Low-temperature aerobic oxidation of decane using an oxygen-free radical initiator

Article abstract of DOI:10.1016/j.jcat.2011.08.003

The direct selective oxidation of long chain alkanes by O₂ is a highly demanding reaction. We have shown that it is possible to oxidise n-decane in the presence of the oxygen-free radical initiator azobisisobutyronitrile. Formation of a range of oxygenated products has been observed under relatively mild conditions (70 °C in air). Although the presence of a catalyst is not essential when the initiator is used, ceria-based catalysts have been found to increase the selectivity to alcohols by modifying the oxyfunctionalisation of decane.

Full text of DOI:10.1016/j.jcat.2011.08.003

Journal of Catalysis 283 (2011) 161–167
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Low-temperature aerobic oxidation of decane using an oxygen-free radical initiator
Rhys Lloyd a, Robert L. Jenkins a, Marco Piccinini a, Qian He b, Christopher J. Kiely b, Albert F. Carley a,
Stanislaw E. Golunski a, Donald Bethell a, Jonathan K. Bartley a, Graham J. Hutchings a,
⇑
a Cardiff Catalysis Institute, School of Chemistry, Cardiff University, Park Place, Cardiff CF10 3AT, UK
b Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA 18015-3195, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 January 2011
Revised 26 June 2011
Accepted 3 August 2011
Available online 15 September 2011
The direct selective oxidation of long chain alkanes by O2 is a highly demanding reaction. We have shown
that it is possible to oxidise n-decane in the presence of the oxygen-free radical initiator azobisisobuty-
ronitrile. Formation of a range of oxygenated products has been observed under relatively mild condi-
tions (70 °C in air). Although the presence of a catalyst is not essential when the initiator is used,
ceria-based catalysts have been found to increase the selectivity to alcohols by modifying the oxyfunc-
tionalisation of decane.
Keywords:
Ó 2011 Elsevier Inc. All rights reserved.
Alkane oxidation
Radical initiator
Cerium oxide
Gold catalysis
1. Introduction
has occurred, the reaction of alkylperoxy radicals with alkane mol-
ecules is rate limiting at low temperatures, whereas the addition of
Catalytic selective oxidations account for a significant proportion
of modern commercial chemical processes [1]. One of the most
important applications of selective oxidation is the functionalisation
of hydrocarbons. With industry moving towards the lower-
environmental-impact and lower-cost processes, the direct oxida-
tion of alkanes has become a key priority [1,2]. Current commercial
selective alkane oxidation processes include the production of
cyclohexanone and cyclohexanol from cyclohexane (KA oil,
106 tons/year) [2], the feedstock chemicals for nylon and maleic
anhydride primarily from butane (106 tons/year) [3]. Another nota-
ble hydrocarbon oxidation is the production of terephthalic acid
from p-xylene (3 ꢀ 106 tons/year), used in the synthesis of poly(eth-
ylene terephthalate) [2].
Direct alkane autoxidation, which avoids an initial dehydroge-
nation step, proceeds by a free-radical chain mechanism producing
hydroperoxides [4], which may undergo consecutive complex reac-
tions forming ketones, alcohols and carboxylic acids [5]. Potential
sources for the reaction initiation include the thermal generation
of radicals or from impurities or initiators [4,6]. Various radical ini-
tiators have been used for the oxidation of alkanes. Twigg [7] used
benzoyl peroxide to initiate the oxidation of decane. Propagation
occurs predominantly by the addition of oxygen to alkyl radicals
to give alkylperoxy radicals that further react with alkane mole-
cules, producing hydroperoxides and alkyl radicals. Once initiation
oxygen to alkyl radicals is rate limiting at high temperatures and
low oxygen concentrations. Consequently, the change in alkylper-
oxy and alkyl concentrations will affect consecutive reactions and
hence the product distribution [5]. Termination may occur by the
combination of radical species or the formation of relatively stable
species (reaction with a radical stabiliser). The hydroperoxides
formed may undergo thermal or catalytic decomposition by free-
radical or ionic mechanisms. Decomposition of the hydroperoxides
produced may therefore both promote the reaction and direct the
product distribution.
Gold catalysts are often used to decompose the radical initiators
to enhance the rate of reaction [8]. Numerous reports on the cata-
lytic decomposition of radical initiators over gold surfaces have
been published, including hydrogen peroxide [9] and organic per-
oxides such as cumene hydroperoxide (CHP) [10]. Recently, the
oxygen-free radical initiator azobisisobutyronitrile (AIBN) has
been reported to enhance significantly both the conversion and
selectivity for the aerobic epoxidation of cyclohexene and 1-octene
over gold nanoparticles supported on nano-ceria or MCM-41
supports [11,12]. Other radical initiators (benzoyl peroxide, tert-
butyl peroxybenzoate, tert-butyl hydroperoxide, cumene
hydroperoxide and 2,20-azobis(2-methylpropionamidine)dihydro-
chloride) were tested with gold on MCM-41, but they were all
found to give lower conversions and selectivities than AIBN. The
addition of 2,2,6,6-tetramethylpiperidin-1-yloxy (TEMPO, which
reacts with other radicals thus shortening their lifetimes) was
found to reduce the conversion dramatically when added together
with AIBN and the gold catalyst. If TEMPO was added at the
⇑
Corresponding author. Fax: +44 (0)29 208 74030.
E-mail address: hutch@cf.ac.uk (G.J. Hutchings).
0021-9517/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2011.08.003

162
R. Lloyd et al. / Journal of Catalysis 283 (2011) 161–167
beginning of the reaction, the selectivity to the epoxide was also
found to be lower, but if it was added after the start of the reaction,
only the conversion was observed to decrease. Corma and co-
workers [11–13] proposed the reaction to proceed by AIBN react-
ing with gold particles to form an alkylgold intermediate that
can catalytically produce hydroperoxides. However, the increase
in selectivity seen with AIBN compared to other radical initiators
indicates a mechanism, such as a direct O-transfer, that by-passes
the usual radical process with hydroperoxides. The results
obtained upon addition of TEMPO suggest that when it is added
at the beginning of the reaction, it inhibits the selective mecha-
nism. Since, if TEMPO is added later, it does not affect the reaction
mechanism, the mechanism(s) must still depend on the generation
of radical species.
5.7 ml H2O) was adjusted to pH 10 by the addition of NaOH
(0.2 M). This solution was added to a slurry comprising nanoCeO2
(1.260 g in 25 ml H2O), and the pH was again adjusted to 10 by
the addition of NaOH (0.2 M). The resulting mixture was stirred
vigorously for 18 h after which the solid was isolated by filtration
and washed thoroughly to remove chloride ions and dried (110 °C,
16 h). The catalyst (Au/nanoCeO2) was not calcined.
Samples for examination by scanning transmission electron
microscopy (STEM) were prepared by dispersing the dry catalyst
powder onto a holey carbon film supported by a 300 mesh copper
TEM grid. STEM high-angle annular dark-field (HAADF) images and
STEM bright field images (BF) of the catalysts were obtained using
an aberration corrected JEOL 2200FS (S)TEM operating at 200 kV.
X-ray photoelectron spectra were recorded on a Kratos Axis
The radical initiator AIBN has been used for the oxidation of
heptane at 100 °C; however, using small amounts of initiator, it
was reported that the molar amounts of ketones and alcohols
formed were significantly less than the amount of initiator used
[14]. The inefficiency of AIBN was attributed to the dimerisation
of radicals formed [15–18] and the formation of 2-cyanopropan-
2-ylperoxy radicals, which act as radical trap [14].
Gold catalysts have previously been used to selectively oxidise
cyclohexane [19–25] and n-hexadecane [26] and have been shown
to be active when supported on silica [19–21], graphite [22],
alumina [23] or ZSM-5 [24]. For the oxidation of cyclohexane, the
selectivity towards alcohols and ketones has been shown to shift
with conversion, alcohols being favoured at low conversions [22].
The results of a recent study suggest that the reaction proceeds
by essentially the same mechanism as in autoxidation, with the
gold playing a role in directing the specific pathway [27]. Many
of these previous studies have been carried out with the more reac-
tive cyclic alkanes, in which all carbon atoms in the molecules are
equivalent. In this study, we have investigated decane as a model
linear alkane that adds a degree of complexity to the regioselectiv-
ity of the reaction over the cyclic systems.
Long chain linear alkanes are relatively inexpensive feedstocks
available from the Fischer Tropsch process [28]. The selective oxi-
dation of alkanes, especially to linear primary alcohols, is of indus-
trial importance for the production of plasticizers and surfactants
for detergents [29]. It is, however, considerably more difficult to
selectively oxidise long chain alkanes, such as decane, than shorter
chain cyclic alkanes (often used as model reactants) or alkenes of
the equivalent length because of their tendency to dehydrogenate
and form coke deposits as well as greater regioselectivity demands
[30].
The aim of this work is to investigate the effect of radical initia-
tors on the aerobic oxidationof n-decane. In particular, the combina-
tion of the radical initiator AIBN with gold supported on nano-ceria
has been studied for alkane oxidation. It is proposed, based on the
previous work of Corma and co-workers [11–13], that the genera-
tion of a sterically hindered hydroperoxide may lead to terminal
selectivity in the H-abstraction from the alkane. The best reported
terminal selectivity for linear alkane oxidation in the literature has
been reported by Zhan et al. for n-hexane oxidation with Mn-
exchanged zeolites [31]. This gave 24% terminal selectivity at
0.007% conversion, which drops to 10% terminal selectivity at 0.1%
conversion. This study was undertaken at low temperature
(690 °C) to limit conversion and determine whether high terminal
selectivity could be achieved using this methodology.
Ultra DLD spectrometer using a monochromatised AlKa X-ray
source (120 W). Spectra were recorded at analyser pass energies
of either 160 eV (survey scans) or 40 eV (detailed scans). Binding
energies are referenced to the C(1s) binding energy of adventitious
carbon contamination, which is taken to be 284.7 eV.
The BET surface area of the prepared Au/nanoCeO2 was mea-
sured by N2 adsorption at ꢁ196 °C using a Micromeritics Gemini
2360 surface area analyser. A Varian 55B AA spectrometer was
used to determine the Au content of the material.
For comparison in addition to the prepared Au/nanoCeO2 and
nano-CeO2 support used, the catalytic properties of a standard cer-
ia powder (Aldrich), a high surface area titania (Degussa, P25) and
ZSM-5 (Zeolyst, SiO2:Al2O3 = 30, calcined at 550 °C in static air for
4 h) were tested.
Reactions were performed at 70 °C under reflux at atmospheric
pressure in air, or at 90 °C and 1.2 MPa O2, both with and without a
radical initiator. The tests at atmospheric pressure were performed
with decane (10 g) and AIBN (50 mg) in 25 ml glass round-bottom
flasks under reflux (70 °C, 20 h) with stirring. High-pressure reac-
tions were performed under conditions similar to those used by
Corma and co-workers [12]. Decane (7.68 g, 54 mmol) or 1-octene
(6.06 g, 54 mmol), catalyst (100 mg) and AIBN or CHP (30 mg) were
placed in 12 ml stainless-steel autoclave reactors. The autoclaves
were flushed twice with O2 and pressurised with 1.2 MPa O2, then
heated to 90 °C for 2 h or 20 h with stirring. Reactions were
quenched using an ice bath.
Reaction products were quantified using a Varian 3380 GC fitted
with a CP-Wax 52CB column, and FID and GC injections were
duplicated and the average reported. For the decane reactions,
1,2,4-trichlorobenzene (150 ll) was used as an internal standard.
Hydroperoxide concentrations of selected reactions were mea-
sured by the reduction method with triphenylphosphine (PPh3)
[32,33]. To an aliquot of the reaction products, 10% PPh3 in acetone
was added (to give an approximate weight ratio of 3:1) and left to
equilibrate for at least 20 min prior to the GC analysis. The hydro-
peroxide species are selectively reduced to alcohols upon addition
of PPh3, as shown in Eq. (1). Assuming in the absence of PPh3, the
hydroperoxides decompose into ketones and alcohols upon GC
analysis in equal proportions then the hydroperoxide concentra-
tions can be estimated.
ROOH þ PPh3 ! ROH þ POPh3
ð1Þ
Decane conversion was calculated from the molar concentrations
using the following equation:
h
i.
X
Conversion ¼
mol productions ꢀ Cn
ꢀh
i
X
2. Experimental
mol reactant remaining ꢀ Cn
h
iꢁ
X
Au was dispersed onto nanoCeO2 (VP AdNanoÒ Ceria 90, Degus-
sa AG) by deposition precipitation, using a previously reported
method [12]. An aqueous solution containing HAuCl4 (0.140 g in
þ
mol products ꢀ Cn ꢀ 100%
where Cn = carbon number (i.e. 10 for decane).

R. Lloyd et al. / Journal of Catalysis 283 (2011) 161–167
163
Carbon balance (C bal) was calculated also using molar concen-
trations using the following equation:
Au particles
A
h
i
X
C bal ¼
mol reactant remaining ꢀ Cn
h
i h
i
X
X
þ
mol products ꢀ Cn =
mol reactant ꢀ Cn
A
ꢀ 100%
where the reactant remaining and product concentrations are calcu-
lated from the gas chromatographic data and the reactant concen-
tration is based on the weight of the sample. Errors in this
analysis can be introduced due to losses incurred during the trans-
fer of solutions and the inaccuracy in using assumed relative
response factors for diones as well as the carbon number for
unidentified products. Response factors for all reported alcohols,
ketones and acids were experimentally determined. GCMS was also
performed, to confirm product identification, using an Agilent 6890
GC also fitted with a CP-Wax 52CB column and a Waters GCT Pre-
mier orthogonal acceleration time-of-flight mass spectrometer.
B
B
Fig. 2. Representative higher magnification STEM–HAADF images of the Au/
nanoCeO2 ‘dried-only’ catalyst. Expanded views of the regions delineated as A
and B show some of the supported Au species in profile view. It is clear that they
have a thickness of two to three atomic layers and are more ‘particle-like’ than ‘raft-
like’ in morphology.
3. Results and discussion
clearly have a thickness of 2–3 atomic layers, suggesting that the
1–2-nm Au nano-clusters are more three dimensional in nature
as opposed to monolayer rafts.
In Table 1, the BET surface area, pore volume and gold content
of the prepared Au/nanoCeO2 catalyst are shown. No significant
change in surface area is observed upon loading gold onto the nan-
oCeO2 support, but the pore volume decreased slightly. The Au/
nanoCeO2 catalyst was found to have a higher gold content than
previously reported materials prepared using this method on
higher surface area ceria supports [12], which suggests that the
ceria support used here has a stronger affinity for the deposition
precipitation of the gold complex from solution. The Au/nanoCeO2
‘dried-only’ catalyst was studied by aberration corrected scanning
transmission electron microscopy in order to elucidate how the Au
was dispersed on the nano-ceria support surface. A representative
lower magnification BF- and HAADF–STEM image pair is shown in
Fig. 1. The support material has a particle size range of about
5–20 nm, and the BF-STEM image showed phase contrast fringes
characteristic of the CeO2 lattice. The Au species can only barely
be discerned by mass-thickness contrast as small dark flecks in
the BF-STEM image. However, in the corresponding HAADF–STEM
image, the Au clusters can be readily distinguished as bright fea-
tures by virtue of their higher Z-contrast (ZAu = 79, ZCe = 58). The
Au clusters were quite monodispersed, being 1–2 nm sized, as
shown in the corresponding particle size histogram. A higher mag-
nification HAADF–STEM image of the same sample is shown in
Fig. 2, in which some of the Au species are imaged in profile. They
XPS analysis of the Au/nanoCeO2 catalyst was carried out to
determine the oxidation state of the gold. The initial Au(4f) acqui-
sition (Fig. 3a) shows that the gold in the catalyst was predomi-
nantly in the 3+ oxidation state, but these species reduced
significantly to Auo after continued exposure to X-rays (Fig. 3b).
The initial state of the surface (at t = 0 irradiation time) is likely
to be almost entirely Au3+
.
No conversion was detected for the blank decane reaction, at
70 °C, under reflux at atmospheric pressure in air for 20 h or at
90 °C and 1.2 MPa O2 for 2 h, without a radical initiator or catalyst.
Oxidation of decane with AIBN under these conditions was found to
produce a range of oxygenated products. AIBN starts to decompose
to form radicals at above 60 °C (Fig. 4), although the rate of decom-
position is slow at these low temperatures and the half-life is in the
order of tens of hours [34]. A summary of the products formed is gi-
ven in Table 2, with a more comprehensive breakdown given in Ta-
ble 3. The raw data in Table 3 has been recalculated based on the
data obtained with and without PPh3 addition to determine the
amount of peroxide species (see the experimental section for de-
tails), which is shown in the summary in Table 2. Decane conversion
using AIBN is found to increase at higher temperature and pressure,
although the selectivity to peroxide species is found to decrease (Ta-
ble 2). The radical initiator is proposed to produce peroxide species
which then form alcohols and carbonyl compounds by consecutive
reactions [4]. At higher temperature and pressure, it is considered
that more peroxides are produced, hence the higher conversion.
However, because at higher temperatures, the product-determining
steps are faster, the selectivity to peroxides is lower.
Table 1
Surface area and Au loading.
Material
[Au] (wt.%) BET surface area (m2/g) Pore volume (ml/g)
NanoCeO2
Au/nanoCeO2 4.3
–
77
78
0.59
0.33
Fig. 1. Representative lower magnification BF (a) and HAADF (b) STEM image pair of the Au/nanoCeO2 ‘dried-only’ catalyst and the corresponding gold particle size
distribution (c).

164
R. Lloyd et al. / Journal of Catalysis 283 (2011) 161–167
opening gives
ther oxyfunctionalisation at the
acyloxy radicals (ꢀOOC (CH2)4COOH) that subsequently form to give
a C5 -formyl radical.
x
-formyl radicals (ꢀCH2(CH2)4CHO) that allow fur-
position. CO2 is eliminated from
x
x
For linear alkanes, short-chain acids are thought to form due to
breaking of C–C bonds adjacent to an alkoxy radical
(CH3(CH2)n(CHOꢀ)(CH2)mCH3) formed by the oxidation of a second-
ary carbon. The bond breaking is analogous to the cyclohexane
mechanism, but the linear alkane will form two new species, an
oxygenated fragment (CH3(CH2)nCHO) and an alkyl radical
(ꢀCH2(CH2)mCH3) that can both be oxidised to carboxylic acids. This
process was studied widely in Germany in the 1940s and Eastern
Europe in the 1960s and 1970s to produce fat and has been used
industrially in the past to oxidise n-butane to two molecules of
acetic acid with cobalt catalysts [35].
Identification of the products by GCMS revealed that the oxida-
tion of decane at 70 °C produced phthalic anhydride, cyclodecane,
hexenone and branched alkenes in addition to the products re-
ported in Tables 2 and 3. AIBN and dimethyl-N-(2-cyano-2-pro-
pyl)-ketenimine (the dimer formed from the radicals generated
from AIBN) were also detected by GCMS, but no other AIBN decom-
position products were found.
Fig. 3. Au(4f) spectra for the dried Au/nanoCeO2 catalyst showing the sensitivity of
the Au oxidation state to X-irradiation: (a) initial measurement, accumulated
irradiation time = 600 s; (b) repeated measurement, accumulated irradiation
time = 3160 s.
At a longer reaction time (20 h, 90 °C, 1.2 MPa O2), a low blank
decane conversion is observed (Table 4). Oxidation of decane with
AIBN under these conditions results in a conversion that is signif-
icantly higher than the theoretical stoichiometric conversion from
the two radicals generated from each AIBN molecule alone (0.68%).
Given the observed conversion and that radicals generated from
AIBN have been reported to readily form dimers [15–18], it is pro-
posed that only a limited proportion of radicals created from AIBN
form oxidation chain reactions.
The Au/nanoCeO2 catalyst was found to be inactive for the oxi-
dation of decane at 90 °C and 1.2 MPa O2, over 2 h, without a rad-
ical initiator. Upon addition of both the Au/nanoCeO2 catalyst and
AIBN to decane, at 90 °C and 1.2 MPa O2, over 2 h and 20 h, there is
a small decrease in conversion, and the selectivity to C10 alcohols
and peroxides increases very slightly compared with the reaction
of AIBN and decane (Tables 2 and 3). Addition of the nanoCeO2 sup-
port with AIBN to decane, at 90 °C and 1.2 MPa O2, over 2 h or 20 h
was found to decrease the conversion even more than
Au/nanoCeO2 and AIBN, under the same conditions, and increase
the ratio of alcohols to ketones (Tables 2 and 3). The nanoCeO2 sup-
port was found to be inactive for the oxidation of decane at 90 °C
and 1.2 MPa O2, over 2 h, without a radical initiator, but over
20 h a low conversion was detected (Table 4).
Fig. 4. Decomposition pathways of AIBN and CHP.
Short-chain carboxylic acids are formed in addition to decanoic
acid. COx was not observed under the reaction conditions em-
ployed, which indicates that the mechanism for the formation of
the short-chain acids is not the same as the one proposed by Her-
mans et al. for cyclohexane conversion [2]. For cyclohexane, ring
Results for the oxidation of decane over either a standard CeO2
powder, TiO2 or ZSM-5 with AIBN, at 90 °C and 1.2 MPa O2 for 20 h
are shown in Table 4. The standard CeO2 powder and AIBN was
found to give the same conversion as using AIBN alone but the
Table 2
Oxidation of n-decane using AIBN, with and without catalyst.
Initiator/catalyst
Conditions
Conversion%
Selectivity%
C10 ketones
C10 internal
alcohols
C10 internal
peroxidesc
C10
C10
diones
C1–C9
acids
Others
terminald
AIBNa
70 °C, 20 h, air, reflux
90 °C, 2 h, 1.2 MPa O2
90 °C, 2 h, 1.2 MPa O2
90 °C, 2 h, 1.2 MPa O2
90 °C, 20 h, 1.2 MPa O2
90 °C, 20 h, 1.2 MPa O2
90 °C, 20 h, 1.2 MPa O2
0.26
0.49
0.46
0.36
1.13
0.87
0.78
16.4
35.2
19.8
26.6
58.1
25.2
15.6
0.5
4.1
6.9
12.0
0.9
19.3
23.7
29.7
15.4
19.4
25.4
12.6
11.7
10.2
0.6
0.1
0.1
1.4
2.1
1.6
1.5
8.4
11.7
10.7
9.6
7.0
11.4
11.0
3.1
3.9
4.3
41.3
29.6
38.8
21.8
6.9
AIBNb
b
b
AIBN + Au/nanoCeO2
b
AIBN + nanoCeO2
3.2
AIBNb
12.3
21.4
21.9
AIBN + Au/nanoCeO2
AIBN + nanoCeO2
9.5
16.2
b
a
b
c
10.0 g decane, 50 mg AIBN.
7.69 g decane, 30 mg AIBN.
Secondary decyl hydroperoxides.
d
C10 terminal includes 1-decanol, 1-decyl hydroperoxide and decanoic acid.

Table 3
Product breakdown for the oxidation of n-decane, with and without AIBN, with and without catalyst, and with and without addition of PPh3 (for selected reactions).
Catalyst/initiator
Conditions
Conv.% Selectivity/%c
C10 5/4-one C10 3-one C10 2-one C10 4-ol C10 3-ol C10 2-ol C10 1-ol C10 acid C9 acid C8 acid C7 acid C5 acid C4 acid C3 acid C2 acid C10 diones Others
C bal.%
AIBNa
AIBNb
70 °C, 20 h, air, reflux
90 °C, 2 h, 1.2 MPa O2
–
0.26
15.0
11.7
7.4
5.8
9.2
6.1
6.8
18.7
3.6
10.0
4.9
14.6
0.3
0.9
0
0
0.4
0.9
0.1
0.3
1.2
1.6
1.2
0
0
0
0
0
0.3
0
8.4
6.6
41.2
23.0
93
86
+PPh3 0.27
–
0.49
20.1
19.6
10.0
9.6
12.7
11.8
4.6
9.3
2.7
5.3
4.6
8.4
0.1
0
0
0
0.3
0.5
0
0
1.3
1.9
1.4
0.4
0.3
0
0
0
0.6
0.2
11.7
11.2
29.6
21.7
83
77
+PPh3 0.45
Au/nanoCeOx + AIBNb 90 °C, 2 h, 1.2 MPa O2
–
0.45
14.0
11.2
6.6
5.9
8.8
7.4
6.6
13.6
3.8
7.8
6.3
11.3
0
0
0
0.1
0.7
0.9
0
0.1
0.8
2.8
1.3
0.3
0.3
0.1
0
0.2
1.2
0.7
10.7
12.2
38.9
25.4
94
81
+PPh3 0.47
NanoCeO2 + AIBNb
90 °C, 2 h, 1.2 MPa O2
–
0.36
17.4
14.0
8.9
6.4
13.1
10.2
11.4
18.7
5.5
10.0
7.8
14.3
0.9
1.1
0.4
1.4
0
0
0
0
1.6
0.7
1.1
0
0.5
0
0
0
0
0
9.6
4.0
21.8
19.1
105
114
+PPh3 0.40
Blankb
AIBNb
90 °C, 20 h, 1.2 MPa O2
90 °C, 20 h, 1.2 MPa O2
–
–
0.028
1.13
21.9
12.2
21.2
6.0
2.6
4.6
0
0
0
0
0
0
0
0
0
0
31.5
95
27.2
23.7
17.9
15.1
19.3
16.3
3.0
5.4
1.7
3.1
2.8
4.9
1.1
0.7
0.8
1.0
0
0
0.3
0.3
2.0
1.0
3.7
2.4
3.5
2.1
2.8
3.2
0
0
7.0
8.7
6.9
12.1
89
99
+PPh3 1.47
Au/
90 °C, 20 h, 1.2 MPa O2
–
0.87
12.5
7.8
10.9
10.3
6.1
8.8
0.9
0.5
0.2
0.6
1.3
5.3
6.4
3.4
4.3
11.4
9.5
79
nanoCeO2 + AIBNb
+PPh3 0.88
11.6
11.1
8.1
5.5
9.7
16.0
8.7
8.7
4.1
11.5
4.9
1.3
2.7
0.8
1.7
0.3
1.0
0.3
0
1.4
0.9
4.1
5.9
3.6
8.1
3.8
8.7
0.5
0
10.7
11.3
7.6
89
b
NanoCeO2
90 °C, 20 h, 1.2 MPa O2
90 °C, 20 h, 1.2 MPa O2
–
–
0.11
0.78
11.1
14.3
107
NanoCeO2 + AIBNb
5.3
11.2
8.0
7.8
11.0
9.3
10.3
13.2
5.9
7.3
9.1
10.2
1.0
1.0
0.4
0.5
0.4
0.3
0.4
0.2
1.0
2.2
4.2
3.0
4.6
3.1
4.3
2.5
7.1
3.2
11.0
10.3
16.0
14.6
70
78
+PPh3 0.79
Aldrich CeO2 + AIBNb 90 °C, 20 h, 1.2 MPa O2
–
–
–
1.13
0.58
0.91
20.0
15.3
25.6
10.2
7.6
15.3
11.7
18.0
7.4
12.8
5.3
4.1
7.0
2.9
5.9
10.0
4.3
0.8
1.1
0.5
0.9
0.8
1.9
0.1
0
0.4
0.2
0.5
3.2
1.5
2.2
5.2
1.4
2.7
3.9
2.3
3.4
2.5
0
0
0
0
10.6
10.3
8.6
9.5
18.0
8.9
107
107
106
TiO2 + AIBNb
90 °C, 20 h, 1.2 MPa O2
90 °C, 20 h, 1.2 MPa O2
ZSM-5 + AIBNb
13.2
0
2.0
Note: Decrease in conversions and selectivity to others after addition of PPh3 considered to be from dilution of products to below the GC detection limits and possible degradation.
a
10.0 g decane, 50 mg AIBN.
7.69 g decane, 30 mg AIBN.
C10-one = decanone, C10-ol = decanol and acid = carboxylic acid.
b
c

166
R. Lloyd et al. / Journal of Catalysis 283 (2011) 161–167
Table 4
Oxidation of n-decane using AIBN with and without catalyst.a
Initiator/catalyst
Conversion%
Selectivity%
C10 ketones
C10 internal alcohols
C10 terminal alcohols
C10 diones
C1–C9 acids
Others
Blank
AIBN
NanoCeO2
AIBN + nanoCeO2
AIBN + Aldrich CeO2
AIBN + TiO2
AIBN + ZSM-5
0.028
1.13
0.11
0.78
1.13
0.58
0.91
55.3
64.4
27.7
24.3
45.5
34.6
56.8
13.2
7.5
0
0
0
7.0
31.5
6.9
14.3
16.0
9.5
1.9
4.4
1.4
1.7
1.9
2.4
12.3
11.3
11.0
15.3
5.4
17.7
25.3
17.4
29.8
12.5
24.6
22.0
10.6
10.3
8.6
18.0
8.9
10.8
a
Decane (7.69 g), AIBN (30 mg), catalyst (100 mg), 20 h, 90 °C, 1.2 MPa O2.
Table 5
OOH
Epoxidation of 1-octene with catalyst and initiator.a
Initiator/catalyst
Conversion%
1,2-Epoxyoctane selectivity%
CHP
Au/nanoCeO2
AIBN
8.5
14.6
9.2
12
10
13
Homolysis
AIBN + Au/nanoCeO2
AIBN + nanoCeO2
OH
O
a
1-Octene (6.06 g), catalyst (100 mg) initiator (30 mg), 2 h, 90 °C, 1.2 MPa O2.
H2O +
HO•
+
alcohols:ketones ratio was increased, although not as much as
using nanoCeO2 and AIBN. Addition of TiO2 with AIBN was found
to have a lower conversion for the oxidation of decane than AIBN
alone, and the alcohols:ketones ratio was higher than using AIBN
alone but not as high as over nanoCeO2 and AIBN. ZSM-5 with AIBN
was found to give only a slightly lower conversion than AIBN alone
for the oxidation of decane, and the alcohols:ketones ratio was also
only slightly higher than over AIBN alone. The results imply that
the accessible surface area of the catalyst exerts an influence on
the product distribution, but that the exact nature of the surface
sites may not be critical. The decrease in conversion over Au/nan-
oCeO2, nanoCeO2 and TiO2 indicates that these materials are inter-
acting with the radicals formed. The change in the product
distribution observed suggests that this is selective for peroxy rad-
icals which are decomposed selectively to alcohols. Corma and co-
workers have previously described radical trapping with Au
[12,13] but the results from this study suggest that Au is not
needed.
R”=O
R’OH
Decomposition
R’OOH
R’•
R-H
O2
R-H
R’OO•
- R’OH
OOH
R’O• +
Fig. 5. Proposed reaction scheme for the oxidation of decane (R–H) with CHP.
R0 = C10H21 or PhC(Me)2, R00 = C10H20. Reduction of CHP by Au/nanoCeO2 also shown
in bold (thick arrow). Consecutive reactions to form diones and acids excluded from
scheme for simplicity.
The Au/nanoCeO2 catalyst was also tested for the less demand-
ing epoxidation of 1-octene, one of the reactions for which Corma
and co-workers [12] reported the catalyst to be active in the pres-
ence of AIBN. Similarly, our catalyst was found to have low activity
until AIBN was added (Table 5). The conversion observed over the
Au/nanoCeO2 with AIBN present was significantly higher than the
blank AIBN reaction, as previously reported [12]. The detected con-
version (14.6%) was, however, higher and the epoxide selectivity
(10.4%) lower than reported by Corma and co-workers under sim-
ilar conditions (conversion ꢂ 4%, selectivity ꢂ 50%). These varia-
tions may be caused by differences in the catalyst preparation
procedure, precursor materials and testing conditions being used.
The nanoCeO2 support was also tested for the epoxidation of 1-oc-
tene and was found in combination with AIBN, not to significantly
change the conversion compared to AIBN alone (Table 5). This lack
of selectivity could indicate that the AIBN-Au compounds formed
in Corma’s study are not replicated here.
Decane oxidation was also performed using the radical initiator
CHP with and without the Au/nanoCeO2 catalyst (Table 6). CHP
was found to be active for the oxidation of decane but not as active
as AIBN. The higher activity of AIBN is attributed to its greater effi-
ciency in producing radicals. AIBN produces more radicals for a
variety of reasons including: it more readily decomposes generat-
ing more radical species under the same conditions [34], and it is
an azo-compound that only decomposes homolytically giving
two C-centred radicals, while CHP can be decomposed by traces
of acids (to give no radicals) and by radical induced decomposition
(one radical gives rise to one radical) as well as by homolysis (giv-
ing 2 radicals); as shown in Fig. 4.
Table 6
Oxidation of decane using CHP with and without Au/nanoCeO2 catalyst.a
Catalyst/initiator
Conversion%
Selectivity%
C10 ketones
C10 internal alcohols
C10 terminal alcohols
C1–C9 acids
Others
CHP
0.15
0.036
24.8
76.5
6.1
0.5
0.2
0
2.5
6.5
66.4
16.5
CHP + Au/nanoCeO2
a
Decane (7.69 g), CHP (30 mg), Au/nanoCeO2 catalyst (100 mg), 2 h, 90 °C, 1.2 MPa O2.

R. Lloyd et al. / Journal of Catalysis 283 (2011) 161–167
167
nanoCeO2 does, however, slightly increase the alcohol selectivity.
Some of the peroxides produced by AIBN are thus considered to
be selectively reduced to alcohols over the nanoCeO2 catalysts.
The highly demanding direct oxidation of decane has been
shown to be possible using radical initiators, resulting in the for-
mation of range of oxygenated products. Ceria-based catalysts
have been found to be beneficial for the selective functionalisation
of hydrocarbons in specific cases. The combination of a gold sup-
ported on nano-ceria catalyst and AIBN has been shown to be ac-
tive for the oxidation of 1-octene, whereas a nano-ceria catalyst
and AIBN has been found to modify the selectivity for the oxidation
of decane.
CN
Me-C-N=N-C-Me
Me CN
Me
R”=O
R’OH
AIBN
Decomposition
- N2 Homolysis
nanoCeO2
R’OOH
R’•
O2
R-H
R’OO•
Fig. 6. Proposed reaction scheme for the oxidation of decane (R–H) with AIBN.
R0 = C10H21 or Me2C(CN), R00 = C10H20. Selective reduction of R0OOH to R0OH on
addition of nano ceria based catalysts also shown in bold (thick arrow). Consecutive
reactions to form diones and acids excluded from scheme for simplicity.
Acknowledgments
This work was sponsored by Johnson Matthey and Sasol. Dis-
cussion and input from Xavier Baucherel and Mike J. Watson (John-
son Matthey) and Jonathan Chetty and Cathy Dwyer (Sasol) are
gratefully acknowledged. Sarwat Iqbal helped with the AAS mea-
surement. Sivaram Pradhan calcined the ZSM-5 catalyst.
Addition of Au/nanoCeO2 with CHP was found to significantly
reduce the decane conversion, as compared to the reaction with
CHP alone. For both the CHP and Au/nanoCeO2 with CHP reactions,
the radical initiator was found to completely decompose during
the reaction. The main decomposition products of CHP from the
reaction without Au/nanoCeO2 were acetophenone (the ketone
product) and 2-phenyl-2-propanol (the alcohol product), in
References
[1] G. Centi, F. Cavani, F. Trifirò, Selective Oxidation by Heterogeneous Catalysis,
Kluwer Academic/Plenum, New York, London, 2001.
[2] I. Hermans, J. Peeters, P. Jacobs, Top. Catal. 48 (2008) 41–48.
[3] Anonymous, Chem. Week 165 (2003) 33.
[4] C.E. Frank, Chem. Rev. 46 (1950) 155–169.
[5] D.M. Brown, A. Fish, Proc. Roy. Soc. Lond. Ser. A 308 (1969) 547–568.
[6] K.U. Ingold, Chem. Rev. 61 (2002) 563–589.
approximately
a 50:50 ratio. However, for the CHP with
Au/nanoCeO2 reaction, the radical initiator was found to decom-
pose almost entirely into 2-phenyl-2-propanol. The Au/nanoCeO2
catalyst therefore selectively reduces CHP to the alcohol that is
consistent with previously reported findings [10]. This reduction
of CHP by Au/nanoCeO2 may prevent the formation of decane per-
oxides and thus explain the lower conversion observed (Fig. 5).
The ability of Au/nanoCeO2 to selectively reduce hydroperox-
ides to alcohols, as observed in the presence of CHP, may also ex-
plain the slight increase in the alcohols:ketones ratio observed
for the oxidation of decane with AIBN and Au/nanoCeO2 as com-
pared to AIBN alone (Table 2 and Fig. 6). However, Au/nanoCeO2
is clearly more efficient at selectively converting CHP than decane
peroxides under the conditions studied.
The radical initiators CHP and AIBN were found to be active for
the oxidation of decane, producing a range of oxygenated products.
AIBN was found to be more active under the conditions studied. At
higher temperature and pressure, in the presence of AIBN, a higher
conversion but lower apparent peroxide selectivities are observed;
more peroxides are thus proposed to have been produced but they
are considered to have shorter lifetimes. The change in ketone and
alcohol selectivities with AIBN under different conditions and com-
[7] G.H. Twigg, Chem. Eng. Sci. 3 (1954) 5–16.
[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, C.J. Kiely,
Nature 437 (2005) 1132–1135.
[9] J. Schwank, Gold Bull. 16 (1983) 103–110.
[10] T. Uchida, M. Wakakura, A. Miyake, T. Ogawa, J. Therm. Anal. Calorim. 93
(2008) 47–52.
[11] M. Álvaro, C. Aprile, A. Corma, B. Ferrer, H. García, J. Catal. 245 (2007) 249–252.
[12] C. Aprile, A. Corma, M.E. Domine, H. Garcia, C. Mitchell, J. Catal. 264 (2009) 44–
53.
[13] C. Aprile, M. Boronat, B. Ferrer, A. Corma, H. Garcia, J. Am. Chem. Soc. 128
(2006) 8388–8389.
[14] A. Goosen, C.W. McCleland, D.H. Morgan, J.S. O’Connell, A. Ramplin, J. Chem.
Soc., Perkin Trans. 1 (1993) 401–402.
[15] G.S. Hammond, C.-H.S. Wu, O.D. Trapp, J. Warkentin, R.T. Keys, J. Am. Chem.
Soc. 82 (1960) 5394–5399.
[16] C.-H.S. Wu, G.S. Hammond, J.M. Wright, J. Am. Chem. Soc. 82 (1960) 5386–
5394.
[17] G.S. Hammond, O.D. Trapp, R.T. Keys, D.L. Neff, J. Am. Chem. Soc. 81 (1959)
4878–4882.
[18] C. Walling, J. Polym. Sci. 14 (1954) 214–217.
[19] L.-X. Xu, C.-H. He, M.-Q. Zhu, K.-J. Wu, Y.-L. Lai, Catal. Lett. 118 (2007) 248–253.
[20] G. Lü, D. Ji, G. Qian, Y. Qi, X. Wang, J. Suo, Appl. Catal., A 280 (2005) 175–180.
[21] G. Lü, R. Zhao, G. Qian, Y. Qi, X. Wang, J. Suo, Catal. Lett. 97 (2004) 115–118.
[22] Y.-J. Xu, P. Landon, D. Enache, A.F. Carley, M.W. Roberts, G.J. Hutchings, Catal.
Lett. 101 (2005) 175–179.
[23] L.-X. Xu, C.-H. He, M.-Q. Zhu, S. Fang, Catal. Lett. 114 (2007) 202–205.
[24] R. Zhao, D. Ji, G. Lv, G. Qian, L. Yan, X. Wang, J. Suo, Chem. Commun. (2004)
904–905.
pared to using CHP may arise from the different decyl, a-cyanoiso-
propyl and cumene peroxy radical pathways (shown in Figs. 5 and
6) or from the various isomeric decyl or decylperoxy radical
pathways.
[25] K. Zhu, J. Hu, R. Richards, Catal. Lett. 100 (2005) 195–199.
[26] L. Chen, J. Hu, R. Richards, J. Am. Chem. Soc. 131 (2008) 914–915.
[27] B.P.C. Hereijgers, B.M. Weckhuysen, J. Catal. 270 (2010) 16–25.
[28] E.D. Mark, J. Chem. Technol. Biotechnol. 77 (2002) 43–50.
[29] J. Knaut, H.J. Richtler, J. Am. Oil Chem. Soc. 62 (1985) 317–327.
[30] C.F. Cullis, M.M. Hirschler, R.L. Rogers, Proc. Roy. Soc. Lond. Ser. A 375 (1981)
543–563.
[31] B. Zhan, B. Modén, J. Dakka, J.G. Santiesteban, E. Iglesia, J. Catal. 245 (2007)
316–325.
[32] G.B. Shul’pin, J. Mol. Catal. A: Chem. 189 (2002) 39–66.
[33] G.B. Shul’pin, D. Attanasio, L. Suber, J. Catal. 142 (1993) 147–152.
[34] J. Meijer, A. Hogt, Acros Org. Rev. (2006) 1–11.
4. Conclusions
The combination of AIBN and Au/nanoCeO2 was found to be ac-
tive for the oxidation of 1-octene as previously reported [12]. How-
ever, the combination of AIBN and Au/nanoCeO2 does not enhance
the conversion in the more challenging oxidation of n-decane com-
pared to AIBN alone. Therefore, either radicals from AIBN do not
complex to the ceria-based catalysts in the alkane medium or the
resulting complex is not active for the oxidation of the alkane.
The oxidation of decane in AIBN with either Au/nanoCeO2 or
[35] K. Weissermel, H.-J. Arpe, Industrial Organic Chemistry, WILEY-VCH,
Weinheim, 2003. ch. 7, pp. 174-176.