Epoxidation of stilbene using supported gold nanoparticles: Cumyl peroxyl radical activation at the gold nanoparticle surface

Article abstract of DOI:10.1039/c3cc48626e

The catalytic epoxidation of cis-stilbene using cumene as a solvent in the presence of supported gold nanoparticles (AuNP) yields a mixture of cis and trans-stilbene oxides. EPR and product distribution studies support a new mechanistic proposal where oxygen centred radicals activate the AuNP surface and form active surface oxygen species responsible for the epoxidation products. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3cc48626e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Epoxidation of stilbene using supported gold
nanoparticles: cumyl peroxyl radical activation
at the gold nanoparticle surface†
Cite this: Chem. Commun., 2014,
50, 2289
Received 12th November 2013,
Accepted 3rd January 2014
Charles-Oneil L. Crites,a Geniece L. Hallet-Tapley,a Marıa Gonzalez-Bejar,ab
´
´
´
J. C. Netto-Ferreira*ac and Juan C. Scaiano*a
DOI: 10.1039/c3cc48626e
www.rsc.org/chemcomm
The catalytic epoxidation of cis-stilbene using cumene as a solvent This work focuses on the reaction between AuNP-generated
in the presence of supported gold nanoparticles (AuNP) yields a radicals and alkenes to promote epoxidation in a system where
mixture of cis and trans-stilbene oxides. EPR and product distribu- epoxidation occurs concurrently with solvent oxidation that
tion studies support a new mechanistic proposal where oxygen follows the autooxidation pathway proposed by Ingold11 in
centred radicals activate the AuNP surface and form active surface the 1960s.
oxygen species responsible for the epoxidation products.
Previous work investigating the role of supported AuNP as
catalysts in epoxidation reactions has suggested that their involve-
Epoxidation plays a very important role in organic chemistry ment is limited to the formation of the peroxyl radicals as part of a
due to the high reactivity and extended versatility of the epoxide normal free-radical chain reaction.12 The epoxidation of stilbene
moiety. Typical peracids, notably meta-chloroperbenzoic acid by AuNP in a solvent that can be easily oxidized (e.g., cumene or
(m-CPBA), can be used as epoxidation agents and are widely methylcyclohexane) can be considered a secondary process
employed at the laboratory scale. Epoxidation using hydro- accompanying oxidation,13 and, therefore, an understanding of
peroxides, such as the Sharpless epoxidation, involves a concerted the primary solvent peroxidation process, solvent is needed.
mechanism in which homogenous Ti complexes are employed as Previous work from our group14 aimed at understanding the role
catalysts.1 Heterogeneous catalysts based on mesoporous materials, of supported AuNP@TiO2 in cumene peroxidation revealed cumyl
used in conjunction with several hydroperoxides as oxygen donors, alcohol, and not cumene hydroperoxide (CHP), as the major
have been frequently employed in epoxidation reactions.2–5 More product.14 This work is aimed at examining the epoxidation
recently, supported AuNP on a variety of materials, including TiO2,6 reaction of cis-stilbene when using supported AuNP@TiO2 as a
have received attention as possible alternate catalysts for alkene catalyst. The role of this catalyst and the TBHP initiator in the
epoxidation. A typical epoxidation reaction involves an aerated or peroxidation process have been examined in some detail.14
oxygenated solution containing a free radical initiator, such as an
Table 1 presents the results obtained when using AuNP@TiO2
azo compound,7 or, more frequently, a hydroperoxide.8,9 A tert- as the catalyst for the epoxidation reaction of cis-stilbene with
butyl hydroperoxide (TBHP) initiator is typically placed in the cumene as the solvent undergoing peroxidation at 80 1C. Specific
presence of a solvent that can be easily oxidized (e.g., cumene or experimental conditions can be found in the ESI.† Briefly the
methylcyclohexane)10 and an alkene, the epoxidation substrate. reaction involved 10 mL of cumene (7.1 M), 94 ml of cis-stilbene
Comparison of experimental results from different laboratories (0.5 mmol), 7.8 ml (0.05 mmol) of tert-butyl hydroperoxide and
is complicated due to the large variability of reaction conditions, tert-butylbenzene as internal standard.
initiators, catalysts and substrates that have been employed.
Table 1 shows that no trans- or cis-stilbene oxide was
obtained in the absence of the catalyst, or when TiO2 alone
a Department of Chemistry and Centre for Catalysis Research and Innovation,
University of Ottawa, Ottawa, Ontario K1N 6N5, Canada.
Table 1 Percent conversion of cis-stilbene and % yield of trans-(TSO) and
cis-stilbene oxide (CSO) after 24 ha
E-mail: scaiano@photo.chem.uottawa.ca
b
´
´
Instituto de Ciencia Molecular/Departamento de Quımica Organica,
Universidad de Valencia, 46980, Paterna, Valencia, Spain
Catalyst
Conversion (%)
Yield (TSO) (%)
Yield (CSO) (%)
c Departamento de Quimica, Universidade Rural de Rio de Janeiro,
Antiga Rio-Sao Paulo km 47, Seropedica, 23851-970, Rio de Janeiro, Brazil.
E-mail: josecarlos@photo.chem.uottawa.ca
Au/TiO2
TiO2
No catalyst
18.7
48.0
0
16.3
0
0
2.6
0
0
† Electronic supplementary information (ESI) available: Experimental details,
TEM data and EPR control experiments. See DOI: 10.1039/c3cc48626e
a
Reaction performed at 80 1C for 24 hours using cumene as solvent.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2289--2291 | 2289

View Article Online
Communication
ChemComm
Table 2 CHP and cumyl alcohol obtained in the epoxidation reaction of
cis-stilbene in cumene using AuNP@TiO2 as catalyst
Catalyst
Au/TiO2
CHP (mM)
Cumyl alcohol (mM)
89
0
549
0
a
TiO2
No catalyst
420
24
a
Phenol was obtained as the only product.
was used. Significant cis-stilbene conversion (48%) was observed
in the presence of TiO2 and has been attributed to mineraliza-
tion products, supported by similar results obtained by Caps
et al.10 for trans-stilbene in the presence of TiO2. However, when
AuNP@TiO2 was employed as catalyst, a mixture of cis- and trans-
stilbene oxide is obtained. Similar results were observed by
Hughes et al.8 when using supported AuNP@graphite.
Fig. 1 Representative EPR spectrum for the oxyl radical/DMPO adduct
obtained in the AuNP@TiO2 catalyzed peroxidation of cumene. The coupling
constants are aH = 9.8 G and aN = 14.9 G.
The size and shape dependent electronic properties of small
AuNP have a significant influence on its ability to directly
activate O2 in oxidation reactions.18 Nanoparticles smaller than
2 nm can exhibit this type of behaviour close to ambient
temperature and, thus, can lead to epoxidation of substrates,
such as styrene.19 Previous work from our group has demon-
strated that peroxyl radicals can be decomposed on the AuNP
surface leading to the formation of surface-oxygen species,14
while EPR spin trap experiments in the current contribution are
consistent with the participation of free radicals. Combined,
our results lead to the proposal that the oxygen-active species
formed on the catalyst support have a direct involvement in the
epoxidation process.
In the proposed mechanism, shown in Fig. 2, the first two
steps involve our previously proposed activation of the cumyl
peroxyl radical on the supported AuNP surface leading to the
formation of a reactive surface-oxygen species. This process is
accompanied by the formation of a cumyloxyl radical (homolysis)
and subsequent H-abstraction to form the observed cumyl alcohol
product. The epoxidation then occurs via reaction between the
alkene and surface oxygen species.
The formation of both cis- and trans-stilbene oxide in the
presence of supported AuNP catalysts suggests the involvement
of a stepwise mechanism in which the formation of an inter-
mediate capable of easy bond rotation is required prior to
completion of the epoxidation process. Table 2 presents the
amounts of CHP and cumyl alcohol formed during the epoxida-
tion of cis-stilbene in the presence of AuNP@TiO2, TiO2 or in the
absence of any catalyst using cumene as solvent showing that in
the absence of catalyst, CHP predominates. Qualitatively com-
paring the information in Tables 1 and 2 can provide valuable
information as the possible pathway for AuNP-mediated epox-
idation. In the absence of catalyst, no epoxidation products were
observed and CHP is the major product. The involvement of free
cumyl peroxyl radicals in the direct epoxidation of cis-stilbene
seems unlikely, as the catalyst is definitely required. When
AuNP@TiO2 was employed as the epoxidation catalyst both
isomers of stilbene oxide were obtained, in conjunction with
the predominant formation of cumyl alcohol. As suggested by
our previous work, the AuNP surface-induced decomposition
of CHP is one of two pathways responsible for the formation of
highly reactive cumyl alkoxyl radicals leading to the formation of
cumyl alcohol.14 In addition, the surface induced decomposition
of the cumyl peroxyl radicals results in the formation of a
reactive oxygen species on the surface of the AuNP that could
be responsible for the products observed in this system.
EPR spin trap experiments employing 3,4-dihydro-2,3-dimethyl-
2H-pyrrole-1-oxide (DMPO) were performed in order to identify better
the nature of the radicals present in the epoxidation reaction mixture.
The radical/trap adduct depends on the nature of the radical involved
and exhibits characteristic EPR spectral features. We were not able to
observe any EPR signal in the absence of catalyst or in the presence of
the TiO2 support, most likely because the amount of radical present is
far below the detection limit of our system (Fig. S2 and S3, ESI†). In
the presence of AuNP@TiO2, the EPR spectrum (Fig. 1) clearly
illustrates the presence of an oxygen centred radical/DMPO adduct,
as determined by comparison with literature precedents.15,16 The
experimental results obtained by EPR spectroscopy suggest that the
major radical species present in solution is oxygen centered, but
unfortunately the data is characteristically ambiguous as to whether
an alkoxyl or peroxyl radical has been trapped.17
Since both products, i.e., cis- and trans-stilbene oxide, are
obtained, the proposed mechanism must account for a transi-
tion state for the reaction on the nanoparticle surface allowing
significant rotational freedom given the stereochemical out-
come of the reaction. The final step of Fig. 2 depicts the
formation of epoxide by way of a reactive oxygen species
residing on the nanoparticle surface. It is important to note
that though the detailed nature of the interaction between the
Fig. 2 Proposed steps for the epoxidation of cis-stilbene involving
reactive oxygen species formed on the AuNP surface.
2290 | Chem. Commun., 2014, 50, 2289--2291
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
supported by product studies that demonstrate the formation
of both cis- and trans-stilbene oxide. EPR spin-trap experiments
were also used to substantiate the formation of an oxygen-
centered radical in the reaction mixture. The presence of an
oxygen-centered radical, coupled with the observed product
distribution of cis/trans-stilbene oxide and lack of epoxidation
in the absence of the supported AuNP catalyst suggests the
involvement of the AuNP surface on product formation by way
of a cumyl peroxyl radical surface decomposition and argues
against the direct involvement of a free peroxyl radical in the
attack and epoxidation of cis-stilbene. An alternative mecha-
nism has also been proposed and is suggested to involve an
electron transfer from the cis-stilbene to the oxygen bound
AuNP adduct, ultimately resulting in ring closure that can also
account for the observed epoxide stereochemistry.
Fig. 3 Possible pathways for the direct reaction a AuNP–oxygen radical
adduct and cis-stilbene.
Thanks are due to the Natural Sciences and Engineering
Council of Canada and the Canadian Foundation for Innova-
stilbene and AuNP surface is not completely understood, tion for generous support. J.C.N.-F. acknowledges the Univer-
coordination of stilbene to the nanoparticle surface could sity of Ottawa for a Visiting Professor fellowship. M. G. B. also
account for the moderate epoxidation yields observed (see thanks MINECO for her Juan de la Cierva contract and FP7-
Table 1); that is, essentially all the surface bound active oxygen PEOPLE-2011-CIG (NANOPHOCAT) for financial support.
leads to epoxidation.
The exact nature of the intermediates involved in this reaction
References
1 T. Katsuki and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102, 5974.
2 K. U. Ingold, D. W. Snelgrove, P. A. MacFaul, R. D. Oldroyd and
J. M. Thomas, Catal. Lett., 1997, 48, 21.
is open to speculation. Based on the previous work of Bruice and
Castellino,20 a mechanism involving both the direct attack of a
nanoparticle-bound radical species on cis-stilbene or a mecha-
nism involving an electron transfer process can be proposed, as
shown in Fig. 3.
3 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.
Pathway a in Fig. 3 describes the direct radical addition of
the reactive oxygen species present on the AuNP surface to cis-
stilbene and the subsequent formation of the corresponding
epoxide; the attachment to the surface may prevent full rando-
mization of the intermediate and thus some retention of the
cis conformation. In contrast, pathway b illustrates an electron
transfer process from cis-stilbene to the AuNP, leading to the
formation of an anionic gold species and the stilbene radical-
cation; its easy isomerization could lead to both isomers, with
the trans configuration clearly dominating due to the absence
4 J. M. Hidalgo, M. J. Jurado, J. M. Campelo, R. Luque and A. A.
Romero, Catal. Lett., 2008, 126, 179.
5 M. J. Jurado, M. D. Gracia, J. M. Campelo, R. Luque, J. M. Marinas
and A. A. Romero, J. Mater. Chem., 2009, 19, 8603.
6 P. Lignier, F. Morfin, S. Mangematin, L. Massin, J.-L. Rousset and
V. Caps, Chem. Commun., 2007, 186.
7 C. Aprile, A. Corma, M. E. Domine, H. Garcia and C. Mitchell,
J. Catal., 2009, 264, 44.
8 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.
9 P. Lignier, F. Morfin, L. Piccolo, J.-L. Rousset and V. Caps, Catal.
Today, 2007, 122, 284.
of any sterical constraints. These two species could then 10 P. Lignier, S. Mangematin, F. Morfin, J.-L. Rousset and V. Caps,
Catal. Today, 2008, 138, 50.
11 K. U. Ingold, Acc. Chem. Res., 1969, 2, 1.
12 V. Mendez, K. Guillois, S. Daniele, A. Tuel and V. Caps, Dalton
undergo a recombination reaction on the surface, forming
the depicted radical intermediate, much the same as in path
a, but with a different stereochemical history. This is followed
by ring closure to the isomeric epoxide products, with trans-
stilbene oxide likely dominating due to increased free rotation
about the central C–C bond in the initial cis-stilbene radical
cation intermediate.8 Evidence of an electron transfer from a
molecule to the AuNP with formation of the corresponding
radical cation was recently proposed for the isomerization of
azobenzene.21 Finally, the last step in both pathways leads to
the formation of a mixture of cis- and trans-stilbene oxides.
Trans., 2010, 39, 8457.
13 K. Guillois, S. Mangemantin, A. Tuel and V. Caps, Catal. Today,
2013, 203, 111.
14 C.-O. L. Crites, G. L. Hallett-Tapley, M. Frenette, M. Gonzalez-Bejar,
J. C. Netto-Ferreira and J. C. Scaiano, ACS Catal., 2013, 3, 2062.
15 Y. Weijun, Y. Bijun, G. Cancheng, T. Ze and Z. Lei, Chin. J. Catal.,
2010, 31, 535.
16 W. Charnulitrat, N. Takahashis and R. P. Mason, J. Biol. Chem.,
1989, 264, 7889.
´
´
17 Q. Guo, S. Y. Qian and R. P. Mason, J. Am. Soc. Mass Spectrom.,
2003, 14, 862.
18 M. Boronat and A. Corma, Dalton Trans., 2010, 39, 8538.
In conclusion, the mechanism for the AuNP-catalyzed epox- 19 W. Gao, X. F. Chen, J. C. Li and Q. Jiang, J. Phys. Chem. C, 2010, 114, 1148.
20 A. J. Castellino and T. C. Bruice, J. Am. Chem. Soc., 1998, 110, 158.
21 G. L. Hallett-Tapley, C. D’Alfonso, N. L. Pacioni, C. D. McTiernan,
idation of stilbene is proposed to involve the formation of a
reactive oxygen species by way of AuNP surface induced decom-
´
´
M. Gonzalez-Bejar, O. Lanzalunga, E. I. Alarcon and J. C. Scaiano,
position of a peroxyl radical intermediate. These results are
Chem. Commun., 2013, 49, 10073.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2289--2291 | 2291