Solvent-Free Aerobic Epoxidation of Dec-1-ene Using Gold/Graphite as a Catalyst

Article abstract of DOI:10.1007/s10562-014-1425-z

The oxidation of dec-1-ene has been investigated using gold nanoparticles supported on graphite in the presence of a radical initiator (α,α-azobisisobutyronitrile) using oxygen from air as oxidant. We have investigated the influence of the reaction temperature (70-100 °C), catalyst mass and reaction time on the epoxide yield. In the absence of a radical initiator the reaction does not proceed, although auto-oxidation can occur at higher temperatures in the range studied. However, in the presence of an initiator, selective oxidation occurs and the initiator propagates the reaction through the formation of a peroxy-radical at the allylic C₃ position. Graphite enhances the formation of the allylic products dec-1-en-3-ol, dec-1-en-3-one, and dec-2-en-1-ol; however, the addition of gold nanoparticles to the graphite, enhances formation of 1,2-epoxydecane. It is suggested that gold suppresses the formation of allylic products via a Russell termination. Graphical Abstract: [Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-014-1425-z

Catal Lett (2015) 145:689–696
DOI 10.1007/s10562-014-1425-z
Solvent-Free Aerobic Epoxidation of Dec-1-ene Using
Gold/Graphite as a Catalyst
•
Upendra Nath Gupta Nicholas F. Dummer Samuel Pattisson Robert L. Jenkins
•
•
•
•
David W. Knight Donald Bethell Graham J. Hutchings
•
Received: 15 October 2014 / Accepted: 2 November 2014 / Published online: 16 November 2014
Ó Springer Science+Business Media New York 2014
Abstract The oxidation of dec-1-ene has been investi-
gated using gold nanoparticles supported on graphite in the
presence of a radical initiator (a,a-azobisisobutyronitrile)
using oxygen from air as oxidant. We have investigated the
influence of the reaction temperature (70–100 °C), catalyst
mass and reaction time on the epoxide yield. In the absence
of a radical initiator the reaction does not proceed, although
auto-oxidation can occur at higher temperatures in the
range studied. However, in the presence of an initiator,
selective oxidation occurs and the initiator propagates the
reaction through the formation of a peroxy-radical at the
allylic C3 position. Graphite enhances the formation of the
allylic products dec-1-en-3-ol, dec-1-en-3-one, and dec-2-
en-1-ol; however, the addition of gold nanoparticles to the
graphite, enhances formation of 1,2-epoxydecane. It is
suggested that gold suppresses the formation of allylic
products via a Russell termination.
1 Introduction
Aerobic epoxidation of a-alkenes, containing allylic pro-
tons remains one of the major challenges in oxidation
catalysis. Epoxides are versatile intermediates that are
useful in organic synthesis and in other chemical technol-
ogies [1, 2]. Terminal alkene epoxides are used in the
manufacture of epoxy resins, as resin modifiers or as
coating materials. Terminal epoxides are also useful for
conversion to hydroxyethers by ring-opening of the epox-
ide with alcohols, and the hydroxyethers. They are
important ingredients in creams, lotions, and ointments for
use in cosmetic and pharmaceutical applications, serving as
oil-soluble bases in which many lipid-soluble solids can be
dissolved [3].
In this study we are concerned with the catalytic
epoxidation of dec-1-ene with air under mild reaction
conditions. There have been a number of previous studies
concerned with this transformation, and most cases
epoxidation is performed using non-environmentally
friendly or expensive stoichiometric oxidants [4–10]. The
use of H2O2 as the terminal oxidant has been used in
many studies; for example, Pescarmona et al. [11] have
studied the epoxidation of dec-1-ene using H2O2 as oxi-
dant at 80 °C. Burgess et al. [12] reported the epoxidation
of a range of substituted or cyclic alkenes using MnSO4
with 30 % H2O2 as oxidant and 99 % epoxide yield was
observed with cyclohexene after 24 h; however with dec-
1-ene no epoxide was observed. Mandelli et al. [13] have
reported the epoxidation of dec-1-ene using Al2O3/H2O2
with 18 % epoxide yield after 17 h however large
amounts of catalyst and oxidant were required. Scotti
et al. [14] have studied the epoxidation of dec-1-ene using
cumyl hydroperoxide as oxidant with a yield of 27 % at
100 °C using CuO/Al2O3 as the catalyst. Ho et al. [15]
Keywords Heterogeneous catalysis Á Alkenes Á
Oxidation Á Epoxidation
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-014-1425-z) contains supplementary
material, which is available to authorized users.
U. N. Gupta Á N. F. Dummer Á S. Pattisson Á
R. L. Jenkins Á D. W. Knight Á D. Bethell Á G. J. Hutchings (&)
Cardiff Catalysis Institute, School of Chemistry, Cardiff
University, Park Place, Cardiff CF10 3AT, UK
e-mail: hutch@cardiff.ac.uk
Present Address:
N. F. Dummer
Faculty of Science, University of Wollongong, Wollongong,
NSW 2522, Australia
123

690
U. N. Gupta et al.
2.2 Catalyst Preparation
have reported 93 % epoxide yield with peracetic acid and
a Mn(ClO4)2 catalyst. Although some of these approaches
give high yields these methodologies use expensive oxi-
dants and can be improved by using oxygen from air. We
have investigated the oxidation of alkenes using supported
gold nanoparticles and our initial work concerned the
reactivity of cycloalkenes [16]. We showed that cyclo-
octene could be selectively epoxidised using air together
with a peroxide initiator. We extended this to a range of
cycloalkenes and showed that the ring size was important
in controlling whether epoxidation or allylic oxidation
was the dominant reaction pathway [17]. For cycloalkenes
with ring sizes of C7 or smaller allylic oxidation was
preferred and for larger ring sizes epoxidation was almost
exclusively observed. This was due to the energetics of
the two reaction pathways [17]. We subsequently exten-
ded our studies to linear internal alkenes and we studied
the oxidation of hex-2-ene using supported gold catalysts
with peroxide initiators [18]. In addition, we showed that
when two oxidisable functional groups were present as in
enols, the alcohol functional group was oxidised in pref-
erence to the alkene and no epoxidation was observed
[19]. Subsequently, and most importantly, we observed
that the alkenes in our studies contained stabilisers to
prevent oxidation and upon their removal we found that
the peroxide initiator was not necessary for the oxidation
of cyclic and internal linear alkenes [20]. However, the
peroxide initiators were still required for the epoxidation
of terminal alkenes [20], which remains a major challenge
in selective oxidation catalysis. Terminal alkenes with
allylic hydrogens exhibit competing pathways for oxida-
tion, namely epoxidation involving the addition of elec-
trophilic oxygen and allylic oxidation involving
abstraction of allylic hydrogen. Allylic oxidation can also
occur in the absence of a catalyst at elevated temperatures
involving the formation of a hydroperoxide from oxygen.
In this paper we explore these reactions using dec-1-ene
and we investigate how the selectivity can be controlled
between epoxidation and allylic oxidation using supported
gold nanoparticles as a catalyst.
The catalyst (1 wt% Au/graphite) used in this study was
prepared via an incipient-wetness impregnation method
[16]. The carbon support (graphite, 2 g) was transferred to
a small evaporating basin. An aqueous solution (0.1 M) of
chloroauric acid (equivalent to 20 mg of of Au) was diluted
with distilled water to make a total gold solution of 2 mL.
The gold solution was added dropwise to the graphite
powder followed by air drying at various intervals
(1.0–1.5 h). After complete addition of the gold solution
the catalyst was dried in static air (25 °C, 16 h; 110 °C,
16 h). The catalyst has been extensively characterised in
our previous studies of the oxidation of cycloalkenes and
the catalyst used in these studies is fully described in a
previous report [16].
3 Oxidation of Dec-1-ene
3.1 Standard Reaction Conditions
All reactions were performed in a stirred glass round bot-
tomed flask (50 mL) fitted with a reflux condenser and
heated in an oil bath. In a typical reaction dec-1-ene
(10 mL (53 mmol)) was heated, typically at 90 °C, for 1 to
96 h under atmospheric air in the presence of AIBN
(35 lmol) and catalyst (0.1 g). The liquid products were
analysed using a Varian gas chromatograph 3300 fitted
with a Zebron ZB-Wax Plus column and a flame ionization
detector. The injector and detector temperatures were
maintained at 250 °C throughout analysis. During the
chromatographic analysis the column was heated from 60
to 200 °C at a heating ramp rate of 10 °C/min and held at
the final temperature for 26 min.
3.2 Catalyst Reuse Experiments
Fresh catalyst (0.5 g) was first stirred with dec-1-ene under
the standard reaction conditions. The catalyst was recov-
ered by centrifugation and washed with solvent (30 ml
acetone or ethanol) and dried (16 h at the 110 °C). After
drying, the recovered catalyst (0.1 g) was reused for dec-1-
ene using the standard reaction conditions.
2 Experimental
2.1 Materials
4 Results and Discussion
Graphite powder (\20 lm, Aldrich), HAuCl4.3H2O
(Sigma Aldrich), dec-1-ene (Alpha Aesar, stabiliser-free),
a,a-azobisisobutyronitrile (AIBN, Molecula), tert- butyl-
hydroperoxide (TBHP, Aldrich, 70 % in water) and ben-
zoyl peroxide (BPO, Merck), tri-phenyl phosphene (Sigma
Aldrich) were used as received.
4.1 Investigation of the Effect of the Catalyst
on the Oxidation of Dec-1-ene
Our initial experiments were aimed at examining the effect
of using the supported gold catalyst for the oxidation of
123

Solvent-Free Aerobic Epoxidation of Dec-1-ene
691
Table 1 Products observed for
the epoxidation of dec-1-ene
Conversion (%)
12.3
Product selectivity (%)
Epoxide
30.9
5.4
5.1
7.2
7.3
1.8
3.8
0.4
1.1
1.6
10.3
1.0
3.3
1.2
0.4
0.5
1.7
8.0a
Dec-1-en-3-one
Dec-1-en-3-ol
Dec-2-enal
Dec-2-en-1-ol
1,2-Decanediol
Dec-2-enoic acid
Dec-2-anone
C8 ? C9 aldehyde
C7 ? C8 alcohols
C7 ? C8 ? C9 acids
Non-3-en-1-ol
Nonan-3-one
Reaction conditions: Dec-1-ene
(53 mmol), AIBN (35 lmol),
1 % Au/graphite (0.1 g), 24 h,
90 °C, air (1 atmosphere)
The unidentified products
comprise ca. 30 minor products
Fig. 1 The influence of temperature on the epoxide yield of dec-1-
ene oxidation; open circle Graphite, x 1 wt% Au/Graphite (points are
coincident with open circle), filled square AIBN, open square
Cyclododecane
Cyclodecanol
Graphite ? AIBN, filled triangle
1 wt% Au/Graphite ? AIBN.
a
Nonalactone
Reaction conditions: dec-1-ene (53 mmol), AIBN (35 lmol), Catalyst
or support (0.1 g), 24 h, air (1 atmosphere)
C16 Epoxide
P
Gaseous products
each with a selectivity\\0.5 %
unknowns liquid
b
Gas products include light
hydrocarbons and COx, all with
products
9.3b
individual selectivities\\0.5 %
dec-1-ene using oxygen from air at atmospheric pressure as
the terminal oxidant. We have shown previously that a
radical initiator is required for the oxidation of linear ter-
minal alkenes [20] and in this case we selected AIBN for
this purpose (10 h half life in toluene at 65 °C [21]).
However, O2 is a diradical in its ground state, and can
therefore initiate radical oxidation reactions of alkenes in
the absence of a catalyst and this non-catalysed process can
be accelerated by the presence of a radical initiator. This
radical auto-oxidation process is favoured at higher tem-
peratures and higher oxygen partial pressures [22]. Hence, it
is important to establish the reaction conditions where the
conversion is at a minimum in the absence of a catalyst and
therefore a series of reactions were performed to study the
effect of temperature on the epoxidation of dec-1-ene in the
presence and absence of the catalyst (Fig. 1). Without
AIBN, oxidation of dec-1-ene does not occur in the pre-
sence of graphite or 1 wt% Au/graphite at temperatures up
to 100 °C. In the presence of AIBN oxidation of dec-1-ene
is observed at temperatures above 80 °C, but this effect is
not enhanced particularly when graphite is present as well.
When 1 wt% Au/graphite is present with AIBN signifi-
cantly enhanced oxidation activity is observed. The full
product selectivity for this catalysed reaction is given in
Table 1 and it is apparent that the selectivity to the epoxide
is ca. 30 % under these conditions; the selectivity to C10
oxidation products is over 60 % of the total products formed
but there is also evidence of C–C cleavage reactions
occuring. For all reactions studied the 1,2 epoxide was
formed; no 2,3- epoxide was observed by detailed GCMS
Fig. 2 Effect of gold on the selectivity to the epoxide and allylic
products as a function of reaction temperature; filled circle epoxide
with AIBN and graphite; open circle allylic products with AIBN and
graphite; filled square epoxide with AIBN and 1 wt% Au/graphite;
open square allylic products with AIBN and 1 wt% Au/graphite.
Reaction conditions: dec-1-ene (53 mmol), AIBN (35 lmol), catalyst
or support (0.1 g), 24 h, air (1 atmosphere)
analysis (see supplementary information). It is interesting to
note that alcohols are formed as by-products during this
reaction and remain stable under conditions where we have
observed previously that alcohols can be oxidised with
123

692
U. N. Gupta et al.
Fig. 4 Effect of reaction time on the formation of the epoxide and the
allylic products Key: filled square conversion (90 °C) open square
conversion (100 °C) filled circle epoxide (90 °C) open circle epoxide
(100 °C) filled triangle allylic products (90 °C) open tiangle allylic
products (100 °C). Reaction conditions: dec-1-ene (53 mmol), AIBN
(35 lmol), 1 wt% Au/graphite (0.1 g), air (1 atmosphere)
Fig. 3 Epoxide yield as a function of reaction time and reaction
temperature for reaction with 1 wt% Au/graphite. Key: filled square
100 °C, open square 90 °C, filled circle 80 °C, open circle 70 °C.
Reaction conditions: dec-1-ene (53 mmol), AIBN (35 lmol), 1 wt%
Au/graphite (0.1 g), air (1 atmosphere)
supported gold catalysts [23]. We will investigate the role of
the alcohol and aldehyde by-products in a future study. A
2-decen-1-ol and 2-decenal) are the predominant products
at the start of the reaction. These products persist for longer
periods when the reaction was performed at lower tem-
peratures but as the reaction temperature is increased the
selectivity to these products decreases markedly, with an
increase in epoxide selectivity.
comparison of the selectivity to the epoxide and allylic
P
products
(
(1-decen-3-one, 1-decen-3-ol, 2-decenal,
2-decen-1-ol) in the presence of AIBN over graphite or 1 %
Au/graphite is shown in Fig. 2. It is apparent that in the
presence of 1 wt% Au/graphite the selectivity towards the
allylic products is suppressed with an increased selectivity
to the epoxide. The epoxide selectivity and hence the yield
increases with an increase in reaction temperature, and
conversely the selectivity of allylic products decreases
(Fig. 2). The presence of the gold catalyst therefore is
directing the reaction preferentially towards the epoxide
and this effect is enhanced at higher conversions facilitated
by higher reaction temperatures.
4.3 Influence of Catalyst Mass on the Oxidation
of Dec-1-ene
As we have shown that the epoxide yield is increased at
increasing conversion we investigated the effect of
increasing the mass of catalysts as this can increase the
conversion. The effect of the amount of catalyst used in the
oxidation of dec-1-ene was studied at 90 °C using AIBN as
radical initiator with a reaction time of 24 h. It is apparent
that both dec-1-ene conversion and epoxide selectivity
increase with increasing catalyst mass (Fig. 5). At this
temperature there is a small conversion observed in the
absence of the catalyst and an epoxide yield of 1 % was
measured when 0.02 g catalyst was used and is marginally
higher than the epoxide yield obtained when the catalyst
was absent and the reaction was performed with AIBN
only. Epoxide yield increases to 3.8 % when the catalyst
amount was increased to 0.1. Further increases in the cat-
alyst mass increased in the epoxide yield to 4.7 % with
0.5 g catalyst. It was observed that when reactions were
performed with 0.02 g of catalyst the product distribution
was similar to the reaction performed without catalyst
(Fig. 5). Nonanal (generated by C–C cleavage), acids and
allylic products (1-decen-3-ol, 1-decen-3-one, 2-decenal,
4.2 Effect of Reaction Time on Dec-1-ene Oxidation
The effect of the reaction time on the oxidation of dec-1-
ene was investigated using 1 wt% Au/graphite in the pre-
sence of catalytic amounts of AIBN as the radical initiator
(Fig. 3). The epoxide yield was observed to increase with
reaction time. For all temperatures investigated an induc-
tion period was observed which shortened with an increase
in the reaction temperature, which could be related to the
thermal decomposition of AIBN. For reactions performed
at 70 °C the rate of reaction was slow and there is an
extended induction period, but at 100 °C the reaction
induction period was decreased to approximately 5 h.
Moreover, the selectivity data (Fig. 4) shows that the
products of allylic oxidation (1-decen-3-ol, 1-decen-3-one,
123

Solvent-Free Aerobic Epoxidation of Dec-1-ene
693
Table 2 Effect of PPh3 reduction of the reaction products
Product Yield (%)
Graphite
1 wt% Au/graphite
PPh3 (g)
0
0.12
0.80
1.68
0.76
0.47
0
0.12
0.97
1.06
0.59
1.26
Dec-2-enal
0.20
0.49
0.84
0.43
0.77
0.70
0.61
1.22
Dec-2-en-1-ol
Dec-1-en-3-ol
Dec-1-en-3-one
Table 3 Effect on the conversion of adding radical scavengers to
selective oxidation of dec-1-ene
Dec-1-ene conversion (%)
Initiator
TBHP
AIBN
BPO
Fig. 5 Effect of 1 wt% Au/Graphite catalyst mass on the conversion
of dec-1-ene and the selectivity to selected products. Reaction
conditions: dec-1-ene (53 mmol), AIBN (35 lmol), 90 °C, 24 h, air
(1 atmosphere)
Scavenger
Graphite
None
1.8
a
None
3.9
b
None
3.6
a
0.04
0.1
0.4
2.0
0.04
0.05
Au/graphite
9.7
12.3
8.8
Reaction condition: dec-1-ene (53 mmol), Catalyst or support (0.1 g),
TBHP (70 lmol), AIBN (35 lmol), BPO (35 lmol), 90 °C, 24 h
2-decen-1-ol) were produced at similar levels as observed
when no catalyst was present. Hence there is a minimum
amount of catalyst required to be present to observe the
enhancement in epoxide selectivity.
a
b
DBPC*BHT, Galvinoxol
the hydroperoxy radical intermediates and this indicates the
pathway by which the gold catalysts secure higher reaction
selectivities to the epoxide.
4.4 Reduction of Reaction Intermediates with PPh3
In oxidation reactions using molecular oxygen it is possible
that hydroperoxides are formed as reaction intermediates,
as demonstrated by Aprile et al. [24]. The hydroperoxide so
formed decomposes readily in the gas chromatograph
injector due to the high temperatures and as a result these
reaction intermediates cannot be identified or quantified
with such a standard gas chromatography analysis. There-
fore, determination of hydroperoxide is achieved by
reduction of these compounds into corresponding products
such as alcohols by reaction with triphenylphosphine
(PPh3) [25–28]. A series of standard reactions were per-
formed for 24 h at 90 °C in presence of AIBN with the
addition of either graphite or 1wt% Au/graphite. The
reaction mixture was filtered and 5 mL of the reaction
mixture was stirred with PPh3 (0.12 g) under a helium
atmosphere at room temperature. Both reaction mixtures
(before and after PPh3 treatment) were analyzed by gas
chromatography and proton NMR spectroscopy. The NMR
spectrum showed no observable differences. Gas chroma-
tography analysis, however, showed an increase in allylic
products (2-decenal 2-decen-1-ol and 1-decen-3-one) after
PPh3 reduction (Table 2). The effects are more marked for
the reactions when graphite was present as compared to
those where 1 wt% Au/graphite was present. This indicates
that the Au nanoparticles are more effective at intercepting
4.5 Reaction with Alternative Radical Initiators
and the Effect of Radical Scavengers
on the Oxidation of Dec-1-ene
We have shown that the autoxidation of dec-1-ene does not
occur until 100 °C without addition of a radical initiator
(Fig. 1). We have contrasted the effects of TBHP and BPO
as alternative radical initiators (Table 3) and it is apparent
AIBN gives the best results for the conversion of dec-1-
ene; in all cases the presence of gold nanoparticles is more
effective than graphite alone. In addition we conducted two
further sets of experiments. In the first equimolar amounts
of radical scavengers [2,6-di-tert-butyl-4-methylphenol,
(DBPC*BHT)) or galvinoxyl] and their respective radical
initiators were added along with reaction mixture.
When reactions were performed in the presence of
graphite or gold/graphite and radical initiator (TBHP,
AIBN or BPO), it is apparent that the oxidation of dec-1-
ene was severely inhibited (Table 3). In a second experi-
ment, to confirm that oxidation of dec-1-ene uses dissolved
oxygen from the air as the terminal oxidant, a standard
reaction was performed under an inert atmosphere (5 bar of
nitrogen in an autoclave reactor) in place of atmospheric
air and we observed no conversion even in the presence of
the catalysts and AIBN.
123

694
U. N. Gupta et al.
reuse. Ethanol and acetone washed catalysts were found to
be active and gave similar dec-1-ene conversion as the
fresh catalyst; however, the epoxide selectivity was
decreased.
32
30
28
26
24
22
20
18
16
14
12
10
8
Conversion
Epoxide
4.7 Comments on the Reaction Mechanism
We have observed that the product is the 1,2-epoxide even
when AIBN reacts with dec-1-ene and no catalyst is
present. This is an interesting finding since the initial
radical abstraction of H would be expected at the 3-
(allylic) position of 1-decene and this is where initial
oxygen attack would be expected to occur, giving C7H15–
CH(O–O.)–CH=CH2; this could possibly lead to the 2,3-
epoxide being formed, which we do not observe even in
trace amounts. Oxygen attachment could be at the 1-posi-
tion in the delocalised allylic radical and the resultant
primary allylic hydroperoxide, on dehydration would pro-
duce C7H15–CH=CH–CHO; this could aid epoxidation, as
noted earlier by Mukaiyama and co-workers [29] by con-
version into the corresponding peracid C7H15–CH=CH–
CO–OOH and the product of peracid epoxidation would be
the 1,2-epoxide exclusively plus an equivalent amount of
dec-2-enoic acid. Indeed, we will investigate the role of
these in situ generated aldehydes on epoxide formation in a
subsequent study. In the presence of the catalyst (Table 1)
6
4
2
0
Fresh
Dried
Acetone
Ethanol
Catalyst treatment
Fig. 6 The influence of catalyst reusability with respect to the
catalyst pre-treatment conditions used on the epoxide yield. Reaction
conditions: dec-1-ene (53 mmol), AIBN (35 lmol), 1 wt% Au/
Graphite (0.1 g), 90 °C, 24 h, air (1 atmosphere)
4.6 Catalyst Reusability
The reuse of the catalyst was investigated after use in a
standard reaction. The used catalysts were either dried in
an oven at 110 °C and/or either washed with acetone or
ethanol (Fig. 6). Drying alone did not facilitate effective
Scheme 1 Proposed
mechanism of 1-decene
epoxidation
123

Solvent-Free Aerobic Epoxidation of Dec-1-ene
695
we do observe products such as dec-1-en-3-one, dec-2-
enoic acid and dec-2-enal, in low amounts in comparison to
the epoxide. When the catalyst/support is present then the
simple radical chemistry described above is likely to be
overwhelmed by the catalysis and the influence of the
surface; the 1,2-epoxide is formed in much higher amounts
compared with the by-products.
without gold more allylic oxidation products are produced
rather than epoxide, perhaps due to a process of Russell
termination (illustrated for the 3-dec-1-enylperoxy radical
in Scheme 2). The peroxy-radical formed after interaction
of an oxygen molecule with the radical from 1-decene,
which contains an a-hydrogen atom, can undergo Russell
termination [30] to produce an alcohol and ketone. Dec-1-
ene conversion is low when the reaction was performed
with graphite only, but with gold 1-decene conversion is
increased three- to four-fold as Russell termination is less
important and further reaction consumes three molecules of
dec-1-ene to produce epoxide, allylic alcohol and a dec-1-
ene radical (Scheme 1). Hence the presence of the gold
surface is critical in directing the reactions towards the
formation of the 1,2-epoxide.
Based on the results obtained, a mechanism is proposed
for the epoxidation of dec-1-ene (Scheme 1) which is
similar to our proposed mechanism for the epoxidation of
cyclooctene using supported gold catalysts [22]. Initially,
as discussed above, the radical initiator removes an allylic
hydrogen forming a dec-1-enyl radical which subsequently
interacts with oxygen to produce a peroxy-radical. This
peroxy-radical reacts with dec-1-ene to produce equimolar
amounts of hydroperoxide and dec-1-ene radical and thus
forming a cycle to produce a peroxy-radical by interacting
with oxygen. In the absence of gold nanoparticles, the
decene hydroperoxide reaction intermediate accumulates in
the reaction mixture and hence we observed more allylic
products when PPh3 was added to the product mixture.
However, when supported gold nanoparticles are present in
the reaction mixture, the hydroperoxide so formed com-
pletes the reaction to produce epoxide and the allylic
products. This involves a complex sequence in which the
hydroperoxide reacts on the surface of the gold nanopar-
ticle to produce a 3-dec-1-enyloxy radical, this is trapped
by a second molecule of dec-1-ene with subsequent frag-
mentation into epoxide and the 3-decen-1-yl radical. A
similar process in which a peroxy radical attaches to the
terminal carbon of 1-decene, followed by fragmentation
(O–O cleavage) can be envisaged and this could yield
epoxide together with a decenyloxy radical. These pro-
cesses appear to require the presence of the gold surface,
which is critical in directing the selectivity towards the
epoxide rather than the allylic products expected from a
normal free radical processes that take place solely in the
homogeneous liquid phase. Hence the intervention of the
gold surface is critical in controlling this free radical
chemistry. As expected when the reaction was performed
5 Conclusions
The epoxidation of dec-1-ene was achieved over gold
supported on graphite in presence of a small amount of
radical initiator (AIBN) using oxygen from air as the ter-
minal oxidant. However, without the addition of radical
initiators no oxidation was observed even when a catalyst
is present in the reaction mixture and the temperature is
raised to 100 °C. Time online studies at different temper-
atures indicated that the induction period of 1-decene
epoxidation at lower temperatures were longer than that at
higher temperatures and this is probably related to the
thermal decomposition of AIBN. Furthermore, at lower
reaction temperatures production of allylic products
(aldehyde, alcohol and ketone) is enhanced; however, the
epoxide selectivity improves at higher reaction tempera-
tures and at higher conversions and reaction times. Reac-
tion of the products with PPh3 indicates that an alkene
hydroperoxide is formed as a reaction intermediate, this
reacts subsequently to produce aldehydes, alcohols and
ketones.
Acknowledgments The authors thank Sasol, Mintek Cardiff Uni-
versity for funding this work.
O
-
O O
O
C
O
O
-
O O
-
-
=
H H
C C
-H
-H
C7
15
C7
-H
C7
C
15
2
15
References
H
C
H
-
-
C
-H
H=
CH
2
C7
C
15
H
=
H
C
H =
CH
2
Russel termination
H
C
C
2
H
1. Othmer K (1994) Encyclopedia of chemical technology, vol 9,
4th edn. Wiley, New York, p 915
2. Franz G, Sheldon RA (1991) In: Elvers B, Hawkins S, Shulz G
(eds) Ullmann’s encyclopedia of industrial chemistry, vol A(18),
5th edn. VCH Publisher, Weinheim, pp 261–311
3. Ansmann A, Kawa R and Neuss M, US Patent 7,083,780 B2,
Aug. 1, 2006, assigned to Cognis Deutschland
-H
C7
H
15
-H
C7
15
O2
H
O
C
O
C
=
H
C
H
C
CH =
CH
2
2
4. Bassler P, Weidenbach M, Goebbel H (2010) Chem Eng Trans
21:571–576
Scheme 2 Russell termination and formation of allylic products
123

696
U. N. Gupta et al.
5. Lines EL, Fairbrother RJ and Herbst JA, US 4157346, 1979,
assigned to Olin Corporation
19. Alshammari H, Miedziak PJ, Morgan DJ, Knight DW, Hutchings
GJ (2013) Green Chem 15:1244–1254
6. Rameswaran M and Cochran RN, US 5081267, 1992, assigned to
Arco Chemical Technology
7. Zajacek JG, Pa D and Crocco GL, US 5214168, 1993, assigned to
Arco Chemical Technology
20. Alshammari H, Miedziak PJ, Davies TE, Willock DJ, Knight
DW, Hutchings GJ (2014) Cat. Sci. Technol. 4:908–911
21. Brandrup J, Immergut EH, Grulke EA (1999) Polymer handbook,
4th edn. Wiley, New York
8. Nijhuis TA, Makkee M, Moulijn JA, Weckhuysen BM (2006) Ind
Eng Chem Res 45:3447–3459
22. Bawaked S, Dummer NF, Bethell D, Knight DW, Hutchings GJ
(2011) Green Chem. 13:127
9. Weissermel K, Arpe HJ (1993) Industrial organic chemical, 2nd
edn. VCH Publishers Inc., New York, pp 141–264
10. Clerici MG, Bellussi G, Romano U (1991) J Catal 129:159–167
11. Pescarmona PP, Jacobs PA (2008) Catal Today 137:52–60
12. Lane BS, Burgess K (2001) J Am Chem Soc 123:2933–2934
13. Mandelli D, van Vliet MCA, Sheldon RA, Schuchardt U (2001)
Appl. Catal. A: Gen. 219:209–213
23. Enache DI, Knight DW, Hutchings GJ (2005) Catal Lett
103:43–52
24. Aprile C, Corma A, Domine M, Garcia H, Mitchell C (2009) J
Catal 264:44–53
25. Shilov AE, Shul’pin GB (1997) Chem Rev 97:2879–2932
26. Shul’pin GB, Druzhinina AN (1992) Bull. Russ. Acad. Sci.
41:346–347
14. Scotti N, Ravasio N, Zaccheria NF, Psaro R, Evangelisti C (2013)
Chem. Commun. 49:1957
27. Shul’pin GB, Nizova GV (1992) React. Kinet. Catal. Lett.
48:333–338
15. Ho KP, Wong WL, Lam KM, Lai CP, Chan TH, Wong KY
(2008) Chem Eur J 14:7988–7996
28. Shul’pin GB, Nizova GV, Kozlov YN (1996) New J Chem
20:1243–1256
16. Hughes MD, Xu YX, Jenkins P, MacMorn P, London P, Enache
DI, Carley AF, Attard GA, Hutchings GJ, King F, Stitt FEH,
Johnston P, Griffin K, Kiely CJ (2005) Nature 437:1132–1135
17. Alshammari H, Miedziak PJ, Knight DW, Willock DJ, Hutchings
GJ (2013) Catal. Sci. Technol. 3:1531–1539
29. Yamada T, Takahashi K, Kato T, Takai T, Inoki S, Mukaiyama T
(1991) Chem. Lett. 20:641–644
30. Sheldon RA, Kochi JK (1981) Metal-catalysed oxidation of
organic compounds. Academic Press, New York
18. Alshammari H, Miedziak PJ, Bawaked S, Knight DW, Hutchings
GJ (2012) Chem.Cat.Chem 4:1565–1571
123