Highly selective olefin epoxidation with aqueous H2O2 over surface-modified TaSBA15 prepared via the TMP method

Article abstract of DOI:10.1039/b706443h

Trialkylsiloxy-modified Ta(v) centers on mesoporous silica exhibit excellent selectivity for epoxide formation (>98% after 2 h) in the oxidation of cyclohexene using aqueous H₂O₂ as the oxidant; the modified catalysts exhibit an incr

Full text of DOI:10.1039/b706443h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Highly selective olefin epoxidation with aqueous H2O2 over surface-
modified TaSBA15 prepared via the TMP method{
Daniel A. Ruddyab and T. Don Tilley*ab
Received (in Cambridge, UK) 30th April 2007, Accepted 6th July 2007
First published as an Advance Article on the web 17th July 2007
DOI: 10.1039/b706443h
is markedly greater than that of comparable Ti(IV)SiO2 catalysts
(10–20%) that have been the focus of much research. Here we
communicate the synthesis, characterization, and catalytic oxida-
tion performance of TaSBA15 and hydrophobic TaSBA15
materials obtained by chemical modification of the silica SBA15
surface, using aqueous H2O2 as the oxidant. In addition, insights
into the reaction mechanism and the structure of the active Ta(V)
site are described.
Trialkylsiloxy-modified Ta(V) centers on mesoporous silica
exhibit excellent selectivity for epoxide formation (.98% after
2 h) in the oxidation of cyclohexene using aqueous H2O2 as the
oxidant; the modified catalysts exhibit an increased lifetime,
retaining high selectivity after 6 h of reaction (.95% epoxide).
The search for catalysts that allow highly selective transformations
of readily available alkanes and alkenes to more synthetically
useful oxygenated compounds such as alcohols, ketones, and
epoxides remains an important research challenge. In particular,
‘‘green oxidants’’ such as hydrogen peroxide have become highly
desirable, since the only by-product (H2O) is environmentally
benign. A number of supported transition metal oxide-based
catalysts have been employed towards these ends.1 However, many
of the highly active oxide-based catalysts are poisoned by
coordination of water to the active site, so non-aqueous organic
oxidants such as tert-butyl hydroperoxide (TBHP) and cumene
hydroperoxide (CHP) are necessary to achieve high activity.2
The chemical properties of the support surface have been shown
to play an important role in determining catalyst activity and
selectivity.3 It was recently demonstrated that surface modification
of TiSBA15 materials, prepared via the thermolytic molecular
precursor (TMP) method, with (N,N-dimethylamino)trialkylsilanes
creates more active epoxidation catalysts and increases the epoxide
selectivity to a maximum of 58% (over about 12% for untreated
catalysts) using aqueous H2O2.4 Experimental and theoretical
studies of titania-silica catalysts have suggested that a Ti(OSi)4
center is more Lewis acidic than a HO–Ti(OSi)3 center, resulting in
a more electrophilic oxygen in the TiOOH intermediate and
enhancing catalytic performance.5 Considering the possible role of
Lewis acidity in titania-silica catalyzed epoxidations, tantala-silica
materials would seem to offer considerable promise as potent
oxidation catalysts. Hydrated tantala and tantala-silica mixed
oxides are known to possess moderate to strong Lewis acidic sites,
but tantala-silica materials have been much less studied as
oxidation catalysts.6,7 A previous report from this laboratory
demonstrated that Ta(V)SiO2 materials prepared via the TMP
method are active catalysts for the epoxidation of cyclohexene with
aqueous H2O2 as the oxidant, and exhibit epoxide selectivities as
high as 43% after 2 h.8 This selectivity with aqueous H2O2 oxidant
The details of the synthesis and grafting chemistry of the
molecular precursor Ta(OiPr)2[OSi(OtBu)3]3 (1) were recently
reported.8 The protonolysis reactions employed to create the
TaSBA15 and surface-modified materials, RcapTaSBA15, are
outlined in Scheme 1a,b. The composition of TaSBA15 was
determined to be 1.80 wt% tantalum by inductively coupled plasma
atomic emission spectroscopy. The degree of surface functionaliza-
tion for each catalyst was determined by combustion analysis and
confirmed by thermogravimetric analysis (TGA) (Table 1). The
surface coverage of the RcapTaSBA15 materials was similar for all
three capping groups, and ranged from 1.1 to 1.4 –SiMe2R groups
per nm2. Following the surface capping reaction, the intensity
of the FT-IR stretch for isolated SiOH groups at 3745 cm21
decreases dramatically and is extremely weak. Unfortunately, no
separate stretching band for a Ta–OH was observed, presumably
due to the low concentration of this species. The parent TaSBA15
material was found to have a BET surface area of 450 m2 g21 with
a pore volume of 0.51 cc g21 (1.8 wt% Ta corresponds to
aDepartment of Chemistry, University of California, Berkeley, Berkeley,
California, 94720-1460, USA. E-mail: tdtilley@berkeley.edu
bChemical Sciences Division, Lawrence Berkeley National Laboratory,
1 Cyclotron Road, Berkeley, California, 94720, USA
{ Electronic supplementary information (ESI) available: Experimental
details; N2 porosimetry isotherms; PXRD data; DSC traces of water
desorption; solid-state 29Si MAS NMR spectra; selected kinetic plots for
cyclohexene oxidation. See DOI: 10.1039/b706443h
Scheme 1 Synthesis of TaSBA15, RcapTaSBA15 and RuncapTaSBA15,
n
where R 5 Me, Bu, or Oc.
n
3350 | Chem. Commun., 2007, 3350–3352
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Table 1 Surface modification of TaSBA15, nitrogen physisorption,
and thermal analysis data
TGA) that were similar to those of the RcapTaSBA15 materials.
Solid-state 29Si magic-angle-spinning (MAS) NMR spectroscopy
of TaSBA15, BucapTaSBA15, and BuuncapTaSBA15 was
employed to investigate the silicon environments of each
material. The spectrum of TaSBA15 reveals Q3 and Q4 Si
environments in the 295 to 2110 ppm range. A resonance at ca.
23 ppm is observed in both the Bucap- and BuuncapTaSBA15
spectra. This is assigned to butyldimethylsiloxy-groups bonded
directly to the silica. The observation of a low intensity, broad
resonance at ca. 28 ppm in the BucapTaSBA15 spectrum is
assigned to the silicon of a nBuMe2SiOTa center. A similar, but
less significant upfield shift was recently assigned to a trimethyl-
siloxy-capped Ti center in TiSiO2.11
Pore
radiusa/ Wt% H2O Desorption
–SiMe2R/ SBET
nm22
/
Material
m2 g21 nm
adsorbed T/uC
SBA15
TaSBA15
—
—
585
450
280
270
190
3.0
2.8
2.6
2.5
2.2
—
31.0
3.32
3.23
3.18
—
68
50
51
51
MecapTaSBA15 1.2
BucapTaSBA15 1.4
OccapTaSBA15 1.1
a
Determined from the adsorption branch of the isotherm.
0.13 Ta nm22). Upon surface modification, both the surface area
and pore volume of the resulting material decreased according to
the size of the silylating agent. The MecapTaSBA15 material had a
BET surface area of 280 m2 g21, whereas the OccapTaSBA15
material had a surface area of 190 m2 g21 (Table 1). The hexagonal
mesoporous structure of the TaSBA15 was maintained after
surface modification, as evidenced by retention of the (100)
reflection in low-angle powder X-ray diffraction (PXRD) patterns,
and by transmission electron microscopy (TEM).
To compare the activities of the heterogeneous catalysts
described above, results were standardized with respect to mass
of catalyst (0.035 g). As previously reported for TaSBA15, no
leaching was detected under the oxidation conditions.8 Table 2
reports the results of cyclohexene oxidation using the TaSBA15
and RcapTaSBA15 catalysts with aqueous H2O2 as the oxidant.
Similar to the results observed with RcapTiSBA15, the surface-
modified TaSBA15 catalysts exhibit a higher activity than the
unmodified catalyst. Use of an RcapTaSBA15 material as the
catalyst with H2O2 yielded 4.3–7.4% of oxidation products after 2 h
(based on hydrogen peroxide), whereas the unmodified TaSBA15
catalyst yielded 3.9%. The effect of surface modification on the
epoxide selectivity is much more significant. The modified catalysts
demonstrate epoxide selectivities greater than 98%, while the
unmodified catalyst is only ca. 32% selective for the epoxide after
2 h, with cyclohexenone being the major product (53%).
To analyze the hygroscopic nature of the surface-modified
materials, they were placed in a sealed container with a saturated
H2O atmosphere for 24 h and then directly examined by TGA.
The water desorption temperature was determined by concurrent
differential scanning calorimetry (DSC). Table 1 lists the wt% of
H2O adsorbed on each material and the desorption temperature,
measured as the minimum of the endothermic transition corres-
ponding to water loss. The unmodified TaSBA15 exhibits a H2O
loss of 31.0 wt% below 150 uC. However, the modified materials
exhibit very little H2O loss (3.2–3.3 wt%) in this temperature range,
confirming the hydrophobic nature of the materials after
modification with trialkylsilyl groups.
As stated above, aqueous conditions generally promote catalyst
deactivation, and result in poor conversions and selectivities
to epoxide. The high epoxide selectivities associated with
RcapTaSBA15 suggest that these are robust epoxidation catalysts
and should produce greater epoxide yields at longer reaction times.
The results of cyclohexene oxidation with H2O2 oxidant at 65 uC
after 6 h of reaction are presented in Table 2. The epoxide
selectivity of the unmodified TaSBA15 catalyst decreases from
31.6% to 10.9%, with a total oxidation product yield of 22.9%.
This results in an epoxide yield of just 2.5%. In stark contrast, the
epoxide selectivities of all RcapTaSBA15 catalysts are above 95%.
In addition, these Ta catalysts are rendered more active after
surface modification. The epoxide yields of 10.0 to 12.8% exhibited
by the RcapTaSBA15 catalysts represent a 4.0- to 5.1-fold increase
in productivity after 6 h versus the unmodified TaSBA15 material,
and this is a direct consequence of the excellent epoxide selectivity
displayed by these materials. This is the highest cyclohexene oxide
selectivity reported by a surface modified catalyst using aqueous
H2O2, and it is markedly greater than that of analogous
Addition of a low concentration of 1 to the SBA15 substrate via
the TMP method is expected to result in isolated Ta centers on the
surface of the material.9 The exact structure of the surface-bound
Ta species is difficult to determine, due in part to the lack of
previous spectroscopic assignments of similar low wt% Ta
materials.7,10 To investigate the structures of the supported Ta
centers, a catalyst was prepared having only the Si–OH (and not
the Ta–OH) sites capped with –SiMe2R groups. This was
accomplished by capping the catalyst surface Si–OH sites after
introduction of 1, but before the calcination procedure that
generates the Ta–OH sites (Scheme 1c). After silylation,
calcination at 200 uC under air decomposes the organic groups
of the supported Ta precursor without affecting the silyl groups,
which decompose at temperatures greater than 250 uC. The
resultant materials, termed RuncapTaSBA15, had –SiMe2R sur-
face coverages (1.0–1.2 nm22) and hydrophobicities (H2O loss by
Table 2 TaSBA15-catalyzed oxidation of cyclohexene with H2O2 oxidant at 65 uCa
Reaction time 2 h
Reaction time 6 h
Epoxide
selectivity (%)
Total yield based
on H2O2 (%)
Epoxide
selectivity (%)
Total yield based
on H2O2 (%)
Epoxide yield based
on H2O2 (%)
Catalyst
TONb
TONb
TaSBA15
31.6
98.2
98.4
3.9
4.3
7.4
6.9
40
47
84
82
10.9
97.0
95.1
22.9
10.3
13.5
10.5
2.50
10.0
12.8
10.4
237
112
152
124
MecapTaSBA15
BucapTaSBA15
OccapTaSBA15
a
.99
.99
b
Reaction conditions: 5 mL CH3CN solvent, 24.7 mmol cyclohexene, 35 mg catalyst. TON 5 mol products/mol Ta at a reaction time of 2
or 6 h.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 3350–3352 | 3351

View Article Online
efficient retention of the siloxy group within the coordination
sphere of tantalum.
To further probe the influence of surface modification at the Ta
center, the RuncapTaSBA15 materials were tested as catalysts. The
cyclohexene oxide selectivities for Me-, nBu-, and nOcuncapTaSBA15
catalysts were similar to the capped catalysts (.95%) after 2 h, but
decreased dramatically to 55–65% after 6 h. This differs from the
analogously prepared RuncapTiSBA15 catalysts, where no increase
in epoxide selectivity was observed after 2 h without siloxy-capped
Ti sites.4 This interesting result suggests that Ta may be less
susceptible to poisoning by water, and supports the idea that the
hydrophobicity of the RuncapTaSBA15 catalyst slows water
coordination to the Ta center and/or preferentially increases the
affinity of cyclohexene for the surface. These conditions appear to
suppress the formation of radicals, but capping of the Ta site is
essential to maintain high epoxide selectivity over several hours.
We gratefully acknowledge the support of the Director, Office
of Energy Research, Office of Basic Energy Sciences, Chemical
Sciences Division, of the US Department of Energy under
Contract DE-AC03-76SF00098. The authors acknowledge sup-
port of the National Center for Electron Microscopy, Lawrence
Berkeley Lab, which is supported by the US Department of
Energy under Contract # DE-AC02-05CH11231. The authors
acknowledge J. J. Ford at the Environmental Molecular Sciences
Laboratory at Pacific Northwest National Lab for solid-state
MAS NMR spectroscopy. We thank A. M. Stacy at the
University of California, Berkeley for use of instrumentation
(PXRD), and H. M. Frei at Lawrence Berkeley Lab for
instrumentation (FT-IR, FT-Raman) and helpful discussions.
Scheme 2 Proposed pathways of cyclohexene oxidation over TaSBA15
and RcapTaSBA15 catalysts with hydrogen peroxide.
RcapTiSBA15 catalysts, which exhibit the highest Ti-based
selectivity with H2O2.
A general cyclohexene oxidation mechanism for the TaSBA15
catalysts is proposed in Scheme 2. Activation of H2O2 results in the
formation of a (SiOsurf)3Ta(g2-O2) intermediate. Direct observa-
tion of a Ta–(O2) moiety via vibrational spectroscopy could not
been achieved due to intense Si–O–Si absorption in the low
wavenumber region of the TaSBA15 materials. However, a recent
report attributes thioether oxidation over tantala-silica mixed
oxides to the presence of a tantalum-g2-peroxo intermediate, and
numerous Ta(g2-O2) complexes have been characterized.12 In
addition, the tetraperoxotantalate complex K3Ta(O2)4, synthesized
according to the literature procedure, was found to be an active
catalyst for the epoxidation of cyclohexene (5.6% yield of
oxidation products, 73% epoxide selectivity after 2 h with aqueous
H2O2 as the oxidant). These values are comparable to the catalytic
performance of the analogous Nb compound,13 and this result is
consistent with electrophilic oxygen transfer from a Ta(g2-O2)
species. This intermediate is likely stabilized through hydrogen
bonding to at least one trialkylsilanol. On the basis of computa-
tional studies, hydrogen bonding interactions involving peroxo
species appear to be critical in Ti-based oxidations, and in oxida-
tions with hydrogen peroxide in fluorinated alcohol solvents.14
It has been suggested that, for metal oxide based catalysts, allylic
oxidation occurs through a radical pathway, whereas epoxidation
occurs through an electrophilic oxygen-transfer pathway as
outlined in Scheme 2.5 To investigate the role of radicals in the
TaSBA15 catalyzed epoxidation of cyclohexene, a low concentra-
tion of a radical trap (1.93 6 1023 M 2,6-di-tert-butyl-4-
methylphenol) was added to the reaction mixture with the
unmodified TaSBA15 catalyst. After 2 h of reaction time, the
yield of cyclohexene oxide remained fairly constant (1.3%), but
the selectivity for epoxide increased to 99%. This strongly suggests
that the tantalum-catalyzed allylic oxidation pathway is similar
to the titanium one, in that radicals play a key role (Scheme 2,
pathway 1). Also similar to the TiSBA15 system, radical chemistry
does not seem to affect the unmodified tantalum-catalyzed epoxi-
dation pathway (Scheme 2, pathway 2).4 With the siloxy-capped
TaSBA15 catalysts, electrophilic oxygen transfer to the olefin is
greatly favored, and the hydrogen-bonded silanol should rapidly
react to re-form a Ta center with the –OSiMe2R cap. It is worth
noting that free trialkylsilanol is not detected in the oxidation
product analysis, which is consistent with a mechanism involving
Notes and references
1 G. I. Panov, A. K. Uriarte, M. A. Rodkin and V. I. Sobolev, Catal.
Today, 1998, 41, 365; J. M. Thomas, R. Raja and D. W. Lewis, Angew.
Chem., Int. Ed., 2005, 44, 6456.
2 R. A. Sheldon, J. Mol. Catal., 1980, 7, 107; J. M. Notestein, E. Iglesia
and A. Katz, J. Am. Chem. Soc., 2004, 126, 16478.
3 A. Corma, M. Domine, J. A. Gaona, J. L. Jorda, M. T. Navarro,
F. Rey, J. Perez-Pariente, J. Tsuji, B. McCulloch and L. T. Nemeth,
Chem. Commun., 1998, 2211; J. Bu and H. K. Rhee, Catal. Lett., 2000,
66, 245.
4 R. L. Brutchey, D. A. Ruddy, L. K. Andersen and T. D. Tilley,
Langmuir, 2005, 21, 9576.
5 J. M. Fraile, J. I. Garcia, J. A. Mayoral and E. Vispe, J. Catal., 2000,
189, 40.
6 T. Ushikubo and K. Wada, Appl. Catal., A, 1995, 124, 19; S. Nemana
and B. C. Gates, Chem. Commun., 2006, 3996.
7 D. Meunier, A. de Mallmann and J. Basset, Top. Catal., 2003, 23, 183;
V. Cimpeanu, V. Parvulescu, V. I. Parvulescu, M. Capron, P. Grange,
J. M. Thompson and C. Hardacre, J. Catal., 2005, 235, 184.
8 R. L. Brutchey, C. G. Lugmair, L. O. Schebaum and T. D. Tilley,
J. Catal., 2005, 229, 72.
9 C. Nozaki, C. G. Lugmair, A. T. Bell and T. D. Tilley, J. Am. Chem.
Soc., 2002, 124, 13194; R. L. Brutchey, I. J. Drake, A. T. Bell and
T. D. Tilley, Chem. Commun., 2005, 3736; D. A. Ruddy, N. L. Ohler,
A. T. Bell and T. D. Tilley, J. Catal., 2006, 238, 277.
10 M. Baltes, A. Kytoekivi, B. M. Weckhuysen, R. A. Schoonheydt, P. Van
Der Voort and E. F. Vansant, J. Phys. Chem. B, 2001, 105, 6211;
T. Tanaka, H. Nojima, T. Yamamoto, S. Takenaka, T. Funabiki and
S. Yoshida, Phys. Chem. Chem. Phys., 1999, 1, 5235.
11 K. Lin, L. Wang, F. Meng, Z. Sun, Q. Yang, Y. Cui, D. Jiang and
F.-S. Xiao, J. Catal., 2005, 235, 423.
12 D. Bayot and M. Devillers, Coord. Chem. Rev., 2006, 250, 2610.
13 L. C. Passoni, M. R. H. Siddiqui, A. Steiner and I. V. Kozhevnikov,
J. Mol. Catal. A: Chem., 2000, 153, 103.
14 M. Neurock and L. E. Manzer, Chem. Commun., 1996, 1133;
A. Berkessel and J. A. Adrio, J. Am. Chem. Soc., 2006, 128, 13412.
3352 | Chem. Commun., 2007, 3350–3352
This journal is ß The Royal Society of Chemistry 2007