MoO2 Formed on Mesoporous Graphene Oxide: Efficient and Stable Catalyst for Epoxidation of Olefins

Article abstract of DOI:10.1071/CH17089

A novel mesoporous MoO2 composite supported on graphene oxide (m-MoO2/GO) has been designed and applied as an efficient epoxidation catalyst. The m-MoO2/GO composite was characterised by X-ray diffraction, Raman spectroscopy, X-ray photoelectron spectrosc

Full text of DOI:10.1071/CH17089

CSIRO PUBLISHING
Aust. J. Chem. 2017, 70, 1039–1047
Full Paper
https://doi.org/10.1071/CH17089
MoO2 Formed on Mesoporous Graphene Oxide: Efficient
and Stable Catalyst for Epoxidation of Olefins
A
A C
,
B
A
Gang Bian, Pingping Jiang,
Kelei Jiang, Yirui Shen,
A
A
A
A
Linggang Kong, Ling Hu, Yuming Dong, and Weijie Zhang
A
Key Laboratory of Synthetic and Biological Colloids, Ministry of Education, School of
Chemical and Material Engineering, Jiangnan University, Wuxi 214122, China.
Wuxi WeiFu Lida Catalytic Converter Co. Ltd, Wuxi 214177, China.
B
C
Corresponding author. Email: ppjiang@jiangnan.edu.cn
A novel mesoporous MoO2 composite supported on graphene oxide (m-MoO2/GO) has been designed and applied as an
efficient epoxidation catalyst. The m-MoO2/GO composite was characterised by X-ray diffraction, Raman spectroscopy,
X-ray photoelectron spectroscopy, Brunauer–Emmet–Teller surface area analysis, field emission scanning electron
microscopy, and transmission electron microscopy. Compared with pure mesoporous MoO2 (m-MoO2) and amorphous
MoO2-graphene oxide (a-MoO2/GO), m-MoO2/GO exhibits the best catalytic activity. The conversion and selectivity for
cyclooctene are both over 99 % in 6 h. Remarkably, the mesoporous structure in m-MoO2/GO which derives from SiO2
nanospheres endows the catalyst better catalytic performance for long chain olefins: the conversion of methyl oleate can be
as high as 82 %. Such a robust catalyst can be easily recycled and reused five times without significant loss of catalytic
activity. This novel catalyst is promising in the synthesis of epoxides with a long carbon chain or large ring size.
Manuscript received: 13 February 2017.
Manuscript accepted: 18 April 2017.
Published online: 19 May 2017.
Introduction
long way to go before applying this epoxidation catalyst to
industrial use.
Considering the wide application of epoxides on manufacturing
epoxy resins, surfactants, and intermediates in organic synthe-
sis,[1–4] the epoxidation of olefins has been an important reaction
in research laboratories and chemical industry. The selective
production of epoxides from abundant olefin starting materials
has received much attention over the past several decades.[5–8]
Among the numerous catalysts, transition metal catalysts have
been chosen as the best candidates owing to their superior
activity and lower cost. In general, titanium silicalites, copper
oxide, molybdenum complexes, tungsten-containing com-
pounds, and titanium-containing molecular sieves[9–13] are the
most commonly used transition metal compounds.
We have been focusing our research on Mo-based catalysts
due to their specific selectivity for epoxides.[14–16] Furthermore,
low oxidation state MoII organometallic complexes have
attracted considerable interest since their simple preparation
and easy recovery from reactions. Poli et al.[17] designed a novel
‘[MoO2(SAP)]’ catalyst and obtained a relatively high selectivity
in the epoxidation of cyclooctene. With the aid of density
functional theory (DFT) calculations, they demonstrated that
the oxidant (tert-BuOOH, TBHP) can be activated by forming
a weakadductstabilisedwitha veryweakMo?O interaction and
a hydrogen bond. Pedro et al.[18] synthesised tremella-like MoO2
nanoparticles (NPs) with high catalytic performance in olefin
epoxidation. In reactions with cyclooctene, the substrate was
almost entirely turned into epoxide product when MoO2 NPs
were applied. However, based on our calculations from this
reference, the TOF (turnover frequency) value of these MoO2
NPs only reaches 4 (hꢀ1), which demonstrates that there is still a
Although organomolybdenum complexes are known for
their excellent selectivity in epoxidation, the problem of sepa-
ration and recycling still exists for these homogeneous catalysts.
To solve the problem, strategies of immobilising the catalysts
onto different carriers like zeolites,[19,20] polymer,[21,22] silica
compounds,[23] graphene nanosheets,[24] and layered double
hydroxide[25] and amine-modified multi-wall carbon nano-
tubes[26] have been reported. Moghadam et al.[27] successfully
grafted MoO2(acac)2 onto multi-wall carbon nanotubes and
applied the heterogeneous catalyst for the efficient epoxidation
of olefins with tert-BuOOH. The prepared robust catalyst can be
reused several times without significant loss of catalytic activity.
In our previous work, Zhang et al.[28] engineered two novel
tridentate Schiff base molybdenum(VI) complexes supported by
modified mesoporous SBA-15 as effective catalysts for epoxi-
dation. After being used five times, the conversion and selectivity
of the mesoporous material were still above 80 and 90%,
indicating a good reusability. In general, the introduction of
carriers can increase the stability of the catalyst and facilitate
recycling. However, some of these catalyst systems still bear
some drawbacks like a tedious preparation process or slow
reaction speed.
Graphene oxide (GO), a new class of 2D carbon nanomater-
ial consisting of conjugated sp2 carbons, is prominent due to its
unique physical, chemical, and electrical properties. The rich
oxygen-containing functionalities such as OH, SO3H, COOH,
epoxide groups, etc., which are introduced during the prepara-
tion of GO, endow this carbon nanomaterial moderate acidic and
Journal compilation Ó CSIRO 2017
www.publish.csiro.au/journals/ajc

1040
G. Bian et al.
(NH4)6Mo7O24
SiO2
GO
SiO2
12 nm
(NH4)6Mo7O24· 4H2O
MoO2
Scheme 1. Schematic depicting the synthesis of the m-MoO2/GO composite.
oxidising properties.[29] These deficient sites are abundant on
the large surface area of GO and support the use of GO as a
versatile catalyst carrier. Nowadays, GO has been employed in
various novel catalytic reactions, such as selective oxidation,[30]
electrocatalytic reactions,[31] and photocatalytic H2 evolu-
tion.[32] Li et al.[33–36] synthesised a series of Schiff base
(12 ꢁ 2nm) were purchased from Shanxi Fuhong Mineral Sub-
stance Co., Ltd. The other reagents ((NH4)6Mo7O24ꢂ4H2O, HF,
etc.) were all obtained from Sinopharm Chemical Reagent Co.,
Ltd and used directly without further treatment or purification.
Synthesis of GO and Mesoporous MoO2-Graphene Oxide
(m-MoO2/GO) Composite
complexes with different metal active centres (Fe2þ, Fe3þ
,
Co2þ, VO2þ, VO4þ, Cu2þ, and Mo6þ) supported on GO and
applied the catalysts to epoxidation reactions. As they claimed,
the catalytic activity of various olefins markedly increased and
this was ascribed to the uniform distribution of metal active sites
above GO. Furthermore, it has been demonstrated that the
addition of GO in the composite can accelerate electron transfer
during the catalytic reactions which all contributes to its superior
electrical conductivity.[37,38] Zhang et al.[39] prepared a ternary
composite UiO-66/CdS/1 % by modifying CdS with a metal–
organic framework (MOF) and reduced graphene oxide (rGO).
During the photocatalytic hydrogen evolution process, they
have found that the introduction of rGO can further accelerate
the transfer of photogenerated electrons and increase the evolu-
tion rate of H2. In short, the property of GO on accelerating the
electron transfer benefits catalytic reactions.
Herein, we demonstrate an efficient approach for the growth
of MoO2 NPs onto GO sheets as shown in Scheme 1. To our best
knowledge, this is the first attempt to enhance the catalytic
epoxidation activity by loading MoO2 NPs onto GO with TBHP
as the oxidant. Moreover, commercially available colloidal
silica nanospheres (12 ꢁ 2 nm) are added into the composite to
create the mesoporous structure which is favourable for epoxi-
dation reactions and imparts beneficial and selective properties
to the composite catalyst.
GO was synthesised from flake graphite with a modified
Hummers method as described in our previous work.[40] The
m-MoO2/GO composite was prepared by a two-step template
agent method. First, GO (200 mg) was dispersed into deionised
water (200 mL) under ultrasonication (40 kHz, 100w) for 1 h to
form a 0.1 % suspension. SiO2 nanospheres (12 ꢁ 2nm, 1 g)
were then added to the GO suspension and ultrasonicated for 1 h.
Ammonium molybdate ((NH4)6Mo7O24ꢂ4H2O, 36.8 mg,
10 wt-%) was gradually added to the mixed solution and stirred
over 1 h, followed by ultrasonication for 30 min. Afterwards, the
resulting aqueous mixture is rotary evaporated at 558C until
the mixture turns to a crisp brown solid (0.9 g). Immediately, the
solid was scraped from the flask and gradually ground into a fine
powder. Second, the mixture was calcined at 9008C for 2 h under
argon gas to obtain m-MoO2/GO. The residual SiO2 nano-
spheres in the composite were removed by centrifugation with
HF acid (three times) and washed with deionised water three
times. Finally, the mixture was dried in an oven at 808C over-
night. As a contrast, mesoporous MoO2 (m-MoO2) and amor-
phous MoO2-graphene oxide (a-MoO2/GO) were also prepared
similarly without the addition of GO or SiO2 nanospheres,
respectively.
Characterisation
Powder X-ray diffraction (XRD) patterns were recorded on a
BRUKER D8-Focus Bragg-Brentano X-ray powder diffrac-
Experimental
Chemicals
˚
tometer equipped with a Cu sealed tube (l 1.54178 A) operating
Flake graphite (44 mm average particle diameter) was purchased
from Qingdao Jinrilai Graphite Co., Ltd, SiO2 nanospheres
at 40 kV and 40 mA. The scanning region 2y was from 58 to 808
with a rate of 4 deg. minꢀ1. Raman spectra of the samples were

m-MoO2/GO as Epoxidation Catalyst
1041
(b)
(a)
GO
m-MoO2/GO
m-MoO2
m-MoO2
m-MoO2/GO
(100)
10°
10
20
30
40
50
60
70
80
1.0
1.5
2.0
2.5
3.0
2q [deg.]
2q [deg.]
Fig. 1. (a) Wide-angle and (b) small-angle XRD patterns of GO, m-MoO2 NPs, and m-MoO2/GO composite.
recorded using a confocal microscopic Raman spectrometer
(Renishow In-Via, USA) with 532 nm laser light irradiation in
the wavenumber range 400 to 3000 cmꢀ1 with a duration time of
10 s. The samples were pressed onto slices before analysis.
X-Ray photoelectron spectroscopy (XPS) was recorded with a
Kratos Ultra DLD imaging spectrometer (UK) using Al Ka
radiation (1486.6 eV) in the TRW research collaboration centre.
The powdered samples were pressed into pellets before XPS
studies. Thermogravimetric analysis (TGA) was performed
using a Mettler Toledo TGA/SDTA 851E analyser between 50
and 8008C at a heating rate of 208C minꢀ1 under an O2 atmo-
sphere. Nitrogen adsorption/desorption isotherms were obtained
at ꢀ1968C with a Quantachrome Quadrasorb SI automated gas
sorption system. Before the measurement, powders were
degassed under vacuum for 3 h at 1508C. The Brunauer–
Emmet–Teller (BET) method was applied to calculate the spe-
cific surface area. The pore volumes and pore size data were
obtained from the adsorption branches of nitrogen isotherms
using the Barrett–Joyner–Halenda (BJH) model. Morphological
studies were carried out by field-emission scanning electron
microscopy (FE-SEM) and scanning electron microscopy–
energy dispersive spectrometry (SEM-EDS) using a S-4800
machine (Hitachi) with an accelerating voltage of 3 kV.
Transmission electron microscopy (TEM) images were recorded
using a JEOL JEM 2100 electron microscope under an
accelerating voltage of 200 kV. High-resolution transmission
electron microscopy (HR-TEM) images were acquired with the
same equipment under 200 kV. Samples for TEM or HR-TEM
were prepared by ultrasonically dispersing the powders in eth-
anol over 2 h, and then adding the mixed solution dropwise onto
a carbon-coated copper grid.
twice, and dried at 608C over night in preparation for the next
run. To reduce the weight loss of the catalyst and make it easy to
be recycled, four times the amount of the recycled catalyst was
collected together so that it could be easily recovered from
reaction solution by centrifugation. Stability and recyclability
were both investigated with the recovered catalyst. The epoxi-
dation products were analysed with a gas chromatograph
(Shimadzu GC-8A), and the conversion or selectivity was cal-
culated by an internal standard method. The TOF values of the
epoxidation products were concluded from the inductively
coupled plasma (ICP) results as listed in the Supplementary
Material (Table S1).
Results and Discussion
Characterisation of m-MoO2/GO Composite
XRD was used to understand the structure of the m-MoO2/GO
composite (Fig. 1). As GO is prepared with a modified Hummers
method, the strong peak at 10.08 provides abundant evidence for
the formation of GO (Fig. 1a).[41] Meanwhile, GO presents a
crumpled and layered structure in the SEM image (Fig. S1a,
Supplementary Material), as well as a shaped profile in the TEM
image (Fig. S1b, Supplementary Material). Comparing the
wide-angle XRD pattern (Fig. 1a) of m-MoO2 with PDF stan-
dard cards, the pattern matches with the monoclinic crystal
structure (PDF#65-1273) that mainly corresponds to the MoO2
phase in the composite. After loading m-MoO2 particles onto
GO, the broad diffraction peaks of GO suppress the feature
peaks from MoO2. Still, the main (110) peak of MoO2 around
26.58 exhibits high intensity, and the (111) peak at 37.58 together
with the (220) peak at 54.18 can also be observed in the XRD
patterns, which all confirms the crystal form of MoO2 in
m-MoO2/GO composite. Furthermore, the small-angle XRD
analysis of m-MoO2 and m-MoO2/GO was carried out to
observe the mesoporous structure in the composites. Both
m-MoO2 and m-MoO2/GO (Fig. 1b) exhibit distinct (100) dif-
fraction peaks at 2y 0.7–1.28, which is solid evidence for the
mesostructure. The slightly reduced peak intensity of m-MoO2
is mainly attributed to particle agglomeration during the calci-
nation process.
Catalytic Tests
Catalytic epoxidation experiments were carried out in a Teflon-
lined screwcap vial (15 mL) with magnetic stirring. The olefin
substrate test-set comprised cyclohexene, cyclooctene, styrene,
a-methylstyrene, 1-phenylcyclohexene, hexene, oct-1-ene, and
methyl oleate. Typical conditions for catalytic epoxidation were
as follows: olefin (2 mmol), CHCl3 (2 mL), chlorobenzene
(2 mmol, a gas chromatography internal standard), and catalyst
(2 mg) were added to the vial. The reaction started with the
addition of TBHP (4 mmol) at 608C under vigorous stirring, and
the routine reaction time was set at 6 h. After the reaction, the
catalyst was compacted by centrifugation, washed with ethanol
The Raman spectrum of m-MoO2/GO is shown in Fig. 2,
together with the spectra of GO and m-MoO2 for comparison.
m-MoO2 is detected with Raman bands at 204, 364, 496, 570,
and 714 cmꢀ1. The finger bands at 204, 364, 496, and 570 cmꢀ1
can be assigned to the phonon vibration modes of MoO2,[42] and

1042
G. Bian et al.
the characteristic bands due to bond vibration modes of molyb-
denum oxide can be clearly confirmed at 570 (O1–Mo) and
741 cmꢀ1 (O2–Mo).[43] Both the m-MoO2/GO and GO show
classical D (1350 cmꢀ1) and G (1600 cmꢀ1) bands[44,45] in the
Raman spectra, which represent the defects and sp2 hybridisa-
tion degree of GO, respectively. The intensity ratio ID/IG of the
m-MoO2/GO composite is found to be ,1.12, higher than that of
pure GO (ID/IG E 0.94), indicating that the doping of MoO2
particles into GO increases the degree of defects in GO.
No obvious characteristic peak of MoO2 can be seen in the
m-MoO2/GO composite, attributable to the low concentration of
MoO2 particles in the catalyst.
The chemical nature of m-MoO2/GO in terms of individual
element’s valence states was obtained by XPS analysis (Fig. 3).
The peaks in Fig. 3a can be ascribed to Mo 3d, O 1s, and C 1s
elements. High resolution spectra of each element are shown in
Fig. 3b–d. The spectrum of Mo 3d shows two distinct peaks at
230.3 and 233.4 eV, which represent the core-levels of Mo4þ
3d[5/246] and corresponds to valence state IV of Mo in MoO2. The
spectrum of O 1s is fitted into three peaks centred at 530.8,
532.3, and 533.7 eV. The first and second peaks at 533.7 and
532.3eV correspond to the oxygen with the single bond in
C–OH[47] and –C=O in –COOH,[48] respectively, which both
derive from the oxygen-containing groups in GO. The highest
intensity peak centred at 530.8 eV is characteristic of Mo–O,[49]
giving strong evidence to the formation of MoO2 in the com-
posite. The C 1s spectrum can also be distributed into three
peaks at 284.8, 285.9, and 288.7 eV. The non-oxygenated
carbon (C–C/C=C) peak of GO is obviously at 284.8 eV,[50]
while the oxygenated carbon peaks are observed at 285.9 and
288.7 eV, corresponding to C–O functional groups and ionic
carboxylate[51] from GO, respectively. The data from XPS
analysis clearly indicate the formation of the m-MoO2/GO
composite.
204
GO
m-MoO2
m-MoO2/GO
364
741
496
570
G
D
500
1000
1500
2000
2500
3000
Raman shift [cmϪ1
]
TGA of m-MoO2 and m-MoO2/GO in an O2 and N2 environ-
ment was carried out to understand the thermal decomposition
behaviour of the constituents. As revealed in Fig. 4a, the weight
Fig. 2. Raman spectra of GO, m-MoO2 NPs, and m-MoO2/GO composite.
(a)
(b)
Experimental data
Simulated
Background
m-MoO2/GO
Mo 3d5/2
230.3 eV
Mo 3d
C 1s
Mo 3d3/2
233.4 eV
O 1s
Mo 3d
238 236 234 232 230 228 226 224 222
Binding energy [eV]
1200
1000
800
600
400
200
0
Binding energy [eV]
(c)
(d)
Experimental data
Simulated
Background
Experimental data
Simulated
Background
O 1s
C 1s
530.8 eV
284.8 eV
532.3 eV
533.7 eV
285.9 eV
288.7 eV
294
292
290
288
286
284
282
540
538
536
534
532
530
528
Binding energy [eV]
Binding energy [eV]
Fig. 3. (a) Survey scan XPS spectra of m-MoO2/GO composite in the binding energy range 0–1200 eV and high-resolution spectra of (b) Mo 3d,
(c) O 1s, and (d) C 1s in m-MoO2/GO.

m-MoO2/GO as Epoxidation Catalyst
1043
of m-MoO2 increases over 3008C, which is ascribed to the
further oxidation of MoIV to MoVI under O2 atmosphere. Weight
calculations from the TGA curves indicated that the mass
increase of m-MoO2 is ,10.17 %, consistent with one O atom
mass proportion in MoO2 (12.51 %), i.e. MoO2 gradually turns
into MoO3. The generated MoO3 starts to melt when the
temperature exceeds the melting point (7958C). On the other
hand, the initial weight loss of m-MoO2/GO up to 1508C
(Fig. 4a) could be attributed to evaporation of water molecules
preferably from GO.[52,53] The major weight loss of m-MoO2/
GO from 300 to 5008C is mainly due to the decomposition of
various oxygen-containing functional groups from GO.[54,55]
Furthermore, m-MoO2/GO also loses its weight either by
releasing CO/CO2 gaseous species or by giving up C from its
skeleton when the temperature exceeds 7008C in an O2 atmo-
sphere, which will reduce MoO3 to MoO2 at the same time.[56,57]
While in the N2 environment (Fig. 4b), the weight of m-MoO2 has
no obvious change, which is ascribed to the stable valence state of
MoVI under a protective gas (N2). When the temperature rises to
7958C, m-MoO2 partially melts under high temperature, leading
to a slight weight loss. The m-MoO2/GO composite sheds its
oxygen-containing functional groups between 200 and 3008C
(Fig. 4b). In addition, the composite still gradually loses its
weight between 300 and 8008C owing to the incomplete
combustion of elemental C and residual oxygen-containing
groups in m-MoO2.
To further confirm the pore structure and morphology,
m-MoO2 and m-MoO2/GO were characterised by SEM, TEM,
HRTEM, and SEM-EDS. Comparing the SEM images of
m-MoO2 (Fig. 6a) and m-MoO2/GO (Fig. 6b), m-MoO2 NPs
are severely agglomerated during calcination without the addi-
tion of GO. The same phenomenon can also be observed in
the TEM image (Fig. 6c); most of the m-MoO2 particles
aggregate into black patches. In contrast, the MoIV activity
centres disperse uniformly over the GO surface as shown in
Fig. 6d, and the MoO2 particles are much smaller. In Fig. 6e, the
enlarged red pane of Fig. 6d, the pores of ,15nm which are
derived from the ,12nm diameter SiO2 spheres are arranged
randomly above GO. The introduction of the mesoporous struc-
ture not only increases the specific surface area of m-MoO2/GO,
m-MoO2
m-MoO2/GO
N2 physisorption experiment results of m-MoO2 and
m-MoO2/GO composites are shown in Fig. 5. Both the isotherms
of m-MoO2 and m-MoO2/GO match well with the type IV BET
classification. The hysteresis loops of two composites can be
classified to type H3 as apparent in Fig. 5, which is associated
with partial aggregation of platelike particles giving rise to slit-
like pores.[58] The larger hysteresis of m-MoO2/GO indicates a
greater extent of mesoporosity above GO sheets. As shown in
Table 1, the specific surface area of m-MoO2/GO is three times
that of m-MoO2, which is attributed to the doping of GO which
has a large surface area (203 m2 gꢀ1). In addition, GO can prevent
the MoO2 particles from agglomerating, which leads to both
smaller sized MoO2 particles and changes in their shape such that
they actively adsorb additional N2 molecules. Furthermore, the
better dispersed metal active centres (Mo) in m-MoO2/GO will
improve the catalytic performance. In addition, the extensive
agglomeration of MoO2 particles caused by calcination generates
part of the macroporous structures, which makes the pore size of
m-MoO2 larger than the m-MoO2/GO composite.
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P0)
Fig. 5. Nitrogen adsorption/desorption isotherms of m-MoO2 NPs and
m-MoO2/GO composite.
Table 1. Physical and structural properties of m-MoO2 NPs and
m-MoO2/GO composites measured by BET
Sample
Surface area
[m2 gꢀ1
Pore volume
[cm3 gꢀ1
]
Pore size [nm]
]
m-MoO2
45.61
0.13
0.08
39.78
15.23
m-MoO2/GO
114.72
(a) 120
m-MoO2
(b)
110
100
90
m-MoO2
m-MoO2/GO
m-MoO2/GO
Ϫ10.17 %
100
80
Ϫ13.18 %
80
Ϫ22.55 %
Ϫ54.58 %
70
60
60
50
40
40
100
200
300
400
500
600
700
800
100
200
300
400
500
600
700
800
Temperature [ЊC]
Temperature [ЊC]
Fig. 4. TGA curves of m-MoO2 NPs and m-MoO2/GO composite under (a) O2 and (b) N2 atmosphere.

1044
G. Bian et al.
(a)
(b)
(g)
m-MoO2
(c)
(d)
Mo-L
O-K
(h)
500 nm
1μm
(e)
(f)
C-K
O-K
m-MoO2-GO
10 nm
50 nm
Mo-L
Fig. 6. SEM images of (a) m-MoO2 NPs and (b) m-MoO2/GO composite. TEM images of (c) m-MoO2 NPs and (d) m-MoO2/GO composite.
(e) Enlargement of red box in (d). (f) Enlargement of yellow circle sections in (d). SEM-EDS element mapping images of (g) m-MoO2 NPs and
(h) m-MoO2/GO composite.
but also enhances the dispersity of the metal active centres. The
periodic lattice fringe space is found to be 0.342 nm as shown in
Fig. 6f, corresponding to the interplanar spacing of the (110)
plane from monoclinic MoO2 (JCPDS: 00-078-1073). The SEM-
EDS element mapping images of m-MoO2 and m-MoO2/GO are
presented in Fig. 6g, h, respectively, to reveal the distribution of
eachelement. After supportingMoO2 particles ontoGO (Fig. 6h),
both the Mo and O elements can uniformly scatter over GO
sheets. The corresponding SEM-EDS data is provided in the
Supplementary Material (Fig. S2).
catalytic activity for the substrate (cyclooctene), which is
attributed to the poor dispersity of the aqueous oxidising agent
in the organic reaction solvent (CHCl3). Although NaIO4 and
iodosobenzene (PhIO) have a satisfactory selectivity, the con-
versions are quite low, owing to the relatively low solubility of
the solid oxidising agents in CHCl3. Only the soluble organic
oxidant tert-BuOOH (TBHP) can dissolve well in CHCl3. Both
the conversion and selectivity surpass 99 % as shown in Fig. 7b.
So tert-BuOOH (TBHP) was chosen as the oxidant in further
epoxidation reactions.
The third factor investigated is the effect of the solvents.
Chlorinated solvents such as CHCl3 and CH2Cl2 gave high
conversions in the presence of TBHP (Fig. 7c). The conversion
of product in CHCl3 solution is over 99 %, which can be ascribed
to the lower dielectric constant of chlorinated solvents.[59]
Meanwhile conversion in CH3COOC2H5 (29 %) is much lower
than in chlorinated solvents, presumably owing to the higher
dielectric constant than chlorinated solvents. The conversions
achieved for cyclooctene in CH3CN and CH3OH are even lower
than for CH3COOC2H5. Catalytic activity is impeded by the
hard coordinate donating atoms (O and N as hard Lewis bases) in
the solvent which strongly link onto the MoVI metal centres.
Consequently, to obtain the optimal catalytic performance,
CHCl3 was chosen to be the reaction solvent.
The last factor consists on the evaluation of catalyst stability.
Cyclooctene was used as the model substrate under the described
reaction conditions in Fig. 7d. The catalyst was recovered by
centrifugation and washed with ethanol three times for subse-
quent experiments (five runs) without adding any fresh catalyst.
A 99% product selectivity was achieved over the five-run test
without significant loss of conversion. Hot-filtration experiments
were carried out to check the possible catalysis of leached metal
species in the reaction solution. Initially, m-MoO2/GO and TBHP
Catalytic Epoxidation Reactions
The m-MoO2/GO composite was applied to olefin epoxidation
using tert-BuOOH as oxidant in CHCl3 solvent. Before the
comprehensive assessment of the catalytic performance for
m-MoO2/GO, four different reaction factors were explored, as
shown in Fig. 7. The first consists of understanding how much
catalyst is optimal. As Fig. 7a reveals, the conversion rises along
with the increasing catalyst concentration, while the selectivity
reaches a maximum upon addition of 2 mg of catalyst. Using
more catalyst only reduces the selectivity of cyclooctene, con-
verting the epoxidation products into other by-products like
aldehydes and alcohol compounds. Hence, the optimum amount
of catalyst was set at 2 mg to achieve both the optimal conver-
sion and selectivity. Meanwhile, the content of Mo active cen-
tres in relation to the amount of olefin substrate used was further
confirmed (0.05 mol-%) through the calculation from ICP
results (Table S1, Supplementary Material) when 2 mg catalyst
was used.
The second factor relies on the evaluation of optimum
oxidant. As shown in Fig. 7b, no epoxidation products can be
detected without oxidant. Simultaneously, H2O2 exhibits a poor

m-MoO2/GO as Epoxidation Catalyst
1045
(a)
Different dosages
(b)
Different oxidants
69
Ͼ99
Ͼ99
Ͼ99
99
99
98
100
98
96
94
92
Ͼ99
TPHP
98
77
95
100
92
80
60
40
20
0
90
90
88
86
84
0
0
6
4
3
0
Blank
H2O2
PhlO
NalO4
1 mg
2 mg
Conversion [%]
5 mg
Selectivity [%]
10 mg
Conversion [%]
Selectivity [%]
Different solvents
Different recycles
(c)
(d)
Ͼ99
Ͼ99
Ͼ99
99
Ͼ99
99
99
99
Ͼ99
99
99
98
97
93
99.0
Ͼ99
99
98.5
98.0
97.5
97.0
96.5
96.0
95.5
95.0
100
80
60
40
20
0
34
29
20
CHCl3
C2H4Cl2
CH3CN CH3CH2OH CH3COOC2H5
Selectivity [%]
1
2
3
4
5
Selectivity [%]
Conversion [%]
Conversion [%]
Fig. 7. Catalytic performance of m-MoO2/GO in epoxidation assessed on four different reaction factors: (a) different concentrations of m-MoO2/GO;
(b) different oxidants; (c) different solvents; (d) different number of cycles. Routine reaction conditions (the explored reaction factor is variable when it is
assessed): cyclooctene (2 mmol), CHCl3 (2 mL, solvent), chlorobenzene (1 mmol, a GC internal standard), catalyst (2 mg), and TBHP as oxidant (4 mmol). The
mixture is continuously stirred at 608C for 6 h.
in CHCl3 were stirred at 608C for 0.5 h. m-MoO2/GO was then
Table 2. Catalytic performance of different heterogeneous catalysts in
removed and the reaction proceeded for another 5.5 h with the
homogeneous filtrate. The GC results demonstrate that no oxida-
tion occurs. Furthermore, SEM, TEM, and ICP-AES elemental
analysis were carried out to investigate the structure and element
content (Mo) changes between the fresh and recycled catalyst. No
obvious structural changes are observed as revealed by SEM
(Fig. S3a, b; Supplementary Material) and TEM (Fig. S3c, d;
Supplementary Material) images, the recycled catalyst retains a
similar morphology except for a slight aggregation of MoO2
particles. Meanwhile, there is no significant change of Mo
elementcontentinthe recycledcatalyst (Table S1, Supplementary
Material), demonstrating the durability of the m-MoO2/GO
structure.
Furthermore, the superior catalytic performance of m-MoO2/
GO has been demonstrated through a comparison with m-MoO2
NPs, amorphous MoO2/GO (a-MoO2/GO), and other similar
catalysts reported in the literature[60–63] as revealed in Table 2.
All catalysts with Mo active centres show high conversion and
selectivity for cyclooctene. Although the incorporation of GO,
(entry 4, m-MoO2/GO) only slightly improves the conversion
and selectivity in comparison with m-MoO2 (entry 2), the TOF
value increases more than 20 times. GO can efficiently acceler-
ate the electron transfer at metal active centres, which promotes
the number of Mo 3d active centres more likely to be activated to
catalyse the substrates. This effect on the acceleration of
electron transfer caused by GO has already been demonstrated
in our previous work[64] and been reported in the literature.[65,66]
Otherwise, pristine GO shows almost the same catalytic activity
(Table 2, entry 3) as the blank experiment on cyclohexene,
namely GO has no catalytic activity in epoxidation, confirming
that the sharply improved catalytic activity is not due to a simple
physical mixture of MoO2 and GO. Comparing m-MoO2/GO
the literature, and comparison with m-MoO2 NPs, a-MoO2/GO, and
m-MoO2/GO composites in this work
Catalyst
O
TBHP, Δ
Entry Catalyst
Conversion [%] Selectivity [%] TOF [hꢀ1
]
A
1
–
23
95
12
86
–
17
–
2
m-MoOA2
GOA
3
25
13
4
m-MoO2/GOA
a-MoO2/GOA
MoO2 NPs[14]
1Mo@SMNP[59]
MSh-bpy-MoVI [60]
MNP30-Si-inic-Mo[61]
Ag NPs[62]
.99
75
.99
90
410
281
4
5
6
99
100
96B
7
–
–
8
94
100
40
–
9
93
–
10
71
83
–
AReaction conditions: cyclooctene (2 mmol), CHCl3 (2 mL, solvent), chlo-
robenzene (1 mmol), catalyst (2 mg), and TBHP (4 mmol). The mixture is
continuously stirred at 608C for 6 h.
BValue refers to epoxide yield as conversion was not listed.
with other Mo type heterogeneous catalysts (entries 7, 8, 9), its
catalytic activity is superior either in conversion or selectivity.
Although the MoO2 NPs (entry 6) exert a higher selectivity than
m-MoO2/GO, their TOF value is much lower. The traditional
noble metal Ag NPs (entry 10) just exhibit a passable catalytic
activity for cyclooctene, which further confirms the superior
catalytic performance of m-MoO2/GO.

1046
G. Bian et al.
Table 3. Epoxidation of various substrates with TBHP over
m-MoO2/GO compositeA
observed from cyclohexene, cyclooctene, a-methylstyrene, and
1-phenylcyclohexene (entries 1, 2, 4, 5). Cyclooctene, which has
a larger size, exerts a better catalytic performance than cyclohex-
ene. Similarly, when the methyl group in a-methylstyrene is
substituted by a cyclic structure with a larger size (entry 5), both
the conversion and selectivity are improved. It appears that
m-MoO2/GO prefers larger size substrates, however no obvious
confinement effect is detected. This hypothesis can be supported
by the TEM image of m-MoO2/GO (Fig. 6e). Most MoO2 NPs
disperse outside the mesopores, rather than inside the porous
channels. Such character is more propitious for long carbon chain
or large size substrates to contact with the Mo activity centres.
Entry Substrate
1
Product
Conversion Selectivity TOF
[hꢀ1
]
[%]
[%]
94
67
259
O
O
2
.99
99
410
3
4
85
71
83
53
292
157
Conclusions
O
In summary, we have demonstrated the synthesis of MoO2
nanoparticles supported on mesoporous graphene oxide
(m-MoO2/GO). The catalytic activity of pure m-MoO2 is signif-
icantly enhanced by combining with GO. The conversion and
selectivity of cyclooctene catalysed with m-MoO2/GO both sur-
pass 99%, which is much higher than those of m-MoO2 and a-
MoO2/GO. In addition, the introduction of mesopores into GO
sheets not only promotes the catalytic activity but also make
m-MoO2/GO more beneficial to catalysing olefins with a long
carbon chain and large ring size: the conversion of methyl oleate
can reach to 82% in 6 h. Moreover, no significant decline in
catalytic activity was observed even after recycling five times,
indicating a high stability. This work provides a brand-new design
strategy for GO-based hybrid catalysts, with high activities to
catalyse long chain and annular olefins to epoxide products.
O
5
83
42
62
98
212
170
O
6
7
O
O
73
82
90
78
271
263
8
O
O
O
O
O
AReaction conditions: olefin (2 mmol), CHCl3 (2 mL, solvent), chloroben-
zene (1 mmol), catalyst (2 mg), and TBHP (4 mmol). The mixture is con-
tinuously stirred at 608C for 6 h
Supplementary Material
Characterisation data of GO, and elemental analysis, ICP anal-
ysis, SEM and TEM images of the composite catalysts are
available on the Journal’s website.
Meanwhile, the role the porous structure plays in m-MoO2/
GO is demonstrated by the controlled experiment of amorphous
MoO2/GO (a-MoO2/GO) in cyclooctene epoxidation. As shown
in Table 2 (entries 4, 5), the conversion of cyclooctene catalysed
with m-MoO2/GO (.99 %) is apparently higher than a-MoO2/
GO (75 %), as well as the selectivity. The mesoporous channels
inside the composite can effectively strut graphene oxide sheets,
which increases the layer distance between GO lamellas and
suppress the aggregation. MoO2 NPs can disperse more evenly
above GO, and active centres (MoIV) will contact with the
epoxidation substrates more efficiently, which finally leads to
a better catalytic performance.
Acknowledgements
The authors gratefully acknowledge the financial support from National
‘Twelfth Five-Year’ Plan for Science & Technology (2012BAD32B03), the
Postgraduate Innovation Project of Jiangsu Province (KYLX15-1156), the
Fundamental Research Funds for the Central Universities (JUSRP51507 and
JUSRP51623A), and MOE & SAFEA, 111 Project (B13025).
References
[1] K. Kamata, K. Yonehara, Y. Sumida, K. Hirata, S. Nojima, N. Mizuno,
Angew. Chem. Int. Ed. 2011, 50, 12062. doi:10.1002/ANIE.
201106064
´
[2] M. Herbert, E. Alvarez, D. J. Cole-Hamilton, F. Montilla, A. Galindo,
Encouraged by the study mentioned above, we assessed the
catalytic activity of m-MoO2/GO under the optimal reaction
conditions with various alkenes. The olefins were all efficiently
oxidised to the corresponding oxides in the presence of
m-MoO2/GO (Table 3, entries 1–8). Compared with the alicy-
clic olefins (cyclohexene and cyclooctene, entries 1 and 2),
aromatic olefins (styrene and a-methylstyrene, entries 3 and 4)
afford relatively low conversion and selectivity. Electron with-
drawing groups connected to the double bond might reduce the
electron cloud density around it, which make it difficult for the
substrate to generate an epoxidation product.
Interestingly, the conversion of methyl oleate (entry 8) is
higher than hexene (entry 6) and oct-1-ene (entry 7) as shown in
Table 3. The longer carbon chain does not suppress the catalytic
activityof m-MoO2/GO. Instead, it makes the catalyst enhance the
conversion of olefin substrates. The same phenomenon can also be
Chem. Commun. 2010, 46, 5933. doi:10.1039/C0CC00462F
[3] S. P. Maradur, C. Jo, D.-H. Choi, K. Kim, R. Ryoo, ChemCatChem
2011, 3, 1435. doi:10.1002/CCTC.201100045
[4] Y. Leng, J. Wang, D. Zhu, M. Zhang, P. Zhao, Z. Long, J. Huang,
Green Chem. 2011, 13, 1636. doi:10.1039/C1GC15302A
[5] W. Zhang, J. L. Loebach, S. R. Wilson, E. N. Jacobsen, J. Am. Chem.
Soc. 1990, 112, 2801. doi:10.1021/JA00163A052
[6] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi,
N. Mizuno, Science 2003, 300, 964. doi:10.1126/SCIENCE.1083176
[7] Y. Lei, F. Mehmood, S. Lee, J. Greeley, B. Lee, S. Seifert, R. E. Winans,
J. W. Elam, R. J. Meyer, P. C. Redfern, D. Teschner, R. Schlo¨gl, M. J.
Pellin, L. A. Curtiss, S. Vajda, Science 2010, 328, 224. doi:10.1126/
SCIENCE.1185200
[8] Y. Liu, H. Tsunoyama, T. Akita, T. Tsukuda, Chem. Commun. 2010,
46, 550. doi:10.1039/B921082B
[9] Y. Zuo, M. Liu, T. Zhang, L. Hong, X. Guo, C. Song, Y. Chen, P. Zhu,
C. Jaye,D.Fischer,RSC Adv. 2015,5, 17897. doi:10.1039/C5RA00194C

m-MoO2/GO as Epoxidation Catalyst
1047
[10] W. Jia, Y. Liu, P. Hu, R. Yu, Y. Wang, L. Ma, D. Wang, Y. Li, Chem.
Commun. 2015, 51, 8817. doi:10.1039/C5CC02480C
¨
[11] M. Bagherzadeh, M. Zare, T. Salemnoush, S. Ozkar, S. Akbayrak,
[38] A. Dey, J. Athar, P. Varma, H. Prasant, A. K. Sikder,
S. Chattopadhyay, RSC Adv. 2015, 5, 1950. doi:10.1039/C4RA10812D
[39] R. Lin, L. Shen, Z. Ren, W. Wu, Y. Tan, H. Fu, J. Zhang, L. Wu, Chem.
Commun. 2014, 50, 8533. doi:10.1039/C4CC01776E
[40] G. Bian, P. Jiang, W. Zhang, K. Jiang, L. Hu, Z. Jian, Y. Shen,
P. Zhang, RSC Adv. 2015, 5, 90757. doi:10.1039/C5RA14813H
[41] Y. Wang, I. Djerdj, B. Smarsly, M. Antonietti, Chem. Mater. 2009, 21,
3202. doi:10.1021/CM9007014
Appl. Catal. A 2014, 475, 55. doi:10.1016/J.APCATA.2014.01.020
[12] M. Amini, M. M. Haghdoost, M. Bagherzadeh, Coord. Chem. Rev.
2014, 268, 83. doi:10.1016/J.CCR.2014.01.035
[13] Z. Deng, Y. Yang, X. Lu, J. Ding, M. He, P. Wu, Catal. Sci. Technol.
2016, 6, 2605. doi:10.1039/C5CY02002F
[14] Y. Leng, J. Liu, C. Zhang, P. Jiang, Catal. Sci. Technol. 2014, 4, 997.
doi:10.1039/C3CY01014G
[42] M. Dieterle, G. Mestl, Phys. Chem. Chem. Phys. 2002, 4, 822.
doi:10.1039/B107046K
[15] Y. Jin, H. Wang, J. Li, X. Yue, Y. Han, P. K. Shen, Y. Cui, Adv. Mater.
2016, 28, 3785. doi:10.1002/ADMA.201506314
[43] Y. Jin, H. Wang, J. Li, X. Yue, Y. Han, P. K. Shen, Y. Cui, Adv. Mater.
2016, 28, 3785. doi:10.1002/ADMA.201506314
[16] J. G. Lopez, M. Zaranek, P. Pawluc, R. M. Gauvin, A. Mortreux, Oil
Gas Sci. Technol. 2016, 71, 20. doi:10.2516/OGST/2015046
[17] J. Morlot, N. Uyttebroeck, D. Agustin, R. Poli, ChemCatChem 2013, 5,
601. doi:10.1002/CCTC.201200068
[18] A. Bento, A. Sanches, E. Medina, C. D. Nunes, P. D. Vaz, Appl. Catal.
A 2015, 504, 399. doi:10.1016/J.APCATA.2015.03.024
[19] C.-C. Yang, S. Yacob, B. A. Kilos, D. G. Barton, E. Weitz, J. M.
Notestein, J. Catal. 2016, 338, 313. doi:10.1016/J.JCAT.2016.03.011
[20] E. Mannei, F. Ayari, M. Mhamdi, M. Almohalla, A. Guerrero Ruiz,
G. Delahay, A. Ghorbel, Microporous Mesoporous Mater. 2016, 219,
77. doi:10.1016/J.MICROMESO.2015.07.036
[44] Q. Tang, Z. Shan, L. Wang, X. Qin, Electrochim. Acta 2012, 79, 148.
doi:10.1016/J.ELECTACTA.2012.06.093
[45] S. Petnikota, N. K. Rotte, M. V. Reddy, V. V. S. S. Srikanth, B. V. R.
Chowdari, ACS Appl. Mater. Interfaces 2015, 7, 2301. doi:10.1021/
AM5064712
[46] C. Shi, S. Zhang, X. Li, A. Zhang, M. Shi, Y. Zhu, J. Qiu, C. Au, Catal.
Today 2014, 233, 46. doi:10.1016/J.CATTOD.2013.10.076
[47] A. A. Savel’ev, A. G. Zheleznyi, D. A. Loktev, Weld. Int. 2012, 26,
787. doi:10.1080/09507116.2011.653162
[48] R. W. Truss, B. Wood, R. Rasch, J. Appl. Polym. Sci. 2016, 133, 43023.
doi:10.1002/APP.43023
[21] S. L. Jain, B. Sain, Adv. Synth. Catal. 2008, 350, 1479. doi:10.1002/
ADSC.200800264
[49] J. Liu, S. Tang, Y. Lu, G. Cai, S. Liang, W. Wang, X. Chen, Energy
Environ. Sci. 2013, 6, 2691. doi:10.1039/C3EE41006D
[22] M. R. Maurya, U. Kumar, P. Manikandan, Dalton Trans. 2006, 29,
3561. doi:10.1039/B600822D
[23] N. Merle, F. Le Que´me´ner, Y. Bouhoute, K. C. Szeto, A. De
Mallmann, S. Barman, M. K. Samantaray, L. Delevoye, R. M.
Gauvin, M. Taoufik, J.-M. Basset, J. Am. Chem. Soc. 2017, 139,
2144. doi:10.1021/JACS.6B11220
[50] K. S. Rao, J. Sentilnathan, H.-W. Cho, J.-J. Wu, M. Yoshimura, Adv.
Funct. Mater. 2015, 25, 298. doi:10.1002/ADFM.201402621
[51] B. De’Ne`ve, M. Delamar, T. T. Nguyen, M. E. R. Shanahan, Appl.
Surf. Sci. 1998, 134, 202. doi:10.1016/S0169-4332(98)00220-7
[52] S. Petnikota, S. K. Marka, A. Banerjee, M. V. Reddy, V. V. S. S.
Srikanth, B. V. R. Chowdari, J. Power Sources 2015, 293, 253.
doi:10.1016/J.JPOWSOUR.2015.05.075
[24] H. P. Mungse, S. Verma, N. Kumar, B. Sain, O. P. Khatri, J. Mater.
Chem. 2012, 22, 5427. doi:10.1039/C2JM15644J
[53] S. Petnikota, V. V. S. S. Srikanth, P. Nithyadharseni, M. V. Reddy,
S. Adams, B. V. R. Chowdari, ACS Sustain. Chem. Eng. 2015, 3, 3205.
doi:10.1021/ACSSUSCHEMENG.5B00791
[25] S.-J. Xia, F.-X. Liu, Z.-M. Ni, W. Shi, J.-L. Xue, P.-P. Qian, Appl.
Catal. B 2014, 144, 570. doi:10.1016/J.APCATB.2013.07.060
[26] M. Moghadam, S. Tangestaninejad, V. Mirkhani, I. Mohammadpoor-
Baltork, N. S. Mirbagheri, Appl. Organomet. Chem. 2010, 24, 708.
doi:10.1002/AOC.1671
[27] F. Esnaashari, M. Moghadam, V. Mirkhani, S. Tangestaninejad,
I. Mohammadpoor-Baltork, A. R. Khosoropour, M. Zakeri,
S. Hushmandrad, Polyhedron 2012, 48, 212. doi:10.1016/J.POLY.
2012.08.084
[28] J. Zhang, P. Jiang, Y. Shen, W. Zhang, X. Li, Microporous Meso-
porous Mater. 2015, 206, 161. doi:10.1016/J.MICROMESO.2014.12.
027
[29] D. R. Dreyer, H.-P. Jia, C.W. Bielawski, Angew. Chem. Int. Ed. 2010,
122, 6965. doi:10.1002/ANGE.201002160
[54] M. Acik, G. Lee, C. Mattevi, M. Chhowalla, K. Cho, Y. J. Chabal, Nat.
Mater. 2010, 9, 840. doi:10.1038/NMAT2858
[55] H. K. Jeong, Y. P. Lee, M. H. Jin, E. S. Kim, J. J. Bae, Y. H. Lee, Chem.
Phys. Lett. 2009, 470, 255. doi:10.1016/J.CPLETT.2009.01.050
¯
¯
[56] A. Oya, M. Kasahara, H. Ohashi, S. Otani, J. Am. Ceram. Soc. 1991,
74, 1439. doi:10.1111/J.1151-2916.1991.TB04125.X
[57] S. Chaudhury, S. K. Mukerjee, V. N. Vaidya, V. Venugopal, J. Alloys
Compd. 1997, 261, 105. doi:10.1016/S0925-8388(97)00212-0
[58] J. Yu, J. C. Yu, M. K.-P. Leung, W. Ho, B. Cheng, X. Zhao, J. Zhao,
J. Catal. 2003, 217, 69. doi:10.1016/S0021-9517(03)00034-4
[59] W. Fan, D. Shi, B. Feng, Catal. Commun. 2016, 74, 1. doi:10.1016/
J.CATCOM.2015.10.022
[30] B. Majumdar, T. Bhattacharya, T. K. Sarma, ChemCatChem 2016, 8,
1825. doi:10.1002/CCTC.201600136
[31] L. Sun, H. Wang, K. Eid, S. M. Alshehri, V. Malgras, Y. Yamauchi,
L. Wang, Electrochim. Acta 2016, 188, 845. doi:10.1016/J.ELEC
TACTA.2015.12.068
[32] R. Ge, X. Li, S.-Z. Kang, L. Qin, G. Li, Appl. Catal. B 2016, 187, 67.
doi:10.1016/J.APCATB.2016.01.024
[33] Z. Li, S. Wu, H. Ding, H. Lu, J. Liu, Q. Huo, J. Guan, Q. Kan, New J.
Chem. 2013, 37, 4220. doi:10.1039/C3NJ00982C
[34] Z. Li, S. Wu, D. Zheng, H. Liu, J. Hu, H. Su, J. Sun, X. Wang, Q. Huo,
J. Guan, Q. Kan, Appl. Catal., A 2014, 470, 104. doi:10.1016/
J.APCATA.2013.10.037
[35] H. Su, Z. Li, Q. Huo, J. Guan, Q. Kan, RSC Adv. 2014, 4, 9990.
doi:10.1039/C3RA47732K
[60] S. Shylesh, J. Schweizer, S. Demeshko, V. Schu¨nemann, S. Ernst,
W. R. Thiel, Adv. Synth. Catal. 2009, 351, 1789. doi:10.1002/ADSC.
200900416
[61] C. I. Fernandes, M. S. Saraiva, T. G. Nunes, P. D. Vaz, C. D. Nunes,
J. Catal. 2014, 309, 21. doi:10.1016/J.JCAT.2013.08.032
[62] C. I. Fernandes, M. D. Carvalho, L. P. Ferreira, C. D. Nunes, P. D. Vaz,
J. Organomet. Chem. 2014, 760, 2. doi:10.1016/J.JORGANCHEM.
2014.01.035
[63] S. Anandhakumar, M. Sasidharan, C.-W. Tsao, A. M. Raichur, ACS
Appl. Mater. Interfaces 2014, 6, 3275. doi:10.1021/AM500229V
[64] G. Bian, P. Jiang, H. Zhao, K. Jiang, L. Hu, Y. Dong, W. Zhang,
ChemistrySelect 2016, 1, 1384. doi:10.1002/SLCT.201600387
[65] H. L. Guo, X.-F. Wang, Q.-Y. Qian, F.-B. Wang, X.-H. Xia, ACS Nano
2009, 3, 2653. doi:10.1021/NN900227D
[36] Z. Li, S. Wu, D. Zheng, J. Liu, H. Liu, H. Lu, Q. Huo, J. Guan, Q. Kan,
Appl. Organomet. Chem. 2014, 28, 317. doi:10.1002/AOC.3127
[37] L. Chen, Y. Zhou, W. Tu, Z. Li, C. Bao, H. Dai, T. Yu, J. Liu, Z. Zou,
Nanoscale 2013, 5, 3481. doi:10.1039/C3NR34059G
[66] C. Zhu, S. Guo, Y. Fang, S. Dong, ACS Nano 2010, 4, 2429.
doi:10.1021/NN1002387