A procedure for the preparation of Ti-Beta zeolites for catalytic epoxidation with hydrogen peroxide

Article abstract of DOI:10.1039/c3gc42534g

Ti-Beta zeolite has been successfully prepared via a reproducible and scalable two-step post-synthesis strategy, which consists of creating vacant T sites with associated silanol groups by dealumination of H-Beta and subsequent dry impregnation of the resulting Si-Beta with titanocene dichloride. The mechanism of Ti incorporation into the framework of Beta is investigated by diffuse reflectance infrared Fourier transform (DRIFT) and multinuclear solid-state nuclear magnetic resonance (SSNMR) spectroscopy. Characterization results obtained from diffuse reflectance ultraviolet-visible (UV-vis) and X-ray photoelectron spectroscopy (XPS) reveal that the majority of incorporated Ti species exist in the form of isolated tetrahedrally coordinated Ti(iv) in the zeolite framework while a minority exists in the form of isolated octahedrally coordinated Ti(vi) at framework or extra-framework positions. The obtained Ti-Beta zeolites are highly active and selective catalysts for the epoxidation of unsaturated ketones, e.g. 2-cyclohexen-1-one, with hydrogen peroxide as an oxidant. A quasilinear correlation between the epoxidation rate and the number of framework Ti(iv) species could be drawn evidencing that these Ti(iv) species are responsible for the epoxidation activity of the Ti-Beta zeolites under study. The impact of preparation parameters and reaction conditions on the catalytic performances of the Ti-Beta zeolites in the epoxidation of unsaturated organic compounds with hydrogen peroxide is discussed in detail. the Partner Organisations 2014.

Full text of DOI:10.1039/c3gc42534g

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
A procedure for the preparation of Ti-Beta zeolites
for catalytic epoxidation with hydrogen peroxide†
Cite this: Green Chem., 2014, 16,
2281
Bo Tang,a Weili Dai,a Xiaoming Sun,a,b Naijia Guan,a Landong Li*a and
Michael Hungerb
Ti-Beta zeolite has been successfully prepared via a reproducible and scalable two-step post-synthesis
strategy, which consists of creating vacant T sites with associated silanol groups by dealumination of
H-Beta and subsequent dry impregnation of the resulting Si-Beta with titanocene dichloride. The mech-
anism of Ti incorporation into the framework of Beta is investigated by diffuse reflectance infrared Fourier
transform (DRIFT) and multinuclear solid-state nuclear magnetic resonance (SSNMR) spectroscopy.
Characterization results obtained from diffuse reflectance ultraviolet-visible (UV-vis) and X-ray photo-
electron spectroscopy (XPS) reveal that the majority of incorporated Ti species exist in the form of isolated
tetrahedrally coordinated Ti(IV) in the zeolite framework while a minority exists in the form of isolated
octahedrally coordinated Ti(VI) at framework or extra-framework positions. The obtained Ti-Beta zeolites
are highly active and selective catalysts for the epoxidation of unsaturated ketones, e.g. 2-cyclohexen-
1-one, with hydrogen peroxide as an oxidant. A quasilinear correlation between the epoxidation rate and
the number of framework Ti(IV) species could be drawn evidencing that these Ti(IV) species are responsible
for the epoxidation activity of the Ti-Beta zeolites under study. The impact of preparation parameters and
reaction conditions on the catalytic performances of the Ti-Beta zeolites in the epoxidation of unsatu-
rated organic compounds with hydrogen peroxide is discussed in detail.
Received 12th December 2013,
Accepted 18th January 2014
DOI: 10.1039/c3gc42534g
www.rsc.org/greenchem
promising way is by the Ti(IV)-modification of Beta zeolites
characterized by 12-ring pores (BEA: 〈100〉 0.66 × 0.67 nm ↔
1. Introduction
Incorporation of metals into zeolites brings about many inter- [001] 0.56 × 0.56 nm), which substantially expands the sub-
esting properties, which can be utilized for heterogeneous strate scope for liquid-phase oxidation on Ti(IV)-containing
catalysis. Of particular importance is the exploitation of TS-1, zeolites.9 Up to now, Ti-Beta zeolites have been successfully
being a Ti(IV)-containing MFI-type zeolite, as a robust catalyst applied in the epoxidation of linear and cyclic olefins, mono-
in green oxidation reactions with hydrogen peroxide, such as terpenes, methyl esters of fatty acids, unsaturated alcohols,
hydroxylation of aromatics, oxidation of alcohols to carboxyl and α,β-unsaturated carbonyls with hydrogen peroxide
compounds, epoxidation of propylene, and cyclohexanone (Scheme 1). In all these reactions, Ti-Beta zeolites act as active
ammoximation.1–4 The success of TS-1 as a milestone has heterogeneous catalysts with high hydrogen peroxide
initiated extensive research on sustainable oxidation reactions efficiency.10
catalyzed by Ti(IV)-containing zeolites. On the other hand, the
The preparation of Ti-Beta is an essential issue to be
catalytic application of TS-1 is limited to relatively small sub- addressed. Generally, Ti can be directly incorporated into the
strate molecules due to its 10-ring pore system (MFI: [100] 0.51 Beta zeolites during the formation of the BEA framework, and
× 0.55 nm ↔ [010] 0.53 × 0.56 nm), and therefore, Ti(IV)-con- different procedures, e.g. conventional hydrothermal crystalli-
taining zeolites with larger pore sizes are required to overcome zation,9 dry-gel conversion,11 and crystallization in fluoride
the diffusion restrictions for reactant molecules.5–8
A
medium,12 have been reported. Although highly active Ti-Beta
zeolites can be obtained via direct synthesis routes, most of
these routes are quite complicated and time-consuming.
Accordingly, a lack of reproducibility in the direct synthesis
procedure of Ti-Beta zeolites may be complained and large-
scale synthesis for industrial applications is yet to be accom-
plished. In this context, alternative post-synthesis strategies
have been developed for the preparation of Ti-Beta zeolites.
Chemical vapor deposition (CVD) offers the possibility of
aKey Laboratory of Advanced Energy Materials Chemistry (Ministry of Education),
College of Chemistry, Nankai University, Tianjin 300071, P. R. China.
E-mail: lild@nankai.edu.cn; Fax: (+)86 22 23500341; Tel: (+)86 22 23500341
bInstitute of Chemical Technology, University of Stuttgart, 70550 Stuttgart, Germany.
E-mail: michael.hunger@itc.uni-stuttgart.de
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c3gc42534g
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 2281–2291 | 2281

View Article Online
Paper
Green Chemistry
industrial production of Ti-Beta zeolites and it can be extended
to the incorporation of other metal ions into a zeolite matrix
providing that a dealumination of the zeolite host is possible
and the organometallic precursor is available. These post-syn-
thesis modified Ti-Beta zeolites are expected to be suitable
catalysts for epoxidation reactions, one of the most important
classes of organic transformations in fine chemical and phar-
macy industries.19–22 Typically, the epoxidation of 2-cyclo-
hexen-1-one, a type of cyclic unsaturated ketone, is selected as
a model reaction to evaluate the catalytic performances of Ti-
Beta zeolites using hydrogen peroxide as an oxidant in the
present study. Furthermore, special emphasis has been laid on
the mechanism of Ti incorporation into the framework of
zeolite Beta and its effects on epoxidation activity.
Scheme 1 Examples of chemical transformations realized by Ti(IV)-
2. Experimental
containing Beta zeolites and hydrogen peroxide.
2.1. Sample preparation
preparing Ti(IV)-containing zeolites with titanium tetrachloride Ti(IV)-containing zeolite Beta, i.e. Ti-Beta, was prepared via a
used as the titanium source deposited on the zeolite host.13,14 two-step post-synthesis procedure consisting of the dealumina-
In addition, post-synthesis grafting of the organometallic pre- tion of a zeolite H-Beta to get a Si-Beta material and sub-
cursor titanocene dichloride (Cp2TiCl2) on zeolite hosts has sequent incorporation of Ti into the Si-Beta framework, as
been reported for obtaining so-called single-site zeolite cata- described in Scheme 2. To obtain a highly siliceous Si-Beta
lysts with
a remarkable activity in the epoxidation of material (nSi/nAl > 1800), commercial H-Beta zeolite (nSi/nAl
=
olefins.15–18 In comparison with titanium tetrachloride, 13.5, Sinopec Co.) was treated with 13 mol L−1 HNO3 solution
Cp2TiCl2 seems to be a more suitable Ti source for the prepa- (20 mL g−1 zeolite) for 20 h and then washed with deionized
ration of Ti(IV)-containing zeolites, because the latter is less water until the pH value of the filtrate was neutral. After drying
reactive, less moisture-sensitive, and easier to handle. at 353 K overnight, the Si-Beta powder was ground with an
However, large amounts of hazardous chlorinated solvents are appropriate amount of Cp2TiCl2 rapidly to obtain an intimate
employed during the organometallic grafting process, which mixture of the two solids. The Si-Beta–Cp2TiCl2 mixture was
may also show negative effects on the incorporation of Ti into then calcined in flowing air at 823 K for 12 h to derive zeolite
zeolite frameworks, resulting in poor reproducibility. More- Ti-Beta. The obtained samples are assigned n wt% Ti-Beta with
over, the ligand removal process may lead to the transform- weight percent (n) of Ti atoms on the dehydrated Si-Beta
ation of Ti species and the formation of unreactive bulk TiO2.
In a comprehensive consideration, a dry impregnation of loadings of 1 wt%, 2 wt%, 3 wt%, and 5 wt% were prepared.
zeolite Beta containing numerous defect SiOH groups with a
For comparison, a microporous TS-1 zeolite and a meso-
zeolite. In this study, four different Ti-Beta zeolites with Ti
Ti source, such as by carefully grinding a dealuminated zeolite porous Ti-grafted MCM-41 (denoted Ti-MCM-41) material were
Beta with a solid organometallic complex and performing a prepared according to the literature.15,23 In a typical synthesis
subsequent calcination of this physical mixture, is regarded as of TS-1, tetraethyl orthotitanate was added dropwise to tetra-
an ideal route for incorporating metal ions into the zeolite ethyl orthosilicate and the mixture was stirred for 2 h to obtain
framework. In this procedure, no hazardous solvents are used a clear solution, to which tetrapropylammonium hydroxide
and the absence of solvents completely prevents the solvation (TPAOH) and deionized water were added under stirring at
shell surrounding the Ti atoms, which could potentially lead room temperature and then heated to 353 K for 5 h to aid
to significant diffusion limitations inside the zeolite pores. hydrolysis. The resulting gel mixture (0.022TiO2 : 1.0SiO2 :
Therefore, we focused on the development of a simple and 0.35TPAOH : 35H2O) was transferred into a Teflon-lined stain-
scalable two-step post-synthesis procedure for preparing Ti- less steel autoclave for static crystallization at 448 K for 48 h.
Beta zeolites based on the dry impregnation of dealuminated The solid products were recovered by centrifuging, washed
zeolite Beta with Cp2TiCl2 as a Ti source. Briefly, vacant T (T = with deionized water, dried at 353 K overnight and calcined in
Al) sites with associated silanol groups are first created by air at 823 K for 12 h. For the synthesis of Ti-MCM-41, siliceous
treating commercial H-Beta zeolite with concentrated nitric MCM-41 (0.4 g)24 was stirred with Cp2TiCl2 (0.02 g) in chloro-
acid solution and the resulting Si-Beta zeolite is subsequently form (10 mL) for 0.5 h. Then, triethylamine (1 mL) was added
ground with the Ti precursor Cp2TiCl2 to produce a physical under stirring and the colour of the mixture gradually changed
Si-Beta–Cp2TiCl2 mixture, followed by calcination to derive a from red to yellow within 2 h, indicating the substitution of
Ti(IV)-containing zeolite Beta. This route is suitable for the the chloride ligands by siloxide ligands. The yellow solid was
2282 | Green Chem., 2014, 16, 2281–2291
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Scheme 2 Schematic representation of the incorporation of tetrahedrally coordinated Ti(IV) species into Beta zeolite.
separated by centrifuging and washed with chloroform three
times, followed by calcination at 823 K for 12 h.
Raman analysis was carried out using a Renishaw InVia
Raman spectrometer and the spectra were obtained with the
Ti(IV)-containing Beta zeolites with a designed Ti loading of green line of an argon-ion laser (514.53 nm) in a micro-Raman
5.0 wt% (5 wt% Ti-Beta) were also prepared via different routes configuration.
to investigate the influence of the preparation procedure on
X-ray photoelectron spectra (XPS) were recorded on a Kratos
the catalytic properties of the various Ti-Beta zeolites. Briefly, Axis Ultra DLD spectrometer with a monochromated Al-Kα
zeolite TiO2/Si-Beta-w was synthesized via wet impregnation, in X-ray source (hν = 1486.6 eV), hybrid (magnetic/electrostatic)
which the Si-Beta support was suspended in cyclohexane and optics and a multi-channel plate and delay line detector
tetrabutyl orthotitanate (TBOT) was added. The mixture was (DLD). All spectra were recorded by using an aperture slot of
refluxed at 341 K for 24 h and the solid was calcined in air at 300 × 700 microns. Survey spectra were recorded with a pass
823 K for 12 h to obtain the final product. Zeolite TiO2/Si-Beta- energy of 160 eV and high-resolution spectra with a pass
m was prepared by mechanically mixing the Si-Beta support energy of 40 eV. Accurate binding energies ( 0.1 eV) were deter-
with TiO2 (Degussa P25), followed by calcination at 823 K for mined with respect to the position of the adventitious C 1s
12 h to obtain the final product. Finally, zeolite Ti-H-Beta was peak at 284.8 eV.
prepared using the same procedure as that for the post-syn-
Diffuse reflectance infrared Fourier transform (DRIFT) spectra
thesis modified zeolite Ti-Beta except that H-Beta zeolite was of samples were measured on a Bruker Tensor 27 spectrometer
used as the zeolite host instead of siliceous Si-Beta.
with 128 scans at a resolution of 2 cm−1. Self-supporting pellets
made of sample material were placed in the reaction chamber
and pretreated in flowing dry air at 473 K for 1 h. The spectra
were recorded in dry air against KBr as the background.
2.2. Sample characterization
X-ray diffraction (XRD) patterns of the sample materials were
Solid-state nuclear magnetic resonance (NMR) experiments
recorded on a Bruker D8 diffractometer with CuKα radiation were performed on a Bruker Avance III 400WB spectrometer at
resonance frequencies of 400.1, 104.3, and 79.5 MHz for 1H,
(λ = 1.54184 Å) from 5° to 40° with a scan speed of 2θ = 6.0°
min−1
.
27Al, and 29Si, respectively. The experimental conditions are as
1
Scanning electron microscopy (SEM) images were obtained follows: single pulse excitation of π/2 for H, π/6 for 27Al, and
1
π/4 for 29Si, with repetition times of 20 s for H and 29Si, and
using a Hitachi S-4800 microscope.
The chemical compositions of the samples under study 0.5 s for 27Al MAS NMR spectroscopy. The 1H and 27Al MAS
were determined by inductively coupled plasma emission spec- NMR spectra were recorded with a sample spinning rate of
8 kHz, while the 29Si MAS NMR spectra were obtained with
trometry (ICP-AES) on a Thermo IRIS Intrepid II XSP atomic
emission spectrometer.
4 kHz. 1H NMR MAS NMR measurements were performed
The surface areas and pore volumes of the calcined using dehydrated samples, which were dehydrated at 673 K at
a pressure below 10−2 Pa for 12 h. 27Al and 29Si MAS NMR
samples were measured by means of nitrogen adsorption at
77 K on a Quantachrome iQ-MP gas adsorption analyzer. measurements were performed using hydrated samples. For
Before the nitrogen adsorption, samples were dehydrated at this purpose, the dehydrated sample materials were exposed to
473 K for 2 h. The total surface area was calculated via the
the vapor over a saturated aqueous Ca(NO3)2 solution at
Brunauer Emmett Teller (BET) equation and the microporous ambient temperature for at least 12 h.
pore volume was determined using the t-plot method.
Diffuse reflectance ultraviolet-visible (UV-vis) spectra were
recorded in the region of 200–700 nm under air conditions
2.3. Catalytic evaluation
and against BaSO4 using
spectrophotometer.
a
Varian Cary 300 UV-vis The catalytic epoxidation reactions were performed in a 25 mL
round-bottom glass flask equipped with a cryogenic-liquid
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 2281–2291 | 2283

View Article Online
Paper
Green Chemistry
condenser under atmospheric pressure. Typically, 5 mmol of
2-cyclohexen-1-one or its derivatives, 0.1 g of catalyst, and
10 mL of acetonitrile (or other solvents) were added to the
vessel and mixed vigorously by a magnetic stirrer. The vessel
was heated to the desired reaction temperature and then
7.5 mmol of H2O2 (31 wt% aqueous solution) were added into
the vessel under stirring. After the reaction, liquid-phase pro-
ducts were qualitatively analyzed by a Shimadzu GC-MS
QP2010 SE and quantitatively determined by a Shimadzu 2010
GC equipped with an Agilent HP-5MS column (30 m ×
0.25 mm × 0.25 μm), while chlorobenzene was used as an
internal standard. Carbon balance of over 95% was obtained
for all experiments. The conversion of the substrate and the
selectivity of the products were calculated as follows:
ðinitial moles ꢀ final molesÞ
Substrate conversion ðmol%Þ ¼
initial moles
ꢁ 100%
moles of the product
moles of all product
Product selectivity ð%Þ ¼
ꢁ 100%
Fig. 1 XRD patterns of H-Beta, Si-Beta, and 5 wt% Ti-Beta.
Recycling tests of Ti-Beta zeolites were performed as
follows. After the reaction at 343 K in acetonitrile for 5 h, the
catalyst samples were recovered by centrifugation and washed
with acetonitrile. After repeating this procedure three times,
the Ti-Beta samples were dried at 353 K overnight and calcined
in static air at 823 K for 6 h.
Table 1 Physicochemical properties of H-Beta, Si-Beta, and Ti-Beta
zeolites
Ti loadinga Surface areab Pore volumec
a
Sample
nSi/nAl
(wt%)
(m2 g−1
)
(cm3 g−1
)
H-Beta
Si-Beta
1 wt% Ti-Beta >1800
2 wt% Ti-Beta >1800
3 wt% Ti-Beta >1800
5 wt% Ti-Beta >1800
13.5
>1800
/
/
590
620
610
598
581
567
0.204
0.210
0.203
0.200
0.198
0.196
1.0
1.9
2.8
4.1
3. Results and discussion
3.1. Evidences for Ti incorporation into the zeolite
a Determined by ICP. b Specific surface area obtained by the BET
method. c Calculated from t-plot.
framework
Possible structure changes of the Beta zeolites due to the post-
synthesis procedures were investigated by X-ray diffraction and loading is calculated to be 4.9 wt% assuming that one Ti atom
the corresponding powder patterns are shown in Fig. 1. replaces one framework Al atom. At relatively low Ti loadings
Typical diffraction lines characteristic for the BEA topology are (1 wt%, 2 wt%, 3 wt%), the actual Ti loading analyzed by ICP
observed for H-Beta, Si-Beta, and Ti-Beta, indicating that the is very close to the desired value. While at the highest Ti
framework structure is well preserved after dealumination and loading of 5 wt%, the actual Ti loading is distinctly lower
Ti incorporation. For 5 wt% Ti-Beta, no obvious diffraction (4.1 wt%), hinting at approaching of saturation of the Ti incor-
lines corresponding to titanium species can be observed, prob- poration and a loss of Ti during the calcination process. The
ably due to the high dispersion of Ti species. The well pre- existence of residual chloride and carbon could be excluded by
served microporous BEA structure after dealumination and Ti XPS and element analysis, respectively.
incorporation is further confirmed by the nitrogen physisorp-
tion (Fig. S1†), in which similar and typical type I adsorption/ are shown in Fig. S2.† Obviously, a similar morphology can be
desorption isotherms can be observed for all samples.
observed for the three samples, ruling out the possibility of
The SEM images of the H-Beta, Si-Beta, and 5 wt% Ti-Beta
The physicochemical properties of H-Beta, Si-Beta, and 1 to morphology changes after dealumination and Ti incorpor-
5 wt% Ti-Beta are summarized in Table 1. All samples exhibit ation. The crystal particles of 5 wt% Ti-Beta are observed to be
similar BET surface areas (567–620 m2 g−1) and micropore uniform nanoparticles with a size of 50–80 nm. Such small
volumes (0.196–0.204 cm3 g−1), confirming that the textural crystal particles are presumed to encounter less diffusion
properties of the Beta zeolite are well preserved after dealumi- limitation in converting bulky substrate molecules.
nation and Ti incorporation. ICP analysis results reveal that
Fig. S3† shows the FTIR spectra in the framework vibration
the nSi/nAl ratio of the parent H-Beta sharply increases from region (700–1400 cm−1) of Si-Beta and 2 to 5 wt% Ti-Beta
13.5 to over 1800 for the dealuminated Si-Beta, indicating that samples. The vibrational band at 948 cm−1 observed for
the latter is essentially free of Al. Theoretically, the maximal Ti Si-Beta is probably due to the presence of a large amount of
2284 | Green Chem., 2014, 16, 2281–2291
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
defect sites, originating from extracting framework Al atoms
The existing states of Ti species in as-synthesized Ti-Beta
during the dealumination process, as confirmed by the SiOH were further investigated by XPS and the results are shown in
bands in the hydroxyl stretching region of the DRIFT spectra Fig. 3. In the Ti 2p region, signals at binding energies of 458.4,
(vide infra, section 3.2). After the introduction of Cp2TiCl2 and 459.3, 464.1, and 465.2 eV appeared for all samples. Generally,
subsequent calcination, the 948 cm−1 band disappears, indi- the binding energy values of 458.4 and 464.1 eV are assigned
cating the binding of Ti atoms at the former defect sites (as to 2p3/2 and 2p1/2 photoelectrons of tetrahedrally coordinated
illustrated in Scheme 2, the Ti incorporation mechanism will framework Ti(IV) species.29,30 Binding energy values of 459.3
be further discussed in the following section). Meanwhile, a and 465.2 eV, on the other hand, hint to 2p3/2 and 2p1/2 photo-
new FTIR band centered at 962 cm−1 appears for Ti-Beta electrons of octahedrally coordinated Ti(VI) species, either in
samples, which is characteristic of titanosilicates as suggested framework or at extra-framework positions.29–31 It is quite
by earlier reports.7,25,26 This can be seen as an evidence of the clear that tetrahedrally coordinated framework Ti(IV) atoms are
successful incorporation of Ti species into the zeolite the dominating species (85–95%) in the Ti-Beta samples, con-
framework.
sistent with the results of UV-vis spectroscopy. On the other
To understand the coordination of the incorporated Ti hand, a significant increase in the percentage of Ti(VI) species
species in the microporous structure, UV-vis analysis was per- from 5 to 15% was observed with increasing Ti loadings from
formed with calcined Ti-Beta zeolites and the corresponding 1 to 5 wt%.
spectra are shown in Fig. 2. Ti-Beta samples with different Ti
Based on the characterization results from XRD, nitrogen
loadings essentially exhibit an intense band centered at physisorption and SEM, the framework structure of Beta zeo-
45 900 cm−1 (218 nm) arising from oxygen to Ti(IV) ligand to lites is well preserved after dealumination and Ti incorpor-
metal charge transfer (LMCT).11,14,27 In addition, a shoulder ation. According to the results of UV-vis spectroscopy and XPS
band centered at 38 700 cm−1 (258 nm), corresponding to octa- analysis, the majority of Ti species are successfully incorpor-
hedral Ti(VI) in framework or extra-framework positions,16,28 ated into the BEA framework. Since all samples have been cal-
can be observed especially at relatively high Ti loadings (3.0 cined at 823 K for 12 h, the calcined Ti-Beta samples are
and 5.0 wt%). The band centered at 45 900 cm−1 increases in expected to be stable materials for applications in hetero-
intensity with increasing Ti loading, indicating that Ti species geneous catalysis.
are mainly incorporated into the framework of Beta. For Ti-
Beta zeolites with different Ti loadings, no band centered at
below 30 300 cm−1 (>330 nm) occurred and the existence of
bulk anatase TiO2 can be excluded. This is further confirmed
by Raman analysis. As shown in Fig. S4,† no Raman active
modes corresponding to anatase TiO2 can be observed in post-
synthesized Ti-Beta even at the highest Ti loading of 5 wt%.
3.2. Mechanism of Ti incorporation into the zeolite
framework
According to the above-mentioned characterization results, we
propose that mainly tetrahedrally coordinated Ti species are
incorporated into the Beta zeolite framework by interacting
with SiOH groups at defect sites formed upon dealumination.
Fig. 2 UV-vis spectra of calcined Ti-Beta zeolites with different Ti load-
ings of 1 to 5 wt%.
Fig. 3 Ti 2p XPS of Ti-Beta with different Ti loadings.
Green Chem., 2014, 16, 2281–2291 | 2285
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Green Chemistry
In order to confirm this assumption, the mechanism of the Ti observed with increasing Ti loading, reveals the incorporation
incorporation into the zeolite framework is further investigated of Ti species mainly into the vacant T sites of the dealumi-
by means of DRIFT and NMR spectroscopy.
nated zeolite framework.
Fig. 4 displays the hydroxyl stretching region of the DRIFT
The results of the solid-state NMR spectroscopic studies are
spectra of calcined H-Beta, Si-Beta, and 5 wt% Ti-Beta. The shown in Fig. 5. The 27Al MAS NMR spectra of H-Beta (left,
parent H-Beta exhibits two major bands at 3740 and bottom) consists of an intense signal at ca. 54 ppm ascribed to
3600 cm−1, corresponding to isolated terminal silanol (SiOH) framework Al atoms in tetrahedral oxygen coordination and an
and bridging hydroxyl groups (Si(OH)Al), respectively.14,32 Two additional weak signal at ca. 0 ppm due to extra-framework Al
bands at 3775 and 3655 cm−1 associated with Al–OH groups of atoms in octahedral coordination.36 Upon treating with con-
extra-framework aluminum species and a broad weak band at centrated HNO3 solution, these two signals disappear comple-
3540 cm−1 corresponding to hydrogen-bonded Si–OH groups tely, indicating that almost all of Al species in the Beta zeolite
can also be observed.32,33 After dealumination with concen- were removed by dealumination (left, top).
trated HNO3 solution, the 3775, 3655, and 3600 cm−1 bands
The 29Si MAS NMR spectra of H-Beta show a series signals
related to Al species could not be detected any more due to at −116, −113, −106, and −103 ppm (Fig. 5, middle). The
complete removal of the Al atoms from the zeolite framework, signals at −116 and −113 ppm are attributed to framework
which is consistent with the ICP analysis (nSi/nAl > 1800). Sim- Si(OSi)4 species (Q4) at two different crystallographic sites,14
ultaneously, an increase in the intensity of the SiOH band at while the signal at −103 ppm is attributed to framework
3745 cm−1 and the appearance of a new, very broad band at Si(OSi)3(OH) species (SiOH groups, Q3) at defect sites,36,37
around 3520 cm−1 are observed, indicating the formation of while the signal at −105 ppm is due to framework Si atoms
defect sites at the zeolite framework. Specifically, the band at with one Al neighbor, i.e. Si(OSi)3(OAl) species (Q4).36–38 As a
around 3520 cm−1 indicates the formation of hydroxyl nests result of the dealumination, the intensities of signals due to
with hydrogen-bonded SiOH groups located at the vacant T Q4 silicon species remain nearly unchanged, while the intensi-
sites of the dealuminated Si-Beta framework (Scheme 2, acid ties of the signal due to SiOH groups (Q3 species) decrease dis-
treatment step), which is in line with earlier assignments.32–35 tinctly. Simultaneously, the signal caused by Si(OSi)3(OAl)
The incorporation of Ti atoms into Si-Beta induces a small species almost disappeared. These results clearly indicate the
decrease in intensity of the band at 3740 cm−1, indicating that removal of framework Al atoms upon dealumination and the
Ti atoms bind to terminal oxygen atoms and replace silanol resulting changes of the SiOH defect sites in the Beta zeolite.
groups to some extent in the case of a high Ti loading (5 wt%). The incorporation of Ti atoms into the Si-Beta framework does
This may result in the formation of some extra-framework Ti not cause obvious changes in the 29Si MAS NMR spectra since
species in Ti-Beta, often as octahedrally coordinated species, the decrease of the Q3 signal (SiOH) is compensated by the
as revealed by UV-vis spectroscopy and XPS (Fig. 2 and 3). In increase of the Si(OSi)3(OTi) signal.
contrast, the strong decrease of the 3520 cm−1 band, also
In the 1H MAS NMR spectrum of H-Beta (Fig. 5, right,
bottom), a strong signal at 1.5 ppm and two weak signals at
2.5 and 3.9 ppm can be observed. These signals are due to
silanol groups at framework defects (1.5 ppm), hydroxyl
groups at extra-framework Al species (2.5 ppm), and bridging
hydroxyl (Si(OH)Al) groups (3.9 ppm),39–41 respectively. The
dealumination process leads to disappearance of hydroxyl
groups associated with framework as well as extra-framework
Al species responsible for the signals at 2.5 and 3.9 ppm,
respectively. On the other hand, the intensity of the signal due
to SiOH groups at framework defects (1.5 ppm) increases
strongly and a new signal at 2.7 ppm emerges. Obviously, deal-
umination of H-Beta creates additional framework defects
covered with SiOH groups. A reasonable explanation for the
relatively high chemical shift of the silanol signal at 2.7 ppm is
hydrogen bonds of some of these SiOH groups with neighbor-
ing framework oxygen species inside silanol nests. That is,
vacant T sites in the zeolite framework are formed during the
dealumination process and hydroxyl nests are formed, which
agrees with the results of DRIFT spectroscopy (Fig. 4). The
introduction of Ti into Si-Beta and subsequent calcination
result in a dehydroxylation of the silanol groups at framework
1
defects and hydroxyl nests, while exclusively the H MAS NMR
signal of non-interacting defect SiOH groups remains, which is
slightly shifted to 1.9 ppm.
Fig. 4 DRIFT spectra in the hydroxyl stretching vibration region of
H-Beta, Si-Beta, and Ti-Beta (5.0 wt%).
2286 | Green Chem., 2014, 16, 2281–2291
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Fig. 5 27Al, 29Si, and 1H MAS NMR spectra of H-Beta, Si-Beta, and 5 wt% Ti-Beta.
Based on the results of DRIFT and solid-state NMR spec-
troscopy, the two-step mechanism of Ti incorporation into the
zeolite Beta framework can be described as follows (Scheme 2):
in the first step, the complete dealumination of zeolite H-Beta
forms a large number of vacant T sites causing hydroxyl nests.
In the second step, the organometallic Ti complex interacts
with silanol nests and isolated defect SiOH groups of Si-Beta
leading to replacement of their hydroxyl protons by Ti species,
which causes Ti incorporation into the zeolite framework
upon calcination.
3.3. Catalytic properties of Ti-Beta in epoxidation
Fig. 6 Alkene conversion (left) and epoxide selectivity (right) in the
2-cyclohexen-1-one epoxidation on Ti-Beta, TS-1, and Ti-MCM-41 with
a similar Ti loading of 2 wt%, plotted as a function of time-on-stream.
Reaction conditions: 5 mmol 2-cyclohexen-1-one, 10 mL acetonitrile,
7.5 mmol H2O2, 0.1 g catalyst, temperature = 343 K.
3.3.1. Effects of structural features of Ti-zeolite. The cataly-
tic activity of the post-synthesis modified Ti-Beta, together
with the reference materials TS-1 and Ti-MCM-41, were investi-
gated using the epoxidation of 2-cyclohexen-1-one as a model
reaction. As shown in Fig. 6, the 2-cyclohexen-1-one conversion
increases with time-on-stream for all Ti-zeolite catalysts with a
similar Ti loading of ca. 2 wt%, while the selectivity to epoxide
decreases. Obviously, Ti-Beta is much more active and selective
than TS-1 and Ti-MCM-41. Typically, a turnover frequency
(TOF) of 88.2 h−1 is obtained for Ti-Beta (calculated at time-on-
stream of 0.5 h, assuming that all Ti species are catalytically
active sites), more than four times higher than that of TS-1
(19.7 h−1) and Ti-MCM-41 (17.6 h−1).
The low activity of TS-1 may be due to steric limitations.
Obviously, there is a strong diffusion hindrance for the reac-
tant 2-cyclohexen-1-one with large kinetic diameter in the
10-ring pores of TS-1 (MFI: [100] 0.51 × 0.55 nm ↔ [010] 0.53 ×
0.56 nm), so that the epoxidation reaction can only take place
on the surface active sites near the pore openings and at a very
small content of Ti sites acting as accessible catalytic sites. Ti-
MCM-41 processes large mesopores of ca. 6 nm in diameter
without diffusion limitations for the reactant 2-cyclohexen-1-
one, and its low activity probably originates from an intrinsic
other nature of the Ti sites in comparison with Ti-Beta. For Ti-
MCM-41, the Ti sites are merely located at the external surface
of MCM-41,15 and these Ti sites are less stable and less active
than framework Ti sites in the epoxidation with hydrogen per-
oxide. As shown in Fig. 6, the 2-cyclohexen-1-one conversion
over Ti-MCM-41 does not change any more beyond time-on-
stream of 2 h, probably due to the quick catalytic deactivation.
These results are in good accordance with those reported by
Corma et al.,17 who suggested that Ti/ITQ-2 (prepared via the
same route as Ti-MCM-41) is a suitable catalyst for the epoxi-
dation of olefins only using organic peroxides in the absence
of water.
Based on the above-mentioned results, it can be stated that
the pore structure of Ti(IV)-containing zeolites do strongly
affect their catalytic performances in epoxidation reactions, in
great contrast to the results reported by Guidotti et al.18 Fur-
thermore, post-synthesis modified Ti-Beta zeolites containing
Ti species, which are mainly incorporated at tetrahedral
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 2281–2291 | 2287

View Article Online
Paper
Green Chemistry
framework positions, are robust catalysts for the epoxidation
with hydrogen peroxide.
3.3.2. Effects of the Ti loading and the preparation
method. The effects of the Ti loading (1–5 wt%) of the Ti-Beta
zeolites on the 2-cyclohexen-1-one conversion and product
selectivity were further investigated and the results are sum-
marized in Table 2. No 2-cyclohexen-1-one conversion could be
observed for Si-Beta as a catalyst, while significant 2-cyclo-
hexen-1-one epoxidation is observed for Ti-Beta with different
Ti loadings, indicating that the catalytic activity of Ti-Beta zeo-
lites is exclusively caused by the framework Ti species. Accord-
ingly, the epoxidation activity of Ti-Beta increases with
increasing Ti loading. Further analysis suggests that a quasi-
linear correlation between the epoxidation rate (in mmol h−1
)
and the number of tetrahedrally coordinated framework Ti(IV)
species (determined by ICP for total Ti number and XPS for
T(IV) content) can be drawn (Fig. 7), evidencing that these
framework Ti species are responsible for the epoxidation
activity. Moreover, the epoxide selectivity (only epoxide and
diol can be detected from epoxidation) also increases slightly
with increasing Ti loading. This should be due to the existence
of remaining silanol nests in the Ti-Beta zeolites with low Ti
loadings, and such weakly acidic surface OH groups are
capable of catalyzing the hydrolysis of formed epoxides to diol.
The catalytic performances of Ti(IV)-containing Beta zeolites
prepared via different routes, but with a similar Ti loading of
5 wt% in the epoxidation of 2-cyclohexen-1-one are compared
in Table 2. The TiO2/Si-Beta zeolites prepared by wet impreg-
nation (-w) and mechanical mixing (-m) show certain catalytic
activity and 2-cyclohexen-1-one conversions of 21.1 and 11.6%,
respectively, after a time-on-stream of 5 h. For these materials,
selectivities to epoxide of 80.9 and 84.0% were observed. The
Ti-H-Beta zeolites appear to be more active than TiO2/Si-Beta
and a 2-cyclohexen-1-one conversion of 32.8% can be obtained
after a time-on-stream of 5 h. However, the selectivity to
Fig. 7 Correlation between the epoxidation rate and the number of
tetrahedrally coordinated framework Ti(IV) species.
epoxide is only 66.2%, and 22.4% diol is obtained caused by
the hydrolysis of epoxide, which is catalyzed by residual acid
sites. Based on the catalytic activity of Ti(IV)-containing Beta
zeolites, it is quite obvious that post-synthesis modified
Ti-Beta zeolites are outstanding catalysts for the epoxidation of
2-cyclohexen-1-one in terms of both catalytic activity and
epoxide selectivity. It is thus demonstrated that the zeolite
host, the organometallic Ti complex and the synthesis pro-
cedures are all very important issues to obtain robust Ti(IV)-
containing Beta catalysts for epoxidation reactions.
In repeated preparations of 5 wt% Ti-Beta zeolites via the
two-step post-synthesis procedure, the obtained Ti loadings
were quite similar to that in the original experiment. Further-
more, the catalytic performances of the compared materials in
the 2-epoxidation, including the 2-cyclohexen-1-one conversion
and epoxide selectivity after a time-on-stream of 5 h were
almost identical (Table 2, see b). These results strongly indi-
cate that the two-step post-synthesis procedure is a feasible
route to obtain Ti-Beta zeolites with excellent reproducibility.
We also tested the regeneration ability of Ti-Beta zeolites pre-
pared via the post-synthesis route, and a slight decrease in
both 2-cyclohexen-1-one conversion and epoxide selectivity
occurred. Typically, a 2-cyclohexen-1-one conversion of 50.8%
and an epoxide selectivity of 94.1% were obtained after the
third use (see c), as shown in Table 2. Since the leaching of Ti
species can be excluded according to ICP analysis, the decrease
in the epoxidation activity should be due to changes in the
nature of the Ti species caused by their interaction with hydro-
gen peroxide or bulky reaction products as suggested by Carati
et al.28 and Hutchings et al.42,43 This is a common problem
when using Ti(IV)-containing zeolites as catalysts for the epoxi-
dation with hydrogen peroxide, and a similar decrease in the
epoxidation activity can also be observed for commercial TS-1
(Fig. S5†).
Table 2 Catalytic properties of the Ti-Beta zeolites under study in the
epoxidation of 2-cyclohexen-1-onea
Selectivity (%)
Ti loading
(wt%)
Conv.
(%)
Sample
Epoxide
Diol
Others
Si-Beta
0
0
/
/
/
1 wt% Ti-Beta
2 wt% Ti-Beta
3 wt% Ti-Beta
5 wt% Ti-Beta
5 wt% Ti-Betab
5 wt% Ti-Betab
5 wt% Ti-Betac
TiO2/Si-Beta-wd
TiO2/Si-Beta-me
Ti-H-Betaf
1.0
1.9
2.8
4.1
4.1
4.2
4.0
4.7
4.9
4.6
48.6
61.6
70.2
75.3
75.5
74.9
50.8
21.1
11.6
32.8
92.9
94.8
96.0
98.9
98.7
99.0
94.1
80.9
84.0
66.2
4.7
3.9
3.5
1.1
1.3
1.0
1.7
5.1
6.1
22.4
2.4
1.3
0.5
0
0
0
4.2
14.0
9.9
11.4
a Reaction conditions: 5 mmol 2-cyclohexen-1-one, 10 mL MeCN,
7.5 mmol H2O2, 0.1 g catalyst, temperature = 343 K, time-on-steam =
5
h. b Repeated experiments. c Third use. d Prepared by wet
impregnation of Si-Beta with TBOT. e Prepared by mechanical mixing
of Si-Beta with TiO2 (Degussa P25). f Prepared by grinding H-Beta with
Cp2TiCl2.
2288 | Green Chem., 2014, 16, 2281–2291
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
3.3.3. Effects of solvents on the epoxidation activity of Ti-
3.3.4. Effects of the substrate on the epoxidation perform-
Beta. It is well known that solvents may show a vital effect on ances. Table 4 shows the catalytic results of the 5 wt% Ti-Beta
the catalytic performances of Ti(IV)-containing zeolites in zeolite in the epoxidation of various unsaturated organic com-
liquid phase oxidation reactions.20,44,45 Such a unique effect is pounds. It is seen that the epoxidation of unsaturated five-ring
generally acknowledged as referring to two different active ketones is relatively slower than the epoxidation of six-ring
species I and II, as shown in Fig. S6.† For species I, a stable ketones, in accord with the earlier results of Tatsumi et al.20,22
five-ring structure is formed by the coordination of ROH to a This should be ascribed to the resonance interaction between
Ti site and hydrogen bondings to the Ti-peroxo complex, the CvC double bond and the electron-withdrawing CvO
which are believed to be the active intermediates in protic groups. The coplanar atoms of the CvC–CvO system result in
alcohol solvents. In the case of aprotic solvents, species II is an electron deficiency at the CvC double bond of 2-cyclopen-
assumed to contribute to the active substrates. In addition, the ten-1-one and its derivatives, which subsequently reduces the
polarity expressed in terms of the dielectric constant, geometry reactivity. In the case of 2-cyclohexen-1-one, the preferential
of the solvent and its interaction with the active centers are conformation excludes the existence of a coplanar CvC–CvO
also of significant importance. Table 3 summarizes the influ- system, and therefore, the conjugative effect does not exist.
ence of solvents on the catalytic performance of the 5 wt% Ti- Actually, the CvC double bond in 2-cyclohexen-1-one is
Beta zeolite for the epoxidation of 2-cyclohexen-1-one with similar to an isolated CvC double bond and high epoxidation
hydrogen peroxide. Typically, protic solvents, e.g. MeOH,
EtOH, and t-BuOH, and aprotic solvents, e.g. MeCN, MeCOMe,
and MeCOEt, are selected as solvents. It is clearly seen that the
Table 4 Epoxidation of various unsaturated compounds over the 5 wt
epoxidation activity increases with increasing solvent polarity
in the case of using both protic and aprotic solvents. However,
the epoxide selectivity decreases with increasing solvent
polarity in the case of using protic solvents, while it increases
with increasing polarity using aprotic solvents. Among protic
alcohol solvents, in contrast to MeOH and EtOH, t-BuOH leads
to distinct lower activity, which could be attributed to a
decrease in the electrophilicity of species I and to serious
steric constrains on its formation.46 Whereas, using MeOH
and EtOH as solvents results in a significant decline in the
2-cyclohexen-1-one epoxide selectivity due to solvolysis effects,
which leads to a sharp increase in the formation of glycol
monomethyl ether or monoethyl ether, as shown in Table 3.
As a whole, aprotic MeCN seems to be the best choice of
solvent in view of substrate conversion and epoxide selectivity.
The favorable effects of MeCN are, on the one hand, associated
with its weak basic nature, which can effectively neutralize the
weak acid sites generated in Ti-Beta, and therefore, suppress
the hydrolysis of epoxide products.47 On the other hand,
MeCN molecules can be activated by the electrophilicity of Ti
species to react with hydrogen peroxide to form the peroxyimi-
dic acid intermediate, CH3–C(vNH)–O–O–H, which is known
to be an active oxidizing agent toward organic substrates.48
% Ti-Beta/H2O2 system
Selectivity (%)
Substrate
Product
Conv. (%)
43.8a
Epoxide
>99.9
Diol
Others
0
0
75.3a
<0.1a
98.9
/
1.1
0
/
/
5.8a
64.7
18.3
17.0
46.8b
48.8b
58.3
23.6
14.9
28.9
26.8
47.5
28.9b
46.5
19.6
33.9
a Reaction conditions: 5 mmol substrate, 10 mL MeCN, 7.5 mmol
H2O2, 0.1 g catalyst, temperature = 343 K, time-on-stream = 5 h.
b Temperature = 333 K, time-on-stream = 2 h.
Table 3 Effect of solvents on the epoxidation of 2-cyclohexen-1-one over the 5 wt% Ti-Beta/H2O2 systema
Selectivity (%)
Solvent
Polarity
Dielectric constant
Conv. (%)
Epo.
Diol
Glycol ether
Others
Methanol
EtOH
t-BuOH
MeCN
MeCOMe
MeCOEt
Protic
Protic
Protic
Aprotic
Aprotic
Aprotic
32.7
24.5
10.9
37.5
20.7
18.5
85.7
82.6
60.7
75.3
53.7
51.8
45.1
48.7
90.2
98.9
84.6
81.7
3.3
2.8
2.1
1.1
15.4
18.3
38.4
36.7
2.5
0
0
0
13.2
11.8
5.2
0
0
0
a Reaction conditions: 5 mmol 2-cyclohexen-1-one, 10 mL solvent, 7.5 mmol H2O2, 0.1 g catalyst, temperature = 343 K, time-on-stream = 5 h.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 2281–2291 | 2289

View Article Online
Paper
Green Chemistry
activity can be achieved using the 5 wt% Ti-Beta zeolite as a
Furthermore, the impacts of the preparation methods and
catalyst. In addition, the dramatic decline in the conversion of Ti loadings on the catalytic performances of Ti-Beta zeolites
methyl substituted substrates, such as 3-methyl-2-cyclopenten- were investigated. The two-step post-synthesis method appears
1-one and 3-methylcyclohexen-1-one, implies that steric con- to be the most feasible route to prepare robust Ti-Beta catalysts
strains on the substrate should be seriously considered. As a for the epoxidation of 2-cyclohexen-1-one with hydrogen per-
whole, it can be concluded that both the electron deficiency oxide. A quasilinear correlation between the epoxidation rate
and steric effect play important roles in the epoxidation of and the number of tetrahedrally coordinated framework Ti
α,β-unsaturated ketones over Ti-Beta/H2O2 system.
species could be drawn for Ti-Beta zeolites, evidencing that
Meanwhile, the rate of solvolysis, which transforms epoxide these Ti species are responsible for the epoxidation activity.
into the corresponding diol, is dependent on the structure of For the epoxidation with hydrogen peroxide, the solvent has a
the epoxide formed. Hence, the effect of substituting methyl great impact on the catalytic performance of the Ti-Beta zeo-
groups on the corresponding epoxide selectivity is worth inves- lites, and aprotic MeCN appears to be the best choice of
tigating. In order to eliminate the electron deficiency and solvent in view of substrate conversion and epoxide selectivity.
steric effect, 1-methyl and 4-methyl substituted cyclohexenes When different unsaturated organic compounds are employed
are selected as model substrates and the results are shown in for the epoxidation with hydrogen peroxide, the electron
Table 4. Compared with 4-methylcyclohexene having a methyl deficiency, steric effects, and substituted groups of the sub-
group located at the position far from the double bond, the strate may influence the catalytic performances of Ti-Beta
hydrolysis rate is relatively higher in the case of a methyl zeolites.
group located on the olefinic double bond, i.e. 1-methylcyclo-
hexene. The higher hydrolysis rate will result in a low epoxide
selectivity, which can be ascribed to the facile ring-opening
Acknowledgements
generating a tertiary carbenium cation followed by the nucleo-
philic attack on the cation via the SN1 mechanism.48
This work was financially supported by the National Natural
Science Foundation of China (21373119) and the Ministry of
Education of China (NCET-11-0251, IRT13022). The support
from 111 Project (B12015) and the Collaborative Innovation
Center of Chemical Science and Engineering (Tianjin) is also
acknowledged. Furthermore, M.H. wants to acknowledge
4. Conclusion
In the present study, a reproducible and scalable two-step
post-synthesis procedure has been developed to prepare Ti(IV)-
financial support by Deutsche Forschungsgemeinschaft.
containing Beta zeolites. In the first step, commercial H-Beta
zeolites are dealuminated to Si-Beta by treating in concen-
trated nitric acid solution and vacant T sites with associated
silanol groups are created. In the second step, the organo-
References
metallic Ti complex Cp2TiCl2 interacts with vacant T sites in
the Si-Beta zeolite and Ti species are subsequently incorpor-
ated into the framework of Beta zeolite upon calcination. This
preparation procedure and the corresponding mechanism are
described according to the characterization results from
DRIFT and multinuclear solid-state NMR spectroscopy (1H
MAS NMR, 27Al MAS NMR, and 29Si MAS NMR). Dominating
tetrahedrally coordinated Ti(IV) species in the zeolite frame-
work, together with a minority of octahedrally coordinated
Ti(VI) species at framework or extra-framework positions, could
be clearly identified by means of UV-vis spectroscopy and XPS.
The above-mentioned post-synthesis procedure can be simply
extended to the preparation of other metal ions incorporated
into zeolite frameworks providing that the dealumination of
the zeolite host is possible and a suitable organometallic
complex is available.
1 M. Taramasso, G. Perego and B. Notari, U.S. Patent, no.
4,410,501, 1983.
2 D. R. C. Huybrechts, L. De Bruycker and P. A. Jacobs,
Nature, 1990, 345, 240.
3 F. Cavanj and J. H. Teles, ChemSusChem, 2009, 2, 508.
4 A. Corma, M. A. Camblor, P. Esteve, A. Martínez and
J. Pérez-Pariente, J. Catal., 1994, 145, 151.
5 A. Corma, P. Esteve, A. Martínez and S. Valencia, J. Catal.,
1995, 152, 18.
6 T. Tatsumi and N. Jappar, J. Phys. Chem., 1998, 102, 7126.
7 P. Wu, T. Tatsumi, T. Komatsu and T. Yashima, J. Phys.
Chem. B, 2001, 105, 2897.
8 P. Wu, Y. Liu, M. He and T. Tatsumi, J. Catal., 2004, 228,
183.
9 M. A. Camblor, A. Corma, A. Martínez and J. Pérez-
Pariente, J. Chem. Soc., Chem. Commun., 1992, 589.
The Ti-Beta zeolites were successfully applied as catalysts in 10 T. De Baerdemaeker, B. Steenackers and D. De Vos, Chem.
the epoxidation of unsaturated ketones, e.g. 2-cyclohexen-1- Commun., 2013, 49, 7474.
one, with hydrogen peroxide as an oxidant. High epoxidation 11 N. Jappar, Q. H. Xia and T. Tatsumi, J. Catal., 1998, 180,
activity as well as high selectivity toward epoxide were simul- 132.
taneously obtained over Ti-Beta, far better than TS-1 and Ti- 12 T. Blasco, M. A. Camblor, A. Corma, P. Esteve, A. Martínez,
MCM-41 reference materials.
C. Prieto and S. Valencia, Chem. Commun., 1996, 2367.
2290 | Green Chem., 2014, 16, 2281–2291
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
13 J. Lu, K. M. Kosuda, R. P. Van Duyne and P. C. Stair, J. Phys. 31 C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder and
Chem. C, 2009, 113, 12412.
14 J. P. Nogier, Y. Millot, P. P. Man, T. Shishido, M. Che and
S. Dzwigaj, J. Phys. Chem. C, 2009, 113, 4885.
G. E. Muilenberg, Handbook of X-ray photoelectron spectro-
scopy: a reference book of standard data for use in x-ray photo-
electron spectroscopy, Perkin-Elmer, MN, Eden-Prairie, 1979.
15 T. Maschmeyer, F. Rey, G. Sankar and J. M. Thomas, 32 L. Li and N. Guan, Microporous Mesoporous Mater., 2009,
Nature, 1995, 378, 159. 117, 450.
16 R. D. Oldroyd, J. M. Thomas, T. Maschmeyer, 33 S. Dzwigaj, J. Janas, J. Gurgul, R. P. Socha, T. Shishido and
P. A. MacFaul, D. W. Snelgrove, K. U. Ingold and M. Che, Appl. Catal., B, 2009, 85, 131.
D. D. M. Wayner, Angew. Chem., Int. Ed. Engl., 1996, 35, 34 J. Janas, T. Machej, J. Gurgul, R. P. Socha, M. Che and
2787. S. Dzwigaj, Appl. Catal., B, 2007, 75, 239.
17 A. Corma, U. Díaz, V. Fornés, J. L. Jordá, M. Domine and 35 S. Dzwigaj and T. Shishido, J. Phys. Chem. C, 2008, 112,
F. Rey, Chem. Commun., 1999, 779. 5803.
18 M. Guidotti, N. Ravasio, R. Psaro, G. Ferraris and 36 X. Wang, W. Dai, G. Wu, L. Li, N. Guan and M. Hunger,
G. Moretti, J. Catal., 2003, 214, 242. Microporous Mesoporous Mater., 2012, 151, 99.
19 T. Punniyamurthy, S. Velusamy and J. Iqbal, Chem. Rev., 37 M. Hunger, S. Ernst and J. Weitkamp, Zeolites, 1995, 15,
2005, 105, 2329. 188.
20 M. Sasidhara, P. Wu and T. Tatsumi, J. Catal., 2002, 205, 38 P. Li, G. Q. Liu, H. H. Wu, Y. M. Liu, J. G. Jiang and P. Wu,
332. J. Phys. Chem. C, 2011, 115, 3663.
21 G. Fioroni, F. Fringuelli, F. Pizzo and L. Vaccaro, Green 39 M. Hunger, S. Ernst, S. Steuernagel and J. Weitkamp,
Chem., 2003, 5, 425. Microporous Mater., 1996, 6, 349.
22 X. H. Shen, W. B. Fan, Y. He, P. Wu, J. G. Wang and 40 M. Hunger, Catal. Rev. Sci. Eng., 1997, 39, 345.
T. Tatsumi, Appl. Catal., A, 2011, 401, 37.
41 Y. Jiang, J. Huang, W. Dai and M. Hunger, Solid State Nucl.
Magn. Reson., 2011, 39, 116.
42 L. J. Davis, P. McMorn, D. Bethell, P. C. Bulman Page,
F. King, F. E. Hancock and G. J. Hutchings, J. Catal., 2001,
198, 319.
23 M. G. Clerici and U. Romano, US Patent 4,937,216, 1990.
24 K. A. Kayano and T. Tatsumi, Chem. Commun., 1996, 145.
25 M. R. Boccuti, K. M. Rao, A. Zecchina, G. Leofanti and
G. Petrini, Stud. Surf. Sci. Catal., 1989, 48, 133.
26 M. Sasidharan and A. Bhaumik, Phys. Chem. Chem. Phys., 43 L. J. Davis, P. McMorn, D. Bethell, P. C. Bulman Page,
2011, 13, 16282.
F. King, F. E. Hancock and G. J. Hutchings, J. Mol. Catal. A:
Chem., 2001, 165, 243.
44 W. B. Fan, P. Wu, S. Namba and T. Tatsumi, J. Catal., 2006,
243, 183.
27 L. Marchese, T. Maschmeyer, E. Gianotti, S. Coluccia and
J. M. Thomas, J. Phys. Chem. B, 1997, 101, 8836.
28 A. Carati, C. Flego, E. P. Massara, R. Millini, L. Carluccio,
W. O. Parker Jr. and G. Bellussi, Microporous Mesoporous 45 W. B. Fan, P. Wu and T. Tatsumi, J. Catal., 2008, 256, 62.
Mater., 1999, 30, 137. 46 M. G. Clerici and P. Ingallina, J. Catal., 1993, 140, 71.
29 J. P. Nogier, Y. Millot, P. P. Man, C. Méthivier, M. Che and 47 A. Corma, P. Esteve and A. Martinez, J. Catal., 1996, 161,
S. Dzwigaj, Catal. Lett., 2009, 130, 588. 11.
30 P. Fejesa, J. B. Nagy, J. Halász and A. Oszkó, Appl. Catal., A, 48 G. Grigoropoulou, J. H. Clark and J. A. Elings, Green Chem.,
1998, 175, 89.
2003, 5, 1.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 2281–2291 | 2291