Stereoselective stilbene epoxidation over supported gold-based catalysts

Article abstract of DOI:10.1039/b610546g

The gold reference catalyst Au/TiO₂ exhibits high activity in the stereoselective epoxidation of trans-stilbene in methylcyclohexane in the presence of 5 mol% TBHP, by taking part in a chain reaction involving the activation of molecular oxygen by a radical produced from methylcyclohexane. The Royal Society of Chemistry.

Full text of DOI:10.1039/b610546g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Stereoselective stilbene epoxidation over supported gold-based
catalysts{
Pascal Lignier, Franck Morfin, Ste´phane Mangematin, Laurence Massin, Jean-Luc Rousset and
Vale´rie Caps*
Received (in Cambridge, UK) 24th July 2006, Accepted 29th September 2006
First published as an Advance Article on the web 24th October 2006
DOI: 10.1039/b610546g
(400 mol%), yields of trans-stilbene oxide are higher in methylcy-
clohexane (MCH) than in cyclohexane (CH), toluene (TOL) and
acetonitrile (ACN), over Au/C as well as Au/TiO2 (see ESI{), even
with a poorer dispersion of the catalytic powder. The epoxide is
formed as the primary product; by-products are essentially
degradation products (see ESI{). Au/TiO2 is more active and
selective than Au/C (Fig. 1). Actually, while 19% of tS is converted
(in 24 h) in the absence of a catalyst with 52% selectivity to the
corresponding epoxide, adding just TiO2-P25 increases the tS
conversion to 31% with similar selectivity. TiO2-PC500 brings the
conversion to 48% with lower selectivity, which is probably due to
the presence of sulfonated (i.e. acidic) surface functions on this
high surface area titania. However, the conversions observed over
very different gold-containing catalysts vary in a short range from
81 to 95%, indicating that the reaction is only slightly influenced by
the nature of the support, unlike some gold-catalysed reactions in
the gas phase, such as CO oxidation.8 Selectivities vary in an
equally short range from 51 to 75% with Au/TiO2 . Au/TiO2-
PC500 y Au/Fe2O3 . Au/C, illustrating the superiority of titania
as a support for epoxidation. Furthermore, Pt/C is more active and
selective than Au/C and nearly as selective as Au/TiO2, indicating
that, in excess TBHP, the reaction is only slightly influenced by the
nature of the metal.
The gold reference catalyst Au/TiO2 exhibits high activity in the
stereoselective epoxidation of trans-stilbene in methylcyclohex-
ane in the presence of 5 mol% TBHP, by taking part in a chain
reaction involving the activation of molecular oxygen by a
radical produced from methylcyclohexane.
Epoxidation of lower alkenes, i.e. with kinetic diameters below
˚
5.5 A, such as 1-hexene, are usually carried out over the
microporous titanium silicalite-1 (TS-1) using a stoichiometric
amount of hydrogen peroxide as the oxidant.1 Larger alkenes,
such as cyclohexene or cis-stilbene can be epoxidised over Ti-
MCM-41 with tert-butylhydroperoxide (TBHP) as the primary
oxidant2 while the epoxidations of 1-octene, vinylcyclohexene,
limonene and monoterpenes have been reported using metal-
substituted polyoxometalates (POM) and excess H2O2.3
Cyclooctene and norbornene can be epoxidised by molecular
oxygen over ruthenium-substituted POM catalysts.4 In all cases,
the mechanisms involved are non radical, which is held as the
reason for the high yields of epoxides achieved.3 Epoxidations with
POM however require the use of environmentally unfriendly
chlorinated solvents.3 On the other hand, Hughes et al. reported in
2005 on the epoxidation of a few bulky alkenes (cyclohexene,
cyclooctene, styrene, cis-stilbene) over Au/C in the presence of
catalytic amounts of TBHP in air at atmospheric pressure.5 In
these reactions, low yields of epoxide were obtained and
mechanistic considerations were limited to the description of
TBHP as the ‘‘initiator of a chain reaction sustained by molecular
oxygen’’ and the presence of an intermediate ‘‘allowing rotation
about the C–C bond’’. Earlier on, it had been shown that gold
complexes could catalyse the epoxidation of cyclohexene, using
molecular oxygen as oxidant, but also in very low yields.6
However, when the TBHP/tS molar ratio is reduced to 0.05,
Pt/C becomes completely inactive, converting 14% tS and yielding
no epoxide, just like in the absence of a solid catalyst (Fig. 2).
Adding TiO2-PC500 increases tS conversion to 44%, but again
without forming epoxide. On the other hand, TiO2-P25 decreases
tS conversion but trans-stilbene oxide is now formed with 63%
selectivity. Over Au/TiO2, the yield of epoxide reaches 53% in 24 h,
showing a real effect of gold in this reaction. The differences in the
catalytic behaviours of the gold reference materials are now
more obvious, the order for activity/selectivity remaining the same
(Au/TiO2 . Au/Fe2O3 . Au/C).
Here, we report on the stereoselective epoxidation of trans-
stilbene (tS) over the reference catalysts Au/TiO2 (P25), Au/C and
Au/Fe2O3 (all provided by the World Gold Council)7 and one
home-made Au/TiO2-PC500. We propose a reaction mechanism
involving free-radicals and we look into the ability of titania-
supported gold to make it highly selective towards the epoxide.
First, unlike Ti-silicates- and POM- catalysed alkene epoxida-
tions,2 gold exhibits higher yield and selectivity towards epoxide in
low polarity solvents. In fact, in the presence of excess TBHP
Institut de Recherches sur la Catalyse (CNRS), 2 avenue Albert
Einstein, 69626 Villeurbanne Cedex, France.
E-mail: caps@catalyse.cnrs.fr
{ Electronic supplementary information (ESI) available: Catalyst prepara-
tion, characterisation and testing, solvent effects, identification and
reactivity of by-products and reactivity as a function of the TBHP/tS
molar ratio. See DOI: 10.1039/b610546g
Fig. 1 tS (1 mmol), gold-based catalyst (Au: 10 mmol), MCH (20 mL),
TBHP (4 mmol), 80 uC, 24 h.
186 | Chem. Commun., 2007, 186–188
This journal is ß The Royal Society of Chemistry 2007

View Article Online
This free-radical mechanism is consistent with the model for the
non-catalytic autoxidation of styrene.9 The initiation step includes
thermolysis of TBHP (1) and production of the methylcyclohexyl
radical (2). The propagation step (3) would consist in production
of the methylcyclohexyl peroxy radical. It can then add to tS (5) or,
to a lesser extent,10 carry out hydrogen abstraction from
methylcyclohexane (4) to regenerate the methylcyclohexyl
radical and form methylcyclohexyl hydroperoxide (compound A,
identified by GC-MS). Addition to (as compared to hydrogen
abstraction from) tS is favored since tS has no allylic hydrogen
that could be abstracted.9 Species B is consistent with the
results obtained in the Au/C-catalysed oxidation of cis-stilbene,
which yields exclusively trans-stilbene oxide. This implies the
presence of an intermediate which allows rotation around the C–C
bond. Addition can be followed by unimolecular decomposition
(6) to form the epoxide and a methylcyclohexoxy radical.
This species will then abstract the tertiary hydrogen from MCH
(7) to produce methylcyclohexan-1-ol (compound C, identified
by GC-MS) and regenerate the methylcyclohexyl radical.
Termination probably occurs via recombination of the peroxo
radicals.11
Fig. 2 tS (1 mmol), gold-based catalyst (Au: 10 mmol), MCH (20 mL),
TBHP (0.05 mmol), 80 uC, 24 h.
The fact that 0.53 mmol tS are converted to trans-stilbene oxide
over Au/TiO2 with only 0.05 mmol TBHP shows that molecular
oxygen from the air has been used during the process. This is
confirmed by the experiment carried out under argon, in which less
than 5% trans-stilbene oxide is yielded. The use of a radical
scavenger (2,6-di-tert-butyl-4-methylphenol, 10 mol%) in the
reaction mixture also results in less than 5% trans-stilbene oxide,
which seems to indicate that oxygen is most probably activated via
free-radical species. The critical solvent effects (acetonitrile and
toluene yielding less than 5% trans-stilbene oxide; see ESI{) show
that, unlike TBHP activation, oxygen activation is tightly related
to the solvent nature, suggesting direct involvement of the solvent
molecule in the reaction mechanism. It is possible indeed that a
methylcyclohexane peroxy radical is produced via hydrogen
abstraction from methylcyclohexane and that the epoxidation of
trans-stilbene involves a cooxidation of MCH and tS (Scheme 1).
?
Of course, although RO are generally less active for addition
than for abstraction,12 addition of hydroxyl radicals to tS cannot
be fully excluded,13 especially in excess TBHP. Addition of tert-
t
butoxy radicals would be slower.14 BuO are more active for
?
hydrogen abstraction.15 It is interesting that H2O2 (4 mmol) leads,
over Au/C, to 30% tS conversion without formation of any epoxide.
This indicates that the conversion of trans-stilbene due to attack
by the hydroxyl radical will lead exclusively to degradation
products, which could explain the lower selectivities obtained in
excess TBHP. Moreover, since 14% trans-stilbene is transformed
while only 0.3% epoxide is produced in the absence of a catalyst
?
(5 mol% TBHP), we believe that this attack of tS by OH is non-
catalytic and guided only by the electrophilicity of the hydroxyl
radical.16
The introduction of TiO2-P25 brings selectivity to the reaction,
while lowering the conversion (8%). It is possible that titania
t
will trap these non-selective hydroxyl radicals and allow BuO to
?
react with MCH, enabling the epoxide-forming reaction to
proceed. The presence of gold on titania P25 further increases
both conversion and selectivity. The actual role of gold is yet
unclear but it seems that it could act in synergy with the titania
support and maybe catalyse hydrogen abstraction from MCH by
t
?
BuO (2) and/or formation of the peroxo radical (3). Its
involvement in the propagation sequence is undoubted since
the yield of epoxide produced after the catalyst has spent 6 h
in the reaction medium before being removed remains at
about 18%, instead of going up to 33% (24 h, Au/C, 0.4 mmol
TBHP). It is also interesting that the produced epoxide is always
in the trans form, whether from trans- or cis-stilbene, suggesting
that selective unimolecular decomposition (6) has occurred. This
is consistent with adsorption onto a solid surface, which creates
sterical constraints, thereby decreasing both symmetry and
accessibility of the molecule and forcing the reaction to proceed
in a specific way. Furthermore, less than 0.04 ppm gold was
detected (ICP-MS) in the reaction medium after removal of the
catalyst, showing that less than 0.04% of supported gold have
leached into the organic phase. On the other hand, 50 ppm of
free gold was found inactive for the production of epoxide
Scheme 1 Proposed mechanism for the epoxidation of trans-stilbene in
methylcyclohexane, in the presence of TBHP (5 mol%) and a gold-based
catalyst.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 186–188 | 187

View Article Online
from trans-stilbene, indicating that the reaction is truly
heterogenous.
Notes and references
1 R. A. Sheldon, M. Wallau, I. W. C. E. Arends and U. Schuchardt,
Acc. Chem. Res., 1998, 31, 485.
XPS studies show two contributions for the Au 4f7/2 peak on the
Au/TiO2 and Au/C catalysts, both before and after reaction: the
main contribution (y90%) is at a binding energy of 83.0 ¡ 0.1 eV
for Au/TiO2 and at a binding energy of 84.1 ¡ 0.1 eV for Au/C,
corresponding to Aud2 and Au0 respectively. The presence of
Aud2 in titania-supported gold catalysts has already been
explained by a charge transfer from the support to the gold
particle.8 The minor contribution (y10%) is at a binding energy of
85.5 ¡ 0.1 eV both for Au/TiO2 and Au/C, before and after
reaction, indicating the presence of Au+, probably at the gold–
support interface. It is interesting that this contribution is
insignificantly increased after reaction, suggesting that gold has
suffered negligible oxidation during the reaction. It is not clear
which oxidation state of gold is actually active in this reaction.
However, if we assume that active gold is involved in interactions
with electrophilic peroxy radicals, we could suggest that Aud2 is
more active than Au0 and Au+, which would be consistent with the
observation that Au/C is less active and selective than Au/TiO2.
2 R. D. Oldroyd, J. M. Thomas, T. Maschmeyer, P. A. MacFaul,
D. W. Snelgrove, K. U. Ingold and D. D. M. Wayner, Angew. Chem.,
Int. Ed. Engl., 1996, 35, 2787; K. U. Ingold, D. W. Snelgrove,
P. A. MacFaul, R. D. Oldroyd and J. M. Thomas, Catal. Lett., 1997,
48, 21.
3 O. A. Kholdeeva, T. A. Trubitsina, M. N. Timofeeva, G. M. Maksimov,
R. I. Maksimovskaya and V. A. Rogov, J. Mol. Catal. A: Chem., 2005,
232, 173; I. C. M. S. Santos, M. M. Q. Simo˜es, M. M. M. S. Pereira,
R. R. L. Martins, M. G. P. M. S. Neves, J. A. S. Cavaleiro and
A. M. V. Cavaleiro, J. Mol. Catal. A: Chem., 2003, 195, 253;
R. Neumann and D. Juwiler, Tetrahedron, 1996, 52, 8781; A. Lambert,
P. Plucinski and I. V. Kozhevnikov, Chem. Commun., 2003, 714.
4 R. Neumann and M. Dahan, Nature, 1997, 388, 353.
5 M. D. Hughes, Y.-J. Xu, P. Jenkins, P. McMorn, P. Landon,
D. I. Enache, A. F. Carley, G. A. Attard, G. J. Hutchings, F. King,
E. H. Stitt, P. Johnston, K. Griffin and C. J. Kiely, Nature, 2005, 437,
1132.
6 A. Fusi, R. Ugo, F. Fox, A. Pasini and S. Cenini, J. Organomet. Chem.,
1971, 26, 417.
7 Gold reference catalysts, Gold Bull., 2003, 36, 24.
8 S. Arrii, F. Morfin, A. J. Renouprez and J. L. Rousset, J. Am. Chem.
Soc., 2004, 126, 1199.
9 F. R. Mayo, Acc. Chem. Res., 1968, 1, 193.
21
Finally, a maximum reaction rate of 150 moltS molAu h21 is
10 A. A. Fokin and P. R. Schreiner, Chem. Rev., 2002, 102, 1551.
11 T. Mill and D. G. Hendry, in Comprehensive Chemical Kinetics vol. 16:
liquid-phase oxidation, ed. C. H. Bamford and C. F. H. Tipper, Elsevier,
Amsterdam, 1980, ch. 1, p. 55; A. Factor, C. A. Russell and
T. G. Traylor, J. Am. Chem. Soc., 1965, 87, 3692.
12 W. B. Motherwell and D. Crich, in Free Radical Chain Reactions in
Organic Synthesis, Academic Press, London, 1992, ch. 1, p. 3.
13 M. P. DeMatteo, J. S. Poole, X. Shi, R. Sachdeva, P. G. Hatcher,
C. M. Hadad and M. S. Platz, J. Am. Chem. Soc., 2005, 127, 7094;
J. K. Thomas, J. Phys. Chem., 1967, 71, 1919.
14 H. Elson, S. W. Mao and J. K. Kochi, J. Am. Chem. Soc., 1975, 97, 335;
T. Mill and D. G. Hendry, in Comprehensive Chemical Kinetics vol. 16:
liquid-phase oxidation, ed. C. H. Bamford and C. F. H. Tipper, Elsevier,
Amsterdam, 1980, ch. 1, pp. 47.
15 H. Paul, R. D. Small, Jr. and J. C. Scaiano, J. Am. Chem. Soc., 1978,
100, 4520.
obtained over Au/TiO2 using 5 mol% TBHP and an optimized
amount of the catalyst.
In conclusion, we have shown that, in methylcyclohexane, trans-
stilbene can be epoxidized stereoselectively in the presence of
Au/TiO2 and 5 mol% TBHP. Both gold and titania seem to take
part in the free-radical reaction mechanism sustained by molecular
oxygen, possibly by trapping the unselective hydroxyl radicals,
catalysing radical formation and stabilising the intermediate
resulting from the addition of the methylcyclohexyl peroxy radical
onto trans-stilbene.
The authors wish to thank the French Ministry of Education
and Research for P. Lignier’s PhD grant, Dr A. M. Sorokin and
Dr L. Piccolo for fruitful discussions and the scientific and
technical services of IRC for valuable support.
16 W. E. Griffiths, G. F. Longster, J. Myatt and P. F. Todd, J. Chem. Soc.
B, 1967, 530.
188 | Chem. Commun., 2007, 186–188
This journal is ß The Royal Society of Chemistry 2007