The influences of Al species and Ti species on the catalytic epoxidation over Si/Ti-pillared MCM-36 synthesized from MCM-22

Article abstract of DOI:10.1039/c5cy00145e

A novel mesopore silica-aluminium material, i.e., Si/Ti-MCM-36 derivatives were prepared with Si/Ti mixed oxide as a pillar. The influences of acidity from Al species and the coordination state of Ti species on the catalytic cylcohexene epoxidation with t-butylhydroperoxide (TBHP) as the oxidant were investigated. The mesoporous structures were confirmed by XRD and N₂ ads/desorption experiment. The acidity and surface polarity were characterized by NH₃-TPD, ²⁷Al 2D MQ MAS NMR and ²⁹Si MAS NMR. The Ti coordination state was identified with UV-Vis spectra. By using ²⁷Al 2D MQ MAS NMR technique, the Al species from the different eight crystallographically T (Si, Al) positions of MWW zeolite were identified, which are associated with bridging hydroxyl groups such as the Bronsted acid site. It was found that the Si/Ti-MCM-36 samples, whether with tetrahedrally (T_d) or octahedrally (O_h) coordinated state Ti species, are all inactive for the cyclohexene epoxidation. A further acid treatment of the Si/Ti-MCM-36 samples can greatly expel Al species from the MWW structure, which largely decrease the acidity and transfer the acid-catalyzed decomposition of t-butylhydroperoxide to Ti catalyzed cyclohexene epoxidation. It shows that not only the amount of Al species but also their framework crystallographic positions have a great influence on the epoxidation activity. Furthermore, the grafting of the tetra-isopropoxide on the acid treated Si/Ti-MCM-36 can increase its hydrophobicity and epoxidation activity. Moreover, the T_d TiO₄ in the units has been confirmed to be the most active site for the epoxidation reaction.

Full text of DOI:10.1039/c5cy00145e

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
The influences of Al species and Ti species on the
catalytic epoxidation over Si/Ti-pillared MCM-36
synthesized from MCM-22†
Cite this: Catal. Sci. Technol., 2015,
5, 3007
c
Fang Jin,ab Shingjong Huang, Soofin Cheng, b Yuanxin Wu,a Chih-Cheng Changb
*
and Yu-Wei Huangb
A novel mesopore silica-aluminium material, i.e., Si/Ti-MCM-36 derivatives were prepared with Si/Ti mixed
oxide as a pillar. The influences of acidity from Al species and the coordination state of Ti species on the
catalytic cylcohexene epoxidation with t-butylhydroperoxide (TBHP) as the oxidant were investigated. The
mesoporous structures were confirmed by XRD and N2 ads/desorption experiment. The acidity and surface
polarity were characterized by NH3-TPD, 27Al 2D MQ MAS NMR and 29Si MAS NMR. The Ti coordination
state was identified with UV-Vis spectra. By using 27Al 2D MQ MAS NMR technique, the Al species from the
different eight crystallographically T (Si, Al) positions of MWW zeolite were identified, which are associated
with bridging hydroxyl groups such as the Brønsted acid site. It was found that the Si/Ti-MCM-36 samples,
whether with tetrahedrally (Td) or octahedrally (Oh) coordinated state Ti species, are all inactive for the
cyclohexene epoxidation. A further acid treatment of the Si/Ti-MCM-36 samples can greatly expel Al spe-
cies from the MWW structure, which largely decrease the acidity and transfer the acid-catalyzed decompo-
sition of t-butylhydroperoxide to Ti catalyzed cyclohexene epoxidation. It shows that not only the amount
of Al species but also their framework crystallographic positions have a great influence on the epoxidation
activity. Furthermore, the grafting of the tetra-isopropoxide on the acid treated Si/Ti-MCM-36 can increase
its hydrophobicity and epoxidation activity. Moreover, the Td TiO4 in the units has been confirmed to be
the most active site for the epoxidation reaction.
Received 29th January 2015,
Accepted 23rd March 2015
DOI: 10.1039/c5cy00145e
www.rsc.org/catalysis
facilitate the diffusion of bulky molecules often encountered
in fine chemicals synthesis. Active catalysts have been
1. Introduction
Titanium (Ti) substituted zeolites, such as TS-1, Ti-beta and
Ti-MWW were reported to have remarkable redox catalytic
activities in selective oxidation reactions with hydrogen perox-
ide as the oxidant at mild reaction temperatures.1–4 The main
inconvenience of these solids for many applications is the
presence of microporosity, which limits the accessibility to
substrates of small kinetic diameter and linear molecules. For
instance, TS-1 is ineffective to catalyse the cyclohexene epoxi-
dation with alkylhydroperoxides such as t-butylhydroperoxide
(TBHP), because of the large molecular size and narrow zeo-
lite pores.5,6 Therefore, in the past two decades, much effort
has been devoted to the preparation of Ti-incorporated meso-
porous silica materials, because their relatively large pores
obtained by treating titanium tetraisopropoxide (TTIP) with
MCM-41 or silica gel,7–9 and one pot co-condensation synthe-
sis of Ti-SBA-15 using tetraethyl orthosilicate (TEOS) as a sil-
ica source and titanium oxychloride or TTIP as a Ti source in
an acidic environment.10,11 However, unlike the hydrophobic
nature and purely tetrahedrally coordinated TiIJIV) species of
the TS-1, the coordination environment of the Ti species and
the hydrophilicity originating from the hydroxyl groups of sil-
ica have great influence on the catalytic activity and selectivity
for the epoxidation.8,10
Silanols are inherent for silica based materials, especially
for layered silicates, the number of which depends on the
conditions of crystallization and post-modification. It is well
known that the surface polarity is considered crucial in
obtaining good catalytic performances.12 Blasco et al.3
reported the synthesis of an Al free Ti-BEA (F) zeolite, which
is a hydrophobic solid and is much more active and selective
than the conventional Ti-BEA (OH) for epoxidation. This
hydrophobicity property could depress the further hydro-
lization of epoxide and decomposition of the epoxidation
agent. Therefore, controlling the polarity of the surface by
a Key Laboratory for Green Chemical Process of Ministry of Education, School of
Chemical Engineering and Pharmacy, Wuhan Institute of Technology, Wuhan
430074, China
b Department of Chemistry, National Taiwan University, Taipei 10617, Taiwan.
E-mail: chem1301@ntu.edu.tw
c Instrumentation Center, National Taiwan University, Taipei 10617, Taiwan
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5cy00145e
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol., 2015, 5, 3007–3016 | 3007

View Article Online
Paper
Catalysis Science & Technology
elimination of surface hydroxyls on the silica surface could
strongly increase the activity and selectivity for the epoxida-
tion. The coordination environment of the Ti species also has
coordination environment of Ti species on the Ti incorpo-
rated mesopore silica catalysts.
a great influence on the catalytic performance. It is generally 2. Experimental methods
accepted that pure TiO2 is not active in the epoxidation reac-
tion, and the tetrahedral (Td) TiO4 units with Ti atoms
2.1. Preparation of materials
The hydrothermal synthesis of aluminosilicate MCM-22 was
substituted Si in the silica framework work as active sites for
the epoxidation reaction.8 Moreover, the highly dispersed
TiO2/SiO2 catalysts in an octahedral (Oh) coordination state
also exhibit high reactivity and high selectivity to epoxide.9,13
However, the interaction of the surface polarity and the coor-
dination environment of the Ti species on the Ti-
incorporated mesopore silica materials for the epoxidation
are still not well understood, due to the lack of fundamental
systematic studies.
carried out by modifying the synthesis procedure reported in
the literature18 using hexamethyleneimine (HMI) as the
organic template and Ludox S40 silica gel as the silica source.
The as-prepared MCM-22 precursor is denoted as MCM-
22(P). The MCM-22(P) powders were introduced to a mixture
of
tetrapropyl-ammonium
hydroxide
(TPAOH)
and
cetyltrimethylammonium bromide (CTMABr) with a relative
weight ratio of TPAOH: 4 CTMABr: 330 H2O, and the mixture
was sealed in a flask and heated at 373 K for 16 h.19,20 The
resulting swollen MCM-22 material was collected by filtra-
tion, washed with distilled water, and dried under vacuum.
The swollen MCM-22 was dispersed in a mixture of tetra-
ethylorthosilicate (TEOS) and tetrabutylorthotitanate (TBOT)
with TEOS/TBOT molar ratios of 5 : 1, 20 : 1, 40 : 1, 80 : 1, 120 : 1,
and ∞ under nitrogen, the solid/liquid weight ratio was 1 : 10,
and the mixture was stirred at 353 K for 25 h. The solid
was collected by filtration and dried under ambient condi-
tions. The hydrolysis of TEOS and TBOT was carried out by
suspending the dried solid in an aqueous solution at 313 K
for 6 h at pH = 9 using 1 M NH3. Finally, the sample collected
by filtration was calcined at 723 K for 3 h in nitrogen and
then at 812 K for 6 h in air (heating rate of 2 K min−1). The
resultant materials are designated as xSi/Ti-MCM-36, where x
is the Si/Ti molar ratio in the pillaring gel. The precursor
before calcination is denoted as xSi/Ti-MCM-36IJP).
The acid treatments of 40Si/Ti-MCM-36 and 40Si/Ti-MCM-
36IJP) were carried out by refluxing the calcined sample in 2
M HNO3 solution at 353 K for 1 or 6 h and subsequent calci-
nation at 833 K. The corresponding acid-treated samples are
denoted as A-40Si/Ti-MCM-36IJy h) and AP-40Si/Ti-MCM-36IJy
h), respectively, where y is the acid treatment hours.
In comparison with A-40Si/Ti-MCM-36, a titanium grafted
sample was prepared according to the procedure reported in
the literature13 by refluxing the calcined A-40Si/Ti-MCM-36 in
an anhydrous toluene-containing titanium tetra-isopropoxide
(TTIP) (0.22 mmol Ti/g SiO2) for 6 h. The solid was collected
by filtration, washed with toluene, and dried at 393 K over-
night, followed by hydrolysis at ambient conditions and calci-
nation at 812 K for 3 h. The prepared sample is denoted as
G-A-40Si/Ti-MCM-36.
The MCM-22 zeolite (IZA code MWW) was first identified
by Leonowicz et al.14 in 1994 to have the crystalline structure
formed by vertically aligning the lamellar zeolite intermedi-
ate, which is a unit cell in thickness. The as-made MCM-22 is
the stacking of sheets of single double layers with repeating
unit dIJ001) of 2.5 nm. Each sheet possesses two independent
pore systems. One system is the 10-membered ring (MR)
sinusoidal channel pores running in the double layer, and
the second system is the 12-MR MWW cage (0.71 nm in
diameter) running perpendicular to the sheet, with the cage
cutting into two halves and the cup type structures with 12-
MR openings exposed to the exterior of the sheet.14 The
lamellar precursor of MCM-22 can be pillared with silica to
generate MCM-36.15 The MCM-36 structure combines zeolitic
microporosity inside its crystalline sheets with slit-like meso-
porosity in the interlayer space generated by pillaring the
layers with silica.15 The 2 nm mesopores formed by pillaring
have made MCM-36 an excellent acid catalyst for the bulky
molecules encountered in reactions such as vacuum gasoil
cracking, alkylation of iso-paraffins or aromatics, etc.16,17
Recently, we have developed a facile and innovative
method for the synthesis of Ti incorporated MCM-36 material
by pillaring an MCM-22 precursor with nanoparticles of Si/Ti
mixed oxide. The prepared Ti incorporated MCM-36 is a cata-
lyst with dual active sites, i.e. the acid sites from the frame-
work Al species and the oxidation sites originating from the
pillared Ti species. This type of catalyst provides a chance for
us to investigate the relationship with the physicochemical
and reactivity/selectivity properties of the surface polarity and
the coordination environment of Ti species on the Ti incor-
porated mesopore silica materials for the epoxidation.
Therefore, this article focuses on the preparation of Ti
incorporated MCM-36 catalysts by intercalating the Si/Ti
mixed oxide. Furthermore, post modification with acid treat-
ment and grafting treatment was applied to modify the acid-
ity and Ti coordination environment. In addition, a detailed
characterization study was carried out to collect information
about the structural characteristics of the mesopore, acidity
and incorporated titanium species and structure. The epoxi-
dation reaction of cyclohexene with TBHP was employed to
investigate the interaction of the surface polarity and the
2.2. Characterization
X-ray powder diffraction patterns were recorded using a
PANalytical X'Pert PRO diffractometer with Cu-Kα radiation
(λ = 1.5418 Å) operated at 40 mA and 45 kV. N2 sorption iso-
therms were measured at 77 K on a Micrometerics TriStar
3000 analyzer after vacuum pretreating the samples at 473 K
for 8 h. Total specific surface areas (SBET) were calculated
using the Brunauer–Emmett–Teller (BET) method,21 and the
3008 | Catal. Sci. Technol., 2015, 5, 3007–3016
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
total pore volume (Vtotal) was evaluated from N2 uptake at a
relative N2 pressure of 0.99. The t-plot was employed to evalu-
ate the volume of the micropores (Vmicro) of MCM-22 and the
volume of the mesopores (Vmeso) for the MCM-36.22 The cor-
responding pore size distribution (PSD) was determined
using the desorption branches of the isotherms by the
Barrett–Joyner–Halenda (BJH) method.23 Diffuse reflectance
(DR) UV-Vis spectra were recorded using a Hitachi U-3310
spectrometer equipped with an integrating sphere detector,
and BaSO4 was the reference. The elemental analyses were
determined using Electron Probe Micro Analysis (EPMA) in a
JEOL JXA-8200 Electron Probe.
chromatograph (GC) equipped with a mass detector. The
quantitative analysis was conducted with a Shimadzu GC-
2014 instrument equipped with a FID detector, where cyclo-
pentanone was used as an internal standard. The cyclohexene
conversion and product selectivities were calculated by the
following equations:
Conversion (%) = (formed derivatives of cyclohexene in
mol/initial cyclohexene in mol) × 100.
Selectivity (%) = (specific product in mol/formed deriva-
tives of cyclohexene in mol) × 100.
NH3 temperature programmed desorption (TPD) was
conducted in a Micromeritics AutoChem2920 equipment with
a quartz reactor and 100 mg sample at atmospheric pressure
with a TCD detector. The system was quenched at 823 K for 1
h in a flow of He 40 ml min−1 (STP). After cooling to 373 K,
NH3 flow was introduced for 40 minutes until saturation. The
sample was quenched again at 373 K for 2 h to remove the
excess NH3. Desorption was then performed under the same
He flow with a ramp of 10 K min−1 to 823 K.
3. Results and discussions
3.1. The influence of Si/Ti ratio of pillaring sol on structure
of the MCM-36
The elemental analysis results in Table 1 show that with
decreasing the Si/Ti ratio of the TEOS and TBOT pillaring sol-
vent, the amount of incorporated Ti increases, which resulted
in the low Si/Ti ratio in the produced Si/Ti-MCM-36. There-
fore, the Si/Ti ratio of the resulting Si/Ti-MCM-36 can be con-
trolled by that of the pillaring TEOS-TBOT moieties.
27Al 2D multiple-quantum (MQ) magic-angle-spinning
(MAS) NMR spectra were acquired on a wide-bore 14.1-T
Bruker AVANCE III NMR spectrometer equipped with a 1.9
mm MAS double-resonance probe head. The sample was
spun at 40 kHz, and a z-filter 3Q MAS sequence was used.
The RF field strength for 3Q excitation and conversion was
100 kHz. The pulse durations for 3Q excitation and conver-
sion were 4 and 1.3 us, respectively. The selective π/2 pulse
duration was set to 10 μs and the hyper-complex was
implemented for quadrature detection in the indirect dimen-
sion. The t1 increment was synchronized with the rotor
period, and 16 steps were acquired with a recycle delay of
0.7 s.
The small-angle and wide-angle XRD patterns of the
pillared MCM-22 samples with different ratios of TEOS to
TBOT in the intercalation mixture are shown in comparison
to the calcinated MCM-22 in Fig. 1. Most of the samples with
the obvious low angle reflection of the (001) peak at 2° con-
firm the effective pillaring with these mixtures, except for the
pure TBOT pillared samples 0Si/Ti-MCM-36. The coalesced
broad peak of (101) and (102) at ca. 8° and 10° for the
pillared MCM-22 characterizes the disordered expanded
structure along vertical alignment stacking in the c-direction.
On the other hand, the positions of intra-layer diffractions
(hk0), such as the prominent narrow peaks at about 7.1°,
25.1° and 26.1° corresponding to the 100, 220, and 310
planes, respectively, remain conspicuously unchanged in the
swollen and pillared samples, indicating that the internal
structure of the MWW layers does not change. All these
results are in accordance with the literature.24
The XRD results allow only qualitative assessment of the
intercalation efficiency, while quantitative appraisal requires
N2 sorption data, as shown in Fig. 2 and Table 1. The sorp-
tion isotherm of the MCM-22 zeolite corresponds to a type I
isotherm in the Brunauer, Deming and Teller classification,
indicating the presence of microporosity. For the Si/Ti-MCM-
36 samples, the increase in the adsorption volume at relative
pressures of 0.2–0.4 corresponds to peak pore diameters
around 2 nm in the pore size distribution (PSD) profiles
(Fig. 2(B)). These mesopores are considered to be generated
by pillaring the interlayer spaces with TiO2–SiO2 clusters.25
The sharp peaks at about 4 nm in the PSD profiles in Fig.
2(B) are an artifact, which are caused by the “tensile strength
effect” created by the sudden collapse of the liquid N2 capil-
lary inside the micropores.26 The quantitative differentiation
of intercalation efficiency of the Si/Ti-MCM-36 samples syn-
thesized by different Si/Ti ratios can be judged by the peak
Solid-state 29Si MAS NMR experiments were carried out at
frequencies of 59.6 MHz, with a pulse width of 10 μs, a pulse
delay of 50 s, and a spinning rate of 4 kHz on a Bruker
DSX300 NMR spectrometer equipped with a commercial 4
mm MASprobe. The MAS frequencies were set at 6 kHz for all
experiments, and the variation was limited to 3 Hz with a
commercial pneumatic control unit. All 29Si chemical shifts
were externally referenced to tetramethylsilane.
2.3. Catalytic reaction
The catalytic activities of Si/Ti-MCM-36 materials were exam-
ined by carrying out liquid phase epoxidation of cyclohexene
with TBHP (1.5 M in decane) at 353 K. Calcined Si/Ti-MCM-
36 samples were pre-dried at 423 K overnight before being
used as the catalysts. In a typical run, a mixture of 10 mmol
cyclohexene, 3 mL decane, 50 mg of the catalyst in powder
form and 5 mmol TBHP was heated in a two necked round-
bottomed flask with a stir bar and a water cooling condenser
at 353 K to start the reaction. At specific times, the reaction
mixtures were collected by removal of the solid catalysts and
then qualitatively analyzed by an Agilent HP 6890 gas
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol., 2015, 5, 3007–3016 | 3009

View Article Online
Paper
Catalysis Science & Technology
Table 1 The physico-chemical properties of the MCM-36 samples prepared with different Si/Ti ratios
Si/Tia
Si/Ala
SBET (m2 g−1
)
Vtotal (cm3 g−1
)
Vmicro (cm3 g−1
)
Vmeso (cm3 g−1
)
b
c
Sample
MCM-22
—
15
—
14
15
16
15
18
479
702
755
789
843
768
782
0.36
0.41
0.42
0.44
0.50
0.58
0.68
0.18
0.08
0.07
0.03
0.03
0.04
0.05
0.19
0.32
0.35
0.41
0.47
0.54
0Si/Ti-MCM-36IJM)
5Si/Ti-MCM-36IJM)
20Si/Ti-MCM-36IJM)
40Si/Ti-MCM-36IJM)
80Si/Ti-MCM-36IJM)
120Si/Ti-MCM-36IJM)
NA
23
29
50
77
81
a
b
c
Elemental composition from EPMA. Vmicro for MCM-22 from the t-plot and that for MCM-36 is calculated by Vtotal − Vmeso
MCM-22 is calculated by Vtotal − Vmicro for the MCM-36 from the t-plot.
.
Vmeso for the
intensity at 2 nm of the PSD profiles in Fig. 2(B) and the
increased SBET and Vmeso values shown in Table 1.
All the prepared samples, characterized by increasing Ti
loadings, were submitted to DR UV-Vis analysis in order to
provide information about the coordination geometry of the
Ti cations, as shown in Fig. 3. Except for 0Si/Ti-MCM-36, all
the DR UV-Vis spectra of the Si/Ti-MCM-36 samples lack an
absorption above 330 nm, indicating no formation of TiO2.
All the pillared samples have two peaks at 220 and 260 nm.
By decreasing the Ti loading, the intensity of the peak at
Fig. 3 DR UV-Vis spectra of (A) MCM-36 with different pillars (a)
0Si/Ti-MCM-36, (b) 5Si/Ti-MCM-36, (c) 20Si/Ti-MCM-36, (d) 40Si/Ti-
MCM-36, (e) 80Si/Ti-MCM-36 and (f) 120Si/Ti-MCM-36.
~260 nm obviously decreases. The 220 nm band, resulting
from the charge transfer from O2− to Ti4+, has been widely
reported for the Ti substituted zeolites and is characteristic
of the tetrahedrally (Td) coordinated Ti highly dispersed in
the framework, while the 260 nm band has been attributed
to the octahedral (Oh) Ti species, belonging to a kind of
extra-framework Ti species.27 For the pillared MCM-22, 120Si/
Ti-MCM-36 shows the main band at 220 nm and an
unobvious shoulder band around 260 nm, which indicates
that most of the Ti cations are incorporated in the pillared
sample with Td coordination, a small number of which are
excited as Oh coordination.
Fig. 1 Small and wide angle XRD patterns of (a) 0Si/Ti-MCM-36, (b)
5Si/Ti-MCM-36, (c) 20Si/Ti-MCM-36, (d) 40Ti/Si-MCM-36, (e) 80Si/Ti-
MCM-36, (f) 120Ti/Si-MCM-36 and (g) MCM-22.
3.2 The influence of sequential treatments on the structure
and Ti coordination state of Si/Ti-MCM-36
The post modification of 40Si/Ti-MCM-36 with 2 M HNO3
was performed at different temperatures and periods. As
shown in Table 2, the acid treatment caused a dramatic
decrease in the Si/Ti and Si/Al ratios, indicating that the Ti
and Al species were leaching from the structure.
The XRD patterns of 40Si/Ti-MCM-36 before and after 1 or
6 h acid treatment are given in Fig. 4. The retention of the
broad X-ray diffraction peak in the small angle region con-
firms the conservation of the pillaring structure. Moreover,
the broad diffraction in the 2θ = 8–10° region is attributed to
the disordered stacking of the MWW layer along the
c-direction, and the well-resolved (100) and (310) peaks
Fig.
2 (A) N2 adsorption/desorption isotherms (B) BJH pore size
distribution of (a) 0Si/Ti-MCM-36 (vertical move 250), (b) 5Si/Ti-MCM-
36 (vertical move 180), (c) 20Si/Ti-MCM-36 (vertical move 160), (d)
40Ti/Si-MCM-36 (vertical move 100), (e) 80Ti/Si-MCM-36, (f) 120Ti/Si-
MCM-36 (vertical move −50) and (g) MCM-22 (vertical move −185).
3010 | Catal. Sci. Technol., 2015, 5, 3007–3016
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Table 2 Physico-chemical properties and catalytic activities of acid treated 40Si/Ti-MCM-36 samples
A-40Si/Ti-MCM-
40Si/Ti-MCM-36 36IJ1 h)
A-40Si/Ti-MCM-
36IJ6 h)
AP-40Si/Ti-MCM-
36IJ6 h)
Mixed
G-A-40Si/Ti-MCM-
Treatment condition
samplesd 36IJ1 h)
Si/Ti molar ratioa
52
16
154
38
168
52
201
37
1.3
0.09
704
0.49
0.04
0.45
0
—
—
—
—
—
—
—
—
0
71
52
Si/Al molar ratioa
Al mass percentage (%)
2.9
0.33
843
0.50
0.03
0.47
0
1.45
0.13
713
0.49
0.05
0.44
20.1
0.98
0.11
646
0.49
0.06
0.43
38.1
1.32
0.34
552
0.48
0.08
0.41
34.8
Ti content (mmol g−1
)
SBET (m2 g−1
Vtotal (cm3 g−1
Vmicro (cm3 g−1
Vmeso (cm3 g−1
)
)
)
)
Cyclohexene conversion
(%)
Selectivities of products
(%)b
CHE
CHD
OH
0
0
0
0
0
74.6
17.0
6.5
2.0
50.1
83.7
11.4
3.3
1.1
113
0
0
0
0
0
0
0
0
0
0
80.3
5.5
12.1
2.1
ONE
TOFc (h−1
)
33.7
a
Reaction conditions: 0.05 g catalyst, 10 mmol cyclohexene, 5 mmol TBHP (60 wt%), 3 mL decane solvent, 353 K, 6 h. Elemental composition
b
c
from EPMA. CHE: cyclohexene epoxide; CHD: 1,2-cyclohexanediol; OH: 2-cyclohexen-1-ol; ONE: 2-cyclohexen-1-one. Turnover frequency
d
(TOF): calculated from average reaction rates per Ti site per hour. The mixed samples of A-40Si/Ti-MCM-36IJ1 h) and AP-40Si/Ti-MCM-36IJ6 h)
with a weight ratio of 1 : 1.
Fig. 4 XRD patterns of (a) 40Si/Ti-MCM-36, (b) A-Si/Ti-MCM-36IJ1 h),
Fig. 5 (A) Nitrogen adsorption/desorption isotherms at 77 K and (B)
(c) A-Si/Ti-MCM-36IJ6 h), (d) AP-Si/Ti-MCM-36IJ6 h) and (e) G-A-Si/Ti-
MCM-36IJ1 h).
the BJH pore size distributions of (a) 40Si/Ti-MCM-36, (b) A-Si/Ti-
MCM-36IJ1 h) (vertical move 80), (c) A-Si/Ti-MCM-36IJ6 h) (vertical
move 150), (d) AP-Si/Ti-MCM-36IJ6 h) (vertical move 210), and (e) G-A-
Si/Ti-MCM-36IJ1 h) (vertical move 280).
corresponding to diffractions from the ab plane are all
retained. It can be concluded that the structures of the MWW
basal layer and the pillared MCM-36 are preserved during
acid treatment. Influence of acid treatment on the textural
properties of 40Si/Ti-MCM-36 sample was examined by N2
physisorption, as illustrated in Fig. 5. The shapes of sorption
isotherms are almost the same as that of pristine 40Si/Ti-
MCM-36. These results infer that the 2 nm mesopores gener-
ated by titanosilicate pillars are still preserved after the acid
treatment.
The UV-Vis spectra of the acid treated samples are shown
in Fig. 6. It is noted that the shoulder peak at 260 nm, attrib-
uted to the Oh Ti species, clearly vanishes after acid treat-
ment at 373 K for 1 h, then the spectra retain the same shape
for longer acid treatment. The remaining narrow 220 nm
band is characteristic of Td coordinated Ti. These results
indicate that acid treatment removed Ti in Oh coordination
and left Td-coordinated Ti in A-Si/Ti-MCM-36.
A further calcination of the acid treated 40Si/Ti-MCM-
36IJP) still results in the same XRD patterns as A-Si/Ti-MCM-
36, as shown in Fig. 4. Moreover, by comparison, AP-Si/Ti-
MCM-36 and A-Si/Ti-MCM-36 are treated at the same acid
conditions; AP-Si/Ti-MCM-36IJ6 h) retains a more mesoporous
structure according to the larger Vmeso and SBET in Table 1,
larger N2 updates in the P/P0 region of 0.2–0.4 and higher
intensity of the 2 nm PSD peak in Fig. 5. It can be deduced
that the remaining organic template and swelling agent in
the 40Si/Ti-MCM-36IJP) protect the structure during the acid
treatment. This phenomenon is in agreement with the report
that the structure-directing agent and swelling agent can pro-
tect the zeolite structure and cause the zeolite to be inert to
Si leaching upon alkaline treatment.28 It also explains the
fact that under the same acid treatment conditions, the
AP-Si/Ti-MCM-36IJ6 h) retains a higher Al content in the
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol., 2015, 5, 3007–3016 | 3011

View Article Online
Paper
Catalysis Science & Technology
Fig. 6 DR UV-Vis spectra of (a) 40Si/Ti-MCM-36, (b) A-Si/Ti-MCM-
36IJ1 h), (c) A-Si/Ti-MCM-36IJ6 h), (d) AP-Si/Ti-MCM-36IJ6 h) and (e)
G-A-Si/Ti-MCM-36IJ1 h).
Fig. 7 NH3 TPD spectra of samples (a) MCM-22, (b) 40Si/Ti-MCM-36,
(c) A-Si/Ti-MCM-36IJ1 h), (d) A-Si/Ti-MCM-36IJ6 h), (e) AP-Si/Ti-MCM-
36IJ6 h) and (f) G-A-Si/Ti-MCM-36IJ1 h).
structure than the A-Si/Ti-MCM-36IJ6 h). However, the incor-
porated structure-directing agent and swelling agent in the
40Si/Ti-MCM-36IJP) has little effect on the Ti leaching during
acid treatment, by comparison of the Si/Ti ratio of the
AP-Si/Ti-MCM-36IJ6 h) and A-Si/Ti-MCM-36IJ6 h) in Table 2
and UV-Vis spectra in Fig. 6. Therefore, it can be deduced
that the swelling agent and structure-directing agent HMI
incorporated in the MWW structure of the 40Si/Ti-MCM-36IJP)
hinders the Al from leaching out. Meanwhile the intercalated
Ti/Si oxide pillars in between the expanded MWW layer can-
not be protected by the retained CTMABr, TPAOH and HMI
agent.
The further grafting of Ti on the A-Si/Ti-MCM-36IJ1 h) has
little influence on the MWW structure as shown in the XRD
figure and N2 adsorption/desorption spectra in Fig. 4 and 5.
But the intercalated 2 nm mesopores are slightly decayed, as
shown by the decreased Vmeso value in Table 1 and the 2 nm
PSD peak intensity in Fig. 5. However, after grafting treat-
ment the increased intensity of the 260 nm peak is obvious,
as shown in the UV-Vis spectra of Fig. 6. It is well known that
Ti-isopropoxide is a tetramer, which is unlikely to produce
the pure tetrahedral sites that a monomer might. Depending
on titanium loading on the SiO2, often indicated as percent-
age by weight of TiO2, different types of titanium species
could be present, ranging from dispersed surface TiO2 spe-
cies, both in Td and Oh coordination, to small TiO2 crystal-
lites. Therefore, this behaviour can be related to the polymer-
ization of grafting Ti and the formation of Oh coordination.
Because the template-containing 40Si/Ti-MCM-36 (P) is vir-
tually inert to Al leaching upon treatment in HNO3 solution,
after 6 h treatment the AP-40Si/Ti-MCM-36 has more retained
acid sites and Al content than the A-40Si/Ti-MCM-36, as
shown by their NH3 TPD spectra in Fig. 7 and Table 2. The
AP-40Si/Ti-MCM-36IJ6 h) and A-40Si/Ti-MCM-36IJ1 h) have
almost the same total acid amount and different acid distri-
bution in Fig. 7. The AP-40Si/Ti-MCM-36IJ6 h) has more
strong acid sites than A-40Si/Ti-MCM-36IJ1 h). This phenome-
non indicates that as the main source of acid sites in zeolite
the retained Al in the AP-40Si/Ti-MCM-36IJ6 h) and A-40Si/Ti-
MCM-36IJ1 h) is almost the same in quantity but do not show
total-equivalence. It can be deduced that the existing
CTMABr, TPAOH and HMI agent in the 40Si/Ti-MCM-36IJP)
changes the selectivity of the Al to be leached by the HNO3 in
the different crystallographically unequivalent Td sites of the
zeolite framework. This deduction is confirmed by the 27Al
MQ MAS NMR spectra in Fig. 8.
3.3 The influence of sequential treatments on the acidity of
the MWW structure
The NH3 TPD spectra of MCM-22, 40Si/Ti-MCM-36 and
A-Si/Ti-MCM-36 samples acid treated at 353 K for 1 and 6 h
are compared in Fig. 7. MCM-22 has two peaks correspond-
ing to two types of acid sites, i.e. the weak sites associated
with the TPD peak at around 350–550 K, and the strong sites
related to the peak at around 550–750 K. After the swelling
and pillaring treatment, the total acid amount of the resulted
40Si/Ti-MCM-36 decreased obviously, especially for the strong
acid sites. The change in acid distribution is similar to the
reported results.29
Fig. 8 27Al 2D MQ MAS NMR spectra of (a) MCM-22, (b) 40Si/Ti-
MCM-36, (c) A-40Si/Ti-MCM-36IJ1 h) and (d) AP-Si/Ti-MCM-36IJ6 h).
3012 | Catal. Sci. Technol., 2015, 5, 3007–3016
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
The 27Al MAS NMR spectrum of MCM-22 (Si/Al = 15) is
composed of two signals (Fig. S1, ESI†). The major resonance
at around 55 ppm−1 is ascribed to the tetrahedral-
coordinated framework aluminum (Alf).30 The additional
peak at around 0 ppm−1 is due to the octahedral extra-
framework aluminum (Alef), which comes from the calcina-
tion process.31,32 The signal corresponding to the Alf species
shows a major resonance at 57.8 ppm along with shoulders
at ca. 50.5 and 63 ppm. However, the interpretation of the Alf
signal in the 27Al MAS NMR spectra is difficult, essentially
because of second-order quadrupolar broadening. The two-
dimensional multi-quanta (2D MQ) sequence method pro-
vides enhanced resolution to separate the contributions of
aluminum species with different isotropic shifts and
quadrupolar constants along the F1 axis, as compared with
ordinary MAS spectra.33,34
intensities of the three signals in the F1 dimension decrease,
especially for the Alef signal at 0 ppm. In Fig. 8(c), the acid
treatment of the 40Si/Ti-MCM-36 causes the shrinkage of the
three peak distribution to spread further. Moreover, the acid
treatment diminishes the Alef signal at the 0 ppm. By com-
parison of the spectra of Fig. 8(c) and (d), it turns out that
the higher and lower frequency peaks at 49 and 61 ppm of
the AP-40Si/Ti-MCM-36IJ6 h) are more spread out than those
of A-40Si/Ti-MCM-36IJ1 h), which indicates that the AP-
40Si/Ti-MCM-36IJ6 h) has more T2 and T6 sites after acid
treatment. On the other hand, although AP-40Si/Ti-MCM-36IJ6 h)
and A-40Si/Ti-MCM-36IJ1 h) have similar Si/Al ratios, the Al
atoms have different distributions in the eight unequivalent
T(Si, Al) positions. According to these results, it can be
deduced that the existing CTMABr, TPAOH and HMI agents
in the 40Si/Ti-MCM-36IJP) prevent the leaching of the Al in
the T2 and T6 sites.
Based on the hexagonal P6/mmm structure, the structure
of MCM-22 has the 8 crystallographically distinct T (Si, Al)
positions (as shown in Scheme 1), their quantity ratio is T1 :
T2 : T3 : T4 : T5 : T6 : T7 : T8 = 1 : 3 : 3 : 1 : 1 : 3 : 3 : 3, which have
been confirmed by 27Al MQ MAS NMR and quantum chemical
calculations.32,35–37 These eight unequivalent T(Si, Al) posi-
tions are associated with bridging hydroxyl groups (SiOHAl)
as the Brønsted acid site.38 From the simulated 27Al MQ MAS
NMR spectrum, it is evident that 8 T sites can be divided into
3 groups: the T1, T3, T4, T5, T7 and T8 sites belong to the sig-
nal at δ 56 ppm, while the T2 and T6 sites correspond to the
signals at δ 61 and δ 49 ppm, respectively.32,39 In accordance
with the conclusion, three peaks at δ = (49, 53), (56, 60) and
(61, 66) ppm of the calcined MCM-22 can be clearly distin-
guished in the 2D MQ MAS NMR spectrum of Fig. 8(a). These
signals are not spread along the anisotropy axis A, which indi-
cates a distribution of chemical shifts and quadrupolar con-
stants, resulting from the random location of aluminum in
the framework.
Solid state 29Si MAS NMR spectroscopy (Fig. 9) was also
employed to investigate the evolution of the eight crystallo-
graphically unequivalent T(Si, Al) positions and the silanol
groups in the post-synthesis process. The spectrum of MCM-22
matches well with those reported previously.25 The Q4 sites in
the region of −104 to 130 ppm could be deconvoluted into
four lines, which should be assigned to several distinctive
crystallographic sites in MWW but with overlapping reso-
nances: −104 ppm (T2), −111 ppm (T3 + T5 + T4), −115 ppm
(T7 + T8), and −119 ppm (T6). In addition, an obvious reso-
nance at −99 ppm is attributed to both the T1 groups and the
Q3 sites, i.e. SiIJOH)IJOSi)3 or SiIJOH)IJOAl)IJOSi)2, which result
from certain structural defects.
Compared with the parent MCM-22, 40Si/Ti-MCM-36
shows very similar resonances due to the Q4 sites but a more
intensive −99 ppm resonance of the Q3 sites and a newly
emerged peak at −94 ppm. The −94 ppm peak originates from
the Q2 groups, i.e. SiIJOH)2IJOSi)2, SiIJOH)2IJOAl)IJOSi), etc.25 The
deconvolutions of the corresponding spectra are plotted
simultaneously by simulation of the spectra (Fig. S2, ESI†) for
the quantitative data. The simulation results of the samples
with four Gaussian lines are presented in Table 3.
Comparing the 2D MQ MAS NMR spectra of
Fig. 8(a) and (b), it was found that the 40Si/Ti-MCM-36 has a
similar T site distribution to that of MCM-22, but the relative
Fig. 9 29Si MAS NMR spectra of (a) MCM-22, (b) 40Si/Ti-MCM-36, (c)
A-40Si/Ti-MCM-36IJ1 h), (d) AP-Si/Ti-MCM-36IJ6 h), and (e) G-A-Si/Ti-
MCM-36IJ1 h).
Scheme 1 Illustration of the eight different T sites of the H-MCM-22
zeolite, reprinted from ref. 35 with permission from Elsevier.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol., 2015, 5, 3007–3016 | 3013

View Article Online
Paper
Catalysis Science & Technology
Table 3 The area percentages of crystallographically unequivalent T(Si, Al)
positions and the silanol groups obtained from deconvolution of the
29Si MAS NMR spectra of the samples
Percentage (%)
Q2
Q3/T1
T2
T3–5
T7, 8
T6
MCM-22
10.4
16.7
13.7
14.1
9.5
18.0
19.8
18.6
23.2
17.5
33.7
30.9
35.2
35.3
40.9
24.5
23.7
24.6
18.2
26.3
13.4
5.5
2.8
4.5
2.6
40Si/Ti-MCM-36
A-40Si/Ti-MCM-36
AP-40Si/Ti-MCM-36
G-A-40Si/Ti-MCM-36
3.4
7.1
4.6
3.1
Scheme
2
Reaction scheme of epoxidation of cyclohexene and
possible by-products.
From MCM-22 to 40Si/Ti-MCM-36 the quantitative data
show that the percentages for the Q4 sites decrease, while the
Q2 and Q3 percentages increase. This is in agreement with
the results that the swelling treatment partially dissolves the
framework silica, breaks the Si–O–Si bond and generates the
silanol groups as the defect sites. After acid treatment the
A-40Si/Ti-MCM-36IJ1 h) has a relatively lower peak intensity of
about −99 ppm; this can be attributed to the leaching of the
Alf in the T1 sites on the outside of the MWW layer. The
decrease percentage of T2 and T6 sites shows the acid treat-
ment is favourable for removing the Alf of these two sites.
However, the AP-40Si/Ti-MCM-36IJ6 h) has a higher percent-
age of T2 and T6 sites than those of A-40Si/Ti-MCM-36IJ1 h).
Therefore, the 29Si and 27Al MAS NMR results corroborate
each other so that the existing CTMABr, TPAOH and HMI
agents in the 40Si/Ti-MCM-36IJP) prevent leaching of the Alf
in the T2 and T6 sites, which corresponds to the AP-40Si/Ti-
MCM-36IJ6 h) having more strong acid sites than A-40Si/Ti-
MCM-36IJ1 h), as shown in Fig. 7.
The further grafting treatment of the TTIP in the
A-40Si/Ti-MCM-36IJ1 h) results in lower acidity, as shown in
Fig. 7. It can be deduced that TTIP is very reactive and likely
to react preferentially with the bridge hydroxyl group of the
Al species. On the other hand, the decreased intensities of
the Q3 and Q2 peaks in the 29Si MAS NMR of Fig. 9 indicate
the decrease of the silanol groups in the defect sites of the
MWW structure. The silanol groups on zeolite generally act
as the acid sites and adsorptive/reactive sites because of their
hydrophilic character.40,41 Therefore, it can be explained by
the consumption of the silanol groups and bridge hydroxyl
groups of Al species with the grafted TTIP, which results in
the lower acidity of the G-A-40Si/Ti-MCM-36IJ1 h) and
decreased hydrophilicity of this catalyst.
epoxide (CHE) is the target product. The hydrolysis of CHE
may be catalyzed by acidic sites, resulting in 1,2-cyclo-
hexanediol (CHD) as a side product. The other allylic oxida-
tion side products, 2-cyclohexen-1-ol (OH) and 2-cyclohexen-
1-one (ONE), are formed through the cyclohexenyl hydroper-
oxide radical.9
When xSi/Ti-MCM-36 samples with different Si/Ti ratios
were used as the catalyst, no conversion of cyclohexene to
any products was observed after 14 h reaction, while TBHP
was totally consumed. This phenomenon does not change
with changing the amount and coordination state of Ti spe-
cies in the MCM-36. The conversion and epoxide selectivity
markedly increased when acid-treated samples were used as
the catalysts instead of pristine 40Si/Ti-MCM-36. Moreover,
the conversion and selectivity seem to increase with the
increase of acid treatment time from 1 to 6 h. Because the
molar ratio of cyclohexene/TBHP in the reactants is 2, the
maximum stoichiometric cyclohexene conversion is 50%.
Moreover, since cyclohexanediol (CHD) is the product of sub-
sequent hydrolysis of CHE, the sum of selectivities of CHE
and CHD reveal the total selectivities of epoxidation prod-
ucts. As a result, total selectivities of epoxidation products up
to 92–95% are obtained over the acid-treated 40Si/Ti-MCM-
36. On the other hand, very limited amounts of the other two
side-products OH and ONE are formed, indicating a low
contribution of the allylic oxidation of cyclohexene, which
can be attributed to the fact that the acid treatment removes
Oh-coordinated Ti species, leaving Td-coordinated Ti as an
active sites for epoxidations other than allylic oxidation.
Stojanova et al.6 have explored the reactivities of TBHP in
the presence of zeolites in their experiment with either
H-MFI or Ga-containing MFI as the catalyst; only light hydro-
carbon gases were observed in the products and no radical
mechanism products such as t-butanol and di-t-butyl peroxide
were detected. This observation is similar to TBHP decompo-
sition in strong acidic solutions, where isobutene and hydro-
gen peroxide are formed.6 Therefore, they proposed that the
highly acidic H-MFI zeolite can effectively catalyze the homo-
lytic decomposition reaction of TBHP, which most likely pro-
ceeds through an electrophilic attack of the zeolite protons on
either oxygen of the peroxide, resulting in the formation of
H2O2 or H2O and the corresponding carbenium cation.
3.4 The influence of sequential treatments on catalytic
properties of Si/Ti-MCM-36
Titanium incorporated silica materials are among the most
important oxidation catalysts for epoxidation of alkenes. Suc-
cess in the preparation of Si/Ti-intercalated MCM-36 encour-
aged us to investigate its catalytic behavior in the oxidation
of bulky molecules. The cyclohexene epoxidation with
tert-butyl hydroperoxide (TBHP) was performed in decane
solution at 353 K. The results after 6 h are shown in Table 2.
Scheme 2 illustrates the possible reaction mechanism of
cyclohexene epoxidation over Ti-MCM-36 zeolite.10 Cyclohexene
The results of NH3 TPD in Fig. 7 confirm that the Si/Ti-
MCM-36 samples contain a large number of acid sites. AlIJIII)
in the framework of the MWW layer should contribute to the
3014 | Catal. Sci. Technol., 2015, 5, 3007–3016
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
acid sites. Therefore, the epoxidation of cyclohexene at the Ti 4. Conclusions
active sites competes with TBHP decomposition at the acid
The MCM-22 was pillared with titanosilicate to generate Si/
Ti-MCM-36 with micropores in the MWW layers and gener-
ated 2 nm mesopores between the layers. The Si/Ti ratio and
the ratio of Td-coordinated to Oh-coordinated Ti in the MCM-
36 matrix can be controlled by adjusting the Si/Ti ratio of the
TEOS and TBOT pillaring moiety.
sites for TBHP consumption over Si/Ti-MCM-36 samples. The
increase in catalytic activities in cyclohexene expoxidation
over acid treated Si/Ti-MCM-36 samples confirms the influ-
ence of Al in contributing to the acid sites in the catalytic
reactions. After leaching Al from the MWW framework during
acid-treatment, a high conversion of cyclohexene is obtained.
The reaction of TBHP is shifted from carbenium-involved
decompositon to radical-involved epoxidaiton. At the same
time, the leading role in the dual active site catalyst is trans-
ferred from the acid sites originating from the Al species to the
oxidative sites of the Ti species. Furthermore, the decrease of
acid sites also suppresses hydrolysis of the epoxide. As a result,
the epoxide selectivity increases with the leaching of Al from
the catalysts.
The acid treatment can greatly increase the epoxidation
activity and suppress the TBHP decomposition except for AP-
40Si/Ti-MCM-36IJ6 h). After 6 h reaction all the TBHP was
decomposed by AP-40Si/Ti-MCM-36IJ6 h). The mixed catalyst
sample of AP-40Si/Ti-MCM-36IJ6 h) and A-40Si/Ti-MCM-36IJ1 h)
with 1 : 1 weight ratio can merely catalyze TBHP decomposi-
tion. These results indicate that the slight differences in Ti con-
tent between these samples can be excluded as the reason for
the missing epoxidation activity for AP-40Si/Ti-MCM-36IJ6 h).
Although the AP-40Si/Ti-MCM-36IJ6 h) has similar Al content
to A-40Si/Ti-MCM-36IJ1 h), the AP-40Si/Ti-MCM-36IJ6 h) retains
higher T2 and T6 Al content with stronger acid sites, as
shown in the results of NH3 TPD in Fig. 7, 29Si MAS NMR in
Fig. 8 and 27Al MQ MAS NMR in Fig. 9. This difference of
Al species results in a lack of epoxidation activity for the
AP-40Si/Ti-MCM-36IJ6 h). Therefore, not only the Al content
but also the position of Al in the T(Si, Al) has a great influ-
ence on the epoxidation activity.
The acidity of the Al species in the Si/Ti-MCM-36 has a
great influence on the cylcohexene epoxidation activity with
TBHP as the oxidant. It is found that the Si/Ti-MCM-36 sam-
ples with a large amount of Al species can only cause TBHP
decomposition instead of cyclohexene epoxidation. Further
acid treatment can greatly eliminate Al species on the MWW
structure and decrease the acidity of Si/Ti-MCM-36, which
suppresses the TBHP decomposition and increases the activ-
ity for cyclohexene epoxidation. Moreover, not only the Al
content but also the position of Al in the T(Si, Al) has a great
influence on the epoxidation activity. The T2 and T6 Al con-
tents of the crystallographic T positions have stronger acidity
and stronger influence on depressing activity for epoxidation.
Finally, the grafting of the tetra-isopropoxide on the acid
treated Si/Ti-MCM-36 can further decrease the hydrophilicity
of this catalyst and increase the epoxidation activity. How-
ever, the most grafting Ti species exist as Oh coordination Ti,
which decreases the TOF for cylcohexene epoxidation. There-
fore, the Td TiO4 units in the framework are the most active
sites for the epoxidation reaction.
Acknowledgements
The financial supports from Ministry of Science & Technol-
ogy, Taiwan (NSC 101-2113-M-002-012-MY3, NSC102-2811-M-
002-136) is gratefully acknowledged. F. Jin also acknowledges
the supports of Natural Science Foundation, China (NSFC
21306143) and Scientific Research Foundation for the
Returned Overseas Chinese Scholars, Education Ministry
(SEM), China. Acknowledgement is also extended to Dr. C. Y.
Kao of Instrumentation Center and Miss Vivien Yeh of depart-
ment of Chemistry, National Taiwan University for the assis-
tance in EPMA and 29Si MAS NMR experiments.
For the sample A-40Si/Ti-MCM-36IJ1 h) after grafting treat-
ment, the cyclohexene conversion and CHE selectivity
increase, while the CHD selectivity decreases. At the same
time, the grafting treatment decreases the acid amount and
the silanol groups, which causes the decrease of hydrophilic-
ity of the sample. Therefore, it can be deduced that the
increased hydrophobicity suppresses the decomposition of
TBHP and hydrolyzation of CHE.
References
The turnover frequency (TOF) listed in Table 2 is defined
as the number of cyclohexene molecules converted per Ti site
per hour, which can be used to compare the catalytic effi-
ciency per Ti site for the conversion of cyclohexene. Although
the G-A-40Si/Ti-MCM-36IJ1 h) has a higher Ti content than
A-40Si/Ti-MCM-36IJ1 h), the former has a lower TOF than the
latter. This can be attributed to the fact that most grafting Ti
species in the G-A-40Si/Ti-MCM-36IJ1 h) exist as the Oh coordi-
nation Ti instead of the pure Td Ti species of A-40Si/Ti-MCM-
36. On the other hand, it can be seen that the higher Oh coor-
dination Ti content in G-A-40Si/Ti-MCM-36IJ1 h) result in
higher selectivity for the allylic oxidation side products of OH
and ONE. Therefore, it can be concluded that Td TiO4 units
are the most active sites for the epoxidation reaction.
1 B. Notari, in Adv. Catal., ed. D. D. Eley, W. O. Haag and B.
Gates, Academic Press, 1996, p. 253.
2 D. R. C. Huybrechts, L. D. Bruycker and P. A. Jacobs, Nature,
1990, 345, 240.
3 T. Blasco, M. A. Camblor, A. Corma, P. Esteve, A. Martinez,
C. Prieto and S. Valencia, J. Catal., 1996, 2367.
4 P. Wu and T. Tatsumi, Chem. Commun., 2001, 897.
5 C. B. Khouw, C. B. Dartt, J. A. Labinger and M. E. Davis,
J. Catal., 1994, 149, 195.
6 M. Stojanova, C. Karshalykov, G. L. Price and V. Kanazirev,
Appl. Catal., A, 1996, 143, 175.
7 Q. Yuan, A. Hagen and F. Roessner, Appl. Catal., A,
2006, 303, 81.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol., 2015, 5, 3007–3016 | 3015

View Article Online
Paper
Catalysis Science & Technology
8 R. Ballesteros, Y. Pérez, M. Fajardo, I. Sierra and I. del
Hierro, Microporous Mesoporous Mater., 2008, 116, 452.
9 J. M. Fraile, J. I. Garcia, J. A. Mayoral, E. Vispe, D. R. Brown
and M. Naderi, Chem. Commun., 2001, 1510.
10 S.-Y. Chen, C.-Y. Tang, J.-F. Lee, L.-Y. Jang, T. Tatsumi and S.
Cheng, J. Mater. Chem., 2011, 21, 2255.
11 F.-T. Kuo, S.-Y. Chen, T.-H. Lin, J.-F. Lee and S. Cheng, RSC
Adv., 2013, 3, 12604.
12 H. Kochkar and F. Figueras, J. Catal., 1997, 171, 420.
13 M. Cozzolino, M. Di Serio, R. Tesser and E. Santacesaria,
Appl. Catal., A, 2007, 325, 256.
26 S. Lowell, J. E. Shields, M. A. Thomas and M. Thommes,
Characterization of Porous Solids and Powders: Surface Area,
Pore Size and Density, Kluwer Academic Publishers, 2004.
27 P. Wu, T. Tatsumi, T. Komatsu and T. Yashima, J. Phys.
Chem. B, 2001, 105, 2897.
28 J. Pérez-Ramírez, S. Abelló, A. Abelló and J. C. Groen, Adv.
Funct. Mater., 2009, 19, 164.
29 S. Maheshwari, C. Martínez, M. Teresa Portilla, F. J. Llopis,
A. Corma and M. Tsapatsis, J. Catal., 2010, 272, 298.
30 M. Sawa, M. Niwa and Y. Murakami, Zeolites, 1990, 10, 532.
31 Y. Zhang, Y. Zhou, A. Qiu, Y. Wang, Y. Xu and P. Wu, Ind.
Eng. Chem. Res., 2006, 45, 2213.
14 M. E. Leonowicz, J. A. Lawton, S. L. Lawton and M. K. Rubin,
Science, 1994, 24, 1910.
15 U. Díaz, Chem. Eng., 2012, 2012, 1.
32 Z. Yu, Q. Wang, L. Chen and F. Deng, Cuihua Xuebao,
2012, 33, 129.
16 A. Miltenburg, J. Pawlesa, A. M. Bouzga, N. Žilková, J. Čejka
and M. Stöcker, Top. Catal., 2009, 52, 1190.
33 C. Fernandez and J. P. Amoureux, Chem. Phys. Lett.,
1995, 242, 449.
17 B. Zhang, Y. Ji, Z. Wang, Y. Liu, H. Sun, W. Yang and P. Wu,
Appl. Catal., A, 2012, 443–444, 103.
34 L. Frydman and J. S. Harwood, J. Am. Chem. Soc., 1995, 117,
5367.
18 A. Corma, C. Corell and J. Prez-Pariente, Zeolites, 1995, 15, 2.
19 W. J. Roth, C. T. Kresge, J. C. Vartuli, M. E. Leonowicz, A. S.
Fung and S. B. McCullen, Stud. Surf. Sci. Catal., 1995, 94, 301.
20 Y. J. He, G. S. Nivarthy, F. Eder, K. Seshan and J. A. Lercher,
Microporous Mesoporous Mater., 1998, 25, 207.
21 S. Brunauer, P. H. Emmett and E. Teller, J. Am. Chem. Soc.,
1938, 60, 309.
22 B. C. Lippens and J. H. de Boer, J. Catal., 1965, 4, 319.
23 E. P. Barrett, L. G. Joyner and P. P. Halenda, J. Am. Chem.
Soc., 1951, 73, 373.
35 L. Chen, F. Deng and C.-H. Ye, Wuli Huaxue Xuebao,
2003, 19, 805.
36 D. Ma, F. Deng, R. Fu, X. Han and X. Bao, J. Phys. Chem. B,
2001, 105, 1770.
37 D. Zhou, Y. Bao, M. Yang, N. He and G. Yang, J. Mol. Catal.
A: Chem., 2006, 244, 11.
38 G. Sastre, V. Fornes and A. Corma, J. Phys. Chem. B,
2000, 104, 4349.
39 P. Mériaudeau, A. Tuel and T. T. H. Vu, Catal. Lett.,
1999, 61, 89.
24 W. J. Roth and D. L. Dorset, Microporous Mesoporous Mater.,
2011, 142, 32.
25 S. Maheshwari, E. Jordan, S. Kumar, F. S. Bates, R. L. Penn,
D. F. Shantz and M. Tsapatsis, J. Am. Chem. Soc., 2008, 130, 1507.
40 M. Trombetta, T. Armaroli, A. D. Gutièrrez Alejandre, J.
Ramirez Solis and G. Busca, Appl. Catal., A, 2000, 192,
125.
41 M. Trombetta and G. Busca, J. Catal., 1999, 187, 521.
3016 | Catal. Sci. Technol., 2015, 5, 3007–3016
This journal is © The Royal Society of Chemistry 2015