB, and then this solid was filtered,
dried and washed in a solution of 0.1 M HCl at 80 °C for 8 h. The
samples were filtered, washed to neutral with de-ionized water
and dried overnight at 120 °C. The alkali metal was entirely
6 8
portionation [28] and both the short n-alkane (n-C and n-C ) and
the long-chain n-alkane (n-C16) isomerization [29] due to the
improved accessibility of active sites and the transportation of
bulky molecules. Similarly, the recrystallized ferrierite materials
showed improved catalytic performance in 1-butene isomeriza-
tion, demonstrating that the enhancing effect of coating mesopore
layers on suppressing the catalyst deactivation [30]. However, the
synthesis of micro/mesoporous ZSM-22 materials with improved
selectivity to multi-branched isomers in long-chain n-alkane iso-
merization has not seen much success. Moreover, the straightfor-
ward explanation for the effect of the uniform mesopores on
long-chain n-alkane isomerization is still not available.
In the present work, series of micro/mesoporous materials with
a varying degree of mesoporosity were prepared by the recrystal-
lization of the H-ZSM-22 zeolite in an alkaline solution with the
assistance of CTAB. The introduced mesoporous phase and the
optimal recrystallization degree are addressed on the basis of the
correlations between the structure, texture, acidity and the cat-
alytic performance. Isomerization of n-dodecane over these cata-
lysts is chosen as a model reaction for long-chain n-alkanes
isomerization to explore the role of the mesopores, for promoting
the distribution of B acid sites and the catalytic selectivity to differ-
ent isomers. Moreover, for comparison, the conventional microp-
orous and desilicated ZSM-22 materials are prepared and
characterized, and the isomerization performance of the corre-
sponding catalysts is discussed.
4 3
removed from the products by exchanging with 0.5 M NH NO
for 3 times. Finally, the samples were dried and calcined in air flow
in a furnace at 550 °C for 8 h.
2.4. Preparation of catalysts
To obtain the catalysts for isomerization tests, the H-ZSM-22,
desilicated sample (DeZEO) and recrystallized materials (samples
RZEO-1, RZEO-2, RZEO-3, and RZEO-4) were extruded by adding
quasi-boehmite (15 wt.%) as the binder, crushed, sieved and
impregnated in a solution of H
2
PtCl
6
Á6H
2
O (1.0 wt.%, pH = 2.3).
Then, the suspension was slowly evaporated to remove moisture
at 80 °C. Finally, the resulting solid was dried at 120 °C for 12 h
and calcined in air at 450 °C for 4 h with a temperature ramp rate
of 1 °C/min. The content of loading Pt species was 0.5 wt.%. The
obtained catalysts were denoted as Pt/H-ZSM-22, Pt/DeZEO, Pt/
RZEO-1, Pt/RZEO-2, Pt/RZEO-3, and Pt/RZEO-4, respectively.
2
. Experimental
2.1. Reagents
The reactants used in the present study are aluminum sulfate,
,6-diaminohexane (DAH), cetyltrimethylammonium bromide
1
2.5. Characterization
(
CTAB) (Sinopharm Chemical Reagent Co., Ltd.), potassium hydrox-
ide, sodium hydroxide, ammonium nitrate (Beijing Chemical Co.,
Ltd.), silica sol (40 wt.% Qingdao Haiyang Chemical Co., Ltd.), and
de-ionized water.
The X-ray diffraction (XRD) patterns were obtained with a Bru-
ker D8 Advance diffractometer with Cu K
a
(c
= 1.5418 Å, 40 kV,
0 mA) radiation in the scan range of 1°–50° with a step size of
.02°.
The morphology of different samples was determined by meth-
4
0
2.2. Synthesis of H-ZSM-22
ods of field emission scanning electron microscopy (FE-SEM)
recorded using an FEI QUANTA 400. Transmission electron micro-
scopy (TEM) images were obtained using an FEI Tecnai G2 F30
operating at 300 kV for high resolution measurements.
The bulk chemical composition was analyzed by inductively
coupled plasma-atomic emission spectrometer (ICP-AES). The
XPS analyses were carried out on a Physical Electronics Company
The ZSM-22 zeolite was synthesized through hydrothermal
treatment of aluminosilicate solution with the chemical composi-
tion of 1.0SiO :0.12K O:0.014Al :0.3DAH:40H O. The synthesis
2
2
2
O
3
2
processes were performed at 160 °C for 48 h in an autoclave under
stirring with the self-generated pressure. The solid zeolite products
were recovered by filtration, washed extensively with de-ionized
water and dried at 120 °C. Finally, the template in the zeolites
was removed by calcination in air at 550 °C, and the final product
was obtained as K-ZSM-22. The K-ZSM-22 zeolites were converted
Quantum-2000 Scanning ESCA Microprobe equipped with Al K
a
source and a multichannel detector to investigate the surface Si/
Al ratio. The binding energies were calculated with respect to the
C–(C,H) component of the C 1s peak of adventitious carbon fixed
at 284.8 eV. The spectra of different samples were conducted using
Shirley background subtraction and Gaussian/Lorentzian (70/30)
product function. Molar fractions were calculated using the nor-
malized peak areas on the basis of acquisition parameters and sen-
sitivity factors (Sisf = 0.82, Alsf = 0.54) provided by manufacturer
and the transmission function.
4 3
to protonic form by ion-exchange with NH NO at 80 °C. The ion-
exchanged samples were dried at 120 °C overnight, and calcined
at 550 °C for 4 h. These ion-exchange treatment steps were
repeated for 3 times, and the final H-ZSM-22 product was formed.
2
.3. Synthesis of micro/mesoporous ZSM-22
29
The micro/mesoporous ZSM-22 materials were prepared by the
Solid-state Si and 27Al magic angle spinning nuclear magnetic
following procedures. The H-ZSM-22 samples were mixed with a
NaOH aqueous solution, and the mixtures were hydrothermally
treated in the presence of CTAB at 100 °C for 24 h. After this step,
the recrystallization was performed by adjusting the pH value of
the suspension solution to right levels at 120 °C for 24 h, leading
resonance (MAS NMR) spectra were performed on a Bruker
Advance 600 NMR spectrometer. Kaolin and Al(NO
as the chemical shift references at À91.5 and 0 ppm, respectively.
Low temperature N
-adsorption was measured at À196 °C
using Micromeritics ASAP 2020 analyzer and 2420 analyzer that
3 3
) were used
2

S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
13
were used to obtain the microporous and mesoporous porosities,
respectively. The total surface area was determined according to
the BET method, and the most probable pore sizes in the ranges
of 0–2 nm and 2–50 nm were calculated by the Horvath–Kawazoe
(
HK) method and the Barrett–Joyner–Halenda (BJH) method,
respectively.
To characterize the acidity, the temperature programmed des-
orption of ammonia (NH
chemisorption analyzer (Micromeritics ASAP 2920). Prior to NH
adsorption at 100 °C, the samples were pretreated in the flowing
helium at 550 °C for 1 h. The physically adsorbed NH was
3
-TPD) was carried out at a dynamic
3
-
3
removed in a flow of helium for 1 h, and then the adsorbed samples
were heated with a rate of 10 °C/min to 600 °C under helium
(
50 mL/min). The signals were recorded by monitoring the des-
orbed ammonia with a TCD detector and a Pfeiffer Omnistar Quad-
rupole mass spectrometer, simultaneously.
Pyridine- and 2,6-dimethylpyridine-adsorbed infrared (Py-IR
and DMPy-IR) spectra were used to discriminate acid type of sam-
ples on a EQUINOX 70 (Bruker, Germany). Wafers of compressed
zeolite samples were mounted in the FTIR cell and degassed in vac-
À2
uum of 10 Pa at 400 °C for 2 h. Samples were saturated with pyr-
idine or 2,6-dimethylpyridine vapor at 30 °C for 10 min. After
equilibration, the samples were evacuated at 200 °C and 350 °C
to remove the excess of probe molecules, respectively, and the IR
À1
spectra were recorded in the range of 1400–1580 cm
and
À1
1
550–1700 cm , correspondingly.
CO-chemisorption was performed in the same apparatus as
-TPD. Prior to the test, catalyst samples were reduced in situ
(50 mL/min) at 400 °C for 4 h and then purged with
NH
in flow of H
3
2
He for 0.5 h at 450 °C. After being cooled to 50 °C, several pulses
of 5% CO/95% He were injected at regular intervals till the sample
was saturated. The average Pt particle size was calculated by
assuming an adsorption of one CO molecule per Pt atom.
2.6. Catalytic experiments
The isomerization of n-dodecane as the model reaction was car-
Fig. 1. Wide (a) and small (b) angle XRD patterns of different samples.
ried out in a continuous flow fixed-bed reactor under the condi-
tions of 2 MPa, the volumetric H
2
/n-alkane ratio of 600 and a
À1
peaks of the RZEO-1 sample indicate that the mild hydrothermal
treatment leads to the enhancement of crystallinity. This result
could be attributed to the dissolution of some amorphous phase
containing in the as-synthesized zeolite crystal structure during
the mild alkaline hydrothermal step. However, the intensity of
diffraction peaks weakens gradually from RZEO-1, RZEO-2 to
RZEO-3 after the hydrothermal treatment at increasing alkalinity.
Further increasing alkalinity leads to the disappearance of
H-ZSM-22 diffraction peaks (RZEO-4) due to the complete dissolu-
tion of the microporous phase. The crystallinity of the desilicated
DeZEO sample decreases obviously with very short treatment in
liquid hourly space velocity (LHSV) of 2.0 h . In each run, 5 mL
of the catalysts was loaded into the reactor. Prior to reaction, the
catalyst was reduced with hydrogen at 400 °C for 4 h and then
the temperature was lowered to the reaction temperature. Subse-
quently, the feedstock of n-dodecane was fed to the reactor by a
microscale pump at a certain flow rate. For the reaction test of each
catalyst sample, 5–6 temperature points were sampled with the
interval of 3 h for each sampling period, in which the temperature
was kept at a constant value. During a sampling period (3 h), the
liquid products were collected and analyzed by an offline gas chro-
matograph (Agilent 7890N) while the gas products by an on-line
gas chromatograph. After a sampling period, the temperature
was increased at a higher value for the next test till the maximum
conversion was reached, which led to the reaction tests on stream
about 20 h for each catalyst sample.
alkaline solutions, showing
dissolution.
a typical result of framework
The small angle range XRD patterns are shown in Fig. 1b. The
absence of the diffraction peaks in the region of 1° < 2h < 6° reveals
that the uniform mesoporous phase is absent in the parent zeolite
(
H-ZSM-22) and the desilicated sample (DeZEO). A single diffrac-
3
. Results and discussion
tion peak assigned to the (100) reflection of mesoporous material
starts appearing at about 2h = 2.35° from RZEO-1 sample, and the
intensity of this peak increases with the increasing recrystalliza-
tion degree for RZEO-2, RZEO-3, and RZEO-4, illustrating the
increasing content of the mesoporous phase in these materials,
for which the appearance of peaks at 2h = 4.04° and 4.66° attribu-
ted to the (110) and (200) reflections further indicates the pres-
ence of hexagonal pore symmetry in composites [32]. The
diffraction peak intensity of RZEO-4 indicates that this recrystal-
lized material is of the MCM-41 type material with high
3.1. Structure
The wide angle range XRD patterns of the parent H-ZSM-22, the
desilicated sample (DeZEO) and the different recrystallized materi-
als (samples RZEO-1, RZEO-2, RZEO-3, and RZEO-4) are plotted in
Fig. 1a. The calculated crystallinity is listed in Table 1. As shown
in Table 1, the as-synthesized H-ZSM-22 zeolite exhibits high crys-
tallinity [31] free from the impurities. The much sharper diffraction

1
4
S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
Table 1
Characteristic of different samples.
Samples
C
NaOH (M)
Si/Al (Chem. anal.)
Si/Al (XPS)
Phase
Crysta (%)
BETb
S
mic
S
ex
H-ZSM-22
DeZEO
RZEO-1
RZEO-2
RZEO-3
RZEO-4
–
31.76
25.14
31.84
32.18
32.35
33.07
25.09
20.14
44.07
51.23
52.40
52.89
TON
TON
TON
TON/MCM-41
TON/MCM-41
MCM-41
100
86
108
88
76
–
165.17
96.58
135.35
114.51
78.35
–
61.72
0.30
0.15
0.30
0.45
0.90
189.87
152.25
179.82
355.01
946.67
a
Relative crystallinity degree of calculated from XRD (normalized intensity of TON (100) peak).
b
2
Obtained by the N -adsorption at À196° using Micromeritic ASAP 2020.
crystallinity with complete disappearance of the parent ZSM-22
structure. Consequently, the XRD results illustrate that the recrys-
tallized micro/mesoporous samples (REZO-1, RZEO-2, and RZEO-3)
consist of both the H-ZSM-22 zeolitic phase and the MCM-41
mesoporous phase formed by the recrystallization treatment.
difference between bulk and surface Si/Al ratios, the Al distribution
in the parent zeolite is uneven, revealing that the aluminum is con-
centrated in the rim, which is in line with those obtained by Haya-
saka et al. [35]. After alkaline treatment, the bulk and surface Si/Al
ratios of the DeZEO sample drop from 32.76 and 25.09 to 25.14 and
2
0.14, respectively, indicating that the alkaline treatment mainly
dissolves silica from the silicon rich framework of H-ZSM-22. The
recrystallization leads to the slightly increase of bulk Si/Al ratios
from original 31.76 to 31.84 for RZEO-1, 32.18 for RZEO-2, 32.25
for RZEO-3, and 33.07 for RZEO-4, respectively, and the surface
Si/Al ratios increase remarkably from original 25.09 to 44.07 for
RZEO-1, 51.23 for RZEO-2, 52.40 for RZEO-3, and 52.89 for RZEO-
3
.2. Morphology and elements distribution
The SEM images are presented in Fig. 2. The as-synthesized H-
ZSM-22 zeolite displays the needle-shaped crystals, showing char-
acteristic morphology of TON-type zeolites [33]. After the
hydrothermal treatment in a mild alkali solution with CTAB, the
crystal of the RZEO-1 sample becomes rough probably due to the
formation of a small amount of mesoporous phase on the outer
surfaces of the original well crystallized H-ZSM-22. With the
increasing alkalinity, the mesoporous crystals become thicker in
the order of RZEO-1, RZEO-2 and RZEO-3. The morphology of
RZEO-4 sample shows lamellar particles of mesoporous material,
suggesting completely dissolving of the original H-ZSM-22 crystals
and the reorganization of structures into the MCM-41 type phase
with the help of CTAB. However, the sample DeZEO after desilica-
tion procedure almost keeps the original needle-like morphology.
More detailed information concerning the structure of different
samples is shown in TEM images (Fig. 3). The TEM image of H-
ZSM-22 shows that the crystal size is uniform with the dimen-
sional size of 700–900 nm in length and 30–50 nm in diameter
and in the absence of any secondary phase. On the contrary, the
image of RZEO-1 displays a thin film of worm-like mesoporous
solid on the outer surface of the H-ZSM-22 crystals. In the case
of the RZEO-2 and RZEO-3 samples, the mesoporous phase grows
and develops into obvious layers of the mesoporous phase, which
fully covers the original crystal rods. These observations provide
the direct evidences that the dissolving of H-ZSM-22 from outer
surfaces and assembling of these dissolved species into meso-
porous phases with the help of CTAB are the major separate pro-
cesses occurring on the outer surfaces of the original needle-like
nanorods of H-ZSM-22. The recrystallized RZEO-4 shows complete
change of the needle-like crystals into the lamellar ones with exis-
tence of highly ordered hexagonal array. In the case of DeZEO sam-
ple, the external surface of individual nanorods roughens, while
the crystals are slimmer. The roughened external surface of the
DeZEO sample is main source of the intra-crystalline mesopores
4
, respectively, indicating that the most of silicon dissolved by
alkaline treatment can be reassembled in the recrystallized process
and the dissolved Al, which is in smaller amount than that con-
tained in the dissolved structure of original zeolites, is also incor-
porated into the meso-pore structures. Combined with the TEM
images and the XPS analysis (with depth of the order of a few
nanometers [36]), it is verified that the mesoporous material has
been formed at the outer-surface of the alkaline treated H-ZSM-
2
2 structure by the dissolved Si and very small amount of Al from
the parent zeolite during the recrystallization step.
3.3. Coordination properties
To confirm the coordination environment for the aluminum, the
Al MAS NMR spectra are performed (Fig. 4a). In the spectra of the
2
7
parent H-ZSM-22 and recrystallized micro/mesoporous materials,
there are two signals with chemical shifts at 54.5 and 0 ppm corre-
sponding to the tetrahedral Al in the framework and the octahedral
form of the extra-framework Al, respectively [37]. The high tem-
perature calcination steps account for the presence of extra-
framework aluminum sites [38]. The decreased intensity of signal
at 0 ppm with the increasing recrystallization degree is the result
of the partial dissolution of tetrahedral and extra-framework Al.
Only one signal at 54.5 ppm is observed in the spectra of RZEO-4
material, revealing that the microporous zeolites were completely
dissolved and then these species reassembled into the high crys-
tallinity mesoporous structure during recrystallization steps with
extreme dissolution conditions. For the RZEO-1, RZEO-2, and
RZEO-3 samples, apart from the clear signals for the tetrahedral
Al in the framework (54.5 ppm), the signals at 0 ppm for the octa-
hedral form of the extra-framework Al are weaker than the original
H-ZSM-22 sample, indicating that the non-framework Al may be
partially dissolved and reorganized in the recrystallization treat-
ment. Furthermore, in the case of DeZEO, the peak at 0 ppm
becomes much weaker due to the partial removal of the extra-
framework Al in the acid-washing treatment [20].
[
34] formed from alkaline treatment, which are completely differ-
ent from the uniform mesoporous structure of the recrystallized
samples. Since the external surface of the recrystallized materials
is affected by alkaline treatment and coated by the new meso-
porous phase as confirmed by SEM and TEM, a large number of
Si species with a small amount of Al simultaneously in the frame-
work have been dissolved in the alkaline solution and then
reassembled with help of CTAB.
2
9
The results of Si MAS NMR are shown in Fig. 4b. The resonance
peaks at À114.5 and À105 ppm are assigned to Si(0Al) and Si(1Al)
respectively [39], suggesting that the parent H-ZSM-22 crystals are
defect-free [40]. The spectra of DeZEO sample are similar to those
The XPS and ICP-AES techniques are used to determine the
surface and bulk Si/Al ratios (Table 1), respectively. From the

S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
15
Fig. 2. SEM images of parent H-ZSM-22, desilicated sample and recrystallized materials.
of parent zeolite, implying no obvious destruction of the chemical
surroundings of framework Si during the desilication process. For
the recrystallized samples, the resonances broaden gradually from
RZEO-1 to RZEO-4. The peaks at around À100.5 ppm for RZEO-3
and at À100.5 ppm and À108.7 ppm for RZEO-4 can be assigned
to the Si(0Al) and Si(1Al) of MCM-41 mesoporous materials
presence of intra-crystalline mesopores. For the recrystallized
samples, the shift of desorption branch to a lower relative pressure
and the increased hysteresis loop suggest a partial loss of
micro-structural organization and the formation of mesopores.
Nevertheless, the RZEO-x samples (x = 1, 2, 3, 4) exhibit a rather
0
sharp isotherm at about P/P ꢀ 0.35, indicating the existence of
[
41,42].
.4. Pore properties
The N -adsorption isotherms of different samples are depicted
uniform mesopores with diameter of 3–4 nm [27].
The pore distribution for all samples is shown in Fig. 6. From the
pore distributions, it can be seen that the well distributed microp-
ores (about 5.0 Å) are presented in the parent H-ZSM-22, the alka-
line treated DeZEO, and the partially dissolved/recrystallized
samples (RZEO-1 to RZEO-3), while the absence of micropores in
REZO-4 is due to complete destruction of the microporous struc-
ture of H-ZSM-22 and the establishment of new mesoporous
phases in the recrystallization process. For the mesopores, the
3
2
0
in Fig. 5. Obviously, all samples exhibit steep raise at P/P < 0.01,
typical for the microporous structure. The H-ZSM-22 zeolite exhi-
bits a typical isotherm of pure microporous material, while the
desilicated DeZEO sample has a larger hysteresis loop due to the

1
6
S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
Fig. 3. TEM images of parent H-ZSM-22, desilicated sample and recrystallized materials.
parent H-ZSM-22 structure shows no mesopore distribution,
implying that the lateral surfaces of the rod shape H-ZSM-22 are
probably inert [43,44] for catalysis interests. The DeZEO from alka-
line treatment shows broad mesopore distribution in the range of
3.5. Acidity
3
The NH -TPD, FTIR spectra of pyridine adsorption (Py-FTIR),
FTIR spectra of 2,6-dimethylpyridine (DMPy-FTIR) adsorption were
used for characterizing the acidity of all samples. The NH -TPD
2
–30 nm with the maximum at about 7.5 nm, indicating that the
3
framework was dissolved irregularly due to the absence of the
template [45] and the alkaline treatment creates significant defects
curves are shown in Fig. 7, and all FTIR spectra and the correspond-
ing quantitative calculation method can be found in Supplemen-
tary Information (SI) (Figs. S1 and S2). The results of total acidity
distributions from these characterizations are listed in Table 2.
The parent H-ZSM-22 has clearly strong and weak adsorption
peaks, and the DeZEO sample exhibits a slight decrease in the both
peaks due to the partial withdrawal of silica species from the
(
irregular holes), which are active for adsorption with probably
catalysis potentials [16], on the lateral surfaces of the rod struc-
ture. The recrystallized samples show sharp mesopore distribu-
tions of increasing narrowing at about 3 nm with increased
intensity in the order of RZEO-1, RZEO-2, RZEO-3, and RZEO-4 in
agreement of conditions used in the post treatment, showing the
obvious difference of the mesopores from those formed in DeZEO
sample. This difference may be due to the absence of the template
3
framework as shown in Fig. 7. The NH -TPD results also show that
the recrystallization at changing conditions (RZEO-1 to RZEO-4)
leads to an enhancing trend of the decrease in the amount and
strength of acid sites with the high temperature desorption peak
shifting to lower temperature and greatly weakening (Fig. 7). This
can well be explained by the dissolving of the parent H-ZSM-22
structure and the reassembling of the new mesoporous structure
with weak acidity.
[
45] during the preparation for DeZEO sample and also the recrys-
tallization of dissolved species, which may start building up
around the defects created in the dissolution procedure.
The results obtained from the textural analysis of all samples
are presented in Table 1. On the basis of the as-synthesized H-
ZSM-22 zeolite with only micropores, the introduction of intra-
crystalline mesoporosity by desilication (DeZEO) leads to the
increase of external surface area and the decrease of microporous
area after alkaline treatment [20]. The RZEO-x samples display
much higher BET surface area, especially the external surface area,
suggesting the formation of the uniform mesoporous structure
with the relative high microporous area kept by the proper control
of recrystallization conditions.
3
The results of weak and strong acid sites obtained from NH -
TPD tests (Table 2) all indicate an obvious trend of decreasing acid-
ity in H-ZSM-22 > DeZEO > RZEO-1 > RZEO-2 > RZEO-3 > RZEO-4.
Further discrimination tests with Py-FTIR and DMPy-FTIR are sum-
marized in Table 2 for the distributions of the different acid sites
[46,47]. The Py-FTIR results reveal that the B acid sites follow a
clear decreasing trend in H-ZSM-22 > DeZEO > RZEO-1 > RZEO-2 >
RZEO-3 > RZEO-4, and the L acid sites follow increasing trend in

S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
17
Fig. 4. 27Al (a) and 29Si (b) MAS NMR spectra of different samples.
Fig. 6. Micropore (a) and mesopore (b) size distribution.
2
Fig. 5. N -adsorption isotherms of parent H-ZSM-22, desilicated sample and
recrystallized materials.
3
Fig. 7. NH -TPD profiles of different samples.
H-ZSM-22 < RZEO-1 < RZEO-2 < RZEO-3 < RZEO-4 < DeZEO.
The
results indicate that both the destruction of microporous struc-
tures in the H-ZSM-22 framework and the formation of meso-
porous structure cause the decrease of B acidity and the increase
of L acidity [23,48,49].
The DMPy-IR results show the variation of the B acidity at out-
side of micropores (at micropore pore mouths and in mesopores)
[50–52]. These quantitative results in Table 2 show that the strong
B acid sites (measured at 350 °C) follow a clear decreasing trend in

1
8
S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
Table 2
The acidity of different samples determined by NH
3
-TPD, Py-FTIR and DMPy-FTIR spectra.
Aciditya
(lmol/g)
Acid typee
(lmol/g)
B acid sites (
f
lmol/g)
Samples
b
c
d
A
w
A
m
A
s
B
2
L
00 °C
350 °C
200 °C
350 °C
200 °C
350 °C
H-ZSM-22
DeZEO
RZEO-1
RZEO-2
RZEO-3
RZEO-4
201
175
188
180
165
88
–
–
–
–
–
65
184
179
168
146
95
61
52
50
48
47
35
59
49
31
26
23
12
60
80
68
70
71
72
46
58
50
53
51
52
29
33
32
38
38
26
30
28
19
19
18
10
–
a
b
c
d
e
f
3
Obtained from the NH -TPD spectra.
Weak acid sites estimated from the relative area of the deconvoluted peak.
Medium acid sites estimated from the relative area of the deconvoluted peak.
Strong acid sites estimated from the relative area of the deconvoluted peak.
Acidity calculated from the peak areas of FT-IR spectra by pyridine desorption at different temperatures.
Brönsted acidity calculated from the peak areas of FT-IR spectra by 2,6-dimethlypyridine desorption at different temperatures.
H-ZSM-22 > DeZEO > RZEO-1 > RZEO-2 > RZEO-3 > RZEO-4, indi-
cating the suppressing effect of the dissolving and recrystallization
treatments on the strong B sites [53,37]. However, for mild B sites
2.20 nm in size, which are too large to penetrate into the microp-
orous channels and thus locate on the external surface of H-ZSM-
22 nanorods and the binders randomly. The introduction of
intra-crystalline mesoporosity by desilication treatment (DeZEO)
hardly has a significant influence on the location and crystal size
of the Pt particles. However, for the reduced Pt/RZEO-2 catalysts,
the Pt particles tend to locate on the uniform mesoporous layers,
and the Pt crystal size decreases obviously, which is in line with
the CO-chemisorption results. The Pt/RZEO-4 catalyst with broad
channels and high surface areas has the smallest metal particles
(
200 °C), samples RZEO-2 and RZEO-3 show the highest adsorption
amount while the RZEO-4 shows the lowest adsorption amount
among all the samples evaluated, implying that the proper control
of the recrystallization conditions can enhance the mild B acidity,
some of which are on the mouths of micropores and some are in
the mesopores.
(
<1.50 nm) confined within the mesopores. It should be noted that
3
.6. Metallic properties
the current work is focusing on the influences of different struc-
tures of acidic supports on the isomerization reaction and more
efforts for improving the balance and contact between acidic sites
and metal sites should be made in further studies
The balance between the (de)hydrogenation function and the
acidity function plays a significant role in a bifunctional isomeriza-
tion catalyst. The Pt dispersion and the average platinum particle
size are determined by CO-chemisorption, and the results are listed
in Table 3. The catalyst samples are designed by loading 0.5 wt.% Pt
on the synthesized support samples, namely Pt/H-ZSM-22, Pt/
DeZEO, Pt/RZEO-1, Pt/RZEO-2, Pt/RZEO-3, and Pt/RZEO-4. From
the CO-chemisorption results, it can be found that the metal dis-
persions for all the preliminary samples are in the increasing order
3
.7. Catalytic performance
The n-dodecane isomerization is tested over the catalysts, Pt/H-
ZSM-22, Pt/DeZEO, Pt/RZEO-1, Pt/RZEO-2, Pt/RZEO-3, and Pt/RZEO-
, which are prepared with the same procedure for all the support
samples simply by impregnation in a solution of H PtCl O. The
4
of
3
Pt/H-ZSM-22ꢀPt/DeZEO < Pt/RZEO-1 < Pt/RZEO-2 < Pt/RZEO-
2
6
Á6H
2
< Pt/RZEO-4 and the particle sizes are in the opposite order, indi-
conversion and isomerization selectivity results of the n-dodecane
reactant under different temperatures are shown in Fig. 9.
cating that the recrystallization treatment for the supports can
enhance the metal dispersion under the same impregnation condi-
tions due probably to the enriched surfaces from the formation of
mesoporous structures. It should be noted that the formation of
the mesoporous structure may be the essential factor for the higher
metal dispersion observed in this study. The formed pores on the
mesopore layer with typical sizes of 3 nm will provide more ideal
space for the small Pt particles 1–2 nm being confined in the meso-
porous structure than the outer surface of the H-ZSM-22, which is
From the conversion results, it is clearly shown that the activi-
ties of all the catalyst samples in Fig. 9a, for n-dodecane isomeriza-
tion/cracking follow the same trends as the B acid site variation,
suggesting that the stronger B acidity accounts for the higher activ-
ity. The recrystallized materials are obviously less active than the
purely microporous H-ZSM-22 zeolites, and require higher temper-
ature to achieve an equivalent conversion level. The reaction tem-
perature of Pt/RZEO-1 catalyst is 5 °C higher than that of Pt/H-ZSM-
rather inert for catalysis purposes indicated by N
analysis.
2
-adsorption
2
2 for the same conversion level, and increases with the enhance-
ment of recrystallization degree from Pt/RZEO-1 to Pt/RZEO-4. The
lower catalytic activity of the recrystallized materials is in agree-
ment with the decrease of the amount of strong B acid sites at pore
mouths, which has been believed to be the active sites for long
chain n-alkane isomerization/cracking [54–56]. It should be noted
that the catalyst Pt/RZEO-4 with completely dissolved and recrys-
tallized support (mesoporous material), on which no strong B site
available, needs very high temperature (360–380 °C) to reach the
high conversion (about 90%). At the high temperature, n-
dodecane isomerization/cracking may have both catalytic and
thermal behaviors, in which the thermal reactions of n-dodecane
indicated by our check test (as shown in Table S1) are mainly
cracking into lighter hydrocarbons. The thermal reactions of the
current feed mainly occur at high temperature above 380 °C with
TEM images of the reduced catalysts are shown in Fig. 8. For the
Pt/H-ZSM-22 catalyst, it is shown that the Pt particles are around
Table 3
Result of CO-chemisorption of different catalysts.
Pt crystal sizea (nm)
Catalysts
Pt loading (wt.%)
0.5
Pt dispersion (%)
Pt/H-ZSM-22
Pt/DeZEO
Pt/RZEO-1
Pt/RZEO-2
Pt/RZEO-3
Pt/RZEO-4
47.29
46.12
54.14
59.61
68.95
71.95
2.1
2.1
2.1
2.0
1.4
1.3
a
Average of Pt particle sizes, assuming homogeneous semispherical Pt particle.

S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
19
Fig. 8. TEM images and particle sizes of four representative catalysts after reduction.
detectible conversions from 1.1% at 380 °C to 5.3% at 420 °C, indi-
cating that in most cases investigated, thermal reactions will not
bias the isomerization results in this study. Compared with the
Pt/RZEO-x catalysts, the catalytic activity of Pt/DeZEO is high, keep-
ing the same level as the catalysts with the original H-ZSM-22 sup-
port, especially above 280 °C. This phenomenon may be explained
by the canceling effect of the suppression of B acidity and creation
of defects (holes) on the lateral surface of the H-ZSM-22 rods by
dissolving of ZSM-22 framework, leading to no obvious loss of
the activity.
The total isomerization selectivity results in Fig. 9b show that
the selectivity to isomerization decreases with the conversion
increase, indicating that higher conversion leads to higher cracking
rate. It is notable that the sample Pt/RZEO-2 shows the highest
total selectivity to isomerization in a very slow decrease trends
with the conversion increase, showing a potential candidate for
an isomerization catalyst. The other samples show a significant
decrease of the total isomerization selectivity with the conversion
increase. In all, the total isomerization selectivity of all samples
follows the order of Pt/RZEO-2 > Pt/RZEO-3 > Pt/RZEO-1 > Pt/
DeZEO > Pt/RZEO-4 > Pt/H-ZSM-22.
The selectivity behaviors of the series of samples are in general
related to the properties of catalysts used in this study. The first
type of catalyst is the Pt/H-ZSM-22 with the highest cracking activ-
ity and low isomerization selectivity due to the strong B acidity
nature of the microporous H-ZSM-22 support. The second type is
the Pt/DeZEO catalyst with improved isomerization selectivity
and without loss of activity due to suppression to the original
strong B acidity, probably formation of the defects and the irregu-
lar mesoporous deposition on the outer (lateral) surfaces of the
original H-ZSM-22 support by the dissolution treatment. The third
type of catalysts is the Pt/RZEO-x (x = 1,2,3) with greatly improved
isomerization selectivity and loss of activity due to the greatly sup-
pression of strong B acidity, formation of the defects, the well-
developed mesoporous layers on the outer (lateral) surfaces of
the original H-ZSM-22 rod shaped support by the dissolution and
the recrystallization treatment. The last type of catalyst is the
Pt/RZEO-4 with slightly improved isomerization selectivity due to
the complete loss of strong B acidity, leaving mild B sites only,
the complete destruction of the original H-ZSM-22 structure
and the formation of the mesoporous structure by the dissolution
and the recrystallization treatment.
In addition, the further results of the selectivities to mono-
branched isomers and to the multi-branched isomers are shown
in Fig. 10a and b, respectively. It is clear that the selectivity to
the mono-branched isomers follows the same order as that of
the total selectivity for the samples Pt/RZEO-2, Pt/RZEO-3, Pt/
RZEO-1, and Pt/DeZEO, while the complete recrystallized sample
Pt/RZEO-4 has the lowest mono-branched isomer selectivity and
the Pt/H-ZSM-22 catalyst has higher mono-branched isomer selec-
tivity than the Pt/RZEO-4. For the multi-branched isomer selectiv-
ity, the order becomes Pt/RZEO-4 > Pt/RZEO-3 > Pt/RZEO-2 > Pt/
RZEO-1ꢀPt/DeZEO > Pt/H-ZSM-22.
3
.8. Structure effect on isomerization performance
Modification of the H-ZSM-22 structure can greatly improve the
isomerization performance of Pt/H-ZSM-22 type catalysts [57], and
this has been revealed systematically in this study. However, it is
desirable that the isomerization catalyst should have both good
activity and appropriate selectivity, not only to the mono-
branched isomers but also to the multi-branched isomers. Accord-
ing to classical bifunctional mechanism of hydro-isomerization/
cracking of hydrocarbon chains, the mono- and multi-branched

2
0
S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
Fig. 10. Selectivity of mono- (a) and multi-branched (b) isomers versus conversion
Fig. 9. Conversion (a) and selectivity (b) of n-dodecane isomerization over different
À1
over different catalysts (Reaction condition: H
2
/n-dodecane = 600, LHSV = 2 h
,
À1
bifunctional catalysts (Reaction condition:
2
H /n-dodecane = 600, LHSV = 2 h
,
P = 2 MPa).
P = 2 MPa).
Table 4
Results of n-dodecane isomerization over different catalysts.
Catalysts
Pt/H-ZSM-22
Pt/H-DeZEO
Pt/H-RZEO-2
Pt/H-RZEO-4
Temperature (°C)
290
290
300
370
Conversion (%)
90.26
47.09
42.77
4.32
90.24
60.70
52.52
8.18
91.32
91.70
73.43
18.27
4.02
84.71
50.29
19.75
30.54
0.65
S
S
S
S
T
(%)
MoB (%)
MuB (%)
MoB/SMuB
9.90
6.42
S (%)
2
3
4
5
6
2
2
2
2
2
2
3
-Methylundecane
15.52
10.84
6.42
7.05
2.94
0.25
0.37
0.38
0.59
0.84
1.48
0.28
0.12
0.01
17.70
14.14
8.30
8.01
4.37
0.69
0.68
0.79
0.92
1.75
1.89
0.59
0.82
0.05
26.63
18.69
10.72
10.04
7.35
1.09
1.56
2.16
2.25
3.87
4.27
0.95
2.01
0.11
5.32
4.59
3.75
3.40
2.69
3.36
3.38
2.91
2.97
3.82
3.94
4.14
4.12
1.90
-Methylundecane
-Methylundecane
-Methylundecane
-Methylundecane
,4-Dimethyldecane
,5-Dimethyldecane
,6-Dimethyldecane
,7-Dimethyldecane
,8-Dimethyldecane
,9-Dimethyldecane
,x-Dimethyldecane
Methylethyl nonane
Other isomers
S
T
, SMoB, SMuB and S corresponded to the selectivity of total isomers, mono-branded isomers, multi-branched isomers and different main products, respectively, at about 90% of
conversion.

S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
21
isomers are formed successively along with the cracking of carbon
chains via carbenium ion mechanism [58], in which the formation
of carbenium ions on typically B acid sites is essentially important
for the catalytic activity for both isomerization and cracking reac-
tions. Apart from the B acid sites, the metal sites are of importance
in the hydro-isomerization/cracking catalysis for an instant hydro-
genation function for fast hydrogenating the unsaturated interme-
diates formed via carbenium ion mechanism over B sites, avoiding
secondary (side) reactions. It is generally understood that the bal-
ance and the intimate contact between metal sites and B acid sites
are essentially important for practical catalysts for the isomeriza-
tion of hydrocarbon chains [59]. In this study, the metal sites are
provided by Pt dispersed on the acidic supports with a same
impregnation procedure for all support samples to reflect the dif-
ferences of catalytic behaviors of the catalysts using different sup-
ports, and the optimization of the balance/contact between acidic
sites and metal sites may need to develop different preparation
procedures and will be the important direction for future studies.
The n-dodecane isomerization products for four representative
catalyst samples, Pt/H-ZSM-22, Pt/DeZEO, Pt/RZEO-2, and Pt/
RZEO-4, are exhaustively analyzed and shown in Table 4. The Pt/
H-ZSM-22, Pt/DeZEO, and Pt/RZEO-2 catalysts exhibit higher selec-
tivity to mono-branched isomers than that of multi-branches,
while the Pt/RZEO-4 catalyst is in the opposite order. The
2
-methylundecane and 3-methylundecane are the main mono-
branched isomers over Pt/H-ZSM-22, Pt/DeZEO, and Pt/RZEO-2
catalysts. The methyl branching preference decreases toward the
center of the chain. This result is in good agreement with the ‘‘pore
mouth” shape selectivity model, which favors branching at 2-C
and/or 3-C positions and is less favorable at more central positions
along the carbon chain due to the partially n-alkane penetration
into one single pore opening [13,14], suggesting the mono-
branched isomers are formed in the pore mouth of zeolite
micropores. For the selectivity of di-branched isomers, the
2
,x-methyl-nonane and 3,x-methyl-nonane isomers are the most
abundantly formed species over these three catalysts, which
implies that the formation of di-branched isomers may be through
a reaction step consecutive to the mono-branched products
according to the bifunctional mechanism. Additionally, the
selectivity to 2,x-methyl-nonane isomers deceases in the order of
2
2
,9-dimethylnonane > 2,8-dimethylnonane > 2,7-dimethylnonane >
,6-dimethylnonane > 2,5-dimethylnonane > 2,4-dimethylnonane,
Fig. 11. Schematic representation of the structure change from original H-ZSM-22
by dissolution and/or recrystallization treatment, forming the treated samples in
this study: DeZEO, REZO-1, RZEO-2, RZEO-3, and RZEO-4.
illustrating that the preferred position for the second branching is
located at the symmetric carbon atoms at the two side of a carbon
chain as explained by the ‘‘key-lock” shape selectivity model,
further inferring that the formation of methyl isomer mainly con-
tributed by the acid sites of the micropores. The Pt/RZEO-4 catalyst
exhibits entirely different distribution of methyl-undecane or
di-methyl-decane positional isomers with respect to that of the
other catalysts, due to the effect of broad channels, resulting in
the probability of formation of methyl-undecane positional
isomers is equivalent.
It is evident that the defects created by dissolution of the
H-ZSM-22 structure in the supports DeZEO and RZEO-x (x = 1,2,3)
with varying suppression of B acidity can improve the mono-
branch selectivity and the formation of the mesoporous structure
on the outer surfaces of the H-ZSM-22 support can promote the
formation of the multi-branch isomers.
At this point, it is necessary to discuss the structure change
along with the variation of B acidity of all the supports. Combining
The Pt/H-ZSM-22 catalyst with the metal dispersed on the
microporous H-ZSM-22 support of strong B acidity shows the high-
est conversion of n-dodecane, the lowest total selectivity and
multi-branch selectivity, implying that the strong B acidity is
responsible for the high cracking activity and thus low isomeriza-
tion selectivity. The higher mono-branch selectivity of the Pt/H-
ZSM-22 catalyst than that of the complete mesoporous supported
Pt catalyst Pt/RZEO-4 with only mild B acid sites may reflect the
contribution of the microporous structure to the mono-branched
isomers with shape selectivity [60,61], while the contribution of
the mesoporous structure in Pt/RZEO-4 to the higher multi-
branch shape selectivity [62,63]. It should be noted that meso-
porous Pt/RZEO-4 catalyst shows too low activity due to its too
weak B acidity.
2
the XRD (Fig. 1), TEM (Fig. 3), MAS NMR (Fig. 4), and N -adsorption
(Fig. 6) results, these changes can be represented schematically by
the following diagram shown in Fig. 11. This model is also partially
supported by the results of Martens et al. [58]. The original H-ZSM-
22 of rod shape (about 30–50 nm in diameter, and 700–900 nm in
length) with only one-dimensional tubular micropores of about
0.5 nm has been known to have B acid sites distributed in the
micropores and on the pore mouths [64,65], forming main active
surfaces, on which Pt is dispersed as metal sites. In such environ-
ment, hydrocarbon chains have limited interaction with active
sites, the end parts of the chains may easily insert into the micro-
pores [66], and the point of the chain on the pore mouth may
access the both the B acid and metal sites, and thus easily undergo
isomerization/cracking reactions, leading to the formation of

2
2
S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
mainly mono-branch isomers or cracking products. The cracking
selectivity is high because of the strong B acidity [57]. The dissolu-
tion sample DeZEO with the same rod shape as the original zeolite
crystals and with both the similar original micropores and the
mesopores, which are mainly the defects or holes formed by alka-
line treatment on the originally inert lateral surfaces of the crystal
rods, has slightly less strong B acidity, greatly improving the mono-
branch selectivity due to lowered cracking activity of the reduced B
acid strength as compared with the original Pt/H-ZSM-22 catalyst.
Due to the existence of the mesoscale holes created by alkaline
treatment on the external/lateral surfaces of the DeZEO rods, the
multi-branch selectivity is improved compared to that of
Pt/H-ZSM-22. The RZEO-1, RZEO-2 and RZEO-3 with both the
micropores in the cores of the rods and the well-established meso-
porous layer (mesopore of 3 nm in diameter) on the outer surfaces
of the microporous core rods have gradually lowered B acidity
located on the micropore mouths and in the micropores and prob-
ably on the mesopore surfaces (mild B acid sites), forming the
major active sites. In such micro/mesoporous environment with
reduced strong B acidity, the hydrocarbon chains will interact with
not only the microporous structure but also the mesoporous
structure. The lowered B acidity suppresses the activity of the n-
dodecane conversion by mainly reducing the cracking possibility
and greatly improves the isomerization selectivity, and, however,
reduction of microporous structure may cause the decrease in
mono-branch selectivity, leading to the existence of an optimal
recrystallization degree (RZEO-2 in this study) for high isomeriza-
tion selectivity. Furthermore, it is interesting that the multi-
branched isomers selectivity is significantly improved by the
enhanced portion of mesoporous structure. One reason is that
the outer layered mesoporous structure would put constrains on
the hydrocarbon reacting with the acid site at the microporous
structure, making the longer residence time of the reactants in
the interconnectivity between mesoporous layers and the new
opening pore mouths of the zeolites external surface and thus
increasing the secondary isomerization opportunity on the micro-
porous pore mouths [67]. Nevertheless, the dispersed Pt sites are
very important for the reactivity of the hydrocarbon conversion.
It is suspected that the formation of the mesoporous layer on the
outer surfaces of the micro-porous structure could provide better
environment for the Pt dispersion, which would improve the con-
tact between Pt sites and acid sites. This should be further
addressed by both theoretical and experimental investigation, for
example, the question on where the Pt particles should be located
with respect to the B acid sites.
mesoporous layers probably accounts for the increment of multi-
branched isomers by providing proper condition to further form
isomers and/or making the longer residence time of reactants in
micropores and thus increasing the secondary isomerization
opportunity on microporous surfaces. This present study highlights
a further understanding of the effect of uniform mesoporous struc-
ture as well as suitable acid sites distribution on the n-alkanes
isomerization.
Acknowledgments
The authors are grateful to Synfuels China Co., Ltd., and Beijing
Key Laboratory of Coal to Cleaning Liquid Fuels for financial
support.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2015.12.009.
References
[
1] A. Primo, H. Garcia, Chem. Soc. Rev. 43 (2014) 7548.
[2] S.J. Miller, Micropor. Mater. 2 (1994) 439.
[3] C. Bouchy, G. Hastoy, E. Guillon, J.A. Martens, Oil Gas Sci. Technol. Rev. IFP 64
(
2009) 91.
[
4] J.H. Gregor, Catal. Lett. 7 (1990) 317.
[5] A. Deklerk, Catal. Today 130 (2008) 439.
[6] F.A.N. Fernandes, U.M. Teles, Fuel Process. Technol. 88 (2007) 207.
[
[
7] N. Batalha, L. Pinard, C. Bouchy, E. Guillon, M. Guisnet, J. Catal. 307 (2013) 122.
8] F. Marques Mota, C. Bouchy, E. Guillon, A. Fécant, N. Bats, J.A. Martens, J. Catal.
3
01 (2013) 20.
[9] G.G. Martens, G.B. Marin, J.A. Martens, P.A. Jacobs, G.V. Baron, J. Catal. 195
2000) 253.
(
[
[
10] Y. Liu, C. Liu, C. Liu, Z. Tian, L. Lin, Energy Fuels 18 (2004) 1266.
11] T.L.M. Maesen, M. Schenk, T.J.H. Vlugt, J.P.d. Jonge, B. Smit, J. Catal. 188 (1999)
403.
[
[
[
12] L. Domokos, L. Lefferts, K. Seshan, J.A. Lercher, J. Catal. 203 (2001) 351.
13] M.C. Claude, J.A. Martens, J. Catal. 190 (2000) 39.
14] M.C. Claude, G. Vanbutsele, J.A. Martens, J. Catal. 203 (2001) 213.
[15] B.C. Gagea, Y. Lorgouilloux, Y. Altintas, P.A. Jacobs, J.A. Martens, J. Catal. 265
2009) 99.
(
[
16] J.A. Martens, D. Verboekend, K. Thomas, G. Vanbutsele, J. Pérez-Ramírez, J.-P.
Gilson, Catal. Today 218–219 (2013) 135.
[17] K. Na, M. Choi, R. Ryoo, Micropor. Mesopor. Mater. 166 (2013) 3.
[18] D.P. Serrano, J.M. Escola, P. Pizarro, Chem. Soc. Rev. 42 (2013) 4004.
[
[
19] M.Y. Kim, K. Lee, M. Choi, J. Catal. 319 (2014) 232.
20] D. Verboekend, A.M. Chabaneix, K. Thomas, J.-P. Gilson, J. Pérez-Ramírez,
CrystEngComm 13 (2011) 3408.
[21] D. Verboekend, J. Perez-Ramirez, ChemSusChem 7 (2014) 753.
[
[
[
[
22] W.C. Yoo, X. Zhang, M. Tsapatsis, A. Stein, Micropor. Mesopor. Mater. 149
(
2012) 147.
23] Y.P. Khitev, Y.G. Kolyagin, I.I. Ivanova, O.A. Ponomareva, F. Thibault-Starzyk, J.
P. Gilson, C. Fernandez, F. Fajula, Micropor. Mesopor. Mater. 146 (2011) 201.
24] I.I. Ivanova, A.S. Kuznetsov, E.E. Knyazeva, F. Fajula, F. Thibault-Starzyk, C.
Fernandez, J.P. Gilson, Catal. Today 168 (2011) 133.
25] V.V. Ordomsky, V.Y. Murzin, Y.V. Monakhova, Y.V. Zubavichus, E.E. Knyazeva,
N.S. Nesterenko, I.I. Ivanova, Micropor. Mesopor. Mater. 105 (2007) 101.
26] I.I. Ivanova, E.E. Knyazeva, Chem. Soc. Rev. 42 (2013) 3671.
27] I.I. Ivanova, I.A. Kasyanov, A.A. Maerle, V.I. Zaikovskii, Micropor. Mesopor.
Mater. 189 (2014) 163.
4
. Conclusions
A series of micro/mesoporous materials with different meso-
[
[
porous layers are successfully synthesized by the hydrothermal
recrystallization of the H-ZSM-22 zeolite in different alkalinity
with template, and their long-chain n-alkanes isomerization per-
formances are studied. The dissolution and recrystallization proce-
dures introduce the well-established mesoporous layer with pores
of about 3 nm in diameter on the outer surfaces of the microporous
H-ZSM-22 zeolites, forming the micro/mesoporous materials,
which have both increased weak B acid sites and reduced amount
of total acid sites locating on the micropore pore mouths and in the
micropores and probably on the mesopore surfaces (mild B acid
sites), forming the major active sites. Consequently, these catalysts
exhibit low cracking activity and rather high isomerization selec-
tivity, while decrease of microporous structure may cause the
decrease of mono-branched isomers selectivity, leading to the exis-
tence of an optimal recrystallization degree (RZEO-2) for high iso-
merization selectivity at even high conversion. The presence of
[28] V.V. Ordomsky, I.I. Ivanova, E.E. Knyazeva, V.V. Yuschenko, V.I. Zaikovskii, J.
Catal. 295 (2012) 207.
[
29] S.V. Konnov, I.I. Ivanova, O.A. Ponomareva, V.I. Zaikovskii, Micropor. Mesopor.
Mater. 164 (2012) 222.
[30] Y.P. Khitev, I.I. Ivanova, Y.G. Kolyagin, O.A. Ponomareva, Appl. Catal. A: Gen.
41–442 (2012) 124.
4
[
31] M.A. Asensi, A. Corma, A. Martınez, M. Derewinski, J. Krysciak, S.S. Tamhankar,
´
Appl. Catal. A: Gen. 174 (1998) 163.
[32] P. Selvam, S.K. Bhatia, C.G. Sonwane, Ind. Eng. Chem. Res. 40 (2001) 3237.
[33] D. Masih, T. Kobayashi, T. Baba, Chem. Comm. (2007) 3303.
[
34] D. Verboekend, K. Thomas, M. Milina, S. Mitchell, J. Pérez-Ramírez, J.-P. Gilson,
Catal. Sci. Technol. 1 (2011) 1331.
[
35] K. Hayasaka, D. Liang, W. Huybrechts, B.R. De Waele, K.J. Houthoofd, P. Eloy, E.
M. Gaigneaux, G. van Tendeloo, J.W. Thybaut, G.B. Marin, J.F. Denayer, G.V.
Baron, P.A. Jacobs, C.E. Kirschhock, J.A. Martens, Chemistry 13 (2007) 10070.
36] K. Raabova, E. Bad’urova, R. Bulanek, P. Eloy, E.M. Gaigneaux, Catal. Sci.
Technol. 3 (2013) 1634.
[
[37] D. Verboekend, J. Perez-Ramirez, Catal. Sci. Technol. 1 (2011) 879.

S. Liu et al. / Journal of Catalysis 335 (2016) 11–23
23
[
[
38] M. Derewinski, P. Sarv, A. Mifsud, Catal. Today 114 (2006) 197.
39] J.J. Williams, Z.A.D. Lethbridge, G.J. Clarkson, S.E. Ashbrook, K.E. Evans, R.I.
Walton, Micropor. Mesopor. Mater. 119 (2009) 259.
[53] Danny Verboekend, Gianvito Vile, Javier Perez-Ramirez, Cryst. Growth Des. 12
(2012) 3123.
[54] Y. Ono, Catal. Today 81 (2003) 3.
[
[
[
[
[
[
40] C.A. Fyfe, G.T. Kokotailo, H. Strobl, C.S. Pasztor, G. Barlow, S. Bradley, Zeolites 9
[55] J.F. Denayer, W. Souverijns, P.A. Jacobs, J.A. Martens, G.V. Baron, J. Phys. Chem.
B 102 (1998) 4588.
[56] J.F. Denayer, G.V. Baron, G. Vanbutsele, P.A. Jacobs, J.A. Martens, Chem. Eng. Sci.
54 (1999) 3553.
[57] G. Wang, Q. Liu, W. Su, X. Li, Z. Jiang, X. Fang, C. Han, C. Li, Appl. Catal. A: Gen.
335 (2008) 20.
[58] H. Deldari, Appl. Catal. A: Gen. 293 (2005) 1.
[59] J.A. Martens, D. Verboekend, K. Thomas, G. Vanbutsele, J.-P. Gilson, J. Pérez-
Ramírez, ChemSusChem 6 (2013) 421.
[60] R. Kenmogne, A. Finiels, C. Cammarano, V. Hulea, F. Fajula, J. Catal. 329 (2015)
348.
(
1989) 531.
41] S.M. Holmes, V.L. Zholobenko, A. Thursfield, R.J. Plaisted, C.S. Cundy, J. Dwyer, J.
Chem. Soc. Fara. Trans. 94 (1998) 2025.
42] R. Takahashi, S. Sato, T. Sodesawa, M. Kawakita, K. Ogura, J. Phys. Chem. B 104
(
2000) 12184.
43] J.A. Martens, W. Souverijns, W. Verrelst, R. Parton, G.F. Froment, P.A. Jacobs,
Angew. Chem. Int. Ed. 34 (1995) 2528.
44] C.S. Laxmi Narasimhan, J.W. Thybaut, G.B. Marin, P.A. Jacobs, J.A. Martens, J.F.
Denayer, G.V. Baron, J. Catal. 220 (2003) 399.
45] R. Chal, T. Cacciaguerra, S. van Donk, C. Gerardin, Chem. Comm. 46 (2010)
7
840.
[61] S.P. Elangovan, M. Hartmann, J. Catal. 217 (2003) 388.
[62] F. Fajula, A. Galarneau, F.D. Renzo, Micropor. Mesopor. Mater. 82 (2005) 227.
[63] K. Chaudhari, T.K. Das, A.J. Chandwadkar, S. Sivasanker, J. Catal. 186 (1999) 81.
[64] J.F. Denayer, A.R. Ocakoglu, W. Huybrechts, J.A. Martens, J.W. Thybaut, G.B.
Marin, G.V. Baron, Chem. Comm. (2003) 1880.
[65] W. Souverijns, J.A. Martens, G.F. Froment, P.A. Jacobs, J. Catal. 174 (1998) 177.
[66] J.A. Martens, G. Vanbutsele, P.A. Jacobs, J. Denayer, R. Ocakoglu, G. Baron, J.A.
Muñoz Arroyo, J. Thybaut, G.B. Marin, Catal. Today 65 (2001) 111.
[67] B. Coasne, A. Galarneau, C. Gerardin, F. Fajula, F. Villemot, Langmuir 29 (2013)
7864.
[
[
[
[
[
46] D. Li, F. Li, J. Ren, Y. Sun, Appl. Catal. A: Gen. 241 (2003) 15.
47] L. Guo, X. Bao, Y. Fan, G. Shi, H. Liu, D. Bai, J. Catal. 294 (2012) 161.
48] S. Abelló, A. Bonilla, J. Pérez-Ramírez, Appl. Catal. A: Gen. 364 (2009) 191.
49] P.Y. Dapsens, C. Mondelli, J. Pérez-Ramírez, ChemSusChem 6 (2013) 831.
50] L. Oliviero, A. Vimont, J.-C. Lavalley, F. Romero Sarria, M. Gaillard, F. Mauge,
Phys. Chem. Chem. Phys. 7 (2005) 1861.
51] F. Thibault-Starzyk, I. Stan, S. Abelló, A. Bonilla, K. Thomas, C. Fernandez, J.-P.
Gilson, J. Pérez-Ramírez, J. Catal. 264 (2009) 11.
52] T. Onfroy, G. Clet, M. Houalla, Micropor. Mesopor. Mater. 82 (2005) 99.
[
[