J. Silvestre-Alberó et al. / Applied Catalysis A: General 507 (2015) 14–25
15
The hydrophobic/hydrophilic character of the siliceous framework
by post-synthetic modification of the silanol groups. The grafted
organic species decrease the concentration of silanol and Ti OH
groups on the surface, thus increasing the hydrophobic charac-
the epoxidation of olefins have shown that silylated Ti-MCM-41
samples exhibits an improved catalytic activity and selectivity to
the epoxide, the catalytic conversion being highly improved when
organic hydroperoxides are used as oxidants [3,22,23]. Although
the improved catalytic behavior on the silylated materials has been
postulated to be due to their larger hydrophobicity and not to a
change in the nature of the Ti active sites, a complete physico-
chemical characterization seems mandatory in order to clarify the
real role of these organic functionalities.
Immersion calorimetry is a very useful technique for the surface
characterization of solids. It has been widely used for the characteri-
zation of microporous solids, mainly microporous carbons [24–27].
When a solid is immersed into a non-reacting liquid there is a heat
evolution called “heat of immersion” or “heat of wetting”. Com-
monly, this heat of immersion can provide information about the
surface area available for a given molecule. However, the existence
of specific interactions between the solid surface and the immer-
sion liquid must also be taking into consideration. Accordingly, an
appropriate selection of the immersion liquid can be used to char-
acterize both the textural and/or the surface chemical properties
of porous solids. In the case of silylated and non-silylated MCM-41
samples, the use of both non-polar molecules, i.e. hydrocarbons,
and polar molecules such as water and alcohols, could give light in
the evaluation of the hydrophobic/hydrophilic character of these
synthesized materials.
With this in mind, we have combined spectroscopic techniques
(e.g. XANES, UV-Visible, 29Si MAS-NMR, etc.) and calorimetric
measurements to better understand the changes in the surface
chemistry of Ti-MCM-41 samples due to a post-synthesis silyla-
tion treatment. This information will be very useful to ascertain
the role of these trimethylsilyl groups in the epoxidation of cyclo-
hexene using tert-butylhydroperoxide as oxidant. The effect of
the silylation degree as well as the influences of both water and
diol molecules contents on the catalytic activity and selectivity
will be analyzed and discussed. Examples of the extended use of
these hydrophobic Ti-mesoporous catalysts together with organic
hydroperoxides for the highly selective and efficient epoxidation
of natural terpenes will be also provided.
(TBHP, 80 wt% in di-tert-butylperoxide/water 3/2, Fluka) was used
as oxidizing agent.
2.2. Synthesis of catalytic materials
The Ti-MCM-41 (≈2.0 wt% of TiO2) mesoporous material
was prepared as reported in Refs. [19,28] starting from 3.11 g
(8.53 mmol) of cetyl-tri-methyl-ammonium bromide (C16TAB,
Aldrich) dissolved in 20.88 g (1160 mmol) of water (Milli-Q).
Then, 5.39 g (14.81 mmol) of tetramethylammonium hydroxide
(TMAOH, 25 wt% aqueous solution, Aldrich) and 0.21 g (0.86 mmol)
of titanium tetraethoxide (TEOT, Aldrich) were added to the above-
mentioned solution, and the system was stirred until the titanium
compound was fully dissolved. Silica (3.43 g, 56.91 mmol) was then
added, giving rise to a gel having the following molar composition:
SiO2:0.015 Ti(OEt)4:0.15C16TMAB:0.26 TMAOH:24.3H2O. The gel
was stirred at room temperature for 1 h at 250 rpm. The result-
ing mixture was placed into autoclaves and heated at 373 K under
autogenous pressure for 48 h. Then, the solid was recovered by fil-
tration, washed thoroughly with distilled water, and dried at 333 K
during 12 h. The solid material was placed in a tubular quartz reac-
tor where the temperature is increased from room temperature to
813 K (under dry N2 flow) followed by a step at 813 K during 6 h
(under dry air flow). Then the solid is cooled at room temperature.
The final catalyst contains 2.0 wt% (expressed as TiO2) based on
,
´
˚
with a narrow pore size distribution centred at 42.3 A, and shows
a band in the UV–Vis spectrum centred at 220 nm.
The post-synthesis silylation of Ti-MCM-41 material repre-
sented in Scheme 1 was performed as follows. Typically, 2.0 g of
Ti-MCM-41 was dehydrated at 373 K and 10−3 Torr for 2 h. The
sample was cooled at room temperature. Then, a solution of hexam-
ethyldisilazane [(CH3)3Si NH Si(CH3)3, HMDS] in 30 g of toluene
was added. The concentrations of HMDS solutions were adapted
to attain SiMe3/SiO2 molar ratios from 0 to 1. The resulting mix-
ture was refluxed at 393 K for 90 min and washed with anhydrous
toluene. The end product was dried at 333 K.
2.3. Catalyst characterization
Phase purity of the catalysts was determined by X-ray diffrac-
tion (XRD) in a Philips X’Pert MPD diffractometer equipped with
a PW3050 goniometer (CuKa radiation, graphite monochromator),
provided with a variable divergence slit and working in the fixed
irradiated area mode. 29Si MAS NMR spectra of Ti-MCM-41 mate-
rials were recorded at a spinning rate of 5 kHz on a Varian VXR
400S WB spectrometer. Diffuse reflectance UV–Vis (DRUV) spectra
of samples were recorded in a Cary 5 Varian spectrometer equipped
with a “Praying Mantis” cell from Harrick.
2. Experimental section
2.1. Reactants
XANES data were collected on XAS-2 station at the Laboratory
for Electromagnetic Radiation Utilization (LURE) of the CNRS in
Orsay (France), by using the synchrotron radiation generated by
a DCI ring (1.85 GeV and 250 mA) and a double crystal Si(311)
monochromator (for Ti K-edge) [29]. The spectra were measured
at room temperature under He flow employing an 8 element solid-
state Canberra detector for measuring the fluorescence yield. The
fluorescence detection mode was used since the materials con-
tained very low amounts of Ti (<2.0 wt% as TiO2). The spectra were
normalized with respect to the first EXAFS oscillation (40–45 eV
above the adsorption-edge). Typically, the materials were previ-
ously compacted to form self-supported wafers of ≈100 mg, and
then dehydrated at 300 ◦C during 2 h. In the case of re-hydrated
samples, the re-hydration treatment was carried out under air dur-
ing 2 h.
Cyclohexene (≥99.5%, Fluka), (1S)-␣-pinene (98%, Aldrich), (R)-
(+)-limonene (97%, Aldrich), terpinolene (94%, ACEDESA, S.A.),
␣-cedrene (99%, Fluka), n-nonane (99%, Aldrich), and trans-1,2-
cyclohexane-diol (98%, Aldrich) were used as received. Acetonitrile
(99.5%, Multisolvent, Scharlau), methanol (99.8%, LiChrosolv,
Merck), tert-butanol (99.5%, ACS, Scharlau), cyclohexane (99%,
Aldrich), dichloromethane (99%, Scharlau), toluene (99%, Schar-
lau), and water (Milli-Q quality, Millipore) were employed without
any previous treatment. For catalysts synthesis the following reac-
tants were used: cetyl-tri-methyl-ammonium bromide (C16TMAB,
Aldrich), tetramethyl–ammonium hydroxide (TMAOH, 25 wt% in
water, Aldrich), titanium tetraethoxide [Ti(OEt)4, TEOT, Alfa], hex-
amethyldisilazane [(CH3)3Si NH Si(CH3)3, HMDS, 97%, Aldrich],
and silica (Aerosil 200, DEGUSSA). Finally, tert-butylhydroperoxide