Selective Catalytic Oxidation of Cyclohexene with Molecular Oxygen: Radical Versus Nonradical Pathways

Article abstract of DOI:10.1002/cctc.201701538

We study the allylic oxidation of cyclohexene with O₂ under mild conditions in the presence of transition-metal catalysts. The catalysts comprise nanometric metal oxide particles supported on porous N-doped carbons (M/N:C, M=V, Cr, Fe, Co, Ni, Cu, Nb, Mo, W). Most of these metal oxides give only moderate conversions, and the majority of the products are over-oxidation products. Co/N:C and Cu/N:C, however, give 70–80 % conversion and 40–50 % selectivity to the ketone product, cyclohexene-2-one. Control experiments in which we used free-radical scavengers show that the oxidation follows the expected free-radical pathway in almost all cases. Surprisingly, the catalytic cycle in the presence of Cu/N:C does not involve free-radical species in solution. Optimisation of this catalyst gives >85 % conversion with >60 % selectivity to the allylic ketone at 70 °C and 10 bar O₂. We used SEM, X-ray photoelectron spectroscopy and XRD to show that the active particles have a cupric oxide/cuprous oxide core–shell structure, giving a high turnover frequency of approximately 1500 h^?1. We attribute the high performance of this Cu/N:C catalyst to a facile surface reaction between adsorbed cyclohexenyl hydroperoxide molecules and activated oxygen species.

Full text of DOI:10.1002/cctc.201701538

Accepted Article
Title: Selective catalytic oxidation of cyclohexene with molecular
oxygen: radical vs. non-radical pathways
Authors: Ilse M Denekamp, Martijn Antens, Thierry K Slot, and Gadi
Rothenberg
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201701538
Link to VoR: http://dx.doi.org/10.1002/cctc.201701538
A Journal of
www.chemcatchem.org

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
Selective catalytic oxidation of cyclohexene with molecular
oxygen: radical vs. non-radical pathways
Ilse M. Denekamp[a], Martijn Antens[a], Thierry K. Slot[a] and Gadi Rothenberg[a]
Abstract: We study the allylic oxidation of cyclohexene with dioxygen under mild conditions in the presence of transition-metal catalysts. The
catalysts are comprised of nanometric metal oxide particles supported on porous nitrogen-doped carbons (M/N:C, where M = V, Cr, Fe, Co, Ni,
Cu, Nb, Mo, and W). Most of these metal oxides give only moderate conversions, whereof the majority of the products are over-oxidation
products. Co/N:C and Cu/N:C, however, give 70–80% conversion and 40–50% selectivity to the ketone product, cyclohexene-2-one. Control
experiments using free-radical scanvengers shows that the oxidation follows the expected free-radical pathway in almost all cases. Surprisingly,
the catalytic cycle in the presence of Cu/N:C does not involve free-radical species in solution. Optimisation of this catalyst gives >85% conversion
with >60% selectivity to the allylic ketone, at 70 °C and 10 bar oxygen pressure. Scanning electron microscopy, X-ray photoelectron
spectroscopy, and X-ray diffraction studies show that the active particles have a cupric oxide/cuprous oxide core/shell structure, with a
correspondingly high TOF of ca. 1500 h–1. We attribute the high performance of this Cu/N:C catalyst to a facile surface reaction between
adsorbed cyclohexenyl hydroperoxide molecules and activated oxygen species.
Introduction
The allylic oxidation of alkenes is an important chemical reaction.
It allows one to keep the double bond while creating a new alcohol
or carbonyl function.[1,2] As such, it is useful across the board, from
bulk chemicals and agrochemicals,[3,4] all the way to fine-
chemicals and fragrances.[5–7] In theory, allylic oxidation is a
straightforward exothermic reaction. It requires only molecular
oxygen, a free, eco-friendly and widely available reagent. But
there is a trade-off: Oxygen has a high activation barrier due to its
resonance stabilisation.[8] Once this barrier is overcome, the
active oxygen species often react with hydrocarbons via free-
radical intermediates. These wreak havoc in solution, causing
side reactions that all-too-often lead to unwanted over-oxidation
products.[9–11]
symmetric molecule, similar to many starting compounds in
chemical synthesis. Second, it is itself industrially important,
participating in the synthesis cycles of key C6 chemicals such as
adipic acid and caprolactone.[13,14] Building on our preliminary
communication on alcohol oxidation,[12] we designed a set of
metal oxide catalysts supported on nitrogen-doped carbons. We
used no noble metals, focussing on abundant transition metal
oxides as catalysts.
Recently we showed that this problem can be solved for the
specific case of oxidising activated alcohols to aldehydes and
ketones, by using a bifunctional catalyst.[12] Yet this oxidation was
“only” a dehydrogenation reaction. It involved the transfer of
protons and electrons, without adding a new oxygen atom to the
substrate. Allylic oxidation is trickier, because it is an oxygenation,
involving the cleavage of at least one C–H bond and the creation
of new C–O or C=O bonds. This requires a direct interaction
between the substrate and an active oxygen species, which must
then be stopped at the allylic alcohol/ketone stage before “burning”
further to carboxylic acids, CO and CO2.
There are several reports on the allylic oxidation of cyclohexene
catalysed by abundant transition metals. In general, using
dioxygen as an oxidant requires additional activation, either by
adding 5–10% of a free-radical initiator such as H2O2 or by using
elevated temperatures. Tang et al.[15] reported a copper catalyst
on expanded graphite which gave 99% conversion and 65%
selectivity to 2-cyclohexene-1-one. Yin et al.[16] and Rossi et al.[17]
used cobalt-based catalysts, obtaining 94% conversion and 44%
selectivity and 90% conversion and 61% selectivity, respectively.
Elsewhere, Peng et al.[18] used metal-free nitrogen-doped carbon
nanotubes as catalysts. They tested 22 organic solvents and
found that acetonitrile gave the best results, with 60% conversion
and 39% selectivity to the 2-cyclohexene-1-one.
Here we examine the catalytic oxidation of cyclohexene with
molecular oxygen under mild conditions (eq 1). Cyclohexene is a
good model compound for two reasons: First, it is a small and
Our initial hypothesis was that the allylic oxidation reaction would
follow a pathway similar to alcohol oxidation, with oxygen
activation at the support surface followed by a reaction at the
oxide particle. Based on the reports in the literature, we expected
the reaction to involve free-radical intermediates.[10,17,19]
Surprisingly, we found that at least in one case, namely when
using copper oxide particles supported on nitrogen-doped carbon,
[a]
I. M. Denekamp, M. Antens, T. K. Slot and G. Rothenberg
Van ’t Hoff Institute for Molecular Sciences
University of Amsterdam
Science Park 904, Amsterdam 1098 XH (The Netherlands)
E-mail: g.rothenberg@uva.nl
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
catalyst gives a combined ketone+alcohol yield of nearly 45% with
a minimum TON > 1400 and TOF > 88 h–1 (the actual TON and
TOF per site are much higher, because most of the copper oxide
is not accessible, see discussion below). We therefore focussed
our investigation on these two catalysts.
there are no free-radicals in the solution. In this paper, we try and
resolve the different pathways leading to allylic oxidation, with the
goal of gaining a better understanding of this important reaction.
Table 1. Cyclohexene oxidation in the presence of different catalysts.a
Results and Discussion
Catalyst synthesis and testing. We began by preparing and
testing a set of nine d-block metal oxides supported on the same
batch of hierarchically porous nitrogen-doped carbons[20] (1.2
mmolg–1 M/N:C, where M = V, Cr, Fe, Co, Ni, Cu, Nb, Mo, and W).
The catalysts were prepared using vacuum-pore impregnation
(see experimental section for details). To this set, we added two
blanks: a clean N:C support and a carbon prepared from a citric
acid precursor (denoted Ccit) having a similar surface area as the
N:C (~1500 m2/g).The latter has the same surface characteristics
as the N:C sample but contains no nitrogen. All catalysts were
then tested in cyclohexene oxidation in an autoclave under 10 bar
O2 and 55 bar Ar, keeping within safe explosion limits. In a typical
reaction, each autoclave was charged with ca. 25 mmol of
cyclohexene, 10 mg of catalyst (a nominal substrate:metal oxide
ratio of 2000:1) and 15 mL of MeCN as solvent. Reactions were
stirred for 16 h at 1000 rpm and analysed by gas chromatography
(GC).
Selectivity (%)
Entry
Catalyst
Conversion (%)
A
B
C
D
other
1
2
None
Ccit
22
23
49
43
44
45
53
58
64
66
71
80
8
2
2
0
0
10
17
5
80
67
64
67
63
69
74
61
63
56
24
44
14
28
17
17
17
23
25
20
32
47
38
3
N:C
3
0
4
W@N:C
Ni@N:C
Mo@N:C
Fe@N:C
Nb@N:C
V@N:C
Cr@N:C
Cu@N:C
Co@N:C
4
0
12
17
11
3
5
3
0
6
5
0
7
3
0
Cyclohexene is oxidised to four main products (Table 1): 2-
cyclohexene-1-one A, cyclohexene oxide B, 2-cyclohexene-1-ol
C and 2-cyclohexene-1-hydroperoxide D (herein the ketone,
epoxide, alcohol and hydroperoxide, respectively). The rest of the
products were over-oxidation products, mainly CO and CO2.
Products A–C were determined directly from the GC analysis. The
hydroperoxide D could not be observed by GC, and was
quantified by reacting each sample with PPh3 (see experimental
section for details). Control experiments confirmed that the
internal standard, cyclohexane, showed no conversion at these
reaction conditions. Further, in the absence of any catalyst the
background reaction at 70 °C gives only 22% conversion, most of
it to CO and CO2 (Table 1, entries 1–3). Adding porous carbon
doesn’t change the conversion, but reduces the amount of over-
oxidation slightly, possibly owing to radical-scavenging by carbon
surface sites.[21] In the presence of pristine N:C, the conversion
more than doubles, to ca. 50%. Moreover, the selectivity to the
ketone A increases to 28%, at the expense of the hydroperoxide
D. Indeed, we showed recently that these porous N:C materials
are excellent oxygen reduction catalysts,[20] yet these results also
point to a N:C-catalysed route from D to A (vide infra).
8
6
0
8
9
15
7
0
2
10
11
12
1
4
9
16
6
4
6
6
[a] Reaction conditions: 10 bar O2; 2,5 mL (24,7 mmol) cyclohexene; 0,5 mL
(1,85 mmol) cyclohexane (IS); 10 mg catalyst; 15 mL MeCN; stirred in an
autoclave (1000 rpm); 70°C; 16 h.
Control reactions using equivalent amounts of cobalt oxide
supported on γ-alumina showed lower conversions and more
side-products, confirming the importance of the N:C support
(Table 2, entry 2). The copper oxide supported on γ-alumina does
show a good conversion, but lower selectivity and more side-
products, therefore making the γ-alumina-supported catalyst less
favourable (entry 6). Note that the difference in surface area
between the carbon and the γ-alumina was corrected for by
increasing the catalyst amount accordingly. To boost the number
of free-radicals at the start of the reaction, we added hydrogen
peroxide (entry 3, 13 mol% relative to substrate).[5,7] Hydrogen
peroxide can decompose into water and oxygen under these
reaction conditions. The water molecules themselves do not
change the conversion and selectivity (the H2O2 solution is
already 90% water), but the decomposition of H2O2 affects the
reaction by releasing free-radicals into the solution. Adding
hydrogen peroxide increased the conversion, but didn’t change
the selectivity significantly. A similar increase was observed when
the reaction was run at 80 °C. With copper oxide, however, adding
H2O2 or increasing the reaction temperature affected both the
conversion and the selectivity (entries 7 and 8). The conversion
increased to 85%, and the combined selectivity to A+C increased
to 70%. Importantly, this increase in selectivity came at the
expense of the over-oxidation products (unlike with cobalt, where
Adding W, Ni, Mo, Fe, or Nb doesn’t change the results
significantly (Table 1, entries 4–8). For some of these catalysts,
the selectivity to A is lower than that of the pristine N:C support,
which may reflect the blocking of labile sites on the support by
metal oxide particles. However, the catalysts containing V, Cr, Cu
and Co showed a significant increase in conversion (entries 9–
12). Vanadium oxide (V/N:C), which is known as a good
epoxidation catalyst,[22,23] indeed gives a higher selectivity to the
epoxide B. The remaining three catalysts are interesting: They are
the only ones that give measureable yields of the alcohol C. All
three give less hydroperoxide compared with the blanks,
indicating a pathway from D to C. Cobalt oxide gives the highest
conversion. But copper oxide gives the highest selectivity to the
ketone product A, with a remarkably low amount of over-oxidation
products. In fact, even at this unoptimised stage, the Cu/N:C
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
there were still a lot of over-oxidation products). To our mind, this
was counter-intuitive: one would assume that adding an initiator
such as H2O2 or increasing the temperature would lead to more
CO and CO2. These results led us to think that perhaps the
copper-oxide catalysed reaction is not a simple free-radical
process. Previous reports comparing the oxidation kinetics of
cyclohexene and cyclohexene-d10 show a clear primary isotope
effect (kH:kD = 8.2), indicating that the rate-determining step
involves C–H bond scission.[10] Moreover, adding a radical
scavenger quenched the reaction.[10,15] To check if this also
applies our system, we ran additional control experiments in the
presence of 6 mol% of 2,6-di-t-butyl-4-methylphenol (BHT, see
details in the experimental section). Adding BHT to the reaction
mixture containing the N:C support or the Co/N:C catalyst stopped
the reaction completely (Table 3, entries 2 and 4, cf. with entries
1 and 3). But when we added BHT to the Cu/N:C-catalysed
reaction, there was only a slight drop in conversion and selectivity
(from 85% to 76% and from 53% to 46%, respectively, see entries
5 and 6). This shows that while the reactions catalysed by metal-
free N:C and by Co/N:C are definitely free-radical processes, the
Cu/N:C-catalysed reaction is not affected by free-radical
scavengers (these experiments were repeated multiple times by
different people to ensure their repeatability and reproducibility).
Therefore, we conclude that in the copper-catalysed system,
there are no free-radicals in solution.
Table 3. Effects of adding free-radical scavengers.a
Selectivity (%)
Entry
Catalyst
Addition
Conversion (%)
A
B
C
other
1b
2 b,c
3
N:C
-
49
0
28
0
3
0
0
0
69
0
N:C
BHT
-
Co@N:C
Co@N:C
Cu@N:C
Cu@N:C
87
0
41
0
12
0
5
42
0
4c
5
BHT
-
0
85
76
53
46
10
10
17
20
20
24
6c
BHT
[a] Reaction conditions: 10 bar O2; 2,5 mL (24,7 mmol) cyclohexene; 0,5
mL (1,85 mmol) cyclohexane (IS); 10 mg carbon catalyst; 15 mL MeCN;
stirred in an autoclave (1000 rpm); 80 °C 16 h. [b] Reaction temperature
70°C. [c] Added 354 mg BHT ( 7mol% based on cyclohexene).
Table 2. Oxidation of cyclohexene with various copper oxide and cobalt oxide
catalysts.a
Selectivity (%)
Temp
(°C)
Conversion
(%)
Entry
Catalyst
A
B
C
D
other
1
2
Co@N:C
Co@Alu
Co@N:C
Co@N:C
Cu@N:C
Cu@Alu
Cu@N:C
Cu@N:C
70
70
70
80
70
70
70
80
80
56
87
87
71
77
86
85
38
20
43
41
47
31
61
53
6
3
6
0
6
3
44
74
40
42
24
60
16
20
3b
4
6
11
5
ndc
ndc
4
12
9
5
16
2
6
7
0
7b
8
15
10
8
ndc
ndc
17
[a] Reaction conditions: 10 bar O2; 2,5 mL (24,7 mmol) cyclohexene; 0,5 mL
(1,85 mmol) cyclohexane (IS); 10 mg carbon catalyst, 73 mg alumina catalyst;
15 mL MeCN; stirred in an autoclave (1000 rpm); 16 h. [b] Added 1,0 mL H2O2
(10 wt%, 3.3 mmol, 13 mol% based on cyclohexene). [c] Not determined.
Figure 1. Scanning electron micrograph of Cu/N:C at ×15.000
magnification (an image with particle measurements is included
in the Supporting Information).
X-ray photoelectron spectroscopy (XPS, Figure 2) showed that
impregnating the N:C surface with copper oxide doesn’t affect the
nitrogen binding energy. This suggests that the copper oxide is
not coordinated to surface nitrogen atoms. The impregnation
does increase the intensity of the O 1s peak, indicating a higher
oxygen content in the sample. For copper, the XPS shows the
typical Cu 2p 1/2 and 2p 3/2 peaks, which can pertain to both Cu+
and Cu2+. Yet the characteristic CuO peak at 945 is absent,
supporting the presence of Cu2O,[24] (metallic copper is unlikely at
such low treatment temperatures[25] and considering the increase
in the oxygen signal). The carbon peak is not affected by
impregnation with copper. However, powder X-ray powder
diffraction (XRD) measurements of the catalyst show CuO as the
major component in the particles (see details in the supporting
information). These results are consistent with a CuO-CuO2 core-
shell structure, as XRD measures the entire particle while XPS
penetrates only a few atomic layers.[26] We therefore suggest that
Scanning electron microscopy (SEM, Figure 1) of the Cu/N:C
catalyst showed copper oxide spherical particles, ca. 200–250 nm
in diameter, distributed evenly across the surface. Unlike the
support, the particles are non-porous. This means that most of the
copper oxide is “inside” the particle and unavailable for catalysis.
If we consider that the active outer shell is about five atomic layers
(roughly 2 nm in thickness), the actual active catalyst comprises
only 0.1 wt%. Accordingly, the actual TON of this catalyst would
be > 24000, with a corresponding TOF > 1500 h–1.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
during the thermal treatment following the impregnation step, the
adsorbed Cu(NO3)2 precursor is first converted to CuO and
NO2(g). As the temperature approaches the final temperature of
300 °C, the Cu2O shell starts forming. Indeed, temperature
programed reduction measurements indicate the presence of
multiple copper oxides (details in Supporting Information).
Similarly, thermogravimetric analysis of the pristine N:C and the
Cu/N:C samples shows that the latter decomposes at a lower
temperature (400 °C vs 500 °C, respectively. This supports the
hypothesis that copper partially oxidises the surface, creating
more labile sites (see details in the supporting information).
disproportionation reaction with another cyclohexene molecule,
giving two molecules of cyclohexene-1-ol
C (Figure 3d).
Compared with the other metal oxides, the scission of the RO–
OH bond on the copper oxide surface is apparently much slower.
This means that fewer free-radicals are released into the solution,
giving enough time for the rearrangement and disproportionation
reactions.
Interestingly, there is a marked difference between the oxidation
of activated alcohols, which we reported earlier,[12] and that of
cyclohexene. With an activated alcohol substrate such as
cinnamyl alcohol, the N:C support is required for oxygen
activation. There, no reaction was observed for copper oxide
particles supported on Ccit, an analogous porous carbon with no
nitrogen dopants. Cyclohexene oxidation, however, does proceed
in the presence of Cu/Ccit, showing that the allylic oxidation in this
case is easier. This is supported by the results of Gray and co-
workers,[10] who showed that the allylic C–H bond scission is rate-
determining, and by the fact that this bond is weaker than the
alcohol C–H bond (83 kcalmol–1 and 96 kcalmol–1,
respectively[29,30]).
Figure 2. XP spectra of the Cu/N:C and pristine N:C samples
showing the oxygen 1s, Cu 1/2p and 3/2p, nitrogen 1s and carbon
1s binding energies. Impregnation of copper oxide on the N:C
increases the oxygen content but does not affect the nitrogen or
carbon peaks. Note that the nitrogen and carbon graphs are
normalised for clarity.
Mechanistic considerations. Based on the above results, we
propose two alternatives for the catalytic allylic oxidation of
cyclohexene with molecular oxygen. The first follows the
traditional free-radical route, and pertains to the Co, Fe, Cr, Mo,
V, Ni, Nb and W/N:C catalysts. Here, oxygen is either activated
thermally or in a redox process on the N:C surface. Insertion of
this activated oxygen into the allylic C–H bond gives the
cyclohexenyl hydroperoxide D. This can then either rearrange to
give the ketone A or undergo scission, giving oxo and peroxo
radicals that propagate a chain oxidation reaction.[2,15] Accordingly,
this pathway, which involves free-radicals in the solution bulk, is
readily quenched when BHT is added.
Figure 3. Proposed reaction pathways for the catalytic oxidation
of cyclohexene with molecular oxygen in the presence of Cu/N:C.
(a) Oxygen activation at the support surface followed by radical
migration into solution. (b) Insertion of activated oxygen into the
allylic C–H bond to give the adsorbed hydroperoxide D followed
by either (c) rearrangement to the ketone A and water or (d)
reaction with another cyclohexene molecule to give two
molecules of the alcohol C.
In all cases, the epoxide B probably forms via another
pathway.[4,31] Cyclohexene molecules can interact with M=O
groups on the particles’ surface, giving the epoxide and a labile
surface site which is then re-oxidised by incoming oxygen.[24]
Conversely, in the presence of Cu/N:C, there are no free-radicals
in the solution bulk. Oxygen can still be activated at the N:C
surface sites, but now there are two options: those small amounts
of short-lived activated oxygen species (e.g. O2–radical anions)
that do travel into solution will be quenched by BHT (Figure 3a, cf.
also the difference in conversion between entries 5 and 6 in Table
3). The BHT molecules are too bulky to enter the micropores.
Therefore, they will quench only the radicals in the solution.
Conversely, the activated oxygen species that are close enough
to diffuse to a supported copper oxide particle,[27] can react there
with cyclohexene, forming an adsorbed hydroperoxide (Figure 3b).
This adsorbed hydroperoxide can undergo two reactions. The first
is rearrangement and dehydration, giving the ketone A and a
Conclusions
The catalytic oxidation of cyclohexene with molecular oxygen can
follow different pathways depending on the type of catalyst. In the
presence of transition-metal oxide nanoparticles supported on
nitrogen-doped carbons, the key step is the insertion of the
dioxygen molecule into the allylic C–H bond, giving the
cyclohexenyl hydroperoxide. This reaction can be enhanced by
molecule of water (Figure 3c).[9,28] The second is
a
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
at 700 °C), and cooled down to room temperature. The different M/N:C
catalysts were prepared from their respective precursors salts:
Co(NO3)2·6H2O, Cu(NO3)2·3H2O, Fe(NO3)3·9H2O, NH4VO3 (dissolved
oxygen activation at the nitrogen-doped carbon surface. In most
cases, the allylic oxidation follows a free-radical pathway.
However, in the presence of Cu/N:C the reaction does not release
free-radicals into solution. This enables a more selective reaction
at the copper oxide surface, which probably involves cuprous
oxide sites.
using
2 equiv of oxalic acid), Cr(NO3)3·9H2O, (NH4)6Mo7O24·4H2O,
Ni(NO3)2·6H2O, C10H5NbO20·xH2O and (NH4)10W12O41·5H2O. The
Co/alumina catalyst sample was similarly prepared starting from (γ-Al2O3,
Ketjen; ground and sieved to 200–400 nm) and using 0.8 mL of 0.55 M of
Co(NO3)2·6H2O stock solution.
Procedure for catalytic oxidation of cyclohexene
Experimental Section
This is a modification of the procedure published by Cao et al.[32] A 75 mL
autoclave lined with a 50 mL Teflon insert was loaded with cyclohexene
(2.5 mL, 24.7 mmol), cyclohexane (0.5 mL, internal standard), acetonitrile
(solvent, 15 mL), catalyst (10 mg M/N:C carbon or 73 mg Co/alumina) and
a stirring bar (30 mm). The autoclave was sealed, flushed with argon and
oxygen twice before applying the final oxygen atmosphere (10 bar) and
argon (55 bar). The autoclave was then heated to 70 °C for 16 h while
stirring at 1000 rpm. After 16 h, the autoclave was cooled down to ambient
temperature. Acetone (5 mL) was added with the sample and the reaction
mixture was filtered using 0.45 μm PTFE syringe filters and analysed by
GC.
Materials and Instrumentation
Gas chromatography analyses were performed on a PerkinElmer Clarus
580 instrument. This system was equipped with a flame ionization detector
and autosampler (G4513A). A 30 m x 32 mm I.D. Rxi-5ms fused silica,
crossbond diphenyl dimethyl polysiloxane column with a film thickness of
0.25 µm was used. For the GC, the injector volume was 1 μl and the flow
was 100 mL/min of He carrier gas. The temperature program was 40 °C,
20 °C/min, 160 °C for 2 min. SEM measurements were performed on a
Verios-460 microscope (FEI) using an accelerating voltage of 5 kV, with a
working distance of 2–5 mm. Powder XRD patterns were obtained with a
MiniFlex II diffractometer using a Ni-filtered CuKα radiation. The X-ray tube
was operated at 30 kV and 15 mA, with a 0.01° step and 1s dwell time.
XPS measurements were performed with a PHI VersaProbe II scanning
XPS microprobe (Physical Instruments AG, Germany), using a
monochromatic AlKα X-ray source with a power of 24.8 W and a beam
size of 100 μm. The spherical capacitor analyser was set at 45° take-off
angle with respect to the sample surface. The pass energy was 46.95 eV,
yielding a full width at half maximum of 0.91 eV for the Ag 3d5/2 peak.
Peaks were calibrated using the C-1s position. Curve fitting was performed
using the XPSPeak 4.1 software package. All chemicals were obtained
from commercial sources (>99% pure) and were used as received.
Temperature programmed reduction (TPR) studies were done by placing
25 mg of sample sandwiched between two quartz wool plugs in a quartz
Testing for the presence of free-radicals in solution was done by adding
2,6-di-t-butyl-4-methylphenol (BHT, 354 mg, 6 mol%) to the reaction at
time zero and then following the above procedure. Reactions were
performed in triplicate, and all GC samples were taken in duplicate.
Procedure for quantifying the cyclohexenyl hydroperoxide D.
The hydroperoxide D cannot be measured directly by gas chromatography
due to its instability. Instead, we quantified it by comparing a control
reaction sample to one where triphenylphosphine (PPh3, 30 mg, 1mol%)
was added. The sample was shaken for 1 min, generating heat in the
process, as the PPh3 reacts with the hydroperoxide D to give PPh3O and
the alcohol C (eq 2). After this reaction, the sample was analysed by GC
and compared to its untreated counterpart. Subtracting the initial amount
of the alcohol formed in the control reaction from the amount of the alcohol
after adding PPh3 gives the amount of hydroperoxide D in the original
sample.[6,33]
tube reactor (4 mm i.d.). After purging with nitrogen,
hydrogen in nitrogen mixture was applied, and the system was allowed
to equilibrate and then heated with at 5 °Cmin−1 to 800 °C (no hold time).
a flow of 5%
Procedure for preparing the nitrogen-doped carbon support
The N:C support samples were prepared following the procedure
published by Eisenberg et al.[20] Briefly, nitrilotriacetic acid (N(CH2COOH)3,
NTA) was mixed at
a
1:1 ratio with magnesium carbonate
((MgCO3)4Mg(OH)2). This was dissolved in de-ionized water, stirred for 10
min at 85 °C, and cooled down to room temperature. The solid was then
precipitated by adding an excess of ethanol and chilling in an ice bath for
2 h. The white solid was scraped out and dried at 40 °C for 48 h and
grinded to a fine white powder. This powder was then pyrolysed in argon
at 900 °C. The MgO particles were washed out with 3 × 500 mL of 0,5 M
citric acid. The resulting crude N:C sample was dried at 120 °C for 2 h and
heat treated under argon at 1000 °C for 1 h.
Acknowledgements
Procedure for preparing M/N:C catalysts
We thank P. Mettraux and Prof. N. Setter (EPFL, Switzerland) for
the XPS measurements and Dr. D. Eisenberg for the SEM
measurements. This work is part of the Research Priority Area
Sustainable Chemistry of the UvA, http://suschem.uva.nl.
This is a modification of the procedure published by Slot et al.[12] The
nitrogen-doped carbon support (100 mg) was placed in a small vial with a
septum. The air was removed carefully using a needle and an aqueous
solution of the desired metal precursor salt (0,2 mL, corresponding to a
nominal loading of 1 mmol/m2) was added to the vial under continuous
stirring. The vial was shaken vigorously for 2–3 min, creating a uniform
solid paste, which was then dried at 85 °C for 12 h. Each catalyst was then
heat-treated at 300 °C under argon (except for Nb/N:C which was treated
Keywords: heterogeneous catalysis • autoxidation • sustainable
chemistry • nanoparticles • nitrogen-doped carbons
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
[1]
[2]
[3]
J. Hao, X. Jiao, L. Han, Q. Suo, A. Ma, J. Liu, X. Lian, L.
Zhang, RSC Adv. 2015, 5, 13809–13817.
[27] M. Bowker, R. A. Hadden, H. Houghton, J. N. K. Hyland,
K. C. Waugh, J. Catal. 1988, 109, 263–273.
K. C. Gupta, A. Kumar Sutar, C.C. Lin, Coord. Chem.
Rev. 2009, 253, 1926–1946.
[28] S. Mukherjee, S. Samanta, A. Bhaumik, B.C. Ray, Appl.
Catal. B Environ. 2006, 68, 12–20.
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, et al., Nature 2005, 437, 1132–1135.
[29] S. J. Blanksby, G. B. Ellison, Acc. Chem. Res. 2003, 36,
255–263.
[30] D. F. McMillen, and D. M. Golden, Ann. Rev. Phys.
Chem. 1982, 33, 493–532.
[4]
[5]
S. M. Islam, A. S. Roy, P. Mondal, N. Salam, J. Inorg.
Organomet. Polym. Mater. 2012, 22, 717–730.
[31] Y. Tsuda, S. Matsui, K. Takahashi, J. Mol. Catal. Chem.
1999, 148, 183–187.
G. Smeureanu, A. Aggarwal, C. E. Soll, J. Arijeloye, E.
Malave, C. M. Drain, Chem. Eur. J. 2009, 15, 12133–
12140.
[32] Y. Cao, H. Yu, F. Peng, H. Wang, ACS Catal. 2014, 4,
1617–1625.
[6]
[7]
[8]
[9]
Y. N. Wei, H. Li, F. Yue, Q. Xu, J. D. Wang, Y. Zhang,
RSC Adv. 2016, 6, 107104–107108.
[33] R. Hiatt, R. J. Smythe, C. McColeman, Can. J. Chem.
1971, 49, 1707–1711.
J. Tong, W. Wang, L. Su, Q. Li, F. Liu, W. Ma, Z. Lei, L.
Bo, Catal. Sci. Technol. 2017, 7, 222–230.
W. T. Borden, R. Hoffmann, T. Stuyver, B. Chen, J. Am.
Chem. Soc. 2017, 139, 9010–9018.
S. M. Mahajani, M. M. Sharma, T. Sridhar, Chem. Eng.
Sci. 1999, 54, 3967–3976.
[10] E. R. Birnbaum, M. W. Grinstaff, J. A. Labinger, J. E.
Bercaw, H. B. Gray, J. Mol. Catal. Chem. 1995, 104,
L119–L122.
[11] T. Liu, H. Cheng, W. Lin, C. Zhang, Y. Yu, F. Zhao,
Catalysts 2016, 6, 24.
[12] T. K. Slot, D. Eisenberg, D. van Noordenne, P.
Jungbacker, G. Rothenberg, Chem. Eur. J. 2016, 22,
12307–12311.
[13] K. Sato, M. Aoki, R. Noyori, Science 1998, 281, 1646–
1647.
[14] I. Nowak, M. Ziolek, Microporous Mesoporous Mater.
2005, 78, 281–288.
[15] R. Zhang, R. Tang, J. Mater. Sci. 2016, 51, 5802–5810.
[16] C. Yin, Z. Yang, B. Li, F. Zhang, J. Wang, E. Ou, Catal.
Lett. 2009, 131, 440–443.
[17] F. P. da Silva, R. V. Gonçalves, L. M. Rossi, J. Mol. Catal.
Chem. 2017, 426, 534–541.
[18] Y. Cao, H. Yu, H. Wang, F. Peng, Catal. Commun. 2017,
88, 99–103.
[19] V. S. Jeyaraj, M. Kamaraj, V. Subramanian, J. Phys.
Chem. C 2015, 119, 26438–26450.
[20] D. Eisenberg, W. Stroek, N. J. Geels, C. S. Sandu, A.
Heller, N. Yan, G. Rothenberg, Chem. Eur. J. 2016, 22,
501–505.
[21] E. V. Ramos-Fernandez, N. J. Geels, N. R. Shiju, G.
Rothenberg, Green Chem 2014, 16, 3358–3363.
[22] T. Itoh, K. Jitsukawa, K. Kaneda, S. Teranishi, J. Am.
Chem. Soc. 1979, 101, 159–169.
[23] E. D. Mihelich, Tetrahedron Lett. 1979, 20, 4729–4732.
[24] D. Tahir, S. Tougaard, J. Phys. Condens. Matter 2012,
24, 175002.
[25] S.K. Ryu, W.K. Lee, S.J. Park, Carbon Lett. 2004, 5, 180–
185.
[26] P. J. Cumpson, Appl. Surf. Sci. 1999, 144, 16–20.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
It’s radical-free: the allylic oxidation
of cyclohexene with dioxygen under
mild conditions is catalysed by copper
nanoparticles supported on nitrogen-
doped carbon, giving >60% selectivity
to the ketone at >85% conversion.
Unlike other transition metals, the
Cu/N:C catalyst system does not
involve free-radicals in solution.
Ilse M. Denekamp, Martijn Antens,
Thierry K. Slot and Gadi Rothenberg*
Page No. – Page No.
Title
Selective catalytic oxidation of
cyclohexene with molecular
oxygen: radical vs. non-radical
pathways.
Layout 2:
FULL PAPER
Page No. – Page No.
((Insert TOC Graphic here; max. width: 11.5 cm; max. height: 2.5 cm))
Title
Text for Table of Contents
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
Supporting Information for Selective catalytic oxidation of cyclohexene
with molecular oxygen: radical vs. non-radical pathways.
Ilse M. Denekamp, Martijn Antens, Thierry K. Slot and Gadi Rothenberg*
Van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Science Park 904, Amsterdam 1098 XH (The Netherlands)
E-mail: g.rothenberg@uva.nl
Figure S1. Scanning electron micrographs of Cu/N:C. The size of the copper oxide is ca. 200-250 nm
Scanning electron microscopy (SEM) of Cu/N:C catalyst with particle measurements. The average size
of the copper oxide spherical particles is ca. 200-250 nm. Furthermore, the clusters are evenly
distributed onto the support surface.
Figure S2. Temperature programmed reduction (TPR) spectrum for Cu/N:C.
The temperature programmed reduction (TPR) shows multiple copper oxide species presence inside the
copper oxide catalyst. At 300 °C copper oxide is reduced, where multiple peaks indicate multiple
copper oxide species. At 600 °C the support N:C is reduced. [1]
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701538
ChemCatChem
FULL PAPER
Figure S3. X-ray powder diffraction (XRD) spectrum for Cu/N:C.
X-ray diffraction showed a broad peak at 25° which is typical for carbon. The small peaks at 36.1° (-
111) and at 39° (111) are typical for CuO.[2]
Figure S4. Thermal gravimetric analysis (TGA) curves of Cu/N:C and plain N:C.
TGA showed the mass loss of N:C at 500°C, whereas the Cu/N:C has a mass loss at a lower temperature.
This difference indicates that the copper oxidises the nitrogen-doped surface during the treatment, thereby
lowering the temperature of the mass loss. The difference in weight percentage at the end of the
measurement around 600 °C is the copper oxide that is present in the Cu/N:C sample.
References
[1] T. K. Slot, D. Eisenberg, D. van Noordenne, P. Jungbacker, G. Rothenberg, Chem. Eur. J. 2016, 22, 12307–12311.
[2] J. Morales, L. Sánchez, F. Martín, J. R. Ramos-Barrado, M. Sánchez, Electrochim. Acta 2004, 49, 4589–4597.
This article is protected by copyright. All rights reserved.