Magnetically recoverable copper oxide catalysts for aerobic allylic oxidation of cyclohexene

Article abstract of DOI:10.1016/j.molcata.2016.07.044

Magnetically recoverable copper oxide catalysts prepared by sol-immobilization method exhibited interesting properties for the allylic oxidation of cyclohexene with molecular oxygen as the sole oxidant. The catalysts were prepared by immobilization of pre-synthesized PVA (polyvinyl alcohol)-stabilized Cu₂O nanoparticles (NPs) on a magnetically recoverable support; the catalyst was further oxidized to CuO NPs after calcination at 600?°C. Both catalysts can selectively oxidize cyclohexene through allylic oxidation to give 2-cyclohexene-1-one as the main product, but CuO was identified as the most active species providing 90% cyclohexene conversion and 96% selectivity for allylic products under 100?°C and 4?bar pressure of O₂ for 6?h of reaction time. The catalysts were magnetically recovered without metal leaching and could be reused in at least six consecutive runs.

Full text of DOI:10.1016/j.molcata.2016.07.044

Accepted Manuscript
Title: Magnetically recoverable copper oxide catalysts for
aerobic allylic oxidation of cyclohexene
Author: Fernanda Parra da Silva Renato V. Gon c¸ alves Liane
M. Rossi
PII:
S1381-1169(16)30294-1
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.07.044
MOLCAA 9978
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
24-6-2016
16-7-2016
25-7-2016
Please cite this article as: Fernanda Parra da Silva, Renato V.Gon c¸ alves,
Liane M.Rossi, Magnetically recoverable copper oxide catalysts for aerobic
allylic oxidation of cyclohexene, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.07.044
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Magnetically recoverable copper oxide catalysts for aerobic allylic oxidation of
cyclohexene
a
b
a
Fernanda Parra da Silva, Renato V. Gonçalves, and Liane M. Rossi,*
aDepartamento de Química Fundamental, Instituto de Química, Universidade de São
Paulo, Av. Prof. Lineu Prestes, 748, 05508-000, São Paulo, SP, Brazil E-mail:
lrossi@iq.usp.br
b
Instituto de Física de São Carlos, Universidade de São Paulo, CP 369, 13560-970, São
Carlos, SP, Brazil
Graphical abstract
Highlights



PVA-stabilized copper oxide NPs were used to obtain heterogeneous catalysts
Aerobic oxidation of cyclohexene into α,β-unsaturated derivatives
Effecient catalyst recovery and negligible metal leaching
Abstract
Magnetically recoverable copper oxide catalysts prepared by sol-immobilization method
exhibited interesting properties for the allylic oxidation of cyclohexene with molecular
oxygen as the sole oxidant. The catalysts were prepared by immobilization of pre-
synthesized PVA (polyvinyl alcohol)-stabilized Cu O nanoparticles (NPs) on a
2
magnetically recoverable support; the catalyst was further oxidized to CuO NPs after
1

o
calcination at 600 C. Both catalysts can selectively oxidize cyclohexene through allylic
oxidation to give 2-cyclohexene-1-one as the main product, but CuO was identified as the
most active species providing 90% cyclohexene conversion and 96% selectivity for
allylic products under 100 °C and 4 bar pressure of O for 6 h of reaction time. The
2
catalysts were magnetically recovered without metal leaching and could be reused in at
least six consecutive runs.
Keywords Magnetic separation; Allylic oxidation; Cyclohexene; 2-cyclohexene-1-one; Copper
oxide .
1
. Introduction
The functionalization of allylic C–H bonds of alkenes represents a powerful approach to
the synthesis of α,β-unsaturated derivatives, which find widespread application as
synthetic intermediates and as products in the commodity chemical, fine chemical, and
pharmaceutical industries [1-4]. The allylic oxidation of alkenes is a valuable organic
transformations for the synthesis of value-added products such as α,β-unsaturated
alcohols, esters and ketones from simple precursors[4]. Classically, these oxidations were
performed using chromium reagents (first stoichiometric or then catalytic with the aid of
oxidants), but due to the radical character, metals that undergo one-electron redox
processes, such as Mn, Rh, Fe, Cu or Co, have been used in the development of
environmentally friendly metal-catalyzed methods [4-8]. Many of these metal-catalyzed
systems are not compatible with the use of molecular oxygen and require unfriendly
oxidants, such as PhIO, NaClO, H
2
2
O , tBuOOH; to achieve catalytic turnover [9-15].
Molecular oxygen should be preferred as an ideal oxidant due to its natural, highly
economical, environmentally benign and abundant characters [16, 17]. Limitations in
2

reaction scope and selectivity as well as safety concerns have hindered the application of
aerobic oxidations in the fine chemical, pharmaceutical, and related industries.
An ideal clean oxidation may use molecular oxygen as the primary oxidant along with
catalysts that can be recycled and non-toxic solvents. Yin et al. [18] and Silva et al. [19]
reported the allylic oxidation of alkenes using molecular oxygen as the sole oxidant with
easily recoverable cobalt catalysts. Copper catalysis also received attention for selective
aerobic oxidative functionalization of C-H bonds [20-22]. The facile aerobic oxidation of
Cu(I) ions to Cu(II) is widely recognized, and a number of important synthetic Cu-
catalyzed aerobic oxidation reactions exist, including industrial applications, with high
catalytic activities and a wide functional groups tolerance.
Herein we explored the preparation of supported copper oxide nanoparticles (NPs) as
catalyst for the allylic oxidation of alkenes with molecular oxygen as the sole oxidant.
Many catalytic processes with copper makes use of metal complexes or enzyme mimetics
[
23-25]; however, due to its excellent conductivity, availability and low cost, the use of
copper as an heterogeneous catalyst receives attention. Metallic copper nanoparticles and
cuprous oxide have been used as catalysts for various reactions, including Ullmann
coupling [26-28], cycloaddition of terminal alkynes and azides [29, 30], hydrogenation
of alkynes [31, 32], oxidative dehydrogenation of alcohols [33], degradation of pollutants
[
34] and oxidation of olefins [35]. Copper tetramers supported on alumina showed
excellent activity in carbon dioxide reduction into methanol at low partial pressure of CO
36]. Allylic oxidation of cyclohexene by copper supported on expanded graphite [22] or
2
[
Cu-MOFs [37] exemplifies the chemistry of copper applied to allylic C-H bond
functionalization.
2
. Materials and methods
3

2
.1. Catalyst preparation
The magnetically recoverable copper catalyst was synthesized by a sol-immobilization
method. Initially the magnetic support, which is comprise of silica-coated magnetite
nanoparticles (Fe
functionalization using (3-aminopropyl)triethoxysilane (APTES) to obtain NH
the surface of the support [19, 38]. Copper oxide nanoparticles were prepared by
reduction of an aqueous solution of copper(II) nitrate (Cu(NO .3H O, Sigma Aldrich)
3
O
4
@SiO
2
), was prepared as previously reported, followed by its
2
groups in
3
)
2
2
-1
in the presence of polyvinyl alcohol (PVA, MW=9000-10000 g.mol , Sigma Aldrich),
according to the literature.[39] To an aqueous solution of copper(II) ions (200 mL, 1.2
-1
mmol L ) was added an aqueous solution of PVA (2.25 mL, [Cu:PVA] = 1:1.5) under
vigorous stirring. Then, a freshly prepared and cold aqueous solution of sodium
-1
borohidrate (12 mL, 0.1 mol L , [Cu:NaBH ]=1:5) was dropwise added. The resulting
4
solution was stirred for 30 minutes and then 1 g of the functionalized support was added,
and the solution was stirred for another 2 h. The solid was magnetically separated using
an external magnet and washed several times with distilled water and acetone, and dried
under vacuum. Part of this sample was used “as-prepared” and part was calcined at 600
oC for 2 h in air before the catalytic studies.
2
.2. Catalytic oxidation experiments
The catalytic reactions were carried out in a modified Fisher-Porter glass reactor (50 mL)
pressurized at the desired O pressure. For each experiment, the glass reactor was loaded
2
with the catalyst (0.04 mmol Cu) and 20 mmol of cyclohexene under inert atmosphere.
The chosen temperature was maintained by an oil bath in a hot-stirring plate connected
to a digital controller (ETS-D4 IKA). The reactor was loaded with O at the desired
2
pressure and kept connected to an oxygen supply reservoir (constant pressure during the
catalytic experiment). The reaction was conducted under magnetic stirring (700 rpm) and
4

after the desired time, the catalyst was magnetically recovered by placing a permanent
magnet in the reactor wall and the products were collected and analyzed by GC, using a
Shimadzu GC-2010 gas chromatograph system with a 30-m RTX-Wax column and
helium as carrier gas. The analyses were performed using the following conditions: initial
-
1
-
temperature 40ºC, rate 10ºC.min , final temperature 200ºC and gas flow of 1.41 mL.min
1.
Conversion and selectivity were calculated based on peak areas of GC calibrated with
an internal standard (biphenyl) added to the organic phase after separation of the catalyst.
The validity of the identification and quantitation of the GC peaks was verified by
comparison of the mass spectra and gas chromatographic retention times with those of
the authentic compounds.
2
.3. Characterization methods
X-ray powder diffraction (XRD) patterns were recorded on a Rigaku Miniflex with Cu
Kα radiation (λ=1.54 Å) at 2θ = 25–80° with a 0.02° step size.
Transmission electron microscopy (TEM) and EDS analysis were performed with the
JEM-2010 microscope operating at 200 kV. Samples for TEM observations were
prepared by placing a drop containing the nanoparticles in a carbon-coated grid.
The quantification of Cu was performed via flame atomic absorption spectroscopy
(
FAAS) using a Shimadzu AA-6300 apparatus.
X-ray absorption near-edge structure (XANES) and extended X-ray absorption
fine structure (EXAFS) measurements were performed at the Cu K-edge in the
transmission mode (channel-cut Si (111) crystal monochromator and three
ionization chambers to detect incident and transmitted photon fluxes) performed at
XAFS beamline of the Brazilian Synchrotron Light Laboratory (LNLS, Campinas,
5

SP) at proposal number XAFS1- 16100. During data collection, the XAS spectrum
of Cu metal foil was simultaneously measured and the energy was calibrated by
aligning the respective absorption edges. All data processing was carried out using
IFEFFIT [40, 41].
The X-ray photoelectron spectroscopy (XPS) experiments were carried out at SXS
beamline of the Brazilian Synchrotron Light Laboratory (LNLS, Campinas, SP) at
proposal number SXS-15172. The XPS spectra were collected using an incident
photon energy of 1840 eV. The operating pressure in the ultrahigh vacuum
–
9
chamber (UHV) during the analysis was 1 × 10 Pa. Energy steps was of 20 eV
with 0.2 eV step energy and 200 ms per point acquisition time. The peak fitting
was performed using a 70% Gaussian type curve and a 30% Lorentzian type curve,
and a shirller nonlinear sigmoid type baseline. The C 1s peak of adventitious
carbon was fixed at 284.8 eV to set the binding energy scale, and the data fitting
was performed using CasaXPS software (Casa Software Ltd., UK).
3
. Results and Discussion
PVA-stabilized copper oxide nanoparticles prepared by a well-known aqueous phase
protocol for the preparation of a many others metal nanoparticles (Pd, Au, Rh, etc.) were
used in this study as component for the preparation of supported catalysts. The
immobilization of the metal nanoparticles in high surface area inorganic solids increases
its stability and recyclability. Furthermore, if the metal nanoparticles are immobilized on
a magnetic support, the handling and separation is tremendously facilitated [42]. Here we
choose a support developed in our research group that is comprised by magnetite
nanoparticles spherically coated with silica (final average size 35 nm) obtained by a
reverse microemulsion method [38, 43]. The magnetic properties of this material were
6

extensively investigated and are adequate for the proposed application due to its
superparamagnetic behavior and high saturation magnetization [43]. The support surface
was functionalized with amino groups in order to facilitate the impregnation of the
colloidal nanoparticles, as also previously discussed [44]. The functionalized support was
added to an impregnation aqueous solution containing the desired amount of PVA-
stabilized copper oxide NPs to give a solid containing 1.2 wt% Cu.
Morphological analysis of colloidal NPs by TEM (Figure 1a) showed that the copper
oxide NPs have an average diameter of (3.1 ± 0.8) nm (Figure 1b), but the analysis of the
supported catalyst (Figure 1c), which discloses the core-shell structure of the support, do
not clearly show the copper oxide NPs on its surface. Two distinct regions were analyzed
by EDS technique (Figure 1d), as highlighted in Figure 1c. EDS analysis of the region 1
shows the presence of the elements Si and Fe corresponding to the magnetic support and
the analysis of region 2 revealed the additional presence of Cu, confirming the
composition of the nanoparticles analyzed (the presence of Ni is due to the TEM grid).
7

Figure 1: (a) TEM image of PVA-stabilized copper oxide nanoparticles; (b) NPs size
distribution histogram; (c) TEM image of the copper catalyst supported on silica-coated
magnetite nanoparticles and (d) EDS analysis of the regions highlighted in (c) using a Ni-
based sample grid.
The crystalline structure of the PVA-stabilized copper oxide NPs was evaluated by X-ray
diffraction analysis. Due to the presence of overlapping peaks between Fe
3
4
O core and
Cu oxide species in XRD, we prepared a sample with the copper oxide NPs supported on
silica spheres without the magnetic core. Figure 2a present the XRD patterns of as-
prepared copper oxide NPs supported on silica. The diffraction peaks identified at (1 1
1
), (2 0 0), (2 2 0) and (3 1 1) crystal planes can be perfectly assigned to the Cu O phase
2
by comparison with JCPDS 84-1286 database. The XRD for the calcined sample at 600
8

°
C for 2 h (Figure 2b) showed strong diffraction peaks at 2θ values 35.54 and 38.70°
corresponding to (1 1 -1) and (1 1 1) planes of Cu O phase (JPCDS-65-3288). We
expected to produce PVA-stabilized Cu(0) NPs, but after reduction of the copper(II)
precursor with NaBH , the characterization confirmed the preparation of a nearly phase
2
4
pure copper(I) oxide NPs.
Figure 2. X-ray diffraction patterns of PVA-stabilized copper oxide NPs supported on
silica: (a) as-prepared and (b) after thermal treatment at 600 °C for 2 h samples.
In order to obtain information on the oxidation state and chemical environment of
the surface of PVA-stabilized copper(I) oxide NPs supported on silica, a XPS analysis
was carried out. Figure 3 shows the XPS data for Cu 2p core level for both as-prepared
and thermal treated at 600 °C for 2 h samples. For the as-prepared sample the fitting peaks
of Cu 2p region in Figure 3a shows clearly the presence of peaks at 951.6 eV (2p1/2) and
1
+
9
31.7 eV (2p3/2) corresponding to Cu specie and peaks at 953.3 eV (2p1/2) and 933.6
(
2p3/2) corresponding to Cu2+ specie [45, 46]. These XPS result showed that the as-
prepared PVA-stabilized copper oxide NPs catalyst surface is composed of a mixture of
9

2
+
CuO and Cu
2
O oxides. The existence of Cu phases in as-prepared sample could not be
detect using XRD analysis probably due to small concentration of this specie presented.
1
+
However, the XPS analysis was enough to identify these two coexisting phase (Cu and
2
+
Cu ) present in the surface of the as-prepared sample. Thus, this result confirms that
combine both methods to understanding the real composition of the sample is very
1
+
important. In the Figure 3a, the peak of 2p 3/2 for Cu is more intense, showing that most
of the surface of the as prepared catalyst is composed of Cu O species. On the other hand,
in the XPS spectra for thermal treated copper oxide NPs supported on silica at 600 °C for
h shown in Figure 3b it was only possible to identify peaks at 954.0 eV (2p1/2) and 934.1
eV (2p3/2) corresponding to CuO, along with a remarkably satellite peak at near 942 eV.
2
2
2
+
This remarkably satellite peak is attributed to the binding energy of Cu suggesting the
1
+
2+
full conversion of Cu in the Cu species after thermal treatment. Therefore, calcination
of the catalyst compound Cu
2
O leads to full oxidation of copper species present on the
surface of the material.
10

Figure 3: XPS spectra and deconvolution of the photoemission line for Cu 2p: (a) as-
prepared PVA-stabilized copper oxide NPs supported on silica e (b) copper oxide NPs
supported on silica after calcination process (600 ºC, 2 h) and XANES spectrum of Cu-K
edge: (c) as-prepared e (d) after calcination process (600ºC, 2 h).
The X-ray Absorption Near Edge Structure (XANES) is a powerful tool for
determining the local atomic structure, giving information about the oxidation state of
compounds. Herein, XANES measurements were used to determine the local atomic
structure of Cu-species present in the as-prepared and after thermal treatment samples.
The XANES spectra revealed that both samples, as-prepared and thermally treated, have
a well apparent shoulder at 8981 and 8985 eV, respectively (Figure 3 c and d). The as-
prepared sample showed a shift in the edge region to lower energy values when compared
with the thermal treated sample at 600 °C for 2 h. The Cu K-edge shift to lower energies
1
+
is characteristic of Cu species [47], which is in good agreement with XRD and XPS
11

results. The XANES spectrum for the sample after thermal treatment presents important
changes in the characteristic shoulder position peak and white line compared to the
spectrum of the sample as-prepared. It is observed that the thermal treated copper catalyst
has a strong absorption at 8997.5 eV (white line) and a shoulder at 8985 eV indicating
that it is comprised of species of CuO, as shown by XRD and XPS results [34].
Figure 4 shows the k-weighted EXAFS signal χ(k) and the corresponding Fourier
transforms (FTs) for the as-prepared and thermal treated samples. The results from the
Cu K-edge are shown in Table 1.
Figure 4: EXAFS spectra of the Cu-K edge for the catalysts before and after the
calcination process. (a) Fourier Transform (FT) space-k and (b) FT space-R of as-
prepared PVA-stabilized copper oxide NPs supported on silica; (c) FT space-k and (d)
12

FT space-R for copper oxide NPs supported on silica after calcination process (600 ºC, 2
h).
The FT of EXAFS oscillations obtained for the as-prepared PVA-stabilized
copper oxide NPs supported on silica shows the presence of a peak at 1.4 Å and a peak at
2
.6 Å (Figure 4b) corresponding to the Cu-O and Cu-Cu bonds, respectively. The
oscillations obtained for the sample obtained after thermal treatment (600 ºC, 2 h)
demonstrates the presence of two main peaks at 1.4 Å and 2.5 Å (Figure 4d),
corresponding to the same type of bindings described above. These results are in
agreement with values reported in the literature for samples of Cu O and CuO,
2
respectively [34, 47, 48]. The coordination sphere parameters (bond distance (R),
2
coordination number (CN) and Debye-Waller factor (σ )) obtained by FT calculations for
the EXAFS oscillations obtained are shown in Table 1. It may be noted that the bond
distance values for both Cu-O and Cu-Cu are slightly lower for the as prepared sample
(
Table 1 - entries 1-2) when compared to the thermally treated sample (Table 1 - entries
-4). These bond distances are actually lower for the cuprous oxide (Cu O) due to the
high symmetry of the cubic structure of this compound, as well as the octahedral
environment of the Cu in Cu O crystal. Finally, the Debye-Waller factor values represent
3
2
2
the structural disorder of the material and, according to the literature, tend to be higher
for small-sized nanoparticles [49]. The values obtained are higher than the ones reported
in the literature for these types of structures, indicating the reduced size of the copper
oxide NPs present in the synthesized catalysts. It is worth mentioning that the reduced
size of the copper oxide NPs is probably responsible for lower Cu-O and Cu-Cu bond
distance values obtained for the catalysts compared to the values reported for bulk Cu
2
O
and CuO [34].
13

Table 1: Structural parameters calculated by EXAFS of synthesized copper oxide
catalysts.
2
2 c
Entry
Sample
Bond
R (Å)a
NCb
σ (Å )
Cu-O
1.89 ± 0.03
2.4 ± 0.1
0.0032
1
2
3
4
As-prepared
Cu-Cu
Cu-O
2.81 ± 0.05
1.90 ± 0.02
2.8 ± 0.1
2.9 ± 0.2
2.7 ± 0.5
0.0292
0.007
0.012
Calcined (600
oC, 2 h)
Cu-Cu
2.9 ± 0.5
a
b
c
bond distance in Angstroms; coordination number; Debye-Waller factor.
The magnetically recoverable copper oxide catalysts were tested in alkene oxidation
reactions, using cyclohexene as a model substrate. Figure 6 shows the possible
cyclohexene oxidation products. Since the mechanisms of allylic oxidation and
epoxidation compete during the reaction, the catalysts were tested for their selectivity for
the generation of allylic products or epoxide products.
Figure 6: Main products of cyclohexane oxidation: (1) 7-oxabicyclo-[4.1.0]-heptane
or cycloexene oxide; (2) 2-cyclohexen-1-ol; (3) 2-cyclohexen-1-one e (4) 7-oxabicyclo-
[
4.1.0]-heptan-2-one.
We first tested the activity and selectivity of the as-prepared PVA-stabilized
copper oxide NPs supported catalyst in the oxidation reaction and its behavior when used
in successive reactions. In a typical experiment, the catalyst and the olefin were added to
the reactor under inert atmosphere in a molar ratio of 500/1 (olefin/catalyst). The reactions
were carried out in solventless conditions. The reactor was then charged with 4 atm of O
2
14

and placed in a silicone oil bath maintained at constant temperature of 100°C. The
reaction was kept under a stirring rate of ~ 700 rpm with the aid of a magnetic stirrer.
After 6 hours, the reactor was cooled, the remaining oxygen pressure relieved and the
catalyst was recovered magnetically with the aid of a magnet placed against the wall of
the reactor. The organic phase was easily separated and analyzed by GC. The reusability
of the solid was measured by adding a new portion of the substrate to the reactor. The
solid after each reaction was separated and dried under vacuum for 10 min. Table 2 shows
the results obtained.
Table 2: Cyclohexene oxidation using as-prepared PVA-stabilized copper oxide NPs
catalyst.a
Selectivity (%)
O
O
OH
O
Entry
Conv. (%)
S
allylic
S
epoxide
O
1
2
3
4
5
6
31.6
46.1
43.7
32.0
40.5
48.4
6.9
5.6
48.1
52.8
55.5
58.6
60.4
66.1
45.0
41.6
44.5
23.8
25.2
30.8
-
-
-
93.1
94.4
99.4
82.4
85.6
96.9
6.9
5.6
0.6
0.6
13.6
12.0
2.0
4.0
2.4
1.1
17.6
14.4
3.1
15

7
8
13.2
13.5
3.1
6.5
63.8
19.7
31.8
72.2
1.3
1.5
95.6
92.0
4.4
8.0
a
Reaction conditions: 20.0 mmol substrate and 0.04 mmol Cu (substrate/catalyst: 500/1);
T = 100°C, 4 bar pressure of O , 6 h. There was no conversion in the absence of catalyst.
2
The catalyst shows high selectivity to the allylic oxidation products formation, and
the main product found was 2-cicloexen-1-one (3), as reported also for other non-noble
metals [18, 19, 50]. The 2-cicloexen-1-one is a versatile substance used as the basis for
the synthesis of a variety of products in the chemical industry, for example,
pharmaceuticals and fragrances [51]. Other products obtained were the epoxide (1),
derived from direct oxidation of cyclohexene and 2-cyclohexen-1-ol (2) is present in
small amounts. It is believed that 2-cyclohexen-1-one is formed by the oxidation of 2-
cyclohexen-1-ol [52, 53], which is also a product of allylic oxidation of cyclohexene. An
interesting finding is the presence of 7-oxabicyclo[4.1.0]heptan-2-one (4) among the
epoxidation products, product of direct oxidation of 2-cicloexen-1-one, which according
to the literature is not usually obtained except in the case of reactions with 100%
conversion [53]. The as-prepared PVA-stabilized copper oxide NPs supported catalyst
could be used for up to six consecutive reactions without losing activity and with high
selectivity for the formation of allylic oxidation products. The conversion increased from
the first cycle (32%) to the second cycle (46%), most probably due to removal of excess
of PVA, and was maintained until the 6th cycle. The conversion was reduced to 13% in
th
th
the 7 cycle and the catalyst lost its activity only in the 9 reaction cycle. The loss of
activity may not be directly related to loss of catalyst mass as copper leaching was not
noticed in the recovered organic phase (<0.01 ppm Cu detected by FAAS).
16

The calcined catalyst, which contains CuO NPs, was used to evaluate the stability
after removing the organic groups present on the catalyst support as well as the stabilizer
(
PVA) present in the as-prepared catalyst. The CuO-based catalyst (calcined at 600°C for
h in air atmosphere) was tested in the cyclohexene oxidation using the same previously
employed conditions (4 atm O pressure, 100 °C and 6 h reaction). The results obtained
2
2
are shown in Table 3. The catalyst was selective for the allylic oxidation products, and
again the preferably obtained product was 2-cyclohexen-1-one (3). We observed an
increase in substrate conversion to 90% in the first run, but there was also a loss of
activity, reaching a conversion compared with the as-prepared catalyst after the 4th run.
This result shows that the removal of organic ligands was favorable for the catalytic
activity; however, another possible explanation for the increase in catalytic activity may
arise from the change in copper oxidation state. The as prepared catalyst contains a
mixture of Cu O and CuO, and the calcined catalyst became CuO, exclusively. Moreover,
2
considering that the calcination step removed the organic stabilizer, morphological
changes or aggregation of the non-protected nanoparticles are possible explanations for
catalyst fast deactivation.
Table 3: Cyclohexene oxidation using calcined CuO-based catalyst.a
Selectivity (%)
O
O
OH
O
Entry
Conv. (%)
S
allylic
S
epoxide
O
1
2
89.9
79.2
4.9
3.9
60.5
63.6
34.6
31.3
-
95.1
94.9
4.9
5.1
1.2
17

3
4
5
6
7
8
61.8
37.2
36.8
19.5
10.2
-
2.5
2.3
3.1
-
66.2
65.9
67.6
68.5
70.2
-
30.3
29.4
28.8
31.5
29.8
-
1.0
2.4
0.5
-
96.5
95.3
96.4
100
3.5
4.7
3.6
-
-
-
-
-
100
-
-
-
a
Reaction conditions: 20.0 mmol substrate and 0.04 mmol Cu (substrate / catalyst:
b
5
00/1); T = 100°C, 4 bar pressure of O ; Selectivity for the main allylic and epoxides
2
products, respectively. There was no conversion in the absence of catalyst.
The influence of temperature on the cyclohexene oxidation with the CuO-based catalyst
(
calcined at 600°C for 2 h in air atmosphere) was investigated by varying the temperature
of 25 ° C to 100 ° C (Table 4, entries 1-4). The conversion and selectivity are highly
dependent on temperature. At temperatures below 75 °C the conversion was very low
compared with that obtained for the reaction carried out at 100 °C. It was also observed
that at room temperature, and direct oxidation of the olefinic bond take place practically
the same extent, thus impairing the selectivity of the catalyst studied for the allylic and
epoxide products. Raising the temperature favors the allylic oxidation mechanism, the
main product obtained cyclohex -2 -en- 1 -one ( 3). The influence of O pressure on the
2
cyclohexene oxidation with the CuO-based catalyst (calcined at 600°C for 2 h in air
atmosphere) was also investigated (Table 4, entry 5-7). The conversion increased with
18

increasing the reaction pressure, but the selectivity was only slightly affected, forming
preferentially the allylic oxidation products at all pressures studied.
Table 4: Oxidation of cyclohexene using calcined CuO-based catalyst at different T and
P.
Selectivity (%)
P O
2
Conv.
(%)
O
O
OH
O
Entry T (°C)
(bar)
Sallylic Sepoxide
O
1
2
25
50
4
4
4
4
1
2
6
3.8
46.9
9.2
26.8
68.6
70.9
60.5
54.6
63.6
26.3
22.2
21.3
34.6
34.6
31.3
-
-
53.1
90.8
92.2
96.5
89.2
90.2
96.3
46.9
9.2
8.7
75
15.7
76.2
28.2
50.2
89.4
5.4
2.4
1.4
-
7.8
3
4
5
6
7
1
1
1
1
00
00
00
00
3.5
3.5
10.8
9.8
10.8
9.8
1.2
3.7
2.7
70.5
25.8
1.0
Conditions: 20.0 mmol cyclohexene and 0.04 mmol Cu each reaction cycle (cyclohexene
copper = 500/1).
/
19

Finally, the reaction conversion and selectivity was investigates at different reaction
times (Figure 5). A typical cyclohexene oxidation reaction using the calcined catalyst at
4
bar of O and 100°C was followed by a longer time interval in order to investigate the
2
distribution of products throughout the reaction.
Figure 5: Profile of allylic oxidation of cyclohexene by calcined CuO-based catalyst
activity as a function of time. Reaction conditions: substrate 500 mol/mol of Cu, 4 bar
pressure of O and 100°C. ( conversion,  cyclohexen-1-one,  cyclohexen-1-ol, 
2
cyclohexen oxide,  oxabicyclo-[4.1.0]-heptan-2-one).
An analysis of the data in Figure 5 reveals that even in 1 hour reaction the catalyst is
selective for the formation of the allylic oxidation products. It is worth to note that after
1
h, there were equivalent amounts of 2-cyclohexen-1-ol (2) and 2-cyclohexen-1-one (3)
being produced, which is in agreement with the proposed mechanism where two
cyclohexenyl peroxy radicals combine, followed by decomposition of the resulting
tetraoxide intermediate into a 1:1 ratio of the allyl alcohol and enone [54]. At longer
reaction times, 2-cyclohexen-1-ol (2) is converted into 2-cyclohexen-1-one (3), which is
the main reaction product. The other observed products were cyclohexene oxide (1) and
7
-oxabicyclo-[4.1.0]-heptan-2-one (4). The first is formed in the beginning of the
reaction, by the direct oxidation of the olefinic bond, and was consumed after long
20

reaction times, by a reaction pathway not yet elucidated. After 4 h of reaction, we also
have noticed the decline of the selectivity of allylic products and an increase in selectivity
of the epoxide product 7-oxabicyclo-[4.1.0]-heptan-2-one (4), which is a product of the
direct oxidation of 2-cyclohexen-1-one (3). It is possible that the heterogeneously copper-
catalyzed oxidation occurs via a radical-chain sequence mechanism with initiation at the
catalyst surface, consistent with the proposed mechanism for the reaction in which some
epoxides products are formed from their corresponding allylic product [52, 55, 56].
Moreover, the radicalar nature of the olefin oxidation mechanism by the copper oxide
catalyst was confirmed by the interruption of an ongoing reaction after adding butylated
hydroxytoluene (BHT). BHT is an antioxidant that reacts with free radicals, delaying
oxidative processes. Here the addition of 1 equiv. of BHT, stopped the reaction and also
did not changed the product distribution. We can also conclude that both mechanisms of
oxidation (epoxidation and allylic oxidation) occur via radicals, as after the addition of
the radical trap the product distribution remained the same.
4
. Conclusions
Magnetically recoverable copper oxides catalysts were prepared by the immobilization of
PVA-stabilized pre-formed NPs, which were fully characterized by TEM, XRD, XPS,
XANES and EXAFS. The as prepared catalyst contains a mixture of Cu O and CuO, and
2
o
the calcined catalyst (600 C, 2 h) was completely oxidized CuO. The catalysts showed
high activity towards cyclohexene oxidation and selectivity to the allylic oxidation
product, which increases when the catalyst was calcined. The magnetic property allowed
a facile and fast product separation with negligible Cu leaching into the products. The
reaction method used has a strong environmental appeal as molecular oxygen is the only
oxidant, and the catalyst is easily recovered. Allylic oxidations are particularly attractive
for the valorization of terpenes through their conversion into allyl alcohols and enones.
21

Acknowledgements
The authors are grateful for the financial support offered by São Paulo Research
Foundation – FAPESP, CNPq and CAPES. Our thanks are also extended to the Brazilian
Synchrotron Light Laboratory (LNLS) for use of the XAFS1 (proposal XAFS1-16100)
and SXS (proposal number SXS-15172) experimental facilities.
5
. References
[
[
[
[
[
[
[
[
[
[
1] K. Godula, D. Sames, Science, 312 (2006) 67-72.
2] T. Jensen, P. Fristrup, Chem. Eur. J., 15 (2009) 9632-9636.
3] W. Yang, H. Chen, J. Li, C. Li, W. Wu, H. Jiang, Chem. Commun., 51 (2015) 9575-9578.
4] V. Weidmann, W. Maison, Synthesis, 45 (2013) 2201-2221.
5] I.I. Moiseev, M.N. Vargaftik, Coord. Chem. Rev., 248 (2004) 2381-2391.
6] B.M. Trost, J. Richardson, K. Yong, J. Am. Chem. Soc., 128 (2006) 2540-2541.
7] J.H. Delcamp, M.C. White, J. Am. Chem. Soc., 128 (2006) 15076-15077.
8] M.S. Chen, N. Prabagaran, N.A. Labenz, M.C. White, J. Am. Chem. Soc., 127 (2005) 6970-6971.
9] M.R. Maurya, P. Saini, C. Haldar, A.K. Chandrakar, S. Chand, J. Coord. Chem., 65 (2012) 2903-2918.
10] A.L. García-Cabeza, R. Marín-Barrios, F.J. Moreno-Dorado, M.J. Ortega, G.M. Massanet, F.M. Guerra, Org. Lett.,
1
6 (2014) 1598-1601.
[
[
[
[
[
[
[
11] E.C. McLaughlin, H. Choi, K. Wang, G. Chiou, M.P. Doyle, J. Org. Chem, 74 (2009) 730-738.
12] H. Choi, M.P. Doyle, Org. Lett., 9 (2007) 5349-5352.
13] B.B. Shingate, B.G. Hazra, D.B. Salunke, V.S. Pore, Tetrahedron Lett., 52 (2011) 6007-6010.
14] A. Dali, I. Rekkab-Hammoumraoui, A. Choukchou-Braham, R. Bachir, RSC Adv., 5 (2015) 29167-29178.
15] S. Serra, Eur. J. Org. Chem., 2015 (2015) 6472-6478.
16] F. Cavani, J.H. Teles, ChemSusChem, 2 (2009) 508-534.
17] E. Roduner, W. Kaim, B. Sarkar, V.B. Urlacher, J. Pleiss, R. Gläser, W.-D. Einicke, G.A. Sprenger, U. Beifuß, E.
Klemm, C. Liebner, H. Hieronymus, S.-F. Hsu, B. Plietker, S. Laschat, ChemCatChem, 5 (2013) 82-112.
[
[
[
[
[
[
[
[
[
[
[
[
[
18] C. Yin, Z. Yang, B. Li, F. Zhang, J. Wang, E. Ou, Catal. Lett., 131 (2009) 440-443.
19] F.P. Silva, M.J. Jacinto, R. Landers, L.M. Rossi, Catal. Lett., 141 (2011) 432-437.
20] A.E. Wendlandt, A.M. Suess, S.S. Stahl, Angew. Chem. Int. Ed., 50 (2011) 11062-11087.
21] S.D. McCann, S.S. Stahl, Acc. Chem. Res., 48 (2015) 1756-1766.
22] R. Zhang, R. Tang, J. Mater. Sci., 51 (2016) 5802-5810.
23] J.-F. Marcoux, S. Doye, S.L. Buchwald, J. Am. Chem. Soc., 119 (1997) 10539-10540.
24] I.E. Markó, P.R. Giles, M. Tsukazaki, S.M. Brown, C.J. Urch, Science, 274 (1996) 2044-2046.
25] J.-q. Xie, J.-z. Li, X.-g. Meng, C.-w. Hu, X.-c. Zeng, S.-x. Li, Transition Met. Chem., 29 (2004) 388-393.
26] J. Hassan, M. Sévignon, C. Gozzi, E. Schulz, M. Lemaire, Chem. Rev., 102 (2002) 1359-1470.
27] M. Samim, N.K. Kaushik, A. Maitra, Bull. Mater. Sci., 30 (2007) 535-540.
28] S. Uk Son, I. Kyu Park, J. Park, T. Hyeon, Chem. Commun., (2004) 778-779.
29] F. Alonso, Y. Moglie, G. Radivoy, M. Yus, Tetrahedron Lett., 50 (2009) 2358-2362.
30] D. Raut, K. Wankhede, V. Vaidya, S. Bhilare, N. Darwatkar, A. Deorukhkar, G. Trivedi, M. Salunkhe, Catal.
Commun., 10 (2009) 1240-1243.
[
[
[
[
[
[
31] B. Bridier, N. López, J. Pérez-Ramírez, J. Catal., 269 (2010) 80-92.
32] B. Bridier, M.A.G. Hevia, N. López, J. Pérez-Ramírez, J. Catal., 278 (2011) 167-172.
33] R. Poreddy, C. Engelbrekt, A. Riisager, Catal. Sci. Technol., 5 (2015) 2467-2477.
34] A. Sharma, M. Varshney, J. Park, T.-K. Ha, K.-H. Chae, H.-J. Shin, RSC Adv., 5 (2015) 21762-21771.
35] J.B. Reitz, E.I. Solomon, J. Am. Chem. Soc., 120 (1998) 11467-11478.
36] C. Liu, B. Yang, E. Tyo, S. Seifert, J. DeBartolo, B. von Issendorff, P. Zapol, S. Vajda, L.A. Curtiss, J. Am. Chem. Soc.,
37 (2015) 8676-8679.
1
[
[
[
37] D. Jiang, T. Mallat, D.M. Meier, A. Urakawa, A. Baiker, J. Catal., 270 (2010) 26-33.
38] M.J. Jacinto, P.K. Kiyohara, S.H. Masunaga, R.F. Jardim, L.M. Rossi, Appl. Catal. A Gen, 338 (2008) 52-57.
39] P.K. Khanna, S. Gaikwad, P.V. Adhyapak, N. Singh, R. Marimuthu, Mater. Lett., 61 (2007) 4711-4714.
22

[
[
[
[
[
[
40] M. Newville, J. Synchrotron Rad., 8 (2001) 322-324.
41] B. Ravel, M. Newville, J. Synchrotron Rad., 12 (2005) 537-541.
42] L.M. Rossi, N.J.S. Costa, F.P. Silva, R. Wojcieszak, Green Chem., 16 (2014) 2906-2933.
43] L.M. Rossi, M.A.S. Garcia, L.L.R. Vono, J. Braz. Chem. Soc., 23 (2012) 1959-1971.
44] M.J. Jacinto, F.P. Silva, P.K. Kiyohara, R. Landers, L.M. Rossi, ChemCatChem, 4 (2012) 698-703.
45] R.V. Gonçalves, R. Wojcieszak, H. Wender, C. Sato B. Dias, L.L.R. Vono, D. Eberhardt, S.R. Teixeira, L.M. Rossi, ACS
Appl. Mater. Interfaces, 7 (2015) 7987-7994.
[
[
[
[
[
[
[
[
[
[
[
46] S. Poulston, P.M. Parlett, P. Stone, M. Bowker, Surf. Interface Anal., 24 (1996) 811-820.
47] P. Khemthong, P. Photai, N. Grisdanurak, Int. J. Hydrogen Energy, 38 (2013) 15992-16001.
48] Y. Tanizawa, T. Shido, W.-J. Chun, K. Asakura, M. Nomura, Y. Iwasawa, J. Phys. Chem. B., 107 (2003) 12917-12929.
49] M.A. Peck, M.A. Langell, Chem. Mater., 24 (2012) 4483–4490.
50] A. Sakthivel, S.E. Dapurkar, P. Selvam, Appl. Catal. A Gen, 246 (2003) 283-293.
51] K.F. Podraza, Org. Prep. Proced. Int., 23 (1991) 217-235.
52] R. Luque, S.K. Badamali, J.H. Clark, M. Fleming, D.J. Macquarrie, Appl. Catal. A Gen, 341 (2008) 154-159.
53] Z.-w. Yang, Q.-x. Kang, H.-c. Ma, C.-l. Li, Z.-q. Lei, J. Mol. Catal. A: Chem., 213 (2004) 169-176.
54] H. Weiner, A. Trovarelli, R.G. Finke, J. Mol. Catal. A: Chem., 191 (2003) 217-252.
55] H. Alshammari, React. Kinet. Mech. Cat, (2016) 1-15.
56] H. Alshammari, P.J. Miedziak, S. Bawaked, .W. Knight, G.J. Hutchings, ChemCatChem, 4 (2012) 1565-1571.
23