Preparation, characterization, and catalytic performance of a novel methyl-rich Ti-HMS mesoporous molecular sieve with high hydrophobicity

Article abstract of DOI:10.1007/s11426-010-3188-8

A novel methyl-rich Ti-containing hexagonal mesoporous silica (Ti-HMS) molecular sieve with high hydrophobicity has been prepared by a two-step method involving co-condensation followed by vapor-phase methyl grafting. The sample was characterized by XRD, N₂ adsorption, FTIR, UV-visible and ²⁹Si NMR spectroscopies, TG, ICP-AES, and hydrophilicity measurements, and its catalytic performance was evaluated using the epoxidation of cyclohexene as a probe reaction. The Ti-HMS material retains a typical mesoporous structure and compared with a co-condensed Ti-HMS prepared in a one-step method possesses more methyl groups and higher hydrophobicity, and also exhibits better catalytic activity and selectivity.

Full text of DOI:10.1007/s11426-010-3188-8

SCIENCE CHINA
Chemistry
•
ARTICLES •
June 2010 Vol.53 No.6: 1337–1345
doi: 10.1007/s11426-010-3188-8
Preparation, characterization, and catalytic performance of a
novel methyl-rich Ti-HMS mesoporous molecular sieve
with high hydrophobicity
1
*
1
2
2
LI XueFeng , XU ZhiHong , GAO HuanXin & CHEN QingLing
1
College of Chemistry and Chemical Engineering, Xuchang University, Xuchang 461000, China
2
Shanghai Research Institute of Petrochemical Technology, Shanghai 201208, China
Received July 25, 2009; accepted December 18, 2009
A novel methyl-rich Ti-containing hexagonal mesoporous silica (Ti-HMS) molecular sieve with high hydrophobicity has been
prepared by a two-step method involving co-condensation followed by vapor-phase methyl grafting. The sample was charac-
2
9
2
terized by XRD, N adsorption, FTIR, UV-visible and Si NMR spectroscopies, TG, ICP-AES, and hydrophilicity measure-
ments, and its catalytic performance was evaluated using the epoxidation of cyclohexene as a probe reaction. The Ti-HMS
material retains a typical mesoporous structure and compared with a co-condensed Ti-HMS prepared in a one-step method
possesses more methyl groups and higher hydrophobicity, and also exhibits better catalytic activity and selectivity.
two-step methyl grafting, high hydrophobicity, methyl-rich, Ti-HMS, epoxidation of cyclohexene
1
Introduction
face hydrophobicity of catalysts can be improved by hy-
droxyl condensation at high temperature [9] and grafting
with organic groups [10]. The grafting methods include in
situ condensation-grafting [10, 11] and post synthesis graft-
ing [7, 10, 12, 13]. The former method is accomplished by
adding an alkoxysilane (e.g. methyltrimethoxysilane) to the
synthesis mixture [10, 11]. The alkoxysilane will hydrolyze
and condense with the silicate ester precursor, and thus the
organic groups of the alkoxysilane will be introduced into
the surface of the pores of the product resulting in the en-
hancement of the surface hydrophobicity [10, 11]. However,
samples prepared by co-condensation have lower crystallin-
ity and weaker framework stability, because the template
surfactant can be only eliminated by solvent extraction and
cannot be removed by calcination at high temperature [10].
The latter method, the so called “grafting” method, involves
organic groups being grafted through condensation of sur-
face SiOH and alkoxy groups by post-synthesis silylation
The discovery of mesoporous molecular sieves marked a
significant milestone in the development of materials sci-
ence [1–3]. Mesoporous molecular sieves have been widely
applied in catalysis, adsorption, and photoelectric devices
due to their characteristics such as large and adjustable
pores, high thermostability, and easy modification [4].
Mesoporous molecular sieves containing transition metal
ions have been shown to possess excellent catalytic activity
in oxidation [4], hydrogenation [5], and desulfurization [6]
reactions. Mesoporous sieves containing titanium have been
mainly used in hydroxylation of aromatic rings and epoxi-
dation of olefins [4]. Previous studies have demonstrated
that the epoxidation activity is associated with the dispersity
of active sites (tetracoordinated Ti) and the surface hydro-
phobicity of the molecular sieve [7, 8]. In general, the sur-
[
7, 12, 13]. Vapor-phase grafting methods are preferable
*
Corresponding author (email: snow_mount@163.com)
since no solvent is needed, a wide range of easily controlled
©
Science China Press and Springer-Verlag Berlin Heidelberg 2010
chem.scichina.com
www.springerlink.com

1338
LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
temperatures can be employed, and the operating proce-
dures are convenient [7]. However, only the surfaces of the
pore entrances can be modified during grafting treatment,
and the internal surfaces cannot be easily modified [10].
The disadvantages of both methods would be overcome if
condensation could be combined with grafting, but despite
its considerable potential significance, such a novel two-
step method has not been reported to date.
In this work, a methyl-rich Ti-containing hexagonal
mesoporous silica (HMS) molecular sieve was prepared by
such a two-step method for the first time. A methyl-con-
taining Ti-HMS was first synthesized by co-condensation
through the introduction of methyltrimethoxysilane to the
raw gel, and then trimethylsilyl groups were vapor-grafted
onto the Ti-HMS. The sample was characterized by a series
of techniques and its catalytic activity was evaluated using
the epoxidation of cyclohexene as a probe reaction.
2.2 Vapor-phase methyl grafting of Ti-Me-HMS [7, 9,
13, 15]
The vapor-phase grafting process was carried out in a 2 cm
quartz tube reactor, with the sample bed being heated using
a tube furnace with a temperature programmed device. 2 g
of a sample of Ti-Me-HMS was put into the quartz tube and
pretreated at 300 °C for 2 h in a nitrogen flow, and then
grafted with a mixture of nitrogen and hexamethyldisilazane
(
HMDS). Finally, a flow of nitrogen was passed over the
sample for 2 h to remove the unreacted HMDS. The
resulting samples are designated as 3Me-Ti-Me-HMS-x.
The preparation process is summarized in Scheme 1.
2.3 Characterization
X-ray powder diffraction (XRD) was performed on a Phil-
ips (the Netherlands) X’Pert MPD diffractometer using Cu
K radiation with  = 0.15418 nm, tube voltage of 40 kV,
tube current of 30 mA, and scanning speed of 2°/min in the

2
Experimental
2
 angle range 1°–8°.
N -adsorption measurements at low temperature were
2
2
.1 Direct synthesis of Ti-Me-HMS by co-condensation
[
2, 11, 14]
performed on an automatic ASAP 2010 instrument from
Micrometrics (USA). Samples were pretreated at 200 °C for
3
Solution A was prepared by mixing 63 g of distilled water
and 32.2 g of ethanol, followed by addition of 7.23 g of
hexadecylamine (HDA). The mixture was stirred and heated
at 50 °C to facilitate dissolution of HDA. Solution B was
prepared by adding a moles of methyltrimethoxysilane
h before adsorption. The specific surface area of samples
was calculated based on the BET formula. The pore volume
and diameter were calculated based on the BJH method.
Scanning electron microscope (SEM) and transmission
electron microscope (TEM) observations were made using
Philips XL 30 and JEOL JEM-2010 microscopes, respec-
tively.
Fourier transform infra-red (FTIR) spectra were recorded
using a Bruker (USA) IFS88 spectrometer and the samples
were analyzed as KBr wafers.
(MTMOS) and 0.68 g of tetrabutyltitanate (TBOT) to a
mixture of 6 g of isopropanol and b moles of tetraethoxysi-
lane (TEOS) with vigorous stirring for 30 min. Solution B
was slowly poured into solution A, followed by stirring for
1
8 h, filtration, washing, and drying at 110 °C for 24 h. Fi-
nally, the as-synthesized sample was refluxed for 1 h in
ethanol containing a small amount of ammonium chloride,
and then filtered, and washed. The refluxing treatment was
carried out 6 times. The sample was dried at 110 °C for 24 h,
and then calcined at 300 °C for 8 h. Samples are designated
as Ti-Me-HMS-x, where x = 0, 0.1, 0.2, and 0.3, is the value
of a(MTMOS)/(a(MTMOS) + b(TEOS)) or the molar ratio
of MTMOS in the silicon source.
2
9
Si nuclear magnetic resonance spectra with cross-
29
polarization and magic-angle spinning ( Si CP/MAS NMR)
were collected on a Bruker (USA) MSL-400 WB spec-
trometer and Si chemical shifts are given with respect to
tetramethylsilane.
Thermogravimetric analysis (TGA) was carried out on a
DuPont instrument using an alumina crucible. The test
29
Scheme 1 Schematic illustration of the procedure for preparing a methyl-rich Ti-HMS with high hydrophobicity by the two-step method.

LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
1339
temperature ranged from room temperature to 1000 °C. The
3 Results and discussion
differential curve of the TG trace is designated as DTG.
The Ti content was determined by inductively coupled
plasma-atomic emission spectrometry (ICP–AES) using a
Thermo Jarrell Ash IRIS Advantage 1000 instrument.
Diffuse reflectance ultraviolet–visible spectroscopy (DR
UV–visible) spectra were recorded on a Perkin Elmer 555
3
.1 XRD
The XRD patterns of the samples (Figure 1) all show a main
peak at low angle (2 = ca. 2°) assigned to the (100) reflec-
tion, and no other diffraction peaks are seen at higher angles.
This indicates that all the samples have long range order but
do not have any short range order; this is characteristic of
HMS mesoporous materials [2, 14]. The intensity of the
double beam spectrophotometer with BaSO as reference
4
and slit width of 2.0 nm. The samples were fast scanned.
Hydrophilicity tests were carried out as follows: A
weighed bottle containing the sample was put in a thermo-
stat containing saturated NaCl solution and weighed again
after adsorbing water for 72 h. The hydrophilicity was cal-
culated using the following equation:
(100) peak decreases with increasing amount of MTMOS
because of the reduction of sample crystallinity associated
with the introduction of methylsilyl groups into the HMS
framework. The 2 angle increases with the increase in
amount of MTMOS indicating a decrease in the unit cell
parameters. The intensity of the 100 peak increases signifi-
cantly after vapor-phase grafting, suggesting that this is
accompanied by an enhancement of the local regularity of
the crystal surface.
H 100%

m  m

mb
a
b
where m and m stand for the sample weight after and be-
a
b
fore adsorbing water, respectively.
3
.2
N
2
adsorption measurements
2
.4 Evaluation of catalytic performance
The N
2
adsorption isotherms and pore diameter distributions
The catalytic performance of the materials was evaluated in
the epoxidation of cyclohexene (CH) using cumene hy-
droperoxide (CHP) as oxidant. All reactions were carried
out in a 100 mL glass reactor. In a typical procedure, 0.3 g
of catalyst were transferred to a reaction mixture of 33%
are displayed in Figures 2 and 3. All the Ti-Me-HMS-x
(
(
w/w) CH, 53% CHP, 11.5% cumene, and 2.5% n-octane
as internal standard), and the mixture was allowed to react
at 70 °C for 2 h. The CHP consumption was determined by
standard iodometric titration. The reaction products were
analyzed using an HP4890 gas chromatograph equipped
with a capillary column (HP-1) and a flame ionization de-
tector (FID). Analysis showed that the reaction products
consisted of a mixture of cyclohexene oxide (CHO), cyclo-
hexanediol (CHD), cyclohexanone, and -hydroxylcyclo-
hexene. However, the contents of cyclohexanone and -
hydroxylcyclohexene accounted for only ~1% of the total
products, and therefore only CHO and CHD are considered
as the products of epoxidation. The conversion of CHP
(
XCHP), efficiency of CHP (ECHP), and the selectivity to
CHO (SCHO) are defined as the following equations:
XCHP 100% CHP  CHP







CHP
0
CHP
0
ECHP 100%


CHO



CHD




CHP0




SCHO 100%

CHO



CHO



CHD


where [CHP]
0
and [CHP] stand for the initial and final con-
centrations of CHP (mol/L), respectively, and [CHO] and
Figure 1 XRD patterns of samples. 1, Ti-Me-HMS-0; 2, Ti-Me-HMS-0.1;
[CHD] stand for the final concentrations of CHO and CHD
mol/L), respectively.
3
, Ti-Me-HMS-0.2; 4, Ti-Me-HMS-0.3; 5, 3Me-Ti-Me-HMS-0; 6, 3Me-
(
Ti-Me-HMS-0.1; 7, 3Me-Ti-Me-HMS-0.2; 8, 3Me-Ti-Me-HMS-0.3.

1340
LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
due to the more efficient introduction of methyl groups into
Ti-Me-HMS-0.3 by direct synthesis (co-condensation) [7].
Moreover, the sharp rises in the isotherms of the grafted
samples at high relative pressure (P/P
0
= ca. 0.99) resulting
from the capillary condensation occur at lower relative
pressures after grafting, indicating the grafting of trimethyl-
silyl groups on the external surface of Ti-Me-HMS mesopor-
ous materials leads to a decrease in pore diameter [7].
3
.3 SEM and TEM
SEM images of Ti-Me-HMS-0.3 and 3Me-Ti-Me-HMS-0.3
are displayed in Figure 4. The uniform microparticles in
directly synthesized Ti-Me-HMS-0.3, as well as the grain
size and uniformity, are all almost unchanged after post-
condensation vapor-phase grafting, suggesting that the
morphologies of the materials are not significantly affected
by the grafting process.
Figure 2
2
N -adsorption isotherms and pore size distributions of Ti-Me-
HMS-0.1 (1) and 3Me-Ti-Me-HMS-0.1 (2).
The TEM images in Figure 5 show that the mesoporous
structure of the sample is retained after grafting because the
“
worm-like” pore structure of Ti-Me-HMS-0.3 is essentially
unchanged [2].
3
.4 FTIR spectroscopy
FTIR spectra of Ti-Me-HMS samples are shown in Figure 6.
Figure 3
2
N -adsorption isotherms and pore size distributions of Ti-Me-
HMS-0.3 (1) and 3Me-Ti-Me-HMS-0.3 (2).
samples show type IV isotherms and most probable pore
diameters of about 2 nm, both typical characteristics of
mesoporous materials [2]. The lower relative pressure (P/P )
0
regions of the isotherms indicate that the samples possess a
number of micropores, and therefore have high specific
surface areas (see Table 1). This is likely to be due to new
porosity resulting from condensation of methylhydroxylsilyl
groups formed as products of methyltrimethoxysilane hy-
drolysis and the lower shrinkage level among pore walls
resulting from calcination at low temperatures [11].
As shown in Figures 2 and 3, the typical IV isotherms of
samples are retained after post-condensation vapor-phase
grafting, while the most probable pore diameter is slightly
reduced and the adsorbed volume is greatly reduced because
grafting of trimethylsilyl groups on the surface of Ti-Me-
HMS leads to a decrease in pore diameter [7]. Furthermore,
the decrease in the adsorbed volume after grafting is less for
Ti-Me-HMS-0.1 than for Ti-Me-HMS-0.3, which may be
Figure 4 SEM images of Ti-Me-HMS-0.3 (a) and 3Me-Ti-Me-HMS-0.3 (b).

LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
1341
1
cm , SiOSi framework symmetric stretching vibrations
1
1
at 803 cm , and silanol bending [18] at 464 cm .
As the methyl content of the materials is increased from
Ti-Me-HMS-0 to Ti-Me-HMS-3 the intensities of bands at
738, 3300–3700, 1630, 955 and 464 cm related to silanol
1
3
groups all decrease, whilst the intensities of bands at 2976,
932, 1413, 1462, 1281 and 842 cm related to methyl
1
2
groups all increase. This confirms that methyl groups have
been successively introduced into Ti-HMS, and that the
hydrophobicity of Ti-Me-HMS increases with increasing
methyl content. Moreover, the band at 955 cm is intense
and sharp for samples with fewer methyl groups, whilst its
intensity is decreased and shifted to 945 cm with increas-
ing incorporation of methyl groups. This is because that the
1
1
band at 955 cm represents the overlap of the 940 cm
1
1
1
band of SiOTi and the 969 cm band of SiOH, and the
band intensity is decreased as Si(OSi) CH groups replace
Si(OSi) OH and the overlapping band is shifted to lower
wavenumbers [17].
3
3
3
The spectra of 3Me-Ti-Me-HMS in Figure 7 show that
the bands ascribed to methyl groups have been distinctly
changed after vapor-phase grafting: the intensities of SiC

1
1
3
stretching bands at 750 cm , SiCH bands at 850 cm ,
1
and CH bands at 2932 and 2976 cm are all increased,
confirming that trimethylsilyl groups have been grafted onto
the surface of Ti-Me-HMS [12, 13, 15].
Figure 5 TEM images of Ti-Me-HMS-0.3 (a) and 3Me-Ti-Me-HMS-0.3 (b).
29
3
.5
Si CP/MAS NMR spectroscopy
29
Figure 8 gives the Si CP/MAS NMR spectra of Ti-Me-
HMS-0, Ti-Me-HMS-0.3 and 3Me-Ti-Me-HMS-0.3. Peaks
4
at 109, 100 and 90 ppm are attributed to Q ((SiO)
Q ((SiO)
4
Si*),
3
2
3 2 2
Si*OH), and Q ((SiO) Si*(OH) ), respectively
[
7, 13]. Peaks at 63 and 54 ppm are attributed to
3
2
3 2
T (MeSi*(OSi) ) and T (Me(OH)Si*(OSi) ) resulting from
Figure 6 FTIR spectra of Ti-Me-HMS-0 (1), Ti-Me-HMS-0.1 (2), Ti-
Me-HMS-0.2 (3) and Ti-Me-HMS-0.3 (4).
Bands are assigned to adsorbed water on hydrogen-bonded
1
silanol groups located at 3300–3700 cm , CH stretching
1
of grafted methyl groups [15] at 2976, 2932 and 1413 cm ,

1
water adsorbed on silanol groups at 1633 cm , SiCH
3
1
stretching [12] at 1281 cm , SiOSi antisymmetric
stretching at 1082 cm , SO and SiOTi vibrations [16,
7] at 950–960 cm , CH bending vibrations [13] at 842
1
Figure 7 FTIR spectra of 3Me-Ti-Me-HMS-0 (1), 3Me-Ti-Me-HMS-0.1
(2), 3Me-Ti-Me-HMS-0.2 (3) and 3Me-Ti-Me-HMS-0.3 (4).
1
1

1342
LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
2
9
Figure 8
Si CP/MAS NMR spectra of Ti-Me-HMS-0 (1), Ti-Me-
HMS-0.3 (2) and 3Me-Ti-Me-HMS-0.3 (3).
different degrees of condensation between methylsilyl and
4
2
silanol groups [11, 16]. As shown in Figure 8, the Q /(Q +
3
4
Q + Q ) ratios of Ti-Me-HMS-0.3 and 3Me-Ti-Me-HMS-
.3 are higher than that of Ti-Me-HMS-0, suggesting the Q
0
4
content increases after methyl grafting. The above results
indicate that methyl groups have been grafted onto the
framework of Ti-HMS. Ti-Me-HMS contains considerable
2
3
3
amounts Q and Q because many lone (Q (SiO)
3
Si*-OH))
Figure 9 TG-DTG curves of Ti-Me-HMS-0.3 (a) and 3Me-Ti-Me-
HMS-0.3 (b).
2
and geminal (Q ((SiO) Si*-(OH) )) silanol groups have
2 2
been retained due to the relatively low calcination tempera-
ture (300 °C). The peak at 14 ppm is characteristic of
Figure 10 shows the relationship between weight loss
and x value (a(MTMOS)/(a(MTMOS) + b(TEOS))) within
the temperature range 400–700 °C for Ti-Me-HMS and
3 3 3
(CH ) Si*O(SiO) groups formed by linkage of trimethyl-
silyl and silanol groups after vapor-phase grafting [7, 13].
3
Furthermore, as shown in Figure 8, the T contents are also
3
Me-Ti-Me-HMS samples. The methyl content of Ti-Me-
greatly increased after vapor-phase grafting of Ti-Me-HMS,
suggesting that large numbers of silanol groups have been
eliminated.
HMS increases with x and their relationship is linear. Mean-
while, the total methyl content decreases with x because
samples with lower values of x, e.g. Ti-Me-HMS-0.1 have
3
.6 TGA (TG–DTG)
Figure 9 shows TG–DTG curves of Ti-Me-HMS-0.3 and
3
Me-Ti-Me-HMS-0.3.
The weight loss below 170 °C is ascribed to the removal
of physisorbed water, that at 230–400 °C ascribed to loss of
chemisorbed water and residual organic template (HDA),
that at 400–700 °C ascribed to thermal decomposition and
oxidation of methyl groups, and that above 700 °C ascribed
to condensation and framework reforming of silanol groups
[
7, 17].
As shown in Figure 9, the weight loss within the range
00–700 °C in the DTG curve of Ti-Me-HMS-0.3 indicates
4
that methylsilyl groups have been introduced into the mo-
lecular sieve framework. Ti-Me-HMS-0.3 has higher weight
loss within the temperature range 400–700 °C than Ti-Me-
HMS-0.3, showing that both methylsilyl and trimethylsilyl
groups have been introduced into the framework [7, 17].
Figure 10 Comparison of weight loss of samples in the temperature
range 400–700 °C. 1, The weight loss trend of Ti-Me-HMS; 2, the weight
loss trend of 3Me-Ti-Me-HMS; 3, the difference between the weight losses
of Ti-Me-HMS and 3Me-Ti-Me-HMS.

LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
1343
more silanol groups to condense with trimethylsilyl groups
whereas the samples with larger values of x have fewer si-
lanol groups and more methyl groups, thus decreasing the
content of grafted trimethylsilyl groups.
These conclusions are consistent with the decrease in the
difference in methyl contents of 3Me-Ti-Me-HMS before
and after vapor-phase grafting. The above results show that
with increasing MTMOS dosage (increasing value of x), the
methyl group contents of Ti-Me-HMS formed by direct
synthesis increase whereas the methyl group contents of
decrease with the increasing value of x and increasing
methyl group content, suggesting that the Ti content in-
creases with increasing methyl group content. This is con-
sistent with the results reported by Igarashi [11] (for
Ti-MCM-41 grafted in situ by phenyl groups) and Bhaumik
[16] (for Ti-MCM-41 grafted in situ by methyl groups).
After vapor-phase grafting of trimethylsilyl groups, the Ti
content decreases because the Si content increases as
trimethylsilyl groups are grafted to the Ti-Me-HMS.
3
Me-Ti-Me-HMS formed by vapor-phase grafting decrease.
3
.9 DR UV–visible spectroscopy
Figures 12 and 13 show the DR UV–visible spectra of
Ti-Me-HMS-0.1 and Ti-Me-HMS-0.3 before and after va-
por-phase grafting, respectively. Ti-Me-HMS-0.1 and Ti-
Me-HMS-0.3 exhibit an intense peak at 217–220 nm as-
cribed to tetracoordinated Ti species and the peak shifts to
212–215 nm after vapor-phase grafting. A similar blue shift
has been attributed by Peña [7] to the opened tetracoordi
nated Ti being transformed to an enclosed tetracoordinated
Ti species. Moreover, the peak intensity decreases after va-
por-phase grafting because of the decrease in Ti content
(increase in Si content) [7].
3
.7 Hydrophilicity measurements
Table 1 shows that Ti-Me-HMS-x (x = 0.1, 0.2, 0.3) samples
have much lower water adsorbing ratios (hydrophilicity)
than Ti-Me-HMS-0 without methyl groups, suggesting that
the hydrophobicity has been significantly improved by in-
troducing methyl groups. The water adsorbing ratios de-
crease gradually with increasing MTMOS dosage, suggest-
ing that the hydrophobicity increases with the content of
methyl groups. This is consistent with the TGA data. The
water adsorbing ratios of 3Me-Ti-Me-HMS are dramatically
reduced after vapor-phase grafting, suggesting that the re-
sidual silanol groups have been eliminated by grafting of
trimethylsilyl groups, which further enhances the surface
hydrophobicity.
Figure 11 shows the physical states of Ti-Me-HMS-0.3
and 3Me-Ti-Me-HMS-0.3 in water and visually reflects the
difference in hydrophilicity/hydrophobicity before and after
vapor-phase grafting. Almost all of 3Me-Ti-Me-HMS-0.3
powder floats on the water surface whereas most of Ti-
Me-HMS-0.3 powders sinks down, suggesting that Ti-Me-
HMS-0.3 has high hydrophilicity and 3Me-Ti-Me-HMS-0.3
has very low hydrophilicity as a result of vapor-phase
grafting.
3
.10 Catalytic performance
As shown in Table 1, CHP conversion over Ti-Me-HMS-x
increases slightly with increasing value of x; this can be
attributed to the corresponding increase in Ti content. Both
conversion and efficiency of CHP are increased after va-
por-phase grafting because the number of acidic OH sites
is reduced after trimethylsilyl groups are grafted onto SiOH
and TiOH (the content of opened Ti sites increases) which
reduces the decomposition side reactions of CHP. Further-
more, CHO selectivities of 3Me-Ti-Me-HMS-x are near to
100% after vapor-phase grafting because the hydrophilicities
of the catalysts have been greatly reduced which prevents
ring opening of epoxides. This results in the selectivity of
the catalysts being improved and the rate of deactivation
being reduced [7, 15]. However, the increases in catalytic
3
.8 ICP–AES
Table 1 shows that Si/Ti ratios in Ti-Me-HMS-x samples
Table 1 Physicochemical properties and catalytic performance of catalysts

1
2
1
a)
b)
3
1
c)
Sample number
Ti-Me-HMS-0
Si/Ti (mol mol )
S
BET (m g )
D (nm)
V
(cm g )
H (%)
X
CHP (%)
E
CHP (%)
S
CHO (%)
d)
79
73


2.10


0.632

19.2
13.5
8.1
64.3
66.4
68.8
70.7
72.1
74.7
77.1
79.6
93.4
93.0
92.8
92.6
99.3
98.0
99.5
98.0
97.5
97.6
97.0
96.5
100
Ti-Me-HMS-0.1
Ti-Me-HMS-0.2
Ti-Me-HMS-0.3
1144.5

68
64
1197.5

2.04

0.664

7.0
3Me-Ti-Me-HMS-0
3Me-Ti-Me-HMS-0.1
3Me-Ti-Me-HMS-0.2
3Me-Ti-Me-HMS-0.3
117
110
103
99
0.08
0.13
0.10
0.10
1060.2

2.08

0.615

99.8
100
820.3
2.04
0.418
100
a) Average pore diameter; b) pore volume calculated from BET measurements; c) hydrophilicity of samples measured on the basis of the method in Sec-
tion 2.3; d)  not measured.

1344
LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
Ti-Me-HMS before grafting are low. Therefore, future work
must focus on how to enhance the Ti contents in molecular
sieve samples prepared by the co-condensation method.
4
Conclusions
The highly hydrophobic 3Me-Ti-Me-HMS mesoporous
molecular sieve catalyst was prepared by co-condensation
of TEOS and MTMOS, in order to introduce methyl groups
in the product (Ti-Me-HMS), followed by grafting of
trimethylsilyl groups for surface modification.
The characterization results show that Ti-Me-HMS has a
typical mesoporous structure and high surface hydrophobic-
ity. The grafting process results in the pore diameter and
specific surface area of 3Me-Ti-Me-HMS being reduced to
some extent compared with Ti-Me-HMS, whilst the mesopor-
ous structure is retained and the hydrophobicity is signifi-
cantly enhanced.
Figure 11 Illustrations of the physical states of 3Me-Ti-Me-HMS-0.3 (1)
and Ti-Me-HMS-0.3 (2) in water.
The activity and selectivity of 3Me-Ti-Me-HMS in the
epoxidation of cyclohexene are higher than Ti-Me-HMS,
showing the beneficial effect of the vapor-phase grafting
process. On vapor-phase grafting, the content of acidic
TiOH and SiOH sites is significantly reduced, and there-
fore the surface hydrophobicity of the catalyst is enhanced
significantly. As a result, there is a decrease in the amount
of decomposition side reactions of CHP and ring opening of
epoxides is avoided, and thus the deactivation process is
dramatically suppressed .
The above results suggest that further research on the
improvement of the two-step method should focus on the
following three aspects. Firstly, the solvent extraction proc-
ess should be improved so that the template can be removed
as completely as possible whilst reducing the loss of Ti. The
thermal treatment should also be improved in order to in-
crease the condensation degree of surface silanol groups and
the molecular sieve framework, which will increase the
regularity of the mesopores. Finally, the two-step method
should be used to introduce other organic groups such as
sulfhydryl, halogen, amino, and epoxy, by means of which
the other novel materials will be obtained.
Figure 12 DR UV-visible spectra of Ti-Me-HMS-0.1 (1) and 3Me-Ti-
Me-HMS-0.1 (2).
This work was supported by the Key Program for Science and Technology
Development of Henan Province (092102210221) and the Program for
Natural Science Research of the Education Department of Henan Province
(2009B150025).
1
2
3
4
Kresge CT, Leonowicz ME, Roth WJ, Vartuli JC, Beck JS. Ordered
mesoporous molecular sieves synthesized by a liquid-crystal template
mechanism. Nature, 1992, 359: 710–712
Tanev PT, Chibwe M, Pinnavaia TJ. Titanium-containing mesopor-
ous molecular sieves for catalytic oxidation of aromatic compounds.
Nature, 1994, 368: 321–323
Zhao DY, Feng JL, Huo QS, Melosh N, Fredrickson GH, Chmelka
BF, Stucky GD. Triblock copolymer syntheses of mesoporous silica
with periodic 50 to 300 angstrom pores. Science, 1998, 279: 548–552
Corma A. From microporous to mesoporous molecular sieve materi-
Figure 13 DR UV-visible spectra of Ti-Me-HMS-0.3 (1) and 3Me-Ti-
Me-HMS-0.3 (2).
activity (CHP conversion) over 3Me-Ti-Me-HMS-x are
slight after vapor-phase grafting because the Ti contents of

LI XueFeng, et al. Sci China Chem June (2010) Vol.53 No.6
1345
als and their use in catalysis. Chem Rev, 1997, 97: 2373–2420
Wang YJ, Wang XL, Xie GQ, Lu JQ, Jin WY, Liu XJ, Luo MF.
Crotonaldehyde hydrogenation over CuO/SBA-15 catalyst (in Chi-
nese). J Catal, 2008, 29: 482–488
Hou DM, Zhou F, Li X, Wang LY, Wang AJ. Hydrodesulfurization
performance of MCM-41 supported Pd catalysts (in Chinese). Acta
Petrolei Sinica (Petroleum Processing Section), 2009, 25: 167–172
Peña ML, Dellarocca V, Rey F, Corma A, Coluccia S, Marchese L.
Elucidating the local environment of Ti(IV) active sites in Ti-MCM-
12 Cagnoli MV, Casuscelli SG, Alvarez AM, Bengoa JF, Gallegos NG,
Crivello M E, Herrero ER, Marchetti SG. Ti-MCM-41 silylation:
Development of a simple methodology for its estimation—Silylation
effect on the activity and selectivity in the limonene oxidation with
5
6
7
H
2 2
O . Catal Today, 2005, 107–108: 397–403
13 Lin KF, Wang LF, Meng FY, Sun ZH, Yang Q, Cui YM, Jiang DZ,
Xiao FS. Formation of better catalytically active titanium species in
Ti-MCM- 41 by vapor-phase silylation. J Catal, 2005, 235: 423–427
14 Tuel A. Modification of mesoporous silicas by incorporation of het-
eroelements in the framework. Micropor Mesopor Mater, 1999, 27:
23–33
4
8: A comparison between silylated and calcined catalysts. Micropor
Mesopor Mater, 2001, 44–45: 345–356
8
9
0
Marchese L, Maschmeyer, T, Gianotti E, Coluccia S, Thomas JM. Prob-
ing the titanium sites in Ti-MCM-41 by diffuse reflectance and photolu-
minescence UV-vis spectroscopies. J Phys Chem, 1997, 101: 8836–8838
15 Sever R R, Alcala R, Dumesic J A, Root T W. Vapor-phase silylation
of MCM-41 and Ti-MCM-41. Micropor Mesopor Mater, 2003, 66:
53–67
16 Bhaumik A, Tatsumi T. Organically modified titanium-rich Ti-
MCM-41, efficient catalysts for epoxidation reactions. J Catal, 2000,
189: 31–39
17 Müller CA, Maciejewski M, Mallat T, Baiker A. Organically modi-
fied titania-silica aerogels for the epoxidation of olefins and allylic
alcohols. J Catal, 1999, 184: 280–293
Li KT, Lin PH, Lin SW. Preparation of Ti/SiO
2
catalysts by chemical
vapor deposition method for olefin epoxidation with cumene hydrop-
eroxide. Appl Catal, 2006, 301: 59–65
Melero JA, Grieken R van, Morales G. Advances in the synthesis and
catalytic applications of organosulfonic-functionalized mesostruc-
tured materials. Chem Rev, 2006, 106: 3790–3812
1
11
Igarashi N, Kidani S, Ahemaito R, Hazuhito H, Tatsumi T. Direct or-
ganic functionalization of Ti–MCM-41: Synthesis condition, organic
content, and catalytic active. Micropor Mesopor Mater, 2005, 81: 97–105
18 Zhang MY, Wang LF, Huang ZT. Synthesis conditions and catalytic
performance of hexagonal mesoporous silica containing copper (in
Chinese). J Catal, 2003, 24: 914–918